advances in food and nutrition research volume 57

ADVISORY BOARDS

KEN BUCKLEUniversity of New South Wales, Australia

MARY ELLEN CAMIREUniversity of Maine, USA

ROGER CLEMENSUniversity of Southern California, USA

HILDEGARDE HEYMANNUniversity of California, Davis, USA

ROBERT HUTKINSUniversity of Nebraska, USA

RONALD JACKSONQuebec, Canada

HUUB LELIEVELDGlobal Harmonization Initiative, The Netherlands

DARYL B. LUNDUniversity of Wisconsin, USA

CONNIE WEAVERPurdue University, USA

RONALD WROLSTADOregon State University, USA

SERIES EDITORS

GEORGE F. STEWART (1
948–1982)
EMIL M. MRAK (1
948–1987)
C. O. CHICHESTER (1
959–1988)
BERNARD S. SCHWEIGERT (1
984–1988)
JOHN E. KINSELLA (1
989–1993)
STEVE L. TAYLOR (1
995– )

Academic Press is an imprint of Elsevier30 CorporateDrive, Suite 400, Burlington,MA01803, USA525 B Street, Suite 1900, San Diego, CA 92101-4495, USA32 Jamestown Road, LondonNW1 7BY, UKRadarweg 29, POBox 211, 1000AEAmsterdam, TheNetherlandsLinacreHouse, JordanHill, OxfordOX2 8DP,UK

First edition 2009

Copyright# 2009 Elsevier Inc. All rights reserved.

No part of this publication may be reproduced, stored in a retrievalsystem or transmitted in any form or by any means electronic, mechan-ical, photocopying, recording or otherwise without the prior writtenpermission of the publisher.

Permissions may be sought directly from Elsevier’s Science & Technol-ogy Rights Department in Oxford, UK: phone (+44) (0) 1865 843830;fax (+44) (0) 1865 853333; email: [email protected]. Alterna-tively you can submit your request online by visiting the Elsevier website at http://elsevier.com/locate/permissions, and selecting Obtainingpermission to use Elsevier material.

NoticeNo responsibility is assumed by the publisher for any injury and/ordamage to persons or property as a matter of products liability,negligence or otherwise, or from any use or operation of any methods,products, instructions or ideas contained in thematerial herein. Becauseof rapid advances in the medical sciences, in particular, independentverification of diagnoses and drug dosages should bemade.

ISBN: 978-0-12-374440-1ISSN: 1043-4526

For information on all Academic Press publicationsvisit ourwebsite at elsevierdirect.com

Printed and bound inUSA

09 10 11 12 10 9 8 7 6 5 4 3 2 1

CONTRIBUTORS

Numbers in parentheses indicate the pages on which the authors' contributions begin.

� Pedro BouchonDepartment of Chemical and Bioprocess Engineering, PontificiaUniversidad Catolica de Chile, Santiago, Chile (209)

� Paul Pui-Hay ButFood and Drug Authentication Laboratory, Department of Biology,The Chinese University of Hong Kong, Shatin, N.T., Hong Kong,P.R. China (1)

� Sheila DuboisBureau of Chemical Safety, Food Directorate, Health Products and FoodBranch, Health Canada, Ottawa, Ontario, Canada (235)

� Kari DunfieldDepartment of Land Resources, University of Guelph, Guelph, Ontario,Canada (155)

� Wei FanDepartment of Food Science, University of Guelph, Guelph, Ontario,Canada (155)

� Zoe GillespieBureau of Chemical Safety, Food Directorate, Health Products and FoodBranch, Health Canada, Ottawa, Ontario, Canada (235)

� Samuel Benrejeb GodefroyBureau of Chemical Safety, Food Directorate, Health Products and FoodBranch, Health Canada, Ottawa, Ontario, Canada (235)

� Alan R. HipkissSchool of Clinical and Experimental Medicine, College of Medicaland Dental Sciences, The University of Birmingham, Edgbaston,Birmingham, United Kingdom (87)

� Ann HuberSoil Resource Group, Guelph, Ontario, Canada (155)

� Ka Ho LingFood and Drug Authentication Laboratory, Department of Biology,The Chinese University of Hong Kong, Shatin, N.T., Hong Kong,P.R. China (1)

ix

x Contributors

� Azadeh NamvarDepartment of Food Science, University of Guelph, Guelph, Ontario,Canada (155)

� Peter D. NicholsCSIRO Marine and Atmospheric Research and Food Futures Flagship,Hobart, Tasmania, Australia; Food and Drug Authentication Labora-tory, Department of Biology, The Chinese University of Hong Kong,Shatin, N.T., Hong Kong, P.R. China (1)

� Olga M. PulidoFaculty of Medicine, University of Ottawa, Ottawa, Ontario, Canada;Bureau of Chemical Safety, Food Directorate, Health Products and FoodBranch, Health Canada, Ottawa, Ontario, Canada (235)

� Mohsin RashidFaculty ofMedicine, Dalhousie University, Halifax, Nova Scotia, Canada;Professional Advisory Board, Canadian Celiac Association, Ottawa,Ontario, Canada (235)

� Karl J. SiebertFood Science & Technology Department, Cornell University, Geneva,New York (53)

� Connie SwitzerFaculty of Medicine, University of Alberta, Edmonton, Alberta, Canada;Professional Advisory Board, Canadian Celiac Association, Ottawa,Ontario, Canada (235)

� Elizabeth VavasourBureau of Chemical Safety, Food Directorate, Health Products and FoodBranch, Health Canada, Ottawa, Ontario, Canada (235)

� Keith WarrinerDepartment of Food Science, University of Guelph, Guelph, Ontario,Canada (155)

� Marion ZarkadasProfessional Advisory Board, Canadian Celiac Association, Ottawa,Ontario, Canada (235)

CHAPTER 1

Contents

Advances in Food and NutISSN 1043-4526, DOI: 10.1

* Food and Drug AuthentiShatin, N.T., Hong Kong

{ CSIRO Marine and Atmo1 Corresponding author

Fish-Induced Keriorrhea

Ka Ho Ling,* Peter D. Nichols,*,†

and Paul Pui-Hay But*,1

I. Introduction 2

A. Fish as food 2

B. Keriorrhea 3

C. Case reports and symptoms 3

II. Fish Incriminated 6

A. Escolar and oilfish 6

B. Harmful effects 9

C. Uses 9

D. Supply 11

E. Mislabeling and mishandling 13

F. Global concerns 14

III. Regulation and Litigation 15

A. Regulation 15

B. Litigation 15

IV. Biochemistry and Toxicity 18

A. Wax esters and their biological roles 18

B. Toxicity 19

C. Animal tests 21

D. Human studies 22

V. Identification and Detection 23

A. Morphological and anatomical analyses 23

B. Protein analysis 25

C. DNA analysis 27

D. Lipid analysis 27

rition Research, Volume 57 # 2009 Elsevier Inc.016/S1043-4526(09)57001-5 All rights reserved.

cation Laboratory, Department of Biology, The Chinese University of Hong Kong,, P.R. Chinaspheric Research and Food Futures Flagship, Hobart, Tasmania, Australia

1

VI. Wax Ester-Rich Fish and Other Potential Hazards 30

A. Gempylidae family 30

B. Other deep-sea fish 30

C. Diacylglyceryl ether (DAGE)-rich fish 39

VII. Discussion and Recommendations 40

A. The rationale: To ban or not to ban? 40

B. Recommendations 41

VIII. Conclusions 44

References 45

Abstract

2 Ka Ho Ling et al.

Many deep-sea fishes store large amounts of wax esters in their

body for buoyancy control. Some of them are frequently caught as

by-catch of tuna and other fishes. The most noteworthy ones

include escolar and oilfish. The accumulation of the indigestible

wax esters in the rectum through consumption of these fish engen-

ders discharges or leakage per rectum as orange or brownish green

oil, but without noticeable loss of water. This physiological

response is called keriorrhea, which is variously described as ‘‘oily

diarrhea,’’ ‘‘oily orange diarrhea,’’ or ‘‘orange oily leakage’’ by the

mass media and bloggers on the internet. Outbreaks of keriorrhea

have been repeatedly reported across continents. Additional symp-

toms including nausea, vomiting, abdominal cramps, and diarrhea

were complained by the victims. They are probably due to anxiety

or panic when suffering from keriorrhea. Escolar and oilfish are

banned from import and sale in Italy, Japan, and South Korea. Rapid

detection of the two fishes is imperative to ensure proper labeling

and safeguarding of the public before and after any keriorrhea

outbreak.

I. INTRODUCTION

A. Fish as food

Fishes are an excellent source of proteins, polyunsaturated fats, vitamins,and other nutrients. The wide range of biodiversity in fishes allows a goodselection of different forms, sizes, colors, tastes, and textures to fit one’sdiet preferences. When served alone or in combination with variousspices, other meats, and vegetables, and prepared by a range of culinarymethods, there are unlimited ways to turn fish into the most enjoyablegourmet item. As compared to other sources of meat, consumption of fishhas additional health benefits, which is most often associated with thepresence of omega-3 long-chain (>C20) polyunsaturated fatty acids (o3LC-PUFA). This provides protection against cancer of the alimentarytract, coronary heart diseases, stroke, and other disorders (Erkkila et al.,

Fish-Induced Keriorrhea 3

2004; Fernandez et al., 1999; He et al., 2004; Hu et al., 2002; Mozaffarian andRimm, 2006; Norat et al., 2005).

Fishes may also occasionally cause harm to health. When incompletelycooked or improperly handled, fishes can become a medium for trans-mission of parasites and diseases (Butt et al., 2004a,b). Allergens such asparvalbumins in fish muscles or even parasites such as Anisakis simplex infish can cause allergic reactions (Du Plessis et al., 2004; Lehrer et al., 2003;Poulsen et al., 2003; Taylor et al., 2004; Wild and Lehrer, 2005). The safetyof fish consumption is now a major consumer worry (Brewer and Prestat,2002; Lyon, 2008; Senkowsky, 2004; Verbeke et al., 2008) and the news ofpoisoning after fish consumption is not infrequent. Certain componentsin fish including tetrodotoxin and ciguatera toxins are notorious for theirtoxic properties (Hashimoto and Fusetani, 1978; Kazuo, 1999; Lawrenceet al., 2007; Lehane and Lewis, 2000; Miyazawa and Noguchi, 2001;Noguchi and Ebesu, 2001; Stommel and Watters, 2004; Ting and Brown,2001). Mercury and other heavy metals and various contaminants, such aspesticides, other organochlorines, and antibiotics, accumulated in fish area serious concern (Du Plessis et al., 2004; Guallar et al., 2002; Hightowerand Morre, 2003; Kostyniak et al., 1998; Senkowsky, 2004).

A common response to fish poisoning is diarrhea, often in the form ofloose andwatery stools accompanied with excessive water loss. However,in some special cases, the uncontrollable urge of bowel movements anddischarges do not involve a noticeable loss of water. In those cases, oil isdischarged or leaked through the rectum, and this type of poisoningresponses is called keriorrhea or keriorrhoea.

B. Keriorrhea

Keriorrhea specifically refers to the pathological symptom of involuntarypassage or leakage of oil, or actually wax esters, through the rectum. Thistermwas coined by Berman et al. (1981) based on the Greek words keri anddiarroia, which mean ‘‘wax’’ and ‘‘to flow through’’, respectively. Morespecifically, they refer to the symptoms observed in cases developed afterconsumption of certain oily fish, wherein the oil discharged appearsorange or brownish green in color, while little water is lost (Fig. 1.1).This ailment is variously described as ‘‘oily diarrhea’’, ‘‘oily orange diar-rhea’’, or ‘‘orange oily leakage’’ by the mass media and bloggers on theinternet.

C. Case reports and symptoms

Outbreaks of keriorrhea have been reported in many continents, includ-ing Africa, America, Asia, Australia, and Europe (Table 1.1). However,few are recorded in the scientific or medical literature. Therefore, the

FIGURE 1.1 Oil discharged after consumption of escolar. Reprinted with permission

from Ruello (2004, Nick Ruello of Ruello and Associates Pty Ltd.).

4 Ka Ho Ling et al.

actual number of affected people over the years are largely underesti-mated as the internet is floating with many more reports or communica-tion about personal experiences of embarrassing oily diarrhea afterconsumption of fish. In most cases, these fishes came into the spotlightbecause of a large outbreak that involved a substantial number of people;otherwise, scattered occurrences are generally neglected.

Australia has documented several keriorrhea outbreaks, allowing fur-ther tracing into the etiology and symptoms in patients. In SouthAustralia,between 1997 and 1999, there were nine cases of gastrointestinal com-plaints after rudderfish consumption. In two episodes that took place in1999, patients complained of diarrhea, often oily and orange colored,within hours of consumption. Through protein fingerprinting, the impli-cated fish was identified as escolar (Givney, 2002).

Two outbreaks of diarrhea in 1999 and 2001 were reported to beassociated with butterfish consumption in Victoria, Australia. The victimscomplained of diarrhea or yellow oily diarrhea. The fish causing theoutbreaks in 1999 was identified as either escolar or rudderfish, and theone in 2001 as escolar (Gregory, 2002)

Another outbreak of gastrointestinal illness occurred among attendeesof a conference lunch in New South Wales, Australia, in October 2001.Analysis of the oil in the fish samples served revealed a high proportion ofwax ester (96–97%) and showed close resemblance to the oil compositionin escolar. A distinctive symptom reported by many ill persons was thepresence of oily diarrhea. Investigators of the outbreak conducted a

TABLE 1.1 Selected cases of keriorrhea outbreak related to escolar and oilfish consumption

Fish

involved Location Date

No. of

people affected Remarks References

Oilfish South Korea May 2007 N/A White tuna sushi at

restaurants

Fuga (2007)

Toronto,

Canada

February 2007 N/A Mislabeled as cod

or sea bass

CBC (2007); Jacquet and

Pauly (2008)

Hong Kong

SAR, China

January 2007 600þ Mislabeled as cod *Ling et al. (2008a,b);

Jacquet and Pauly

(2008)Victoria,

Australia

August 2001 5 Mislabeled as

butterfish

**Gregory (2002)

Sydney,

Australia

January 2001 9 Fish curry in a

canteen

Leask et al. (2004)

Victoria,

Australia

November 1999 11 Mislabeled as

butterfish

**Gregory (2002)

Victoria,

Australia

November 1999 �10 Mislabeled as

butterfish

**Gregory (2002)

Escolar California,

USA

11 August 2003 42 Served in a buffet Feldman et al. (2005)

New South

Wales,

Australia

October 2001 20 Mislabeled as

rudderfish or

butterfish

Yohannes et al. (2002)

South

Australia

October 1999 N/A Mislabeled as

rudderfish

Givney (2002)

Cape Town,South Africa

1989 N/A Mislabeled asrudderfish

Berman et al. (1981)

* Escolar could be possibly involved in a few cases.** The fish involved were reported as escolar but under the scientific name Ruvettus pretiosus.

6 Ka Ho Ling et al.

telephone interview of the cohort of conference attendees using astandard questionnaire. Out of 44 attendees, 20 (46%) became ill followingthe conference. The median incubation period was 2.5 h (range 1–90 h).The most common symptoms reported were diarrhea (80% including38% reporting oily diarrhea), abdominal cramps (50%), nausea (45%),headache (35%), and vomiting (25%). None of the food or beveragesconsumed was significantly associated with the illness. However, allindividuals who consumed fish became sick, but not those who did not(four persons). Among those who consumed fish, the following potentialrisk factors did not have a significant association with the illness: bodymass index (BMI), age, health status, and the amount of fish consumed(Yohannes et al., 2002).

II. FISH INCRIMINATED

A. Escolar and oilfish

Outbreaks of keriorrhea are reported in consumers who admitted havingconsumed various fishes (e.g., Atlanta cod, butterfish, cod, ruddercod, orrudderfish). However, so far, almost all episodes can be traced totwo varieties: escolar and oilfish.

1. BiologyEscolar and oilfish belong to the Gempylidae (snake mackerel) family inthe order Perciformes (Alexander et al., 2004; Nakamura and Parin, 1993).There are currently 24 species under 16 genera in Gempylidae, and theyare all found in the marine environment (FishBase, 2008). All species inthis family usually occur in very deep waters in tropical and subtropicalseas (Nakamura and Parin, 1993). They have elongated and compressedbody with isolated finlets after the anal and dorsal fins (Nakamura andParin, 1993).

Escolar (Lepidocybium flavobrunneum Smith) is also called black oilfish(Fig. 1.2A and B). It is a large fish (up to 2-m long, but usually 150-cmwide) covered with small cycloid scales and smooth skin (Fishbase, 2008).It has a faint but highly undulating lateral line on the semifusiform body,which changes color from dark brown to almost black with age.In addition, a prominent keel flanked with two small oblique ridges, oneon each side of the keel, is present on the caudal peduncle (Nakamura andParin, 1993; Pauline, 1980). A single species of escolar is generally recog-nized. However, Brendtro et al. (2008) recently analyzed the mitochondrialcontrol region and flanking tRNA sequences for 225 escolar specimenscollected from six sites at different locations of the Atlantic andPacific Oceans. Their results revealed two distinct clades, one for the

A

B

C

D

10 cm

10 cm

FIGURE 1.2 (A) Escolar (Regulatory Fish Encyclopedia, U.S. FDA, 1993–2008, reprinted

with permission); (B) Line drawing of escolar (Nakamura, 1995, Food and Agriculture

Organization of the United Nations, reprinted with permission). (C) Oilfish (Regulatory

Fish Encyclopedia, U.S. FDA, 1993–2008, reprinted with permission); (D) Line drawing of

oilfish (Schneider, 1990, Food and Agriculture Organization of the United Nations,

reprinted with permission).


Atlantic and the other for the Pacific (plus four from South Africa) samples.A previous study by Collette et al. (1984), interestingly, also noteddifferences between escolar in the Atlantic and the Indo-Pacific regions,their vertebral counts being 31 and 32, respectively. There is a possibilitythat two species or subspecies of escolar may be warranted (Brendtroet al., 2008).

Oilfish (Ruvettus pretiosus Cocco) is also called castor-oil fish (Fig. 1.2Cand D). It is a large fish (up to 3-m long, but usually 100–150 cm wide)possessing a dark brown semifusiform body and a single, usuallyobscure, lateral line. Its lower jaw slightly protrudes beneath the upperjaw (Bettoso and Dulcic, 1999; Fishbase, 2008). Rows of spiny tubercles onthe cycloid scales make the skin of oilfish very rough. A layer of porousblubber-like tissue is observed after the scales are removed (Gudger andMowbray, 1927). In addition, a rigid and scaly abdominal keel is locatedon the ventral contour of the fish (Bettoso and Dulcic, 1999; Bone, 1972;Nakamura and Parin, 1993).

Escolar and oilfish are commonly found at depths between 100–800 mand 200–1100 m, respectively (Nakamura and Parin, 1993). As in many

8 Ka Ho Ling et al.

deep sea fish, both escolar and oilfish lack a swim bladder. Their buoy-ancy in water is achieved by the storage of large amounts of low-densitylipid, particularly in the dermis, flesh, and the bones of the skull (Bone,1972). The stored lipid provides a sufficient lift to make them neutrallybuoyant. Oilfish, at rest, hangs in the water heads up at 45� to thehorizontal (Bone, 1972).

2. ChemistryBoth escolar and oilfish possess considerable lipid in the body, accountingfor approximately 20% of their wet weight (Cox and Reid, 1932; Gudger,1925; Mori et al., 1966c; Nichols et al., 2001; Ukishima et al., 1987). Themajor components contributing to more than 90% of the total oil contentare indigestible wax esters (Alexander et al., 2004; Ruiz-Gutierrez et al.,1997; Yohannes et al., 2002). Other lipid classes (hydrocarbon, triacylgly-cerol, sterol, and polar lipid) are present in small or negligible quantities.Moreover, both escolar and oilfish have high levels of histidine (8–11 mg/g)in their muscles (Feldman et al., 2005; Kan et al., 2000; Leask et al., 2004).

A large and essentially unpredictable variability in the total oil andwax ester content of escolar was found with no clear correlation betweenthe fish size and oil content. Also, no significant difference was foundbetween the oil content of fish on the west coast and east coast ofAustralia, at least for specimens collected in summer (Ruello, 2004).

3. Effect of cooking on oil content and compositionRuello & Associates Pty. Ltd. (Ruello, 2004) conducted a study for theAustralian Government Department of Health and Ageing. A compara-tive study was made on baking and grilling escolar. Far more water thanoil is lost during cooking as the total oil content on a wet-weight basisactually increases. Moreover, the cooking method has little effect on theoil composition; wax esters remain as the predominant oil class in thesamples tested. The perception for the reduction of the volume or potencyof the oil/wax ester in escolar by heavy grilling or other normal cookingmethod is unfounded. The notion of correct and incorrect cookingmethods (with regards to keriorrhea) and that one cooking methodis better than another (e.g., grilled vs. baked) is equally unfounded.Battering and frying are unlikely to raise the risk of keriorrhea. Therewas no evidence to support the common perception that wax esters can be‘‘grilled out’’ of the fish or otherwise substantially reduced by normalcooking methods. Cooking actually ‘‘concentrates’’ the oil as water isexpelled from the flesh. Freezing the (raw) fillet for as long as 9 monthsalso failed to noticeably reduce the capacity of escolar to inducekeriorrhea (Ruello, 2004).


B. Harmful effects

Escolar and oilfish have long been known to possess purgative effectsbecause of their high oil content, accounting for approximately 20% oftheir wet weight (Cox and Reid, 1932; Gudger, 1925; Mori et al., 1966c;Ukishima et al., 1987). Substantial amounts of indigestible wax esters havebeen incriminated as the cause of keriorrhea and other acute gastrointes-tinal symptoms, such as abdominal cramps, nausea, headache, andvomiting in susceptible individuals (Alexander et al., 2004; Ruiz-Gutierrez et al., 1997; Yohannes et al., 2002). Therefore, the wax esters inthese two fish are regarded as a natural toxin called gempylotoxin (FDA,2001a). Moreover, both escolar and oilfish have high levels of histidinein their muscles (Feldman et al., 2005; Kan et al., 2000; Leask et al., 2004).If they are refrigerated improperly, bacteria can multiple and convert thehistidine into histamine also termed scombrotoxin (FDA, 2001b), whichcan lead to cardiovascular, gastrointestinal, and neurological disorders(FDA, 2001b; Feldman et al., 2005; Leask et al., 2004).

C. Uses

Escolar and oilfish have been variously used as food. Oilfish has beentraditionally used by Polynesians and Melanesians as a purgative medi-cine (Cox and Reid, 1932; Gudger, 1925). In the Union of the Comoros, anisland nation in the Indian Ocean, oilfish is targeted as a food source andcaught by local people regularly (Helfman et al., 1999; Stobbs and Bruton,1991). In Bermuda, oilfish is used as food and is acclaimed as an excellenteating (Gudger and Mowbray, 1927). In the Canary Islands and otherseafaring regions in Spain, the fish is also used as a folk medicine indrinking broth made from the bones to relieve constipation (Raisfeldand Patronite, 2006; Ruiz-Gutierrez et al., 1997). About 10 tons of escolarand oilfish were sun-dried and consumed annually in Japan before theirsale was prohibited (Mori et al., 1966c). Although described by healthauthorities as a purgative in Taiwan, they are openly sold and served assashimi and fish steaks under the name You-yu (literally oilfish) (Fig. 1.3).Their roes (Fig. 1.4) are pressed and dried in Tungkang (Dunggang),Taiwan, and hailed as one of the three treasures of that little port as anew delicacy (Li, 2007). Compliments to the fish have come from manygastronomes. Berman et al., (1981) commended that escolar had a veryagreeable flavor with a soft, butter-like texture. Some people regard thefish as a delicacy; for instance, Bykov (1983) remarked in his book that‘‘The taste qualities of this fish (oilfish) are high. It is an excellent tablefish.’’ Raisfeld and Patronite (2006) praised it (escolar) as a dream fish, liketoro (tuna), but cheaper.

FIGURE 1.3 You-yu sold as skinless and boneless fillet at fish market in Tungkang,

Taiwan. The sample was identified as escolar using DNA sequencing by the authors

(photo provided by authors).

10 Ka Ho Ling et al.

The two fish are banned in Italy, Japan, and the Republic of Korea.In the rest of the world, they are generally not regarded as suitable forcatering. As a result, they are frequently marketed and labeled as othermore expensive commercial fish but at lower prices.

There are various suggestions for their exploitation, with the hope thattheir unusually high levels of wax esters could be utilized for the good ofmankind. Wax esters derived from orange roughy and oreo dories havebeen included in various industrial processes like cleaning and degreas-ing (Nichols et al., 2001). Wax esters enriched in o3 LC-PUFA can beabsorbed by rats, and wax esters are less prone to oxidation and can bebetter formulated than liquid o3 derivatives (Gorreta et al., 2002). Thus,wax esters enriched in o3 can be a food supplement as there is increasingevidence that a diet high in o3 LC-PUFAmay help prevent coronary heartdiseases (Iacono and Dougherty, 1993), and high level of wax esters inescolar and oilfish can act as a potential source for this purpose.Wax esters from escolarwere processed by deacidification, decolorization,

FIGURE 1.4 You-yu roe retailed in Tungkang, Taiwan. One of the samples was identified

as escolar using DNA sequencing by the authors (photo provided by authors).


hydrogenation, and distillation to obtain a semisolid wax at roomtemperature; the refined wax was tested safe and useful as a base formedicine and cosmetics (Ukishima et al., 1987). A further value-addedprocess involved removal of most of the lipid from escolar meat byrepeated alkaline washing to make a gel, which was tested and found tobe a better raw material than the commercial bigeye snapper used cur-rently for surimi (ground fish meat) production (Pattaravivat et al., 2008).

Additionally, on the internet, there are discussions on the possibility ofusing escolar and oilfish for slimming or weight reduction. The valueof this application is doubtful as only the oil (wax esters) wouldbe discharged.

D. Supply

As escolar and oilfish are widely distributed in the tropical and temperateseas, they are frequently caught and marketed as a result of by-catch withother commercially important species (Mori et al., 1966c; Tserpes et al.,2006). It was reported that about ten tons of by-catch oilfish and escolarwere sun-dried and consumed annually in Japan before the sale prohibi-tion was implemented (Mori et al., 1966c). Up to 400 tons of escolar arecaught annually in Australia (Shadbolt et al., 2002). In 2003, escolar catchaccounted for up to 16,501 tons of total by-catch species in longline fishingconducted by the Southern and Western Tuna and Billfish Fishery


(Lynch, 2004). In a report on the by-catch of tuna longliners from a SouthKorean observer program, escolar and oilfish ranked the second (20.8%)and third (15.6%) most common by-catch fish species, respectively (Yanget al., 2005). According to the US National Marine Fisheries Services,annual landings of escolar varied between 40 and 80 tons from 1999 to2007, while that of oilfish increased from 32.4 tons in 1999 to 216.3 tonsin 2007 (Figs. 1.5A and B) (NMFS, 2008). Taiwan has a decade-long recordof oilfish harvest, and the annual catch increased 15-fold from 2,700 tonsin 1999 to near 42,000 tons in 2007 (Fig. 1.6).

90

80

A

B

70

60

50

40

30

20

10

02000 2001 2002 2003 2004 2005 2006

Quantity (in ton)Price per ton (100US$/ton)

1999 2007

250

200

150

100

50

02000 2001 2002 2003

Year

Year

2004 2005 2006

Quantity (in ton)Price per ton (100US$/ton)

1999 2007

FIGURE 1.5 (A) The landing quantity and price of escolar in the US from 1999 to 2007.

(B) The landing quantity and price of oilfish in the US from 1999 to 2007. Data from

NMFS (2008), National Marine Marine Fisheries Services Annual Landings Database.

Quantity (in ton)

Price per ton (NT$/ton)

45,000

40,000

35,000

30,000

25,000

20,000

15,000

10,000

5000

02000 2001 2002 2003

Year2004 2005 20061999 2007

FIGURE 1.6 The landing quantity and price of oilfish in Taiwan from 1999 to 2007. Data

from Taiwan Fisheries Agency (1999–2007, Taiwan Fisheries Yearbook).


In the market, escolar is often deep skinned, and the skin and redmuscle are discarded. The longitudinal portion of muscle tissues is cutparallel to the backbone and then into chunks or blocks of white skinlessand boneless fillets. In sushi bars or fish markets, escolar chunks are cutinto slices and served as sashimi in Hong Kong and Taiwan under thenames of Bai-yu-tuan and You-yu, respectively. Escolar of smallersizes may be cut into transverse sections through the backbone, and thecutlets are retailed with the skin on (Ruello, 2004). Oilfish, on the otherhand, is more often retailed as cutlets. The scales on the outside of the fishare removed, leaving big grayish quadrangular patterns of the skinremaining on the cutlets.

E. Mislabeling and mishandling

Escolar and oilfish are of low commercial values because of their kerrior-rheic properties. They are considered as ‘‘not suitable for catering’’ oreven banned from sale in various countries. However, they are commonlymarketed as a result of their substantial by-catch with tuna and swordfish(Shadbolt et. al., 2002; Tserpes et al., 2006). According to the EuropeanCommunities (Labelling of Fishery and Aquaculture Products) Regula-tions 2003 (S.I. No. 320 of 2003), L. flavobrunneum and R. pretiosus must bemarketed as escolar and oilfish, respectively, and no other commercialnames can be used alternatively. Yet, both species are usually mislabeledas sea bass, butterfish, rudderfish, white tuna, or codfish either


intentionally or accidentally. Under these circumstances, outbreaks ofkeriorrhea associated with consumption of escolar or oilfish have beenrepeatedly reported in several continents (Berman et al., 1981; Feldmanet al., 2005; Givney, 2002; Gregory, 2002; Jacquet and Pauly, 2008; Leasket al., 2004; Ling et al., 2008a,b; Shadbolt et al., 2002; Waldman et al., 2006).

Besides wax ester poisoning, escolar and oilfish are common candi-dates for histamine poisoning, which has a greater affect than keriorrheaand can cause more serious manifestations. Escolar and oilfish containhigh levels of histidine in their muscles (Feldman et al., 2005; Kan et al.,2000; Leask et al., 2004). If these fish are inadequately refrigerated, bacteriacan multiply and convert histidine into histamine, also termed scombro-toxin (FDA, 2001b). This conversion often happens when large numbersof unsold fish steaks are stocked over time to avoid food inspection in caseof a related keriorrhea outbreak. The fish may reappear later in markets,but at that time, the steaks may be contaminated and not suitable forconsumption. Scombrotoxin, like wax esters, is heat stable and is notdestroyed by cooking; it can lead to cardiovascular, gastrointestinal, andneurological disorders (FDA, 2001b; Feldman et al., 2005; Leask et al.,2004). Therefore, detection of wax esters could help prevent more severefood poisoning from happening once any keriorrhea outbreak is reported(Ling et al., 2008a,b).

F. Global concerns

Australia, Canada, the United States, and a majority of member states ofthe European Union have issued special guidelines toward trading andconsumption of the two fish. Italy and Japan have banned their importand sale (Alexander et al., 2004) before 2007. However, outbreaks ofkeriorrhea are still reported occasionally across continents. Recently, anoutbreak of over 600 cases of keriorrhea occurred in Hong Kong towardthe end of 2006 (Chong, 2007; Chung, 2007a; Connolly et al. 2007; Jacquetand Pauly, 2008; Ling et al., 2008a,b). The packages of oilfish cutlets weremislabeled as codfish (Chong, 2007; Ling et al., 2008a,b). Escolar, on theother hand, was found offered as sushi or sashimi under the name ofsnowfish or white tuna (Chung, 2007b; Ling et al., 2008a,b; Mok, 2007).In February 2007, similar fish cutlets were found in Chinatown in Tor-onto, Canada, and resulted in a keriorrhea mini-epidemic there. Threemonths later (May 2007), the oily fish was found sold as white tuna sushiin South Korea and several cases of illness were reported, leading to theSouth Korean government prohibiting the use of oilfish and escolar forhuman food and banning any import of the fish in August 2007(Stenhouse, 2007). Escolar- or oilfish-related illness has long been recog-nized, yet the problems have never been eradicated and still occur repeat-edly (Shadbolt et al., 2002). The public always express great concern over


this problem; for instance, the oilfish scandal in 2006 awoke theHong Kong public to this food safety issue and there were strongdemands for regulating and identifying these two fish (Goh, 2007).Indeed, voices to completely ban the fish are on the rise. The industry,however, holds an opposite opinion and insists that the two fish are nottoxic and are suitable for catering provided that certain guidelines arefollowed. In addition, there are great individual differences in terms ofsusceptibility. Some people consider the fish as a delicacy and enjoy thefish without problems, while others experience frequent keriorrhea(Ruello, 2004). Therefore, it is a dilemma for the responsible agencies asto whether to ban the two fish. As a result, different countries havedifferent rules and the rules also change with time making it a nightmareto food safety and control agencies in case of further outbreaks.

III. REGULATION AND LITIGATION

A. Regulation

Because of the differential impact of escolar- or oilfish-related problemsaround the world, and also variation in individual susceptibility, differentgovernments continue to promulgate only modest regulation on both fish.Only three countries, Japan, South Korea, and Italy, completely banthe trading and import of the two fish, while other countries only issuespecial guidelines or warnings toward them (Table 1.2). It thereforeremains that the majority of countries do not have any regulations forthe two fish.

B. Litigation

Although outbreaks of keriorrhea have repeatedly occurred worldwide, itis rare to find cases of prosecution or litigation. However, recently inHong Kong, a supermarket chain was fined over selling oilfish mislabeledas codfish (Lau, 2007; Wong and Lam, 2007). In that case, a total of14 complaints were received by the Hong Kong Centre for Food Safetyregarding diarrhea or serious stomach upsets after consumption of fishcutlets labeled as cod purchased from the supermarket chain. It was laterconfirmed that the so-called cod fish was actually oilfish. The Food andEnvironmental Hygiene Department, HKSAR, later initiated prosecutiontoward the supermarket chain. Finally, the supermarket chain pleadedguilty of selling food not of the substance expected by consumers and wasfined HK$45,000 for mislabeling oilfish as cod and ordered to pay the costof laboratory tests. In Hong Kong, sale of oilfish is not regulated because itis not considered poisonous. The magistrate, however, still condemned

TABLE 1.2 Polices on escolar and oilfish in various countries or cities

Country/city Authority Action/recommendation References

Australia Queensland Health,

Queensland

Government

No ban. Recommended that the fish are

not suitable for catering

Queensland Health

(2008)

Canada Canadian Food

Inspection Agency

No ban. Recommended to choose smaller

portion sizes and prepare the fish in a

way to reduce oil content

CFIA (2007)

Denmark Danish National Food

Administration

No ban. Cautioned the Danish fish

importing companies and issued

cooking and storage recommendations

Alexander et al. (2004)

European Union

member states

European Food Safety

Authority, European

Union

No ban. Issued opinion from advisory

group. Recommended a notification to

public of the potential health risks and

proper preparation practices


German German federal Institutefor Risk Assessment

No ban. Published informationconcerning potential problems in

connection with the consumption

of the fish


Hong Kong Centre for Food Safety No ban. Issued guidelines on labeling and

handling the fish

WGNCO (2007)

Italy N/A Banned import and sale Alexander et al. (2004)

Japan Japanese Ministry of

Health and Welfare

Banned import and sale Kawai et al. (1985)

Macau Department of Health No ban. Recommended the fish are not

suitable for catering

Macau Disease

Control Centre

(2007)

Singapore Agri-Food and

Veterinary Authority

No ban. Issued notification of the

potential health risks and required

proper labels

Chua (2007)

Sweden Swedish National Food

Administration

No ban. Cautioned the Swedish National

Fish Trade Association and issuedcooking and storage recommendations


United Kingdom Food Standards Agency No ban. Issued notification of the

potential health risks and mislabeling

problems of the fish

Statham (2003)

USA Food and Drug

Administration

Reversed to no ban. Advised against the

sale of the fish in intrastate/interstate

commerce, and requested warning

labels

Yohannes et al. (2002)


the supermarket chain for contravening food safety regulation(Public Health and Municipal Services Ordinance, Hong Kong LawsChapter 132) by failing to ensure its products were safe and labelingaccurate, and pointed out that the supermarket had committed veryserious offence in selling, without warning, a product that was generallyunsuitable for human consumption.

IV. BIOCHEMISTRY AND TOXICITY

A. Wax esters and their biological roles

Wax esters are carboxylic esters consisting of a fatty acid esterified to afatty alcohol (Fig. 1.7), wherein both the acids and alcohols can be eithersaturated or unsaturated (Kolattukudy, 1976). Wax esters are present indifferent organisms, from the seeds of jojoba to the head oil of spermwhale (Busson-Breysse et al., 1994; Spencer et al., 1977; Takagi et al., 1976).Wax esters serve a variety of biological functions; for instance, they areused as energy reserve in seeds and roes, provide buoyancy in dinofla-gellates and pelagic invertebrates, and prevent water loss as in the waxylayer on the cuticle of insects ( Joh et al., 1995; Nelson et al., 2000; Phleger,1998). Although wax esters can be found across different taxa, the majormuscle lipid components of most fish species, including many commer-cially important fish, are triacylglycerols and phospholipids. Wax esters,in contrast, are considered less common lipid components, and wherethey occur in deep-sea fish species provide a way to enhance buoyancy(Bone, 1972; Lee and Patton, 1989). The source of such a high level of waxesters (up to 20% of wet weight) in escolar and oilfish is still unknown, butmay possibly be formed by similar mechanisms as in another wax ester-rich fish, orange roughy (Hoplostethus atlanticus Collett). The ability tosynthesize large amounts of long-chain alcohols is the key to determinewhether a marine animal produces wax ester or triacylglycerols as itsmajor neutral body lipid (Lee and Patton, 1989). In a study of wax esterssynthesis in Euchaeta norvegica, which is a wax ester-rich zooplankton,radio-labeled glucose or alanine was given to the organism and most ofthe radioactivity in the wax esters was detected in the alcohol moiety,implying that fatty alcohols are synthesized de novo from nonlipid pre-cursors, while the fatty acids in wax esters are sourced from dietary fatty

O

R C O R’

FIGURE 1.7 The basic structure of a wax ester (R ¼ fatty acid chain; R0 ¼ fatty

alcohol chain).


acids (Henderson and Sargent, 1980). Sargent et al. (1983) in their studyof orange roughy suggested that the wax esters could be produced by(a) de novo biosynthesis of 20:1 and 22:1 fatty acids that are then reduced tofatty alcohols, (b) chain elongation and desaturation of shorter chaindietary fatty alcohols and fatty acids to yield long chain fatty acids thatare finally reduced to alcohols, or (c) or modification of dietary fattyalcohols and acids. In orange roughy, wax esters are stored extracellularly(Phleger and Grigor, 1990). Extracellular wax esters serve only for buoy-ancy. The storage of wax esters could be superior to that of triacylglycer-ols under certain physiological situations. A unit volume of wax estersprovide approximately 70% more upthrust than the same volume oftriacylglycerols in seawater with a density of 1.025 g/cm3 (Sargent, 1978).In addition,wax esters are essentially noncompressible, and are superior toa gas-filled swim bladder, during vertical migration (Phleger et al., 1999).Escolar and oilfish, which both lack a swim bladder, could travel verticallyat depths between 100–800 m and 200–1100 m, respectively (Nakamuraand Parin, 1993), and wax esters provide them with better buoyancycontrol. High concentrations of wax esters in the fish skin, like the functionof wax ester-rich blubber in whale, help insulate them from the freezingdeep-sea environment. Indeed, oilfish, similar to orange roughy, has thehighest oil content (32.3%) in the integument (Bone, 1972).

B. Toxicity

It was reported that most people experience keriorrhea without bowelcramps or abdominal discomfort, implying that the frequent passage ofoil in most people is caused by the lubricant effect of the oil, but not by anirritant effect as in the case of toxic substances in ordinary diarrhea(Du Plessis et al., 2004). Oilfish was previously called ‘‘castor-oil fish’’based on an erroneous report that its oil was composed of 13% hydro-xyoleic acid (Gudger, 1925) and a mistaken test that falsely indicated asimilar purgative ability between the fish oil and castor oil (Cox and Reid,1932). Unlike wax esters, hydroxyoleic acid, which is the purgativechemical in castor oil, causes diarrhea by an irritant effect on the bowelinstead of the lubricating and pooling effects of wax esters to the rectum(Berman et al., 1981). Indeed, studies revealed that hydroxyoleic acid isunlikely to be present in oilfish (Mori et al., 1966c; Nevenzel et al., 1965).Wax esters are not destroyed or decomposed during cooking. Resistanceto digestive enzymes, such as lipase, and a lowmelting point (in oily stateat human body temperature) results in pooling of large amount of theselubricant wax esters in the rectum leading to keriorrhea.

In a strict sense, wax esters are not completely indigestible in mam-mals. Wax esters are hydrolyzed by lipases at a very slow rate and theproducts, especially fatty alcohols are only slowly absorbed. Therefore,


dietary wax esters are little absorbed and the majority is excreted in amixture of wax esters (altered or unaltered) and fatty alcohols. Rats thatwere fed with single small doses of cetyl palmitate (major wax ester inspermaceti) had triacylglycerols present in the lymph (Munk andRosenstein, 1891). Large doses of mutton-bird oil (mainly as cetyl andoleyl oleates) fed to rats were partially absorbed (Carter and Malcolm,1927). It was also true for cats that cetyl esters and cetyl alcohols werefound in the feces (Carter and Malcolm, 1927). Moreover, rats on a dietwith 15% jojoba oil (rich in wax esters) absorbed 70% and excreted theremainder as wax esters and free alcohols, while a purgative effect wasobserved for this type of diet (Savage, 1951). Again, partly absorbed andpartly excreted wax esters and free alcohols were observed in rats fedwith feeds containing 15% of oleyl palmitate (Hansen and Mead, 1965).

There is no specific wax ester digestive lipase in mammals. Hydrolysisof wax esters is carried out by the same lipase that also acts on triacylgly-cerols. Savary (1971) showed that purified mammalian pancreatic lipaseis 10–50 times slower in hydrolyzing wax esters than triacylglycerols. Thereasons for a much slower hydrolysis of wax esters by lipase are productinhibition and hydrophobicity. The hydrolyzed products, fatty alcoholsand fatty acids, form an oil or solid phase in water. The products diffusedonly slowly out of the oil-water interface and the bulk of insoluble reac-tion mixture cause ‘‘product smothering’’ to the water-requiring lipase(Lee and Patton, 1989). Moreover, the large amount of slowly absorbed,thus accumulated, fatty alcohols could reverse the reaction and result insynthesis of wax esters. Hansen andMead (1965) showed that synthesis ofwaxes can occur in the intestine of rats fed with oleyl alcohol (fattyalcohol). The hydrophobic nature of wax ester molecules can hampertheir interaction with the active site of lipase (Lee and Patton, 1989).All of these properties engender a very ineffective hydrolysis of waxesters, and if a large dose of wax esters is consumed, much of the waxesters are passed through the intestines without digestion and absorption.Therefore, a pooling of large amounts of these lubricant wax esters in therectum leads to keriorrhea.

Data from one escolar-associated outbreak found no correlationbetween BMI, age, health status, and amount of fish consumed to theseverity and occurrence of symptoms, while other factors, such as varia-bility in wax ester content in different fillet cut depths, could be relevant(Yohannes et al., 2002). Unlike some fish species, such as herring, whichhave uniform muscle oil content along the body (Brandes and Dietrich,1953), muscle oil content in escolar and oilfish is not evenly distributed.Bone (1972) found that muscle oil content in oilfish increases from 14.5%(near vertebral column) to 24.7% (near the skin). A similar trend was alsoobserved by Ruiz-Gutierrez et al. (1997) with higher oil content found insubcutaneous muscles than the periosteum. However, the lipid profile for


dorsal and ventral muscle portions is much the same (Ruiz-Gutierrezet al., 1997). Moreover, there is a tendency that muscle oil content is higherat the anterior part and decreases toward the posterior end (Bone, 1972).The highest oil content (32.3%) is found in the integument (Bone, 1972).Wax ester content in a fish species also could vary with environmentalfactors, such as catch location, diet of the fish and physiological factors,such as gender and reproductive stage (Nichols et al., 2001). For instance,wax ester content in mullet increased sharply in the roes during thereproductive season (Iyengar and Schlenk, 1967). Seasonal variation ofwax ester content may account for the fact that a majority cases of oilfish-related outbreaks in Australia occurred between August and November(Doyle, 2002).

Although painless, keriorrhea was frequently reported as the onlysymptom associated with escolar consumption (Berman et al., 1981;Ruello, 2004). Other reports, however, recorded much severer symptoms,such as abdominal cramps, nausea, headache, and vomiting, after escolarconsumption (Yohannes et al., 2002). Yohannes et al. (2002) indicated thatscombroid (histamine) poisoning was unlikely the reason for the severersymptoms in the 2002 outbreak, given that the onset was very rapid(45 min) while symptoms common for scombroid poisoning, includingfever, flushing, and rapid pulse rate, were not detected.

C. Animal tests

A limited number of studies have been performed with animals. In onecase, two cats were fed 20 g of escolar flesh each. The smaller cat weighing530 g exhibited diarrhea with frequent watery stools 4 h after consump-tion, while the larger animal weighing 810 g did not show any diarrhea(Mori et al., 1966c). Due to the limited sampling size and the experimentaldesign, the divergent responses could be due to a difference in sensitivityor dosage level; the smaller cat consumed approximately 50% more flesh(0.038 g flesh/g body weight) than the larger one (0.025 g flesh/g bodyweight) in terms of the flesh consumed per unit body weight.

Toxicity of the wax esters from escolar and oilfish was also assessed inrats (Mori et al., 1966c). A total of 20 rats were divided into five groups.The test groups were fed daily with rice, casein, yeast, and salt mixturetogether with escolar flesh (7.5 g), oilfish flesh (7.5 g), escolar oil (1.5 g), oroilfish oil (1.5 g), while the control group was fed the same without theflesh or fish oil. Signs of seborrhea, with oil smudging on hairs, mouth,and belly, was observed on the second day of feeding in all, except thecontrol, groups. All 16 rats fed with either flesh or fish oil showeddiarrhea and 13 of them died within 10 days.

Seborrhea is another long-term side effect of eating the oily fish. Waxesters are released through the sebaceous gland of the skin, blocking the


pores and potentially interfering with metabolism. Seborrhea and acutefatality was reported in animals fed with various wax esters (Matsuo,1962), but not by Hansen and Mead (1965), where rats were fed with oleylpalmitate.

Two studies conducted on orange roughy, another deep-sea fish rich inwax esters, can be illuminating. Rats were fed for 28 days a diet containingvarious amounts of flesh from orange roughy. Rats on a low dose diet ofless than 360 g orange roughy/kg bodyweight (�30 gwax esters/kg bodyweight) displayed no observable difference from the control groups onawax ester-free diet. In the group fedwith 540 g orange roughy/kg (�44 gwax esters/kg), persistent oiliness around the anus and fecal smudges onfood pots were observed. The hairs over the whole body became heavilycoated with oil and the excretion of oily feces was observed in the groupson a diet of 720 and 1430 g orange roughy/kg. All rats on 2870 g orangeroughy/kg (�233 g wax esters/kg) diet died within 11 days ( James andTreloar, 1984). The second studywas on pigs fedwith orange roughy fleshat 6.7 g/kg/day (equivalent to a daily fish meal of 500 g for an averageperson of 75 kg body-weight) over an extended period (starting from liveweight at 20 kg until 80 kg). All pigs stayed healthy and gained weightthroughout the whole period. No purgative effect was observed ( Jamesand Body, 1986). Based on these results, the two reports concluded thatnormal consumption of small portions of orange roughy by humansgenerally would not cause any serious health problem. An estimate ofthe safety window can also be made by taking into consideration the factthat escolar and oilfish have three to four times more wax esters thanorange roughy (Nichols et al., 2001).

D. Human studies

No clinical studies are available, except three reports that describedvolunteer tests on the responses after consumption of escolar or oilfish(Berman, et al., 1981; Cox and Reid, 1932; Ruello, 2004). In the study byCox and Reid (1932), one of the authors ingested an ounce (about 28 g) ofextracted oilfish oil (equivalent to about 140 g of oilfish flesh) andreported no symptoms. In the second study (Berman et al., 1981), two ofthe authors consumed about 500 g of baked escolar flesh. They found thefish to have a very agreeable flavor with a soft, butter-like texture. After 12symptom-free hours, oil began to be passed per rectum. It was difficult tocontain the oil that was pooling in substantial quantities in the lowerrectum, and therefore frequent evacuation was necessary. Approximately10 ml of oil were passed on one occasion. The oil was clear orange orgreen in color, inoffensive in odor, and not on most occasions contami-nated significantly by fecal material. The authors did not experience anybowel cramps or visceral discomfort, suggesting that the discharges were


caused by the lubricant effect of the oil, rather than by any irritant action.The oil passed per rectum was further analyzed and confirmed as waxesters, suggesting that wax esters were not digested in humans andpassed through the gastrointestinal tract unchanged.

In the third report (Ruello, 2004), the author and his wife consumed140–260 g of baked or grilled escolar flesh on seven occasionswith remark-ably different experiences and varying degrees of painless keriorrhea. Theauthor experienced a verymild to heavy formof keriorrhea after eachmealof fresh or frozen fish, while his wife experienced only one episode ofmildkeriorrhea. According to the author, it was typical keriorrhea and painlesspassage of oil per rectum, with no diarrhea (watery liquid feces) at anystage. He also highlighted the potential for embarrassment from stainedclothing arising from the unanticipated passage of oil as an aerosol withflatulence (or when at the toilet). Although the limited and informalrecords may not be conclusive, the author did emphasize that the onelarger portion of 260 g did lead to ‘‘an unsettled lower intestine’’ anddiscomfort at 6 h after consumption and stronger oil flow, whereas thesmaller portions triggered milder keriorrhea with an onset time at 13 hpost ingestion. The author also added that, in a casual trial by nineinformed volunteers on 150–180 g of escolar fillet, only three reportedmild doses of orange oil discharge for about two days after the meal.

Based on the three volunteer trial reports and the summaries of out-break episodes, it may be concluded that typical keriorrhea resulting fromconsumption of reasonable portions of escolar or oilfish fillet will lead tooily discharges generally without warning 1–36 h after ingestion. Fre-quent urges for bowel movements occur due to the lubricant qualities ofthe indigestible wax esters accumulated in the rectum. These symptomsare generally not accompanied by other discomfort nor diarrhea. Suchresponses after intake of escolar or oilfish flesh are not obligatory and varyamong people due to individual sensitivity. There is a possibility thatother symptoms such as stomach cramps, loose bowel movements, diar-rhea, headache, nausea, and vomiting, as reported in various outbreaks,are incidental or due to embarrassment, anxiety, and panic induced bykeriorrhea. It is also likely that diarrhea is a wrong descriptive used byconfused victims suffering from keriorrhea.

V. IDENTIFICATION AND DETECTION

A. Morphological and anatomical analyses

Oilfish is usually sold as cutlets with integument and bone still attached(Fig. 1.8). Bone (1972) studied the musculature of oilfish and found that itpossesses a high proportion of white muscle (80%) and little red muscle in

FIGURE 1.8 Oilfish cutlet. (photo provided by authors).

FIGURE 1.9 Oilfish fillet. (Regulatory Fish Encyclopedia, U.S. FDA, 1993–2008, reprinted

with permission).


the myotomes (Figs. 1.8 and 1.9). Oil stored in bone is not infrequent infish, but the high oil content in oilfish skeletal elements (21.1% in vertebralbone and 30.5% in frontal bone) is remarkable (Bone, 1972). The bonestructure is indeed a girder system enclosing oil sacs, making the bonesignificantly less dense than water (Bone, 1972). Consumers are advisednot to suck the bones because of the notably high oil content (Nordhoff,1928). The integument of oilfish is covered with characteristic ctenoidscales with scattered pores of various sizes (Fig. 1.10) (Bone, 1972). Thepores are connected to a large system of subdermal space, which can beobserved in the transverse section of the fish (Bone, 1972). The character-istic integument of oilfish, if not removed, can assist a trained eye to

FIGURE 1.10 (A) Oilfish integument; (B) close up of oilfish integument (photos provided

by authors).


identify the fish. In contrast, escolar is often sold as fillets without skin(Fig. 1.11), making it difficult to reach a definitive identification. When thewhole fish is available, the highly undulating lateral line in escolar is adistinguishing characteristic (Figs. 1.2B and 1.12), which was used foridentification purposes in an outbreak of histamine poisoning (Kanet al., 2000). An additional characteristic for the two fish is that they areexceptionally oily to the touch.

B. Protein analysis

Polyacrylamide gel electrophoresis (PAGE) and cellulose acetate mem-brane electrophoresis (CAME) were applied to distinguish escolar andoilfish from 27 commercial fish based on muscle protein differences(Ochiai et al., 1984). Myogen fractions from the muscles were subjected

FIGURE 1.11 Escolar sashimi purchased in Japanese restaurant in Hong Kong (photo

provided by authors).

FIGURE 1.12 Fillet from escolar (Regulatory Fish Encyclopedia, U.S. FDA, 1993–2008,

reprinted with permission).


to either PAGE with Coomassie brilliant blue staining or CAME withPonceau 3R staining to visualize the protein profile. The gel was alsostained for lactate dehydrogenase (LDH) and malate dehydrogenase(MDH) activities to look for characteristic patterns to identify escolarand oilfish. Ochiai et al. (1984) concluded that the myogen protein couldbe used to distinguish escolar and oilfish from other fish, while bothdehydrogenase (LDH and MDH) did not give species-specific pattern.Moreover, PAGE is said to be better than CAME for the purpose. How-ever, if fish is processed by cooking or sun-drying, the species-specificproteins could be denatured (Carrera et al., 1999).


C. DNA analysis

In contrast to the heat-labile proteins, DNA is relatively stable and can betested in samples heated up to 120 �C (Lenstra, 2003). It is, unlike proteins,less affected by physiological conditions, environmental factors, storage,and processing (Shaw et al., 2002). DNA sequences are now widely usedfor species identification in DNA barcoding (Ratnasingham and Hebert,2007). Species identification can be made by sequence searches on publicsequence databases, such as GenBank (www.ncbi.nlm.nih.gov) andBOLD (www.barcodinglife.org). DNA sequencing, which determinesthe actual nucleotide types and arrangements in amplified DNAfragments, is used to differentiate escolar and oilfish from othercommonly marketed fish (Ling et al., 2008a,b). Four mitochondrial DNAregions, namely 12S rRNA gene, 16S rRNA gene, cytochrome b gene, andcytochrome oxidase subunit I (COI) gene, were sequenced and the resultsconfirmed that some codfish samples on the market were actually oilfish.The four regions were used to construct four neighbor-joining (NJ) treesand they were all useful in distinguishing escolar and oilfish anddifferentiating the two fish from other fish (Fig. 1.13). DNA sequencingwas also successful when applied to cooked oilfish samples.

D. Lipid analysis

Escolar and oilfish contain a mixture of wax esters with different carbon-chain length, mainly C32, C34, C36, and C38, formed by combiningdifferent fatty acids and fatty alcohols. The dominant fatty acids in escolarand oilfish wax esters are the monounsaturated fatty acids (Table 1.3),namely oleic acid (18:1o9) and eicosenoic acid (20:1o9), while the domi-nant fatty alcohols are saturated and monoenoic fatty alcohols (Table 1.4),known as cetyl alcohol (16:0) and oleyl alcohol (18:1o9). PUFA, which aretrace components in muscle wax esters, are commonly found in waxesters from roe, they include 20:4o6, 20:5o3, 22:5o3 and 22:6o3. Thesedifferences could be due to the functional role in muscle for providingbuoyancy, while that of roe is to store energy and key essential PUFA forfry development (Lee and Patton, 1989).

Wax esters are more hydrophobic than analogous triacylglycerols. Foreach hydrocarbon chain in triacylglycerols there is one hydrophilic estergroup, whereas in wax esters there is one-half of an ester group (Lee andPatton, 1989). Thus, long-chain wax esters are classified as nonpolar lipidswith triacylglycerols being somewhat more polar. This small difference inpolarity is applied to separate wax esters from triacylglycerols and otherlipids using chromatography (Lee and Patton, 1989). The unusually highlevels of wax esters in escolar and oilfish allow thin-layer chromatogra-phy (TLC) or gas chromatography to be applied to differentiate the

http://www.ncbi.nlm.nih.gov

http://www.barcodinglife.org

R1

C4

E2

100

100

86

88

100

100

100

100

100

70

0.02

99

A3

C3

Oilfish group

Escolargroup

A2

A1

R2

E3

E5

E6

E7

A6

A4

A5

G3

W1

H2

H1

E4

M1

C1

C2

C5

B1

B2

G1

J1

S1

S2

Othercommonlymarketedfishsamples

G2

T1

T2

E1

FIGURE 1.13 The neighbor-joining (NJ) trees for 12S rRNA gene sequences from 34 fish

samples. Escolar and oilfish could be distinguished from other commonly marketed fish

(including cod, sea bass, salmon, catfish, swordfish, halibut etc.) (Ling et al., 2008a,b; data

from authors).


two fish. Ling et al. (2008a,b) used TLC to rapidly differentiate escolar andoilfish from other wax ester-absent fish. In that study, lipid was extractedby hexane and applied to silica gel plates. The plate was developed in a

TABLE 1.3 The major fatty acids in the wax esters of escolar and oilfish muscle

Fatty acids

(R chain in Fig. 1.7)

Escolar (%)a(a and b)

Oilfish (%)a(a and c)

16:1 Palmitoleic acid 1.1/2 1.8/3.7

18:1 Oleic acid 64.2/80 72.1/47.720:1 Eicosenoic acid 24.7/15 10.9/6.9

a Data as in a: Mori et al. (1966c); b: Berman et al. (1981) and c: Ruiz-Gutierrez et al. (1997).

TABLE 1.4 The major fatty alcohols in the wax esters of escolar and oilfish muscle

Fatty alcohols

(R0 chain in Fig. 1.7)

Escolar (%)a(a, b, and c) Oilfish (%) a(a)

14:0 Myristyl alcohol 3.1/3/2.6 2.5

16:0 Cetyl alcohol 33.7/43/52.6 48.1

16:1 Palmitoleyl alcohol 4.1/5/3.9 7.4

18:0 Stearyl alcohol 10.4/10/8.9 5.3

18:1 Oleyl alcohol 24.6/16/24.9 29.5

20:1 Eicosenol 11.2/15/2.7 1.5

a Data as in a: Mori et al. (1966c); b: Berman et al., (1981); and c: Nichols et al., (2001).


glass tank lined with filter paper and saturated with xylene, which wasthe mobile phase for resolving nonpolar lipids. After development, theplate was oven-dried and sprayed with 40% sulfuric acid in ethanol:anisaldehyde (9:0.1), and heated at 100 �C until color was observed.A characteristic spot at Rf ¼ 0.6, which belongs to the nonpolar waxesters, was found only in escolar and oilfish (Fig. 1.14). Because wax estersare heat-stable, cooked oilfish samples showed an identical TLC spot tothe untreated oilfish samples. Nichols et al. (2001) analyzed the nonsapon-ifiable lipids of escolar and oilfish by gas chromatography and TLC-flameionization detection and revealed that the lipid class profiles and the waxester-derived fatty alcohol profiles readily distinguish the two fish fromother wax ester-rich fish, such as orange roughy and six species of deep-sea oreos. The lipid class and fatty alcohol profiles of escolar and oilfishare very similar, although oilfish has higher levels of 18:1o9 and 16:1o7,and lower levels of 18:0 than escolar (Nichols et al., 2001). Based on thesubtle differences in lipid class and fatty alcohols profiles, two unknownfillet samples associated with an outbreak of keriorrhea was traced toescolar (Nichols et al., 2001).

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

FIGURE 1.14 Thin layer chromatogram of oil extracted from 26 fish samples observed

under visible light. Only oilfish (lanes 1–7, 11) and escolar samples (lanes 8–10) showed a

characteristic spot at Rf ¼ 0.6 (Ling et al., 2008a,b; data from authors).


VI. WAX ESTER-RICH FISH AND OTHERPOTENTIAL HAZARDS

A. Gempylidae family

Escolar and oilfish are the only species of their respective genera(Alexander et al., 2004; Nakamura and Parin, 1993), and there are another22 species in the same family, Gempylidae (Table 1.5). Species in thisfamily share similar characteristics and thus these species may containindigestible wax esters in their muscle. There is evidence that the presenceof wax esters is an environment-based characteristic rather than aphylogeny-based character; for example, the deeper-living members inMycophidae have higher wax ester contents than the epipelagic species ofthe same family (Nevenzel et al., 1969). However, examination of the lipidcontent and composition in other species of the Gempylidae family,except escolar and oilfish, is limited, and further investigation appearswarranted to both inform industry, health authorities, and governmentagencies and also to safeguard the public.

B. Other deep-sea fish

Fish with more than 10% wax esters in the total lipids of body tissues areuncommon. When higher levels of wax esters are found in epipelagic fishspecies, they are mainly stored in roe and the body lipids of these fish

TABLE 1.5 A list of the 24 species under the family Gempylidae

Species name Common name Remarksa

1 Diplospinus

multistriatus

Striped escolar

2 Epinnula magistralis Domine Consumed as food3 Gempylus serpens Snake mackerel Sold frozen, as

sausages or

fish cake

4 Lepidocybium

flavobrunneum

Escolar Flesh oily and may

have purgative

properties

5 Nealotus tripes Black snake

mackerel6 Neoepinnula

americana

American sackfish

7 Neoepinnula

orientalis

Sackfish

8 Nesiarchus nasutus Black gemfish Consumed as food

9 Paradiplospinus

antarcticus

Antarctic escolar Of no fishery

interest

10 Paradiplospinus

gracilis

Slender escolar Limiteddistribution near

Namibia and

western South

Africa

11 Promethichthys

prometheus

Roudi escolar Reports of ciguatera

poisoning

12 Rexea alisae N/A

13 Rexea antefurcata Long-finned escolar A bycatch of deep-water prawn

trawl fishery in

Australia

14 Rexea bengalensis Bengal escolar

15 Rexea brevilineata Short-lined escolar

16 Rexea nakamurai Nakamura’s escolar

17 Rexea prometheoides Royal escolar Utilized as a food

18 Rexea solandri Silver gemfish Good edible qualityand especially

tasty when

smoked

(continued)


TABLE 1.5 (continued )

Species name Common name Remarksa

19 Rexichthys

johnpaxtoni

Paxton’s escolar

20 Ruvettus pretiosus Oilfish The flesh is veryoily, with

purgative

properties

21 Thyrsites atun Snoek Highly commercial,

marketed fresh

22 Thyrsitoides marleyi Black snoek

23 Thyrsitops

lepidopoides

White snake

mackerel

Rarely caught since

1980. Good forsmoked fish and

fish and chips

24 Tongaichthys

robustus

Tonga escolar

a According to information listed in FishBase (2008).


have no or negligible levels of wax esters. Wax esters, however, are storedin the muscle and other body tissues in deep-sea fish. Wax esters havelower specific gravities than triacylglycerols, and their viscosities aremuch less influenced by temperature and pressure variations; these prop-erties make wax esters superior to triacylglycerols or the presence of aswim bladder for buoyancy control in deep-sea fish. Therefore, it is notunusual to find high levels of wax esters in deep-sea fish. For example,orange roughy, which is a deep-sea species with high levels of wax esters(Table 1.6) (90–97% of total lipids), is commonly available in the market(Fig. 1.15). Wax esters are mainly found in the skin of orange roughy, andthe removal of skin and superficial flesh (deep skinning) significantlyreduces the amount of oil present. However, deep-skinned orange roughystill contain 5.5% total lipids of which as much as 93% is indigestible waxesters (de Koning, 2005). Ruello (2004) mentioned that an informant hadoily discharge after eating orange roughy, and he himself experiencedmild keriorrhea 38 and 60 h after consuming 300 g of this fish. A note waspublished in the Hong Kong Medical Journal to alert the medical profes-sionals to this fish when dealing with sensitive patients (But et al., 2008).In Myctophidae (lantern fish family), many members contain largeamounts of wax esters in the body (Table 1.6). Species in this family arewell-known for their diel vertical migrations, traveling between 10 m(at night) to 3000 m (at day time), and they are abundant and small in

TABLE 1.6 Wax ester-containing fish species

Familya Species name [valid name]bBody part

analyzedcWax esters (%

of total lipid) References

Arripidae Armpits trutta [Arripis

trutta]

Roe 26 Bledsoe et al. (2003)

Bathylagidae Bathylagus antarcticus Muscle 5 Reinhardt and van Vleet

(1986)

Carangidae Seriola aureovittata [Seriola

lalandi]

Roe 42 Joh et al. (1995)

Chlamydoselachidae Chlamydoselachus

anguineus

Liver 58 Shimma and Shimma

(1970)

Coryphaenidae Coryphaena hippurus Roe 36 Lee and Patton (1989)

Gadidae Melanogramus aeglefinus

[Melanogrammus

aeglefinus]


Gempylidae Lepidocybium

flavobrunneum

Muscle 89 Matsumoto et al. (1955),

Nevenzel et al. (1965),Mori et al. (1966c)

Ruvettus pretiosus Muscle 92 Cox and Reid (1932),

Nevenzel et al. (1965),

Mori et al. (1966c)

(continued)





Gonostomatidae Cyclothone acclinidens Whole body 29 Lee and Patton (1989)

Cyclothrone ataria Whole body 70 Kayama and Nevenzel

(1974)

Cyclothrone pallidae

[Cyclothone pallida]Whole body 22 Lee and Patton (1989)

Cyclothrone pseudopallidae

[Cyclothone

pseudopallida]

Whole body 17 Lee and Patton (1989)

Cyclothrone signata Whole body 33 Lee and Patton (1989)

Gonostoma gracile [Sigmops

gracilis]

Whole body 20 Kayama and Nevenzel

(1974)

Howellidae Howella sp. Whole body 16 Patton et al. (1977)Muscle 93

Latimeridae Latimeria chalumnae Adipose tissue 97 Nevenzel et al. (1966)

Lotidae Lota lota Roe 80–85 Kaitaranta and Ackman

(1981)

Lutjanidae Lutjanus campechanus Roe 18 Lee and Patton (1989)

Merlucciidae Macruronis novaezelandiae

[Macruronus

novaezelandiae]


Merluccius capensis Roe 25 Mori and Saito (1966a)

Merluccius hubbsi Roe 28 Mendez et al. (1992)

Moridae Laemonema morosum Muscle 50 Komori and Agawa (1954,1955), Ueno et al. (1955)

[Laemonema longipes] liver 60

Lotella phycis Liver 30 Komori and Agawa,

(1953)

Podonema longipes

[Laemonema longipes]

Liver 25 Hayashi and Yamada

(1976)

Pseudophycis bacchus

[Pseudophycis bachus]


Mugilidae Mugil cephalus Roe 67 Iyengar and Schlenk

(1967), Spener and Sand

(1970)

Mugil japonicus [Mugil

cephalus]

Roe 70 Mori and Saito (1966a)

Myctophidae Benthosema glaciale Whole body 55 Lee and Patton (1989)

Centrobranchus

chaerocephalus

Whole body 15 Patton et al. (1977)

Electrona antarctica Muscle 62 Reinhardt and van Vleet

(1986)

Electrona carlsbergi Whole body 7 Reinhardt and van Vleet

(1986)

Gonichys barnesi

[Gonichthys barnesi]

Whole body 12 Patton et al. (1977)

Gymnoscopelus braueri Whole body 61–90 Phleger et al. (1999)

Gymnoscopelus nicholsi Muscle 20 Reinhardt and van Vleet(1986)

(continued)





Krefftichthyes anderssoni

[Krefftichthys anderssoni]

Whole body 56–94 Phleger et al. (1999)

Lampanyctus ritteri

[Nannobrachium ritteri]

Muscle 87 Nevenzel et al. (1969)

Myctophum nitidulum Whole body 12 Patton et al. (1977)

Neocyttus helgae Muscle 30 Bakes et al. (1995)

Neocyttus rhomboidalis Muscle 22 Bakes et al. (1995)

Oreosoma atlanticum Muscle 9 Bakes et al. (1995)

Protomyctophum bolini Muscle 8 Reinhardt and

van Vleet (1986)

Stenobrachius leucopsarus Muscle 90 Nevenzel et al. (1969)

Symbolophorus evermanni Whole body 10 Nevenzel et al. (1969)Triphoturus mexicanus Muscle 74 Nevenzel et al. (1969)

Nomeidae Cubiceps gracilis Whole body 47 Lee and Patton (1989)

Notosudidae Scopelosaurus sp. Whole juvenile 22 Lee and Hirota (1973)

Nototheniidae Pleuragramma antarcticum Muscle 48 Reinhardt and

van Vleet (1986)

Oreosomatidae Allocyttus niger Muscle 52 Bakes et al. (1995)

Allocyttus verrucosus Muscle 76 Mori et al. (1966b)

Paralepididae Pseudocyttus maculates Muscle 62 Bakes et al. (1995)Paralepsis rissoi [Arctozenus

risso]

Muscle 85 Ackman et al. (1972)

Percidae Perca fluviatilis Roe 80–85 Kaitaranta and Ackman

(1981)

Rachycentridae Rachycentron canadum Roe 36 Lee and Patton (1989)

Sciaenidae Cynoscion nebulosus Roe 40 Iyengar and Schlenk

(1967)

Scombridae Euthynnus alletteratus Roe 31 Lee and Patton (1989)Scomber australasicus Roe 26 Bledsoe et al. (2003)

Sternoptychidae Argyropelecus hawaiiensis

[Argyropelecus sladeni]

Whole body 41 Lee and Hirota (1973)

Stomiidae Astronesthes sp. Whole body 20 Patton et al. (1977)

Eustomias sp. Whole body 10 Lee and Hirota (1973)

Tetragonuridae Tetrogonurus cuvieri

[Tetragonurus cuvieri]

Whole body 40 Lee and Patton (1989)

Trachichthyidae Hoplostethus gilchristi

[Hoplostethus atlanticus]Muscle 97 Mori et al. (1978)

Hoplostethus islandicus

[Hoplostethus atlanticus]

Muscle 90 Kaufmann and

Gottschalk (1954)

Part of the table modified from Lee and Patton (1989, Table 1) with updated information.a According to information listed in FishBase (2008).b Species name cited as the one listed in the original literature, valid name is based on FishBase (2008).c Whole body included the whole specimen, but some studies excluded guts to eliminate lipids in food.

FIGURE 1.15 Orange roughy fillet retailed in Hong Kong (photo provided by authors).


size (FishBase, 2008). They are occasionally found in fish markets. Otherdeep-sea fish families that have high levels of wax esters in their muscle,such as Oreosomatidae (oreo family) and Gonostomatidae (bristlemouthfamily), are mostly of limited fishery interest (FishBase, 2008).

The predominant lipid components in fish roes, like muscles, aretriacylglycerols or phospholipids. Yet certain fish species have high levelsof wax esters in their roes but not in muscles (Table 1.6) (Bledsoe et al.,2003). Wax esters are present in the roe oil, but not in other muscle andintestine tissues in amber fish (Seriola aureovittata) ( Joh et al., 1995). Waxesters are specifically located in mullet roes (Mugil caphalus) and nowhereelse in the fish (Lyengar and Schlenk, 1967). The wax esters in roes, likethose in muscles, may play a role in buoyancy, permeability control,


insulation, or as an energy reserve (Kaitaranta and Ackman, 1981). How-ever, these wax esters may also be, specifically in roe, acting as a fatty acidreservoir for modifying structural lipids after fertilization (Bledson et al.,2003). The functional differences between the wax esters in muscles androes could be attributed to the structural differences of wax esters in roes,in which a much higher content of PUFA are present in roes in compari-son to those in muscles (Iyengar and Schlenk, 1967; Kalogeropoulos et al.,2008). Indeed, fish roes, such as mullet roes, has long been consumed(Bledson et al., 2003; Kalogeropoulos et al., 2008) and the roes from escolarand oilfish are advertised as high-priced souvenirs from Tungkang,Taiwan (Fig. 1.4). However, the high level of wax esters in these fishroes may cause keriorrhea if too much is consumed.

C. Diacylglyceryl ether (DAGE)-rich fish

It was reported in Japan that students in a primary school suffereddiarrhea and oil leakage after consumption of Stromateus maculatus (Iida,1971; Sato et al., 2002). Sato et al. (2002) analyzed the lipid composition ofS. maculatus and found DAGE as the major lipid class in muscle (55% oftotal lipids), but no wax ester was detected. They further conducted anacute toxicity test and found dose-dependent toxicity responses. On day2 of administration, the results showed significant reduction in bodyweight (4.4 � 1.7 g), high diarrhea rate (43%), and high mortality rate(4/7) in mice given DAGE equivalent to 1/40 of their body weight. Thetoxicity was even higher when DAGE and triacylglycerol was giventogether. However, this dose is equivalent to 0.43 g of DAGE in a mouseof 17 g and 1.5 kg of DAGE in a man of 60 kg. The dose tested is farbeyond normal human consumption.

DAGE is widely distributed in various fish (Mori et al., 1972). Somefish, however, possess extraordinary high levels of DAGE in their muscles(Table 1.7) (Endo et al., 2001; Iida, 1971; Mori et al., 1972; Nichols et al.,2001; Sato et al., 2002). DAGE is likely used in deep-sea fish to achievebuoyancy (Endo et al., 2001). Little is reported on the toxicity of DAGE.Capsules of DAGE derived from the livers of deep sea dogfish, however,are widely marketed as neutraceuticals for human consumption (Nicholset al., 2001).

DAGE is not wax. In a strict sense, DAGE excretion should not beclassified as keriorrhea, which is coined specifically for wax excretion,and thus is beyond the coverage in this review. Further studies, however,are recommended concerning Stromateus maculatus and its DAGE-richlipid and possible incidences of illness. If a broader descriptive is neededfor medical description of oil leakage including wax esters and other oils,ladiorrhea, where ‘‘ladi’’ stands for oil in Greek, can be used.

TABLE 1.7 Diacylgylceryl ether (DAGE)- or gylceryl ether (GE)-rich fish species

Family Species name

DAGE/GE

(% of total

lipid) References

Centrolophidae Centrolophus

niger

14.7–92.5 Nichols et al.

(2001)

Centrolophus sp. 27.7 Mori et al. (1972)

Tubbia sp. 2.2–82.1 Nichols et al.(2001)

Nomeidae Cubiceps gracilis 25.4 Mori et al. (1972)

Stromateidae Stromateus

maculatus

20.3 Mori et al. (1972)


VII. DISCUSSION AND RECOMMENDATIONS

A. The rationale: To ban or not to ban?

The incidence of keriorrhea is unlike other food-poisoning cases. Onlysome people have a reaction after eating the fish. Escolar and oilfish havebeen traditionally used for food and as a purgative medicine (Cox andReid, 1932; Gudger, 1925; Gudger and Mowbray, 1927; Helfman et al.,1999; Raisfeld and Patronite, 2006; Ruiz-Gutierrez et al., 1997; Stobbs andBruton, 1991). Some connoisseurs of good foods even gave them highcompliments as dream foods (Bykov, 1983; Raisfeld and Patronite, 2006).Their flesh and roes are regarded as a delicacy in Taiwan (Li, 2007).

Despite the shocking animal study that saw 13 out of 16 rats fed dailywith high doses of escolar or oilfish die within 10 days (Mori et al., 1966c),it is very unlikely that normal human consumption at any single timecould be life-threatening. Indeed, the responses to eating the fish arehighly variable and unpredictable due to differences in individual sus-ceptibility and variation in wax ester content between servings. However,the rationale is that people do not all tolerate the same food in the samemanner, and some people may have medical conditions, such as foodallergies, that preclude them from consuming certain items. Therefore,countries that consider the two fish dangerous and ban them shouldcheck fish on the market constantly to avoid illegal sale, while countriesthat consider the two fish safe should check for mislabeling to ensure thatpeople know what they consume and, at the same time, require a properwarning sign be shown. In either case, food safety is of utmost concernwhether the fish is banned or not. The highest food safety standards can


be achieved through education and inspection, together with appropriatepenalties that deter either illegal sale or mislabeling.

B. Recommendations

1. LabelingThe problem of misidentification and mislabeling of escolar and oilfishoccurs throughout the entire supply chain. Escolar and oilfish are nameddifferently in different places. Moreover, the fish are mislabeled differ-ently across countries, for example, as butterfish and rudderfish inAustralia and as sea bass in Britain. In response to keriorrhea outbreaks, atechnical advisory group was established in Australia to assist the fishingindustry and consumers in identification and labeling. The commonnames of ‘‘escolar’’ and ‘‘oilfish’’ are endorsed for L. flavobrunneum andR. pretiosus, respectively. The advisory group also distributed pictures offish species responsible for keriorrhea to the industry and included bothfish in the Australian Seafood Handbook. Similarly, the Hong KongCentre for Food Safety issued Guidelines on Identification and Labelingof Oilfish/Cod with a view to regulating the naming, labeling andhandling of the two fish (hereafter known as HK Guidelines; WGNCO,2007). In the HK Guidelines, retailers are advised to indicate designatedcommon names on their packages, in which case R. pretiosus and L. flavo-brunneum should both be labeled as ‘‘oilfish’’ and no other names areallowed. In Britain, both L. flavobrunneum and R. pretiosus are labeled thesame as ‘‘escolar’’ according to UK Food Standards Agency. To facilitate aproper communication between authorities in different countries, standardcommon names should be designated for L. flavobrunneum and R. pretiosus,as has occurred in Australia.

2. Risk assessment and educationMany food science and public health professionals are not well-informedof fish-induced keriorrhea. The public and health professionals should beeducated on the potential health threat related to exotic food, such asescolar and oilfish. User-friendly brochures or leaflets are useful to thegeneral public, especially vulnerable groups including pregnant womenand people suffering from cardiovascular and gastrointestinal illness, onthe possible consequences arising from escolar and oilfish consumption.After all, well-informed consumers are less likely to panic once affectedand in turn, they could better describe their symptoms and take properaction, such as seeking help and reporting to pertinent authorities, oncesymptoms are noted. A prompt response to any keriorrhea outbreakallows physicians and sensitive consumers to be alerted to possiblehazards and facilitates rapid diagnoses. Time is an essence to the problembefore it becomes widespread. The seafood and catering industry is


another key player in preventing such outbreaks from becoming moreprevalent. Industry should be educated on identifying problematic fishand use correct seafood names with proper labels and warning signs.Moreover, the catering industry should be educated on the appropriatemanner to handle the fish so as to reduce the level of offensive wax estersand to avoid bacterial growth, which will lead to scrombrotoxin. In short,any education program needs to include the industry, consumers, healthprofessionals, and food safety agencies in order to establish a social normfor appropriate preventive and remedial measures.

3. Warnings and handlingIn countries where escolar and oilfish are not banned from sale, thedisplay of warnings and proper handling procedures should be madecompulsory whenever the fish are available. In the HK Guidelines(WGNCO, 2007), retailers are advised to display supplementary warningstatements, such as: (i) this fish can cause digestive discomfort to someindividuals; (ii) if you are pregnant, have bowel problems, or malabsorp-tion, you are advised not to consume this fish; (iii) if you are eating thisfish for the first time, consume only a small portion; (iv) if you experiencegastrointestinal symptoms after eating this fish, do not consume the fishin future; and (v) seek medical advice if symptoms persist. Both British(Food Standards Agency) and Singaporean (Agri-Food and VeterinaryAuthority) authorities recommend cooking methods, such as grilling, inorder to reduce the oil content. In 1999, the Swedish and the DanishNational Food Administrations required cooking recommendations,including cooking these fish in such a way that most of the fat could beseparated from the dish and the cooking liquid must not be used forpreparation of sauce, to be available where the fish are offered for sale(Alexander et al., 2004). Ruello (2004), however, did not find much differ-ence by grilling or baking. Moreover, people are advised to remove oravoid sucking the bones because of their notably high oil content(Nordhoff, 1928). It was also recommended to keep the storage timeshort because of the high content of histidine (Alexander et al., 2004).

For health professionals in consultations with patients complaining ofdiarrhea, it is pertinent to ask ‘‘Is it oily?’’ and ‘‘Have you consumedfish?’’ If answers are confirmative, the case should be reported to foodinspection agencies. To follow up on the alert, food inspection agenciesshould collect the expelled oil sample, a residual sample of the cookedfish (if available), and any uncooked fish from the same source to checkfor wax esters. If wax esters are detected, DNA analysis should beapplied, if possible, to identify the source species. Concurrently duringthe inspection process, food safety agencies should trace back to thesupply source of the fish to see if proper labels were in place. If not,


warnings should be issued to the public, healthcare professionals andfishery industry.

4. Detection and inspectionAny sound policy needs proper inspection to facilitate implementation.A good authentication method is clearly necessary to check against sub-stitution, misidentification, and mislabeling. This is particularly impor-tant at an initial stage of a keriorrhea outbreak, to stop further circulationof incriminated samples at outlets and sources and to prevent the situa-tion from escalating- to political uproar. The recent situation in HongKong was illuminating. In fall 2006, oilfish cutlets were labeled as codfishbut at lower prices. Initially, scattered cases of keriorrhea and othercomplaints surfaced through the mass media. In response to publicinquiry, the agency responsible for food monitoring merely advised citi-zens to buy from reliable sources and recommended the public to becautious in purchasing fish. That recommendation could hardly curbthe spread of keriorrhea and soon over 600 cases were recorded bymid-January 2007. The general public in Hong Kong was panicking andvarious political parties organized demonstrations demanding for propergovernment actions (Fig. 1.16). To help round up mislabeled items andoffer a tool to the fish industry, the team led by the senior authorannounced in early February 2007 a rapid TLC detection method validatedby molecular sequencing and GC-MS (Goh, 2007; Ling et al., 2008a,b).The uproar then subsided.

Authentication methods, including morphological, anatomical, pro-tein, DNA, and lipid analyses, can be utilized in differentiating escolarand oilfish from other commonly marketed fish. Accuracy is of utmostimportance. Cost and time are also important factors in screening largenumbers of fish samples, either for routine inspection or after an out-break. The simple, rapid, and inexpensive TLC method developed byLing et al. (2008a,b) serves as a good example for differentiating escolarand oilfish from other commonly marketed fish, which contain no waxester. It is applicable to cooked or excreted samples as wax esters are heat-stable and indigestible by humans. The whole process, from lipid extrac-tion to staining, took less than 30 min while the equipment involved isreadily available in most analytical laboratories. Because each TLC platecan accommodate more than 20 samples, and multiple plates can be runsimultaneously, the method is space-saving and suitable for screeninglarge sample sets. The method, however, cannot distinguish escolar fromoilfish nor differentiate the two fish from other wax ester-containing fish.Other more thorough protocols, such as DNA sequencing, can be the nextstep in confirming the precise sample identity.

FIGURE 1.16 Demonstration demanding immediate government actions against oilfish-

induced keriorrhea epidemic in Hong Kong (photo of the Democratic Alliance for the

Betterment and Progress of Hong Kong, reprinted with permission).


VIII. CONCLUSIONS

The escolar- and oilfish-related problems are global in scope and respectno national boundaries. Differences in opinions, lack of clinical data,confusing labeling systems, and expensive detection methods are allfactors that have contributed to the prevalence of the problem overdecades. Albeit the uncertainty, like playing Russian roulette, it is clearthat some people consume the fish without any notable response whilesome others experience serious keriorrhea. It is important to regulate thefish, either by complete banning or designated labeling, and educate thegeneral public for the potential risks regarding escolar and oilfish con-sumption. Keriorrhea is likely not restricted to escolar and oilfish, andother wax ester-rich fish, known or unknown, may also pose a threat toconsumers. It is therefore crucial to inform health professions of thedistinctive symptoms of expelled oil droplets. Lastly, the academia andresearch agencies should continue to investigate the lipids of various fishfor the possible identification of those species with unusually high contentof wax esters, given that fish are the largest group of vertebrates and asignificant part of human diet.


REFERENCES

Ackman, R. G., Hooper, S. N., Epstein, S., and Kelleher, M. (1972). Wax esters of barracudinalipid: A potential replacement of sperm whale oil. J. Am. Oil Chem. Soc. 49, 378–382.

Alexander, J., Autrup, H., Bard, D., Carere, A., Costa, L. G., Cravedi, J.-P., Di Domenico, A.,Fanelli, R., Fink-Gremmels, J., Gilbert, J., Grandjean, P., Johansson, N., et al. (2004).Opinion of the scientific panel on contaminants in the food chain on a request from thecommission related to the toxicity of fishery products belonging to the family of Gempy-lidae. Eur. Food Saf. Authority J. 92, 1–5.

Bakes, M. J., Elliott, N. G., Green, G. J., and Nichols, P. D. (1995). Variation in lipid composi-tion of some deep-sea fish (Teleostei: Oreosomatidae and Trachichthyidae). Comp. Bio-chem. Physiol. 111B, 633–642.

Berman, P., Harley, E. H., and Spark, A. A. (1981). Keriorrhea—The passage of oil per

rectum—After ingestion of marine wax esters. S. Afr. Med. J. 59, 791–792.Bettoso, N. and Dulcic, J. (1999). First record of the oilfish Ruvettus pretiosus (Pisces:

Gempylidae) in the northern Adriatic Sea. J. Mar. Biol. Assoc. UK 79, 1145–1146.Bledsoe, G. E., Bledsoe, C. D., and Rasco, B. (2003). Caviars and fish roe products. Crit. Rev.

Food Sci. Nutr. 43, 317–356.Bone, Q. (1972). Buoyancy and hydrodynamic functions of integument in the castor oil fish,

Ruvettus pretiosus. Copeia 1, 78–87.Brandes, C. H. and Dietrich, R. (1953). The distribution of fat in the bodies of herrings.

Veroffentl Inst Meeresforsch Bremerhaven 2, 109–121.Brendtro, K. S., McDowell, J. R., and Graves, J. E. (2008). Population genetic structure of

escolar (Lepidocybium flavobrunneum). Mar. Biol. 155, 11–22.Brewer, M. S. and Prestat, C. J. (2002). Consumer attitudes toward food safety issues. J. Food

Saf. 22, 67–83.Busson-Breysse, J., Farines, M., and Soulier, J. (1994). Jojoba wax: Its esters and some of its

minor components. J. Am. Oil Chem. Soc. 71, 999–1002.But, P. P. H., Ling, K. H., and Cheng, S. W. (2008). Orange roughy is rich with indigestible

wax esters. Hong Kong Med. J. 14, 246.Butt, A., Aldridge, K., and Sander, C. (2004a). Infections related to the ingestion of seafood.

Part I: Viral and bacterial infections. Lancet Infect. Dis. 4, 201–212.Butt, A., Aldridge, K., and Sander, C. (2004b). Infections related to the ingestion of seafood.

Part II: Parasitic infections and food safety. Lancet Infect. Dis. 4, 294–300.Bykov, V. P. (1983). ‘‘Marine Fishes: Chemical Composition and Processing Properties.’’

Amerind Publishing, New Delhi, India.Carrera, E., Garcia, T., Cespedes, A., Gonzalez, I., Fernandez, A., Hernandez, P. E., and

Martin, R. (1999). Salmon and trout analysis by PCR-RFLP for identity authentication.J. Food Sci. 64, 410–413.

Carter, C. L. and Malcolm, J. (1927). Observations on the biochemistry of ‘‘mutton bird’’ oil.Biochem. J. 21, 484–493.

CBC (Canadian Broadcasting Corporation). (2007). Canadians Fall Ill After EatingMislabeled Oily Fish. Canadian Broadcasting Corporation, Canada. http://www.cbc.ca/health/story/2007/02/23/oilfish.html. Accessed Nov. 5, 2008.

CFIA (Canadian Food Inspection Agency). (2007). ‘‘Facts on Escolar.’’ Canadian FoodInspection Agency, Canada. http://www.inspection.gc.ca/english/fssa/concen/specif/escoe.shtml. Accessed Nov. 5, 2008.

Chong, W. (2007). Sale of oilfish to be curbed. The Standard 26 Jan.Chua,L.H. (2007). ‘‘PotentialHealth IssuesAssociatedwithConsumptionofEscolarandOilfish.’’

Agri-Food and Veterinary Authority, Singapore. http://www.ava.gov.sg/NR/rdonlyres/191531A0-5689-4DD9-9648-AE283FD78656/14747/Cir_FoodTradersEscolarFish1.pdf.Accessed Nov. 5, 2008.

http://www.cbc.ca/health/story/2007/02/23/oilfish.html



http://www.inspection.gc.ca/english/fssa/concen/specif/escoe.shtml



http://www.ava.gov.sg/NR/rdonlyres/191531A0-5689-4DD9-9648-AE283FD78656/14747/Cir_FoodTradersEscolarFish1.pdf




Chung, C. (2007a). Label mistake revealed in oilfish saga. The Standard 29 Jan.Chung, C. (2007b). Legislation to regulate fish products on the table. The Standard 3 Feb.Collette, B. B., Pottho, V. T., Richards, W. J., Ueyanagi, S., Russo, J. L., and Nishikawa, Y.

(1984). Scombroidei: Development and relationships. In ‘‘Ontogeny and Systematics ofFishes’’ (H. G. Moser, W. J. Richards, D. M. Cohen, M. P. Fahay, A. W. Kendall, Jr., andS. L. Richardson, Eds.), pp. 591–620. American Society of Ichthyologists and Herpetolo-gists, Special Publication 1, Allen Press, Lawrence.

Connolly, N., Wong, M., and Fong, L. (2007). ParknShop withdraws oilfish, cod as 14 fall ill.The South China Morning Post 24 Jan.

Cox, W. M. and Reid, E. E. (1932). The chemical composition of oil of Ruvettus pretiosus, the‘‘castor oil fish’’. J. Am. Chem. Soc. 54, 220–229.

de Koning, A. J. (2005). Phospholipids of marine origin: The orange roughy (Hoplostethus

atlanticus). S. Afr. J. Sci. 101, 414–416.Doyle, E. (2002). Too much fish oil. UW-Madison Food Res. Inst. Newsl. 14, 6.Du Plessis, K., Lopata, A. L., and Steinman, H. (2004). Adverse reactions to fish. Curr. Allergy

Clin. Immunol. 17, 4–8.Endo, Y., Tagiri-Endo, M., Seo, H. S., and Fujimoto, K. (2001). Identification and quantifica-

tion of molecular species of diacylglyceryl ether by reversed-phase high-performanceliquid chromatography with refractive index detection and mass spectrometry. J. Chro-matogr. A 911, 39–45.

Erkkila, A. T., Lichtenstein, A. H., Mozaffarian, D., and Herrington, D. M. (2004). Fish intakeis associated with a reduced progression of coronary artery atherosclerosis in postmeno-pausal women with coronary artery disease. Am. J. Clin. Nutr. 80, 626–632.

Feldman, K. A., Werner, S. B., Cronan, S., Hernandez, M., Horvath, A. R., Lea, C. S.,Au, A. M., and Vugia, D. J. (2005). A large outbreak of scombroid fish poisoningassociated with eating escolar fish (Lepidocybium flavobrunneum). Epidemiol. Infect. 133,29–33.

Fernandez, E., Chatenoud, L., La Vecchia, C., Negri, E., and Franceschi, S. (1999). Fishconsumption and cancer risk. Am. J. Clin. Nutr. 70, 85–90.

FDA (Food andDrugAdministration). (2001a). Natural toxins. In ‘‘Fish and Fishery ProductsHazards and Controls Guidance,’’ 3rd ed., pp. 73–82. Center for Food Safety and AppliedNutrition, Office of Seafood, Washington, DC.

FDA (Food and Drug Administration). (2001b). Scombrotoxin (histamine) formation. In‘‘Fish and Fishery Products Hazards and Controls Guidance,’’ 3rd ed., pp. 83–102. Centerfor Food Safety and Applied Nutrition, Office of Seafood, Washington, DC.

FishBase. (2008). ‘‘FishBase: World Wide Web Electronic Publication’’ (R. Froese andD. Pauly, eds.). www.fishbase.org. Accessed Nov. 5, 2008.

Fuga, S. (2007). ‘‘Oilfish sold as ‘‘Sushi’’ in Korea.’’ Korea Sparkling Worldpress. http://koreasparkling.wordpress.com/2007/05/04/oilfish-sold-as-sushi-in-korea. AccessedNov 5, 2008.

Givney, R. C. (2002). Illness associated with rudderfish/ escolar in South Australia. Commu-

nicable Disease Intelligence 26, 440.Goh, L. (2007). Importers can use fast test for oilfish, say researchers. The South ChinaMorning

Post 7 Feb.Gorreta, F., Bernasconi, R., Galliani, G., Salmona, M., Tacconi, M. T., and Bianchi, R. (2002).

Wax esters of o-3 polyunsaturated fatty acids: A new formulation as a potential foodsupplement—Digestion and absorption in rats. Lebensmittel-Wissenschaft und Technologie

35, 458–465.Gregory, J. (2002). Outbreaks of diarrhoea associated with butterfish in Victoria. Communica-

ble Disease Intelligence 26, 439–440.

http://www.fishbase.org

http://koreasparkling.wordpress.com/2007/05/04/oilfish-sold-as-sushi-in-korea




Guallar, E., Sanz-Gallardo, I., van’t Veer, P., Bode, P., Aro, A., Gomez-Aracena, J., Kark, J. D.,Riemersma, R. A., Martın-Moreno, J. M., and Kok, F. J. (2002). Mercury, fish oils, and therisk of myocardial infarction. N. Engl. J. Med. 347, 1747–1754.

Gudger, E. W. (1925). A new purgative, the oil of the ‘‘castor oil fish,’’ Ruvettus. Boston Med.

Surg. J. 192, 107–111.Gudger, E. W. and Mowbray, L. L. (1927). The oilfish, Ruvettus pretiosus, at Bermuda. Science

65, 145–146.Hansen, I. A. and Mead, J. F. (1965). The fate of dietary wax esters in the rat. Proc. Soc. Exp.

Biol. Med. 120, 527–532.Hashimoto, K. and Fusetani, N. (1978). Toxins occurring in commercially important marine

organisms. In ‘‘Section I Report of the Session and of the Indo-Pacific Fishery Commis-sion Symposium on Fish Utilization Technology and Marketing in the IPFC Region,’’pp. 416–424. Indo-Pacific Fishery Commission Proceedings 18th Session, Manila,Philippines, 8–17 March 1978; Indo-Pacific Fishery Commission, FAO Regional Officefor Asia and the Pacific, Bangkok, Thailand.

Hayashi, K. and Yamada, M. (1976). The lipids of marine animals from various habitatdepths. V. Composition of wax esters and triglycerides of the gadoid fish, Podonemalongipes. Bull. Faculty Fisheries Hokkaido Univ. 26, 356–366.

He, K., Song, Y., Daviglus, M. L., Liu, K., Van Horn, L., Dyer, A. R., and Greenland, P. (2004).Accumulated evidence on fish consumption and coronary heart disease mortality.A meta-analysis of cohort studies. Circulation 109, 2705–2711.

Helfman, G. S., Collette, B. B., and Facey, D. E. (1999). ‘‘The Diversity of Fishes.’’ BlackwellScience, Malden, MA.

Henderson, R. J. and Sargent, J. R. (1980). Biosynthesis of neutral lipids of Euchaeta norvegica.Mar. Biol. 56, 1–6.

Hightower, J. M. and Moore, D. (2003). Mercury levels in high-end consumers of fish.Environ. Health Perspect. 111, 604–608.

Hu, F. B., Bronner, L., Willett, W. C., Stampfer, M. J., Rexrode, K. M., Albert, C. M.,Hunter, D., and Manson, J. E. (2002). Fish and omega-3 fatty acid intake and risk ofcoronary heart disease in women. J. Am. Med. Assoc. 287, 1815–1821.

Iacono, J. M. and Dougherty, R. M. (1993). Effects of polyunsaturated fats on blood pressure.Ann. Rev. Nutr. 13, 243–260.

Iida, H. (1971). Glyceryl ethers found in Stromateus maculates. Bull. Jpn. Soc. Sci. Fisheries 37,338.

Iyengar, R. and Schlenk, H. (1967). Wax esters of mullet (Mugil cephalus) roe oil. Biochemistry6, 369–402.

Jacquet, J. L. and Pauly, D. (2008). Trade secrets: Renaming and mislabeling of seafood.Mar.

Policy 32, 309–318.James, K. A. C. and Body, D. R. (1986). A nutritional evaluation of orange roughy (Hoplos-

tethus atlanticus) using growing pigs. NZ J. Technol. 2, 219–223.James, K. A. C. and Treloar, B. P. (1984). Comparative effects of orange roughy (Hoplostethus

atlanticus) and snapper (Chrysophrys auratus) in the diets of growing rats. NZ J. Sci. 27,295–305.

Joh, Y. G., Brechany, E. Y., and Christie, W. W. (1995). Characterization of wax esters in theroe oil of amber fish, Seriola aureovittata. J. Amer. Oil Chemists Soc. 72, 707–713.

Kaitaranta, J. K. and Ackman, R. G. (1981). Total lipids and lipid classes of fish roe. Comp.Biochem. Physiol. 69B, 725–729.

Kalogeropoulos, N., Nomikos, T., Chiou, A., Fragopoulou, E., and Antonopoulou, S. (2008).Chemical composition of Greek avgotaracho prepared from mullet (Mugil cephalus):Nutritional and health benefits. J. Agric. Food Chem. 56, 5916–5925.


Kan, K., Ushiyama, H., Shindo, T., Uehara, S., and Yasuda, K. (2000). Outbreak of histaminepoisoning due to ingestion of fish (Lepidocybium flavobrunneum). J. Food Hyg. Soc. Jpn(Shokuhin Eiseigaku Zasshi) 41, 116–121.

Kaufmann, H. P. and Gottschalk, E. (1954). The body oils of Hoplostethus islandicus. FetteSeifen Anstrichm 56, 512–518.

Kawai, N., Nakayama, Y., Matsuka, S., and Mori, T. (1985). Lipid composition of varioustissues of Lepidocybium flavobrunneum. Yukagaku (Jpn Oil Chem. Soc.) 34, 25–31.

Kayama, M. and Nevenzel, J. C. (1974). Wax ester biosynthesis by midwater marine animals.Mar. Biol. 24, 279–285.

Kazuo, S. (1999). Toxins in marine animals. J. Jpn Assoc. Acute Med. 10, 4–27.Kolattukusy, P. E. (1976). Introduction to natural waxes. In ‘‘Chemistry and Biochemistry of

Natural Waxes,’’ pp. 1–12. Elsevier, Amsterdam.Komori, S. andAgawa, T. (1953). The new alcohol of liver oil of Lotella phycis. J. Chem. Soc. Jpn.

74, 1025–1026.Komori, K. and Agawa, T. (1954). Liver oil of Laemonema morosum Matsubara. II. Studies on

docosenol fractions. J. Chem. Soc. Jpn. 75, 1051–1054.Komori, K. and Agawa, T. (1955). Oil on Laemonema morosum. I. Research on the docosenol

fraction. J. Amer. Oil Chem. Soc. 32, 525–528.Kostyniak, P. J., Stinson, C., Greizerstein, H. B., Vena, J., Buck, G., and Mendola, P. (1998).

Relation of Lake Ontario fish consumption, lifetime lactation, and parity to breast milkpolychlorobiphenyl and pesticide concentrations. Environ. Res. 80, S166–S174.

Lau, N. (2007). ParknShop hit with ‘‘mild’’ $45,000 fine for mislabeling oilfish as cod.The Standard 18 Dec, p. 10.

Lawrence, D. T., Dobmeier, S. G., Bechtel, L. K., and Holstege, C. P. (2007). Food poisoning.Emerg. Med. Clin. North Am. 25, 357–373.

Leask, A., Yankos, P., and Ferson, M. J. (2004). Fish, so foul! Foodborne illness caused bycombined fish histamine and wax ester poisoning. Commun. Dis. Intell. 28, 83–85.

Lee, R. F. and Hirota, J. (1973). Wax esters in tropical zooplankton and nekton and thegeographical distribution of wax esters in marine copepods. Limnology Oceanography 18,227–239.

Lee, R. F. and Patton, J. S. (1989). Alcohol and waxes. In ‘‘Marine Biogenic Lipids, Fats, andOils’’ (R. G. Ackman, Ed.), Vol. 1, pp. 73–102. CRC Press, Boca Raton.

Lehane, L. and Lewis, R. J. (2000). Ciguatera: Recent advances but the risk remains. Int. J. FoodMicrobiol. 61, 91–125.

Lehrer, S. B., Ayuso, R., and Reese, G. (2003). Seafood allergy and allergens: A review. Mar.Biotechnol. 5, 339–348.

Lenstra, J. A. (2003). DNAmethods for identifying plant and animal species in food. In ‘‘FoodAuthenticity and Traceability’’ (M. Lees, Ed.), pp. 34–53. CRC Press, Boca Raton.

Li, J. L. (2007). Taiwanese Fish Expert’s First Lecture to Seafood. IF Publishing, Taipei.Ling, K. H., Cheung, C. W., Cheng, S. W., Cheng, L., Li, S. L., Nichols, P. D., Ward, R. D.,

Graham, A., and But, P. P. H. (2008a). Rapid detection of oilfish and escolar in fish steaks:A tool to prevent keriorrhea episodes. Food Chem. 110, 538–546.

Ling, K. H., Cheung, C. W., Cheng, S. W., Cheng, L., Li, S. L., Nichols, P. D., Ward, R. D.,Graham, A., and But, P. P. H. (2008b). Erratum to ‘‘Rapid detection of oilfish and escolarin fish steaks: A tool to prevent keriorrhea episodes’’. Food Chem. 111, 795.

Lyengar, R. and Schlenk, H. (1967). Wax esters of mullet (Mugil cephalus) roe oil. Biochemistry6, 396–402.

Lynch, A.W. (2004). ‘‘Southern andWestern Tuna and Billfish Fishery Data Summary 2003.’’Logbook Program, Australian Fisheries Management Authority, Canberra.

Lyon, P. (2008). Consumer confidence, food safety, and salmon farming. In ‘‘Aquaculture,Innovation and Social Transformation’’ (K. Culver and D. Castle, Eds.), Vol. 17,


pp. 297–305. The International Library of Environmental, Agricultural and Food Ethics,Springer-Verlag, Netherlands.

Macau Disease Control Centre. (2007). ‘‘Effects of Escolar and Oilfish on Health’’. Depart-ment of Health, Macau. http://www.ssm.gov.mo/foodsafety/doc/d1.pdf. AccessedNov. 5, 2008.

Matsumoto, T., Sone, H., and Niiya, I. (1955). The lipide of a fish, Xenogramma carinatum.J. Jpn. Oil. Chem. Soc. 4, 131–133.

Matsuo, N. (1962). Nutritional effects of oxidized and thermally polymerized fish oils.In ‘‘Lipids and Their Oxidation’’ (H. W. Schultz, E. A. Day, and R. O. Sinnhuber, Eds.),p. 321. Avi Publishing Co., Westport.

Mendez, E., Fernandez, M., Pazo, G., and Grompone, M. A. (1992). Hake roe lipids: Compo-sition and changes following cooking. Food Chem. 45, 179–181.

Miyazawa, K. and Noguchi, T. (2001). Distribution and origin of tetrodotoxin. J. Toxicol.Toxins Rev. 20, 11–33.

Mok, D. (2007). Wellcome sold sushi and sashimi made from oilfish. The South Morning Post

2 Feb.Mori, M. and Saito, T. (1966a). The occurrence and composition of wax in mullet and

stockfish roe. Bull. Jpn. Soc. Sci. Fisheries 32, 730–736.Mori, M., Saito, T., and Nakanishi, Y. (1966b). Occurrence and chemical properties of wax in

the muscle of an African fish, Allocyttus verrucosus. Bull. Jpn. Soc. Sci. Fisheries 32, 668–672.Mori, M., Saito, T., Nakanishi, Y., Miyazawa, K., and Hashimoto, Y. (1966c). The composition

and toxicity of wax in the flesh of castor oil fishes. Bull. Jpn. Soc. Sci. Fisheries 32, 137–145.Mori, M., Hikichi, S., Kamiya, H., and Hashimoto, Y. (1972). Three species of teleost fish

having diacyl glyceryl ethers in the muscle as a major lipid. Bull. Jpn. Soc. Sci. Fisheries 38,56–63.

Mori, M., Yasuda, S., and Nishimuro, S. (1978). Two species of teleosts having wax esters ordiacyl glyceryl ethers in the muscle as a major lipid. Bull. Jpn. Soc. Sci. Fisheries 44,363–367.

Mozaffarian, D. and Rimm, E. B. (2006). Fish intake, contaminants, and human health:Evaluating the risks and the benefits. J. Am. Med. Assoc. 296, 1885–1899.

Munk, I. and Rosenstein, A. (1891). Cetyl palmitate metabolism in the rat. Virchows Archiv furPathologische Anatomie und Physiologie und fur Klinische Medizin 123, 230.

Nakamura, I. (1995). Gempylidae. In ‘‘Guia FAO para identification de Especies para lo Finesde la Pesca. Pacifico Centro-Oriental’’ (W. Fischer, F. Krupp, W. Schneider, C. Sommer,K. E. Carpenter, and V. Niem, Eds.), Vol. 3, pp. 1106–1113. Food and AgricultureOrganization of the United Nations, Rome, Italy.

Nakamura, I. and Parin, N. V. (1993). FAO species catalogue: Snake mackerels and cutlass-fishes of the World. In ‘‘FAO Fisheries Synopsis No. 125’’ (I. Nakamura and N. V. Parin,Eds.), Vol. 15, pp. 29–30, 52–53. Food and Agriculture Organization of the UnitedNations, Rome, Italy.

Nelson, D. R., Freeman, T. P., and Buckner, J. S. (2000). Waxes and lipids associated with theexternal waxy structures of nymphs and pupae of the giant whitefly, Aleurodicus dugesii.Comp. Biochem. Physiol. B 125, 265–278.

Nevenzel, J. C., Rodegker, W., and Mead, J. F. (1965). The lipids of Ruvettus pretiosus muscleand liver. Biochemistry 4, 1589–1594.

Nevenzel, J. C., Rodegker, W., Mead, J. F., and Gordon, M. S. (1966). Lipids of the livingcoelacanth, Latimeria chalumnae. Science 152, 1753–1755.

Nevenzel, J. C., Rodegker, W., Robinson, J. S., and Kayama, M. (1969). The lipids of somelantern fishes (family Myctophidae). Comp. Biochem. Physiol. 31, 25–36.

Nichols, P. D., Mooney, B. D., and Elliott, N. G. (2001). Unusually high levels of non-saponifiable lipids in fishes escolar and rudderfish: identification by gas and thin-layerchromatography. J. Chromatogr. A 936, 183–191.

http://www.ssm.gov.mo/foodsafety/doc/d1.pdf

http://www.ssm.gov.mo/foodsafety/doc/d1.pdf


NMFS (National Marine Fisheries Services). (2008). ‘‘National Marine Fisheries ServicesAnnual Landings Database.’’ http://www.st.nmfs.noaa.gov/st1/commercial/landings/annual_landings.html. Accessed Nov. 5, 2008.

Noguchi, T. and Ebesu, J. S. M. (2001). Puffer poisoning: Epidemiology and treatment.J. Toxicol. Toxins Rev. 20, 1–10.

Norat, T., Bingham, S., Ferrari, P., Slimani, N., Jenab, M., Mazuir, M., Overvad, K., Olsen, A.,Tj�nneland, A., Clavel, F., Boutron-Ruault, M., Kesse, E., et al. (2005). Meat, fish, andcolorectal cancer risk: The European prospective investigation into cancer and nutrition.J. Nat. Cancer Inst. 97, 906–916.

Nordhoff, C. B. (1928). Fishing for the oilfish. NY Nat. Hist. 28, 40–45.Ochiai, Y., Watabe, S., Hashimoto, K., Narita, H., Ukishima, Y., and Nara, M. (1984).

Biochemical identification of two gempylid fishes causative of a food poisoning. Bull.Jpn. Soc. Sci. Fisheries 50, 721–725.

Pattaravivat, J., Morioka, K., Shirosaki, M., and Itoh, Y. (2008). Effect of washing conditionson the removal of lipid from the fatty fish escolar (Lepidocybium flavobrunneum) meat.J. Biol. Sci. 8, 34–42.

Patton, J. S., Abraham, S., and Benson, A. A. (1977). Lipogenesis in the intact coral Pocilloporacapitata and its isolated zooanthellae: Evidence for a light-driven carbon cycle betweensymbiont and host. Mar. Biol. 44, 235–247.

Paulin, C. D. (1980). First record of Lepidocybium flavobrunneum (Pisces: Gempylidae) fromNew Zealand. NZ J. Mar. Freshwater Res. 14, 405–407.

Phleger, C. F. (1998). Buoyancy in marine fishes: Direct and indirect role of lipids. Am.

Zoologist 38, 321–330.Phleger, C. F. and Grigor, M. R. (1990). Role of wax esters in determining buoyancy in

Hoplostethus atlanticus (Beryciformes: Trachichthyidae). Mar. Biol. 105, 229–233.Phleger, C. F., Nelson, M. M., Mooney, B. D., and Nichols, P. D. (1999). Wax esters versus

triacylglycerols in myctophid fishes from the Southern ocean. Antarct. Sci. 11, 436–444.Poulsen, L. K., Hansen, T. K., N�rgaard, A., Vestergaard, H., Skov, P. S., and Bindslev-

Jensen, C. (2003). Allergens from fish and egg. Allergy 56, 39–42.QueenslandHealth. (2008). ‘‘Food Industry Fact Sheet 9: Escolar and Rudderfish not Suitable

for Catering.’’ Queensland Government, Australia. http://www.health.qld.gov.au/ph/Documents/ehu/23004.pdf. Accessed Nov. 5, 2008.

Raisfeld, R. and Patronite, R. (2006). The ‘‘It’’ fish. New York Magazine. www.nymag.com/restaurants/features/16899. Accessed Nov 5, 2008.

Ratnasingham, S. and Hebert, P. D. N. (2007). BOLD: The barcoding of life data system.Mol. Ecol. Notes 7, 355–364.

Regulatory Fish Encyclopedia, U.S. Food and Drug Administration. (1993–2008). ‘‘Photo-graphy: B. Tenge, D. Mellen.’’ http://www.cfsan.fda.gov/~frf/rfe0.htmlAccessed.Dec. 11, 2008.

Reinhardt, S. B. and van Vleet, E. S. (1986). Lipid composition of twenty-two species ofAntarctic midwater zooplankton and fish. Mar. Biol. 91, 149–159.

Ruello, N. (2004). Report on the oil content, composition, and the consumption of escolarprepared for the Australian Government Department of Health and Ageing. Ruello andAssociates Pty Ltd, Australia (unpublished report).

Ruiz-Gutierrez, V., Perez-Zarza, M. C., Muriana, F. J. G., and Bravo, L. (1997). Lipid and fattyacid composition of muscle and internal organs from Ruvettus pretiosus. J. Fish Biol. 50,1353–1357.

Sargent, J. R. (1978). Marine wax esters. Sci. Prog. Oxford 65, 437–458.Sargent, J. R., Gatten, R. R., and Merrett, N. R. (1983). Lipids of Hoplostethus atlanticus and

H. mediterraneus (Beryciformes: Trachichthyidae) from deep water to the west of Britain.Mar. Biol. 74, 281–286.

http://www.st.nmfs.noaa.gov/st1/commercial/landings/annual_landings.html



http://www.health.qld.gov.au/ph/Documents/ehu/23004.pdf



http://www.nymag.com/restaurants/features/16899

http://www.nymag.com/restaurants/features/16899

http://www.cfsan.fda.gov/~frf/rfe0.html

http://www.cfsan.fda.gov/~frf/rfe0.html


Sato, T., Seo, H. S., Endo, Y., and Fujimoto, K. (2002). Diacylglyceryl ether as themajormusclelipid in Stromateus stellatus and its hydrolyzability by lipase and oral acute toxicity onmice. Jpn. Soc. Fisheries Sci. 68, 569–575.

Savage, E. S. (1951). A comparative study of the utilization of jojoba and cottonseed oil in therat. PhD thesis, Department of Biochemistry and Nutrition, University of SouthernCalifornia, Los Angeles, CA.

Savary, P. (1971). The action of pure pig pancreatic lipase upon esters of long-chain fattyacids and short-chain primary alcohols. Biochim. Biophys. Acta 248, 149–155.

Schneider, W. (1990). ‘‘FAO Species Identification Sheets for Fishery Purposes. Field Guideto the CommercialMarine Resources of the Gulf of Guinea’’, p. 268. Food andAgricultureOrganization of the United Nations, Regional Office for Africa, Rome, Italy.

Senkowsky, S. (2004). Fear of fish: The contaminant controversy. BioScience 54, 986–988.Shadbolt, C., Kirk, M., and Roche, P. (2002). Editorial: Diarrhea associated with consumption

of escolar (rudderfish). Commun. Dis. Intell. 26, 436–438.Shaw, P. C., Wang, J., and But, P. P. H. (Eds.). (2002). ‘‘Authentication of Chinese Medicinal

Materials by DNA Technology.’’ World Scientific, Singapore.Shimma, H. and Shimma, H. (1970). Studies on liver oil of a frill shark. Bull. Jpn. Soc. Sci.

Fisheries 36, 1157–1162.Spener, F. and Sand, D. M. (1970). Neutral alkoxylipids and wax esters of mullet (Mugil

cephalus) roe. Comp. Biochem. Physiol. 34, 715–719.Spencer, G. F., Plattner, R. D., andMiwa, T. (1977). Jojoba oil analysis by high pressure liquid

chromatography and gas chromatography/mass spectrometry. J. Am. Oil Chemists Soc.

54, 187–189.Statham, D. (2003). ‘‘Potential Health Issues Associated with the Consumption of Escolar’’.

Food Standards Agency, UK. http://www.food.gov.uk/multimedia/pdfs/escolarletter.pdf. Accessed Nov. 5, 2008.

Stenhouse, F. (2007). ‘‘Market Access Advice: Prohibition of Oilfish and Escolar Exports toSouth Korea (Aug 13).’’ Australian Quarantine and Inspection Service, Australia. http://www.daff.gov.au/aqis/export/fish/fish-notices/2007/FISH0708_Oilfish_escolar_market_ban_to_South_Korea.pdf. Accessed Nov. 5, 2008.

Stobbs, R. E. and Bruton, M. N. (1991). The fishery of the Comoros, with comments on itspossible impact on coelacanth survival. Environ. Biol. Fishery 32, 341–359.

Stommel, E. W. and Watters, M. R. (2004). Marine neurotoxins: Ingestible toxins. Curr. Treat.Options Neurol. 6, 105–114.

Taiwan Fisheries Agency. (1999–2007). ‘‘1999–2007 Taiwan Fisheries Yearbook.’’ Council ofAgriculture, Taiwan.

Takagi, T., Itabashi, Y., Ota, T., and Hayashi, K. (1976). Gas chromatographic separation ofwax esters based on the degree of unsaturation. Lipids 11, 354–356.

Taylor, S. L., Kabourek, J. L., and Hefle, S. L. (2004). Fish allergy: Fish and products thereof.J. Food Sci. 69, R175–R180.

Ting, J. Y. S. and Brown, A. F. T. (2001). Ciguatera poisoning: A global issue with commonmanagement problems. Eur. J. Emerg. Med. 8, 295–300.

Tserpes, G., Tatamanidis, G., and Peristeraki, P. (2006). Oilfish and shark by-catches of theGreek swordfish fishery in the E. Mediterranean: A preliminary analysis applied to‘‘presence-absence’’ data. Collect. Vol. Sci. Papers ICAAT 59, 987–991.

Ueno, S., Hidaka, T., and Okamoto, T. (1955). The occurrence of highly unsaturated, triethe-noid, and diethenoid alcohols in the liver oil of Laemonema morosum, a preliminary report.J. Jpn. Oil Chemists Soc. 4, 26–27.

Ukishima, Y., Masui, T., Matsubara, S., Goto, R., Okada, S., Tsuji, K., and Kosuge, T. (1987).Wax components of escolar (Lepidocybium flavobrunneum) and its application to base ofmedicine and cosmetics. Yakugaku Zasshi 107, 883–890.

http://www.food.gov.uk/multimedia/pdfs/escolarletter.pdf



http://www.daff.gov.au/aqis/export/fish/fish-notices/2007/FISH0708_Oilfish_escolar_market_ban_to_South_Korea.pdf





Verbeke, W., Vanhonacker, F., Frewer, L. J., Sioen, I., De Henauw, S., and Van Camp, J.(2008). Communicating risks and benefits from fish consumption: Impact on Belgianconsumers’ perception and intention to eat fish. Risk Manage. 28, 951–967.

Waldman, W., Chodorowski, Z., and Kujawska, H. (2006). Acute intoxication with oilfish(Ruvettus pretiosus). Przeglad Lekarski 63, 462–464.

WGNCO (Working Group on Naming of Codfish/Oilfish). (2007). ‘‘Guidelines on Identifi-cation and Labeling of Oilfish/Cod.’’ Hong Kong Centre for Food Safety, Hong KongSAR.

Wild, L. G. and Lehrer, S. B. (2005). Fish and shellfish allergy. Curr. Allergy Asthma Rep. 5,1529–7322.

Wong, C. and Lam, A. (2007). Chain ‘failed Hong Kong citizens’ ParknShop fined over oilfishscandal. South China Morning Post Dec 18, CITYC1.

Yang, W. S., Moon, D. Y., Kim, S. S., and Koh, J. R. (2005). ‘‘Report on the Bycatch from aKorean Observer on the Korean Tuna Longliner in the Indian Ocean in 2004.’’ NationalFisheries Research and Development Institute, Busan.

Yohannes, K., Dalton, C. B., Halliday, L., Unicomb, L. E., and Kirk, M. (2002). An outbreak ofgastrointestinal illness associated with the consumption of escolar fish. Commun. Dis.Intell. 26, 441–445.

CHAPTER 2

Haze in Beverages

Karl J. Siebert

Contents I. The Physics of Haze 54

II. Visual Perception of Haze 57

III. Causes of Hazes in Beverages 58

IV. Diagnosing Haze Problems 59

A. Microscopy 59

B. Chemical analysis 60

C. Enzyme treatment 60

V. Protein–Polyphenol Haze 60

A. Nature of haze-active (HA) protein 60

B. Nature of HA polyphenols 63

C. Nature of protein–polyphenol interaction 65

D. Effects of conditions on particle size and

haze intensity 70

E. Particle size effects on sedimentation

and filtration operations 72

F. The effects of pH and alcohol on haze 72

G. Time course of haze formation 73

H. Beverage differences 74

VI. Analyses Related to Protein–Polyphenol

Haze Formation 75

A. Predictive haze tests 75

B. HA protein 75

C. HA polyphenol 75

VII. Preventing or Delaying Haze Development 76

A. Cold maturation 76

B. Ultrafiltration 77

Advances in Food and Nutrition Research, Volume 57 # 2009 Elsevier Inc.ISSN 1043-4526, DOI: 10.1016/S1043-4526(09)57002-7 All rights reserved.

Food Science & Technology Department, Cornell University, Geneva, New York

53

C. Adsorbents 77

D. Enzymes 80

VIII. Summary 81

References 82

54 Karl J. Siebert

Abstract Beverages such as beer, wine, clear fruit juices, teas, and formulated

products with similar ingredients are generally expected by con-

sumers to be clear (free of turbidity) and to remain so during the

normal shelf life of the product. Hazy products are often regarded

as defective and perhaps even potentially harmful. Since consu-

mers are usually more certain of what they perceive visually than of

what they taste or smell, the development of haze in a clear

product can reduce the likelihood of repeat purchasing of a prod-

uct and can have serious economic consequences to a producer.

Hazes are caused by suspended insoluble particles of colloidal or

larger size that can be perceived visually or by instruments. Hazes in

clear beverages can arise from a number of causes, but are most

often due to protein–polyphenol interaction. The nature of

protein–polyphenol interaction and its effect on haze particles,

analysis of haze constituents, and stabilization of beverages against

haze formation are reviewed.

I. THE PHYSICS OF HAZE

The phenomenon of haze or turbidity in beverages occurs when lightpassing through a sample is deflected or scattered by suspended particu-late matter. An observer perceives the scattered light and, as a result, thesample appears turbid. While particles larger than colloidal size canscatter light, these usually settle out and do not form stable systems.Stable systems must either have particles of a density similar to thesuspending liquid or have particles that are sufficiently small for theambient thermal energy to keep them suspended indefinitely. The latterare called colloids, which by most definitions range between 1 and1000 nm (largest dimension). Ambient temperatures provide energy thatresults in Brownian motion. This produces random collisions of solventmolecules with particles that are sufficient to keep small particles withsomewhat greater density than a solvent in suspension indefinitely.

While light scattering can be observed with a photometer (an instru-ment in which a light source, a sample, and a detector are in a straightline; see Fig. 2.1), light that fails to reach the detector could either havebeen scattered or absorbed. For this reason, scattering is typicallyobserved with an instrument in which the detector is placed at someangle to the incident light beam (see Fig. 2.2). The two most frequently

Lamp Lens Samplecell

Forwardscatter

detector

90� Detector

FIGURE 2.2 The light path in a turbidimeter.

Lamp Monochromatoror filter

Samplecell

Transmittedlight

detector

FIGURE 2.1 The light path in a photometer.

Haze in Beverages 55

used types of static light scattering instruments have detectors at a narrowangle (generally in the 11–25� range) or at 90�. The latter are also callednephelometers. In the case of instruments designed to measure very highturbidities, there may also be a backscatter detector; however, the levels ofturbidity typically found problematic in clear beverages are considerablylower than this level. With most light scattering instruments, the amountof light scattered in colored solutions is underestimated because some ofthe scattered light is absorbed. In order to make color-corrected hazemeasurements or haze-corrected color measurements, an instrumentthat simultaneously measures scattering and transmission is required.A ratio between the two observations is then used to produce the desiredresult.

Turbidimeters can employ white light, light passed through an opticalfilter, or monochromatic (laser) light. Because the relationship betweenthe wavelength of light and the size of the particles affects scattering,instruments that use different light sources (e.g., white light vs. whitelight passed through a green filter) inevitably give different results. Mostphotometric instruments employ filters or monochrometers to select lightof a narrow wavelength range; this certainly impacts the results.

The intensity of scattered light depends on a number of parametersincluding the size, shape, and concentration of the suspended particles,

56 Karl J. Siebert

the angle at which the scattering is observed, the wavelength of light inrelation to the particle diameter, and the refractive indices of the particleand the solvent. The physics of this for spheres much smaller than thewavelength of light was described by Strutt (who later inherited the titleof Lord Rayleigh) in 1871 (Strutt, 1871). Rayleigh later developed the formshown in Eqs. (1) and (2), which express the scattered light intensity atany angle relative to the incident beam, where Iy is the intensity of thescattered light at angle y, r is the particle radius, and l0 is the wavelengthof light used (Thorne and Svendsen, 1962).

Iy ¼ 3

z3ðsin z� z cos zÞ

� �2(1)

z ¼ 4pr

l0 sin

y2

� �(2)

The Mie theory (actually Mie’s solution to Maxwell’s equations forspheres) can be applied to spherical particles that are smaller than, similarin size to, and larger than the wavelength of light used (Mie, 1908). Withparticles much larger than the wavelength, the Mie theory can be simpli-fied to the Fraunhofer theory. The mathematics of scattering is compli-cated for other than spherical shapes, and that is why the assumption thatparticles are spherical is often made.

Many anecdotal accounts state that turbidity measured with narrow-angle scattering is oversensitive to large particles, while that measured at90� is oversensitive to small particles (Siebert, 2008). This can lead to‘‘invisible hazes’’ that are perceptible visually but not with a turbidimeter,and vice versa. According to both the Rayleigh and Mie theories, lightscattering intensity is very strongly influenced by the relationship ofparticle size to the wavelength of light used, with larger particles scatter-ing light much more intensely than small particles at narrow angles(Gales, 2000; Siebert, 2008). With 90� scattering, small particles scattersubstantially more intensely than larger ones. Light scattering results arehighly influenced by the method of instrument calibration (Gales, 2000;Siebert, 2008). This is frequently done using formazin, which is verydifferent in size and shape from yeast (often the largest particles encoun-tered in a sample) or fine-particle haze (the smallest). Commercial cultureyeasts are usually polyploid and larger than diploid strains, with ovalshapes and mean diameters on the order of 10 mm. Fine particle hazes infiltered beverages are typically below 1 mm. Process samples or those thatdevelop haze in package can have intermediate-sized particles. Calibrationof a turbidimeter with particles of the same size as those measured shouldgive correct results at any angle. However, the sizes of particles found inbeverage samples are frequently not known in advance and may be bi- or


even trimodal. For example, both yeast and fine-particle haze are presentin samples of fermented beverages during processing.

For a fixed number of colloidal-size spherical particles, 90� scatteringintensityappears tobeessentiallyproportional to theparticle radius squared(Siebert, 2000); this was attributed to the particle cross-sectional area.

Temperature can affect haze in several ways. Lowering temperaturecan result in reduced solubility of marginally soluble substances and maylead to a higher concentration of particles. This is responsible for thephenomenon known as ‘‘chill haze.’’ Typically, warming a sample willdispel most of the turbidity provoked by chilling. On the other hand,elevated temperatures can speed interactions between substances thatform insoluble particles, leading to more rapid haze development.

II. VISUAL PERCEPTION OF HAZE

As with instrumental turbidity measurements, the conditions underwhich humans observe light scattering also impact results. As expected,the geometry of the viewing system (the angle between the light beam andthe observation) has a large influence (Gales, 2000). Visual observationsare nearly always made with white light, but differences in the lightsource (e.g., incandescent, photoflood, or fluorescent lamps) presumablyhave some effect on the results.

It has been observed that the particle size and concentration as well asthe illumination intensity, solution color, and viewing background influ-ence visual perception of turbidity. Studies were carried out with asensory panel using polymer spheres with a number of diameters in therange 0.15–10.3 mm, each suspended in different colored solutions.Thresholds were determined using the Ascending Method of Limits.When expressed as weight or number concentration, the thresholdsvaried greatly, but when expressed as turbidity measured at 90� theywere quite similar, regardless of the particle size or solution color(Carrasco and Siebert, 1999; Fleet and Siebert, 2005). With bright illumi-nation, thresholds ranged from 0.21 to 2.19 nephelometric turbidity units(NTU). Surprisingly, reducing illumination intensity led to generallylower thresholds (greater sensitivity) up to a point, but further reductionsproduced higher thresholds (Fleet and Siebert, 2005). It appears that lowerillumination results in less reflection from the sample container, making iteasier to perceive the scattered light. Using light-colored rather than blackviewing backgrounds led to much higher thresholds (Fleet and Siebert,2006). This appears to be due to the difficulty in seeing scattered whitelight against a light-colored background.

Suprathreshold particle suspensions were evaluated using MagnitudeEstimation (ME) and Sensory Descriptive Analysis (Carrasco and Siebert,

58 Karl J. Siebert

1999). Equations predicting ME or instrumental turbidity as a function ofsample characteristics were developed. Principal Components Analysiswas applied to the Descriptive Analysis results and this indicated thatpanelists responded to only two fundamental properties, degree of cloud-iness and homogeneity/nonhomogeneity. Only the larger particles stud-ied caused much change on the second axis, leading to the conclusion thatwhen modest numbers of large particles are present, samples take on anonuniform appearance (Carrasco and Siebert, 1999; Siebert, 2008).

III. CAUSES OF HAZES IN BEVERAGES

Hazes in clear beverages (beer, wine, clear fruit juices, tea, etc.) can becaused by a variety of phenomena. Processing problems can lead toparticles from filter media (such as diatomaceous earth) or adsorbents.These are not normal occurrences and can usually be readily discoveredand their cause addressed.

Grape juice andwine can contain tartrateparticles that arise from tartaricacid in grapes forming saltswith various cations. Often, this leads to regularcrystals. Cool storage and additions of seed crystals or salts facilitatessettling out of tartrate precipitates during processing (Jackson, 1994).

Grains typically contain oxalates. As a result, grain-based beverages(such as beer) can develop oxalate crystals (mainly calcium salts). Thestandard practice of adding gypsum (CaSO4) to brewing water (for anumber of reasons) leads to crystallization and precipitation of calciumoxalate during processing (Rehberger and Luther, 1999).

Fragments of plant rawmaterials (e.g., grape skins or fruit pulp) can insome cases pass through a process and enter the final product. If thisoccurs due to a processing problem, it is generally transitory and can beaddressed by refiltering the product.

Microorganisms (yeast or bacteria), which may be either culture organ-isms added intentionally or contaminants, if not removed by filtration orsedimentation can lead to turbidity. The organisms themselves, in suffi-cient numbers, scatter light. The growth of some organisms alters productchemistry and may cause formation of unsightly hazes, ropes, or strings.

In some cases, microbial cell fragments may arise and can be particu-larly problematic. For example, the disc centrifuges often used to removeyeast after brewery fermentations are known to produce shearing forcesthat break off yeast cell wall fragments (Siebert et al., 1987). Agitation ofyeast by other means is also problematic (Lewis and Poerwantaro, 1991;Stoupis et al., 2003). In beer, the resulting particles resist sedimentationand impair filtration.

Sucrose syrups from either beet or cane origin are used in someformulated products or added to coffee beverages at the point of sale


(in the form of flavored sugar syrups). Occasionally, these productsdevelop flocs (large gauzy-appearing structures that float in the product).A number of causes have been associated with these, but most authorshave attributed this to associations between positively charged proteinsand negatively charged polysaccharides that form either under acidicconditions or in products containing ethanol (Clarke et al., 1978; Foonget al., 2002; Morel du Boil, 1997).

Polysaccharide–protein interaction has also been reported in applejuice, between arabinogalactan and protein (Brillouet et al., 1996).

A number of polysaccharides have been associated with beveragehazes or flocs. These include arabinans in red wine (Belleville et al.,1993), starch and mannan in wheat beers (Delvaux et al., 2000), beta-glucans in beer (Jackson and Bamforth, 1983), and retrograded starch inapple juice (Beveridge, 1997).

Polyphenols have been implicated in hazes of many beverages includ-ing white wine (Somers and Ziemelis, 1985), apple juice (Beveridge, 1997;Van Buren and Way, 1978), and beer (Gramshaw, 1969; Steiner andStocker, 1969).

Proteins have been associated with hazes in beer (Asano et al., 1982;de Clerck, 1969), red and white wine (Dizy and Bisson, 1999; Hsu et al.,1989; Pocock and Rankine, 1973; Sitters and Rankine, 1980; Waters et al.,1995), apple juice (Beveridge et al., 1998; Hsu et al., 1989), grape juice (Hsuand Heatherbell, 1987; Hsu et al., 1987), pear juice (Hsu et al., 1990), andkiwifruit juice (Wilson and Burns, 1983).

While most of the previously mentioned causes of haze can createproduct defects, they do not normally occur if a process is carried outproperly. The most frequent cause of haze in clear beverages is protein–polyphenol interaction (Bamforth, 1999; Siebert, 1999). This occurs nor-mally and even when a beverage is properly stabilized, protein–polyphenol haze usually develops eventually. The objective is to delayits onset so that any haze produced is imperceptible until after a product’sintended shelf life.

IV. DIAGNOSING HAZE PROBLEMS

A. Microscopy

Light microscopy can be used to detect particles with regular shapes (e.g.,crystals) and microbes like yeast and bacterial cells (Glenister, 1971).Microscopy can also detect some irregular particles such as diatomaceousearth or adsorbents (Glenister, 1974). It is much less informative withamorphous particle hazes. The use of specific stains can, however, pro-vide useful information. An excellent book by Glenister (unfortunately no

60 Karl J. Siebert

longer in print) describes stains and microscopic techniques that areuseful for characterizing beer hazes of various origins (Glenister, 1975).

A fluorescent tag (fluorescein isothiocyanate) attached to the lectinConcanavalin A is useful in staining yeast cell wall fragments (Siebertet al., 1981). Concanavalin A specifically binds to mannan, which isprominent in yeast cell walls.

B. Chemical analysis

Chemical analysis of haze materials isolated from a beverage must beinterpreted with caution because composition is often not well-related tocause. For example, beer hazes typically contain a high proportion ofcarbohydrate, with a modest amount of protein, and little polyphenol(Belleau and Dadic, 1981; Siebert et al., 1981). In order to prevent or delayhaze formation, however, it is not necessary or helpful to remove carbohy-drate. Reducing the amount of either protein or polyphenol typically hasthat effect. As a result, it appears that the large amount of carbohydratefound in the haze was coagulated with or adhered in some way to theprotein–polyphenol haze backbone.

C. Enzyme treatment

Treatment with specific enzymes has sometimes been used to diagnosehaze problems (Siebert et al., 1981). Conclusions of the effects must betempered by considering that enzyme preparations may have smallamounts of unspecified enzyme contaminants.

V. PROTEIN–POLYPHENOL HAZE

A. Nature of haze-active (HA) protein

Only proteins that contain proline bind polyphenols. Asano et al. (1982)demonstrated that the haze-forming activity of a protein is roughly pro-portional to the mole percentage of proline it contains (see Fig. 2.3). DNAhas codes for exactly 20 amino acids. If each of these were equally presentin a protein, there would be 5 mol% of each one. In fact, most proteinshave much less proline than this. There are a few exceptions. Casein hasabout 8 mol% proline and the grain prolamins (proline-rich, alcohol-soluble proteins) are even higher. Hordein, the barley prolamin, containsabout 20 mol% proline. As a result, it readily forms haze with polyphe-nols and is themain beer haze-active (HA) protein. Hordein contains evenmore glutamine (Q) than proline (P), and often these amino acids areadjacent in the protein (see Fig. 2.4). In fact, the sequence P-Q-Q-P occurs

120

100

80

60

Haz

e fo

rmin

g ca

paci

ty40

20

00 20 40

Mole % proline60 80 100

FIGURE 2.3 The relationship between the proline content of a polypeptide and its

haze-forming activity with catechin based on data from Asano et al. (1982).

Q Q Q P F P Q Q P I P Q Q P Q P Y P Q-Q P Q P Y P Q Q P F P P Q Q P F P Q Q-P V P Q Q P Q P Y P Q Q P F P P Q Q P-F P Q Q P P F W Q Q K P F P Q Q P P F-G L Q Q P I L S Q Q Q P C T P Q Q T P L-P Q -

FIGURE 2.4 Partial amino acid sequence of barley hordein (source of haze-active

protein in beer); P ¼ proline and Q ¼ glutamine.


repeatedly. The adjacent location of proline and glutamine appears toprovide unusually strong polyphenol binding. Some relatively proline-rich proteins (PRPs) have been found in apple juice (5 mol%) (Wu andSiebert, 2002) and grape seeds (9.5 mol% proline) (Wu and Lu, 2004).

Even higher proline contents have been reported in salivary PRPs;these can contain 40–45 mol% proline and also have a substantial amountof glutamine. This protein binds ingested polyphenols, which precipitatesthe PRPs and removes the lubrication these normally provide. The resultis the sensation of astringency (Green, 1993; Haslam and Lilley, 1988).

The number of different amino acids actually found in proteins isgreater than the 20 in the DNA code. This is because some of them areproduced by posttranslational modification. Hydroxyproline, for exam-ple, is not coded in DNA. Proline is inserted in the peptide chain and thehydroxy group is later added to the peptidically liked proline to formhydroxyproline. Although hydroxyproline is found in some proteins thatare known to be haze active, such as gelatin, polyhydroxyproline(the synthetic homopolymer of hydroxyproline) forms no haze with

62 Karl J. Siebert

polyphenols (Siebert et al., 1996c). The polyphenol binding of gelatin (andsimilar proteins) appears to be entirely due to the proline they contain.Free proline does not form haze with polyphenols (Siebert et al., 1996a);it appears that only peptidically linked proline can do that.

Proline differs from all the other coded amino acids in having asecondary amine group that participates in a peptide bond (see Figs. 2.5and 2.6). Because of its ring system, proline is more rigid than most aminoacids and cannot form an alpha-helix structure. As a result, it leads to amore open, less compact protein, which provides better access topolyphenol-binding sites than a more compact protein structure(Hagerman and Butler, 1981). Only one other amino acid (but not one ofthe coded 20), when peptidically linked, has been shown to bind poly-phenols. This is sarcosine (N-methyl glycine), which has a secondaryamine like proline (see Fig. 2.7), but lacks a ring system. Polysarcosine hasbeen shown to bind to polyphenols and produce haze (Hagerman andButler, 1981; Siebert and Lynn, 2008). So it appears that the essential

R

RO

O

NH

NH

HN

FIGURE 2.5 Segment of a peptide composed of alpha-amino acids.

N

N

NO O

FIGURE 2.6 Segment of a peptide composed of proline.

NN

N

O

O

FIGURE 2.7 Segment of a peptide composed of sarcosine.


feature for polyphenol binding is a peptidically linked secondary aminerather than a nitrogen-containing ring system.

B. Nature of HA polyphenols

The polyphenol molecular features that lead to attachment to proteins aregenerally understood. Simple phenols, that is, phenols containing a singlehydroxy group on an aromatic ring, essentially do not bind to proteins(Eastmond and Gardner, 1974). In experiments in which various poly-phenols were combined with bovine serum albumin, the energy releasedupon binding was observed (McManus et al., 1985). The binding energywas weak with m-diphenol, moderate with o-diphenol, and strong withthe vicinal triphenol. So, two or more hydroxy groups on an aromatic ringare required and the binding is stronger when they are adjacent and whenthere are more hydroxy groups.

One aromatic ring with two or more hydroxy groups constitutes onebinding moiety. In order to cross-link two protein molecules, a polyphe-nol needs to have two such binding groups. ‘‘Single-ended’’ polyphenols(with only one binding moiety) can bind to proteins and have been shownto compete with HA polyphenols for binding sites in proteins under someconditions, inhibiting haze formation (Siebert and Lynn, 1998).

The polyphenols in beer, fruit juices, and tea are typically members ofthe flavan-3-ols (see Fig. 2.8) and the proanthocyanidins constructed fromthem.

Each of the flavan-3-ols has two asymmetric centers, at positions2 and 3 (on the C ring). The more naturally prominent members of theflavan-3-ols are (þ)-catechin and (�)-epicatechin (see Fig. 2.9). These twomolecules are remarkably similar, differing only in the orientation of thehydroxy group at position 3; both are expected to have one moderatelystrong binding end (the B ring) and one weak binding end (the A ring).

HO 78

A C

B

OH

OH

65 4

3

OH

2O

OH

FIGURE 2.8 The basic flavan-3-ol structure.

HO

OH

OH

OH

OH

O

HO

OH

OH

OH

OH

O

FIGURE 2.9 The structures of (þ)-catechin (top) and (�)-epicatechin (bottom).

64 Karl J. Siebert

However, one study showed substantial differences between them, with(þ)-catechin producing more haze than (�)-epicatechin when combinedwith polyproline (the synthetic homopolymer of proline) at 25 �C (Siebertand Lynn, 1998). Also prominent in some beverages are gallocatechin andepigallocatechin; in these compounds, an additional hydroxy group islocated on the B-ring vicinal to the two already there (see Fig. 2.10). Thesemolecules are expected to have one strongly binding end (the B ring withthree vicinal hydroxy groups) and one weakly binding end.

Proanthocyanidins are formed from flavan-3-ol ‘‘building blocks’’;although not truly polymers, it is convenient to refer to these as dimers,trimers, etc., indicating the number of flavan-3-ol ‘‘monomers’’ they con-tain. As the size and complexity of proanthocyanidins increase, theirhaze-forming activity increases (Asano et al., 1984; Hagerman andButler, 1981; Mulkay and Jerumanis, 1983) and their solubility and likeli-hood of surviving processing decrease (Asano et al., 1984). As a result,beer and clear fruit juices have significant amounts of monomers anddimers but little of trimers (and higher). The prominent proanthocyanidin

HO

OH

OH

OH

OH

OH

O

HO

HO

HO

HO

HO

HOO

FIGURE 2.10 The structures of gallocatechin (top) and epigallocatechin (bottom).


‘‘dimers’’ found inmost fruit juices and beer are formed of twomonomersconnected from position 8 (on the A ring) of one monomer to position 4(on the C ring) of the other. The prominent dimer in grape juice andwine is procyanidin B1 (catechin–epicatechin; see Fig. 2.11). In applejuice, procyanidin B2 (two epicatechins) predominates. The prominentdimers in beer are procyanidin B3 (two catechins joined together) andprodelphinidin B3 (a catechin and a gallocatechin joined together; seeFig. 2.11). Because of the additional hydroxy group in prodelphinidin B3,this compound is expected to be somewhat more haze active than pro-cyanidin B3, and that has been shown to be the case (Mulkay andJerumanis, 1983).

C. Nature of protein–polyphenol interaction

The basic mechanism is that a polyphenol molecule with at least twobinding sites attaches to two proteins and bridges them together. Addi-tional polyphenol molecules attach this structure to additional proteinmolecules and eventually the complex grows so large that it is no longersoluble. At this point, it becomes a colloidal particle and scatters light. The

66 Karl J. Siebert

particle may continue to grow until it is so large that Brownianmotion canno longer suspend it; at that point it starts to sediment.

Because warming can often disperse protein–polyphenol hazes, it isclear that covalent bonding is not involved in their formation. Asano et al.demonstrated that protein–polyphenol haze formation is inhibited by thenonpolar solvent dioxane and the hydrogen bond acceptor dimethylforma-mide (DMF), but not by a solution of sodium chloride (Asano et al., 1982).

HO

HOOH

OH

OH

OH

OH

OH

OH

OH

O

O

HO

HO

OH

OH

OH

OH

OH

OH

OH

OH

O

O

FIGURE 2.11 Continued

HO

HOOH

OH

OH

OH

OH

OH

OH

OH

O

O

HO O

O

OH

OH

OH

OH

OH

OH

OH

OH

OH

HO

FIGURE 2.11 The structures of the procyanidin ‘‘dimers’’ prominent in (from top to

bottom) grape juice (procyanidin B1), apple juice (procyanidin B2), and beer (procyanidin

B3 and prodelphinidin B3).


They concluded that the interaction involves hydrophobic and hydrogenbonding, but not ionic bonding. Siebert et al. demonstrated that preformedhaze could be dispelled by adding dioxane or DMF, but adding sodiumchloride solution actually increased the haze (Siebert and Lynn, 2008).Increased binding strength with increased ionic strength is a known effect

68 Karl J. Siebert

of hydrophobic bonding (Oh et al., 1980). Hagerman and coworkers carriedout a study in which two quite different types of polyphenols were eachcombined with bovine serum albumin; in one of these cases, the interactionwas attributed to hydrogen bonding (Hagerman et al., 1998). Bianco et al.(1997) used NMR to measure the energy involved in polyphenol interac-tions with caffeine, a surrogate compound for peptidically linked proline.Based on the binding energy observed, the authors suggested that p-bond-ing occurs (in the case of peptidically linked proline, this would be mani-fested as stacking of the flat aromatic ring of the polyphenol with therelatively flat proline ring).

A response surface model of the effects of HA protein concentration(gliadin, the wheat prolamin), HA polyphenol concentration (tannic acid,TA), alcohol, and pH on the amount of haze formed was constructedusing a buffer model system (Siebert et al., 1996a). Figure 2.12 shows theeffects of protein and polyphenol on haze predicted by the model at fixedlevels of pH and alcohol. The model indicates that as protein increases atfixed polyphenol levels, the haze rises to a point and then starts to decline.Similarly, when polyphenol increases at a fixed protein level, the hazeincreases to a maximum and then declines.

A conceptual model that accounted for this behavior was proposed(see Fig. 2.13) (Siebert et al., 1996c). It is assumed that there are a fixednumber of polyphenol-binding sites in an HA protein, presumablyrelated to the proline content, and an HA polyphenol molecule can attachto at the most two proteins (this is likely for steric reasons, even if a

10080

Tannic acid(mg/L)

Gliadin

(mg/L)

Haz

e (N

TU

)

6040

20 100200

300400

500

20

40

60

80

100

120

140160

FIGURE 2.12 Response surface model predictions of the effects of HA protein (gliadin)

and HA polyphenol (TA) on the haze intensity in a model system at fixed levels of pH

and alcohol. Reprinted with permission from Siebert et al. (1996a). Copyright 1996

American Chemical Society.

[polyphenol] = [protein]

[polyphenol] < [protein]

[polyphenol] > [protein]

Polyphenol molecule

Protein molecule with fixednumber of polyphenol binding

sites (i.e., haze-active)

FIGURE 2.13 Concept of influence of protein–polyphenol proportion on particle

size and haze. Reprinted with permission from Siebert et al. (1996c). Copyright 1996

American Chemical Society.


polyphenol has more than two parts of the molecule that can bind toprotein). In some cases, a polyphenol may form an intramolecular bridgebetween two parts of a protein molecule, but this would not lead to haze.When there are similar numbers of polyphenol-binding sites in proteinsand polyphenol-binding ends present in a system, a large network willform, corresponding to large particles and a lot of light scattering. Whenthere is a high proportion of protein to polyphenol, polyphenols will have

70 Karl J. Siebert

no difficulty finding sites in proteins to attach to and can readily join twoprotein molecules together. However, there will be few additionalpolyphenols available to join these protein ‘‘dimers’’ or ‘‘sandwiches’’together. This will produce small particles with relatively little lightscattering. With a polyphenol-rich system, most of the attachment sitesin the proteins will be occupied by one end of an HA polyphenol mole-cule; however, few sites in other proteins will be available for the otherend of the polyphenol to attach to. Small particles will again result, withlittle light scattering.

This conceptual model was later verified with particle size analysis(Siebert and Lynn, 2000) (see Fig. 2.14). When each of the several concen-trations of gliadin was combined with a fixed amount of TA in a modelsystem, the particle sizes changed, and the largest particles were seenwith a gliadin-to-TA concentration ratio (by weight) of 5:1, with smallerparticles at higher and lower ratios. A similar pattern was seen when afixed amount of gliadin was combined with various levels of TA. Onceagain, the largest particles were seen with intermediate ratios. Thechanges were striking in that they were not gradual shifts of a monomo-dal distribution. Rather, particles of one or two discrete sizes werepresent, depending on the protein-to-polyphenol ratio.

D. Effects of conditions on particle size and haze intensity

A more detailed study was carried out with many more levels of proteinand polyphenol than were used to construct the initial response surfacemodel (Siebert and Lynn, 2000). The results (see Fig. 2.15) indicated that

16

14

12

10

8

Num

ber

%

6

4

2

00.1 0.6 3.8

Particle diameter (µm)

23.6 145.6

FIGURE 2.14 Particle sizes measured with 100 (&), 200 (d), 300 (▲), 400 (l), and

500 (▼) mg/L protein (gliadin) concentrations added to 40 mg/L TA. Reprinted with

permission from Siebert and Lynn (2000). Copyright 2000 American Society of Brewing

Chemists.

500Haze (NTU)

400

300

Glia

din

(mg/

L)

200

10040 60 80

5050

Tannic acid (mg/L)120100

100

150

150

5:1

2:1100

10050

140

FIGURE 2.15 Response surface predictions from haze intensity observations made with

30 combinations of gliadin and TA at pH 4.5. Reprinted with permission from Siebert

and Lynn (2000). Copyright 2000 American Society of Brewing Chemists.


the basic concept of the relationship of protein-to-polyphenol ratio to hazeintensity was correct, but that there was fine structure. Ridges of greaterhaze intensity were seen at 2:1 and 5:1 concentration weight ratios ofgliadin to TA; these correspond to TA:gliadin molar ratios of 15:1 and6:1, respectively. These ratios coincided with the larger size particlesobserved with particle size analysis. This indicates remarkably quantizedbehavior. With 30 different protein:polyphenol ratios, particles of only afew sizes were seen and changes were in the proportions of particles ofdifferent size.

72 Karl J. Siebert

E. Particle size effects on sedimentationand filtration operations

The dramatic changes in haze particle size seen with alterations inprotein-to-polyphenol ratio in a model system, would, if this also occursin real beverages, have profound effects on both sedimentation (e.g., coldmaturation in a tank or centrifugation) and filtration operations.

F. The effects of pH and alcohol on haze

The previously described response surface model was used to predict theeffects of ethanol and pH on haze intensity at fixed levels of protein andpolyphenol (Siebert et al., 1996a) (see Fig. 2.16). Changing the ethanolconcentration at the pH of grape juice and wine (near 3) would appearto have little effect on haze intensity. Slightly above pH 4, increasingethanol first decreased haze intensity and then increased it slightly. Thisis within the pH range of apple juice and beer. This effect was attributedto the known suppression of protein–polyphenol interaction caused bynonpolar solvents. Ethanol, which is semipolar, has been shown todecrease protein–polyphenol precipitation (Hagerman et al., 1998).At higher ethanol levels, haze tends to increase. This could be due to thewell-known effect of ethanol in reducing protein solubility by competingfor water.

The effect of pH on haze intensity was striking (Siebert et al., 1996a)(see Fig. 2.16). When the pH rose from near 3 to slightly above 4, the hazeintensity increased by a factor of 7 with the same amounts of protein andpolyphenol. At higher pH, the haze intensity declined. While proteins

Alcohol(%v/v)

FIGURE 2.16 Response surface model predictions of the effects of pH and ethanol on

haze intensity in a model system at fixed levels of protein and polyphenol. Reprinted

with permission from Siebert et al. (1996a). Copyright 1996 American Chemical Society.


often have least solubility near their isoelectric points, the calculated pI ofgliadin is near 8 (Siebert, 2006) and the pKa of polyphenols is also high,typically of the order 9 or 10. The interaction of saliva protein andpolyphenol also demonstrates a sharp maximum slightly above pH 4(Siebert and Chassy, 2003). So presumably the behavior has somethingfundamental to do with the nature of the protein–polyphenol interaction.

G. Time course of haze formation

The time course of protein–polyphenol haze development in many pack-aged clear beverages has a two-phase pattern (see, for example, Fig. 2.17).At first no observable change occurs for some time. After this, hazeformation begins and follows an essentially linear development rate.This phenomenon has been reported in beer (McMurrough et al., 1992)as well as apple juice, grape juice, and cranberry juice cocktail (Siebert,1999, 2006).

While it is possible that soluble protein–polyphenol complexes couldbe formed during the initial stage and that these only grow large enoughto become insoluble particles (which scatter light) during the secondphase, the pattern has typically been attributed to changes in the poly-phenol component that affect the development of the protein–polyphenolhaze (see Fig. 2.18). Various authors have proposed that oxidation orpolymerization of polyphenols enhances their combination with proteinsand thus their haze-forming activity. The evidence here, however, issomewhat contradictory. Oxygen 18 added to beer reportedly ended upin the polyphenol fraction (Owades and Jakovac, 1966); this indicates that

140

120

100

80

Haz

e (N

TU

)

60

40

20

00 20 40 60

Time (days)80 100 120

FIGURE 2.17 The haze development pattern in cranberry juice cocktail stored at 37 �C.Reprinted with permission from Siebert (1999). Copyright 1999 American Chemical

Society.

Simple phenolics Preformed complex phenolics

Complex phenolics (”tannins”)

+ complex nitrogenousmaterial

Tannin-nitrogen complexes

Haze material

+ complex nitrogenous material

Haze material

Combination and�growth� Further �activation�

and �growth�

�Activation�(e.g., by oxidation)

�Active� phenolics

�Phenol-nitrogen complexes”

Polymerization(various mechanisms)

FIGURE 2.18 Possible mechanisms of polyphenol polymerization or activation

leading to haze development based on concepts from Gardner and McGuinness (1977).

74 Karl J. Siebert

polyphenol oxidation is occurring. It has been reported, however, thatdimeric proanthocyanidins depolymerized rather than polymerized inwort and beer (Derdelinckx and Jerumanis, 1987). Radiolabeled epicate-chin did not polymerize in beer to form dimers or trimers; however,labeled dimeric catechin was readily incorporated into beer haze(McGuinness et al., 1975). Some authors who looked for, but failed tofind evidence of polymerization, attributed it instead to ‘‘activation’’ ofsome sort (Gardner and McGuinness, 1977). McMurrough et al. (1992)showed that reducing polyphenol concentration by treatment with poly-vinylpolypyrrolidone (PVPP), a well-known polyphenol adsorbent (seelater), led to a longer time before the start of haze formation and to a lowerhaze development rate once the process began. It is quite clear that somechange in the polyphenols occurs that leads to the increase in the rate ofhaze formation.

H. Beverage differences

In general, beer tends to be rich in HA protein and poor in HA polyphe-nol, while apple juice tends to have the opposite pattern (Siebert et al.,1996a). Grape juice is fairly low in HA protein and variable in HApolyphenol.

White wines were uniformly low in HA protein, while red wines werequite variable (Siebert et al., 1996b).Vitis viniferawhite wines had very lowlevels of HA polyphenols, while Vitis labruscawhite wines had higher andvinifera–labrusca hybrids had intermediate levels (Siebert et al., 1996b). Allred wines had high levels of HA polyphenols, and most had low levels ofHA protein; the two exceptions were both hybrids.


VI. ANALYSES RELATED TO PROTEIN–POLYPHENOLHAZE FORMATION

A. Predictive haze tests

Most producers of clear beverages employ forcing tests in which packagesof the beverage are stored at elevated temperature for some time, oftenwith agitation or periods of chilling (Bamforth, 1999; Berg, 1991). After thestorage period, samples are withdrawn and their hazes are measured,either as is or after chilling. The test conditions are usually designed toproduce results similar to those expected after much longer periods in thetrade. A more rapid predictive test for beer is based on the addition ofalcohol followed by chilling and thenmeasuring haze (Chapon, 1973); thishas generated useful results (McCarthy et al., 2005; Moll et al., 1976).

B. HA protein

A number of approaches have been used to determine the amount of HAprotein in a sample. Themost successful of these is based on adding a fixedamount of TA to a sample (Thompson and Forward, 1969); after incuba-tion, the turbidity is measured and the increase in turbidity observed ispresumed to be proportional to the amount of HA protein in the sample.Thismethod has the advantage that only substances able to formhazewithpolyphenols respond. The saturated ammonium sulfate precipitation limit(SAPL) method has also been widely used, but is far inferior in providinguseful information (Berg et al., 2007; Siebert et al., 2005).

C. HA polyphenol

A variation on the Thompson and Forward method was developed inwhich a HA peptide or peptide-like material (e.g., gelatin, gliadin, polypro-line, or soluble polyvinylpyrrolidone) is added to a sample to induce hazein proportion to the amount of HA polyphenol it contains (Siebert et al.,1996a). This gives little response in beer, which contains very little HApolyphenol, and causes much larger haze increases in fruit juices and wine.

Both of the tests based on provoking haze with a single addition havethe disadvantage that the endogenous amounts of the complementarymaterial influence results. For example, when adding TA to induce hazewith HA protein, the amount of endogenous HA polyphenol will influ-ence results. And similarly, when adding a protein-like material to pro-voke a response with HA polyphenols, the endogenous HA protein willaffect results. The effects are small when measuring high concentrationsof one species in the presence of small amounts of the other. In most beers,for example, which are high in HA protein and low in HA polyphenol, the

76 Karl J. Siebert

results of a single-addition method are fairly accurate for HA protein andinaccurate for HA polyphenol. Titration provides an approach to mea-surement that is less affected by endogenous amounts of HA species.Methods that carry out titration manually or automatically have beenwidely used. The Tannometer instrument developed by Chapon (1993),which carries out automated turbidimetric titrations, has been widelyused. The P-T Standard instrument of Schneider has been applied tobeer (Schneider and Raske, 1997).

VII. PREVENTING OR DELAYING HAZE DEVELOPMENT

It is normal to employ beverage-processing steps that lead to a reductionin the likelihood of haze formation, or at least a delay the onset of hazedevelopment beyond the intended shelf life of a product.

A. Cold maturation

A traditional approach in many cases is prolonged cold storage followedby a sharp filtration, also carried out cold. Cold storage encouragesformation and settling out of insoluble complexes. Reduced temperaturedecreases the solubility of some potential haze material and also reducesthe energy from ambient heat that keeps particles suspended. As a result,some of the haze material is precipitated and left on the floor of thestorage tank or taken out by the filter.

The effect of cold maturation can be enhanced by the use of finingagents. These facilitate the formation of haze and precipitation of sub-stances that, if not removed, could later give rise to haze. A number ofsubstances have been used as fining agents, including the HA proteinsgelatin and isinglass (Harding, 1979; Hickman et al., 2000), HA polyphe-nols such as TA, and some fine particles such as bentonite (Duncan, 1992)and colloidal silica (Hahn and Possmann, 1977) or silica sol (Goertges andHaubrich, 1992). Gelatin is often used to fine fruit juices (Bannach, 1984)and wine (Baldwin, 1992). Also often used for fruit juices and wine arebentonite or silica. Mixtures of two or more fining agents are frequentlyused. Gelatin, isinglass, TA, and colloidal silica are used in beer fining.

Both isinglass and gelatin are derived from collagen proteins. Gelatinis largely from bovine or porcine skins, while isinglass is from the swimbladders of certain tropical fishes. Collagen is rich in both proline andhydroxyproline. Gelatin is generally thought to contain about 12% prolineand a similar amount of hydroxyproline. While hydroxyproline does notparticipate in binding polyphenols (see earlier), it does facilitate a veryopen molecular structure, and this presumably aids access to thepolyphenol-binding sites in the protein.


TA is not a single well-defined compound, but rather a family ofrelated compounds with some common structural features (Haslam,1974). All TAs have some number of gallic acid (3,4,5-trihydroxybenzoicacid) moieties attached to a glucose molecule by ester linkages. Some ofthe galloyl groups can be attached to other galloyl groups (also by esterlinkage) rather than the glucose. The structure of a particular TA (numberand location of galloyl groups) depends on its natural source. Becausegalloyl groups have three vicinal hydroxy groups on an aromatic ring,they bind very strongly to HA proteins. It was shown that the strength ofTA binding to proteins is a function of the number of terminal galloylgroups; that is, those with all three hydroxy groups available (Siebert,1999). Apparently interior galloyls (those with one of the hydroxy groupsoccupied in an ester linkage) are not available to bind to proteins, proba-bly for steric reasons. TA is a very strong HA polyphenol. This hasbenefits for use both as a reagent to measure HA protein and as a finingagent in beverage stabilization.

B. Ultrafiltration

The removal of macromolecules by ultrafiltration has often been used inthe production of clear fruit juices andwine (Girard and Fukumoto, 2000).This treatment removes both proteins and polysaccharides. Ultrafiltrationthrough a 10,000 Da cut-off membrane has been shown to stabilize winesagainst haze formation (Flores, 1990).

Because proteins are involved in beer (Evans and Sheehan, 2002) andchampagne foams (Senee et al., 1999), and these are desirable properties,ultrafiltration is not a suitable treatment for these products. Adsorbentsthat indiscriminately remove protein are unsuitable for the same reason.

C. Adsorbents

Adsorbents that remove proteins or polyphenols are used to treat a numberof beverages to delay the onset of haze formation. Protein adsorbentsinclude bentonite and silica. Bentonite removes protein nonspecifically(see Fig. 2.19) and so is unsuitable for stabilizing beverages where foam isdesirable (beer and champagne). Silica, on the other hand, has remarkablespecificity for HA proteins while virtually sparing foam-active proteins inbeer (Siebert and Lynn, 1997b) (see Fig. 2.20). Silica removes approximately80% of the HA protein from unstabilized beer, while leaving foam-activeprotein nearly untouched at commercial treatment levels.

This was shown to occur because silica binds to the same features inpolypeptides that polyphenols do (peptidically linked proline; Siebertand Lynn, 1997b) (see the concept in Fig. 2.21). In contrast, in unstabilizedapple juice, silica removes only on the order of 20% of the HA protein

160

120

80

Foa

m-a

ctiv

e pr

otei

n (m

g/L)

HA

pro

tein

(N

TU

)

40

00 1 2

Bentonite added (g/L)3 4 5

0

20

40

60

80

100

FIGURE 2.19 The effects on foam active (&) and HA (d) protein of treating

unstabilized beer with bentonite. Reprinted with permission from Siebert and Lynn

(1997b). Copyright 1997 American Society of Brewing Chemists.

160

120

80

Foa

m-a

ctiv

e pr

otei

n (m

g/L)

40

00 1 2 3

Silica gel added (g/L)4 5

100

80

60

40

Haz

e-ac

tive

prot

ein

(NT

U)

20

0

FIGURE 2.20 The effects on foam active (&) and HA (d) protein of treating unstabi-

lized beer with silica hydrogel. Reprinted with permission from Siebert and Lynn (1997b).

Copyright 1997 American Society of Brewing Chemists.

78 Karl J. Siebert

even with very high treatment levels (Siebert and Lynn, 1997a). Thisdifference is accounted for in Fig. 2.22; silica has limited effectiveness inpolyphenol-rich beverages (especially fruit juices), where most of thepolyphenol-binding sites in proteins are occupied by polyphenols, leav-ing few for silica to attach to (Siebert and Lynn, 1997a).

Polyphenol adsorbents are mainly polyamides (Dadic, 1973). At onetime various nylons were used, but PVPP is most frequently used today(McMurrough et al., 1997). The structure of PVPP (see Fig. 2.23) resemblesthat of polyproline (Fig. 2.6); both have five-membered, saturated,nitrogen-containing rings and amide bonds.

As a result, it appears likely that PVPP binds to polyphenols in asimilar manner to that of HA proteins. As with gliadin–TA binding,

Protein molecule with nopolyphenol binding sites(e.g. foam-active protein)

Polyphenol molecule

Protein molecule with fixednumber of polyphenol bindingsites (haze-active)

Silica gel typeadsorbent

FIGURE 2.21 Concept of silica binding to HA protein in beer. Reprinted with

permission from Siebert and Lynn (1997b). Copyright 1997 American Society of Brewing

Chemists.


DMF and dioxane both impede polyphenol binding to PVPP, while NaClenhances it (Siebert and Lynn, 2008). So, both hydrogen and hydrophobicbonding appear to be involved, but not ionic bonding. While PVPP workseven in protein-rich beverages, it is far less effective than in polyphenol-rich beverages. PVPP removed at most half of the HA polyphenol fromunstabilized beer (Siebert and Lynn, 1997b), but at high doses it took out100% of the HA polyphenol from unstabilized apple juice (Siebert andLynn, 1997a). This appears to be because much of the HA polyphenol inprotein-rich beverages is attached to proteins at both ends and inaccessi-ble to PVPP (see Fig. 2.24). In order for PVPP to bind to the polyphenol,the complex with protein must come apart, at least at one end. If PVPPbinds to one end of a polyphenol molecule that is attached to protein atthe other end, then the PVPP treatment could remove some HA proteinand there is some evidence that this occurs (Siebert and Lynn, 1997b). Inhigh-polyphenol, low-protein beverages, the vast majority of the HApolyphenol is readily accessible (see Fig. 2.25).

Polyphenol molecule

Silica gel type adsorbent


sites (i.e. haze-active)

FIGURE 2.22 Concept of silica action in a polyphenol-rich beverage. Reprinted with

permission from Siebert and Lynn (1997a). Copyright 1997 American Chemical Society.

N N O

O

FIGURE 2.23 Structure of a segment of polyvinylpyrrolidone or PVPP.

80 Karl J. Siebert

PVPP is commonly used to remove undesirable brown or pink pig-ments from wine (Jackson, 1994). However, because much of the color ofred wine is due to polyphenolic compounds, treatment with PVPP orother polyamides can diminish the red color and so must be carefullycontrolled. Additions of gelatin or egg white (egg albumin has about3.6 mol% proline) have traditionally been used to more gently removesome polyphenol from red wines to ‘‘soften’’ astringency.

D. Enzymes

At one time, broad spectrum proteolytic enzymes (mainly papain andbromelain) were widely used to delay or minimize haze formation inbeer (de Clerck, 1969). The enzymes cleaved protein chains, that when

Polyphenol molecule

Polyphenol adsorbent



FIGURE 2.24 Concept of PVPP action in beer. Reprinted with permission from Siebert

and Lynn (1997b). Copyright 1997 American Society of Brewing Chemists.


cross-linked by polyphenols, led to smaller and more soluble complexesthat resulted in less haze. These enzymes were inexpensive and effectivein stabilizing beer against haze formation. Unfortunately, the enzymesalso attacked foam proteins, seriously impairing beer foam. This often ledto the use of a foam stabilizer (typically propylene glycol alginate) to atleast partially offset the damage. Most major brewers replaced enzyme(and foam stabilizer) use with adsorbent treatments.

Recently, proteolytic enzymes that cleave peptide bonds only adjacentto proline were introduced (Lopez and Edens, 2005). Since proline isinvolved in the polyphenol-binding sites and there is little proline in thefoam-active proteins, these enzymes are specific for haze proteins and dolittle damage to foam proteins.

VIII. SUMMARY

This review has summarized knowledge of the phenomena of haze devel-opment in clear beverages. The most frequent cause of haze is from theinteraction of PRPs with polyphenols that have at least two binding

Polyphenol molecule

Polyphenol adsorbent



FIGURE 2.25 Concept of PVPP action in a polyphenol-rich beverage. Reprinted with

permission from Siebert and Lynn (1997a). Copyright 1997 American Chemical Society.

82 Karl J. Siebert

locations. Beverages are generally stabilized against haze formation withfining agents or adsorbents that remove one or another of the HA species,or with enzymes that attack the HA proteins. The nature of a beverage(protein-rich or polyphenol-rich) impacts the effectiveness of particularprotein and polyphenol adsorbents.

REFERENCES

Asano, K., Shinagawa, K., and Hashimoto, N. (1982). Characterization of haze-formingproteins of beer and their roles in chill haze formation. J. Am. Soc. Brew. Chem. 40, 147–154.

Asano, K., Ohtsu, K., Shinagawa, K., and Hashimoto, N. (1984). Affinity of proanthocyani-dins and their oxidation products for haze-forming proteins of beer and the formation ofchill haze. Agric. Biol. Chem. (Jpn.) 48, 1139–1146.

Baldwin, G. (1992). Filterability can be improved by fining. Aust. Grapegrower Winemaker 344,21–22.

Bamforth, C. W. (1999). Beer haze. J. Am. Soc. Brew. Chem. 57, 81–90.Bannach, W. (1984). [Food gelatin in the beverage industries - an important help in juice and

wine fining.] Speisegelatine in der Getranke -Industrie- Ein wichtiger Partner bei der Saft-und Weinschonung. Confructa Studien 28, 198–206.

Belleau, G. and Dadic, M. (1981). Beer hazes. II. Further analyses of basic components by highperformance liquid chromatography. J. Am. Soc. Brew. Chem. 39, 142–146.


Belleville, M. P., Williams, P., and Brillouet, J. M. (1993). A linear arabinan from a red wine.Phytochemistry 33, 227–229.

Berg, K. A. (1991). A method for the rapid evaluation of colloidal stability in beer. Tech. Q.

Master Brew. Assoc. Am. 28, 163–167.Berg, K. A., Ding, L. L., and Patterson, R. E. (2007). The dangers of the SASPL test in

chillproofing evaluation. Tech. Q. Master Brew. Assoc. Am. 44, 29–31.Beveridge, T. (1997). Haze and cloud in apple juices. Crit. Rev. Food Sci. Nutr. 37, 75–91.Beveridge, T., Veto, L., and Harrison, J. E. (1998). Formation of chain-like structures in apple

juice haze. Lebensm. Wiss. Technol. 31, 74–77.Bianco, A., Chiacchio, U., Rescifina, A., Romeo, G., and Uccella, N. (1997). Biomimetic

supramolecular biophenol–carbohydrate and biophenol–protein models by NMR experi-ments. J. Agric. Food Chem. 45, 4281–4285.

Brillouet, J. M., Williams, P., Will, F., Mueller, G., and Pellerin, P. (1996). Structural charac-terization of an apple juice arabinogalactan–protein which aggregates following enzymicdearabinosylation. Carbohydr. Polym. 29, 271–275.

Carrasco, A. and Siebert, K. J. (1999). Human visual perception of haze and relationshipswith instrumental measurements of turbidity. Thresholds, magnitude estimation andsensory descriptive analysis of haze in model systems. Food Qual. Pref. 10, 421–436.

Chapon, L. (1973). [Foam and chill haze.]. Bios. 4, 117–125.Chapon, L. (1993). Nephelometry as a method for studying the relations between polyphe-

nols and proteins. J. Inst. Brew. 99, 49–56.Clarke, M. A., Roberts, E. J., Godshall, M. A., and Carpenter, F. G. (1978). Beverage floc and

cane sugar. Int. Sugar J. 80, 197–202.Dadic, M. (1973). Comparative adsorption of beer constituents by Nylon-66 and Polyclar AT.

Tech. Q. Master Brew. Assoc. Am. 10, 60–64.de Clerck, J. (1969). The use of proteolytic enzymes for the stabilization of beer. Tech. Q.

Master Brew. Assoc. Am. 6, 136–140.Delvaux, F., Delvaux, F. R., and Delcour, J. A. (2000). Characterisation of the colloidal haze in

commercial and pilot scale Belgian white beers. J. Inst. Brew. 106, 221–227.Derdelinckx, G. and Jerumanis, J. (1987). Depolymerization of the proanthocyanidins: Iden-

tification of intermediate compounds. In ‘‘Proceedings, European Brewery Convention21st Congress,’’ pp. 577–582. EBC, Madrid.

Dizy, M. and Bisson, L. F. (1999). White wine protein analysis by capillary zone electropho-resis. Am. J. Enol. Vitic. 50, 120–127.

Duncan, B. (1992). Varietal differences in white grape protein: Implications for bentonitefining. Aus. New Zealand Wine Ind. J. 7, 189–193.

Eastmond, R. and Gardner, R. J. (1974). Effect of various polyphenols on the rate of hazeformation in beer. J. Inst. Brew. 80, 192–200.

Evans, D. E. and Sheehan, M. C. (2002). Don’t be fobbed off: The substance of beer foam—A review. J. Am. Soc. Brew. Chem. 60, 47–57.

Fleet, C. F. and Siebert, K. J. (2005). Effect of illumination intensity on visual perception ofturbidity. Food Qual. Pref. 16, 536–544.

Fleet, C. F. and Siebert, K. J. (2006). Effect of viewing background on visual perception ofturbidity. J. Sens. Stud. 21, 34–53.

Flores, J. H., Heatherbell, D. A., and McDaniel, M. R. (1990). Ultrafiltration of wine: effect ofultrafiltration on White Riesling and Gewuerztraminer wine composition and stability.Am. J. Enol. Vitic. 41, 207–214.

Foong, K. K., Amal, R., Doherty, W. O. S., and Edye, L. A. (2002). A review on acid beveragefloc. Zuckerindustrie 127, 382–388.

Gales, P. W. (2000). A comparison of visual turbidity with turbidity measured by commer-cially available instruments. J. Am. Soc. Brew. Chem. 58, 101–107.

84 Karl J. Siebert

Gardner, R. J. andMcGuinness, J. D. (1977). Complex phenols in brewing—A critical survey.Tech. Q. Master Brew. Assoc. Am. 14, 250–261.

Girard, B. and Fukumoto, L. R. (2000). Membrane processing of fruit juices and beverages:A review. Crit. Rev. Biotechnol. 20, 109–175.

Glenister, P. R. (1971). Applications of optical staining in the examination of beer sediment.Proc., Am. Soc. Brew. Chem. 29, 192–199.

Glenister, P. R. (1974). Some useful techniques for the study of beer sediments. II. Character-izing particles of absorbent materials: Particle counting. Proc., Am. Soc. Brew. Chem. 32,11–12.

Glenister, P. R. (1975). ‘‘Beer Deposits. A Laboratory Guide and Pictorial Atlas for the Studyof the Various Particles Found in the Deposits of Beer and Ale’’. Miles Laboratories,Chicago.

Goertges, S. and Haubrich, H. (1992). [Fining agents and their effects on the treatment of fruitjuice and wine.] Schonungsmittel und ihre Effekte bei der Saft-und Weinbehandlung.Flussiges Obst. 59, 462–466.

Gramshaw, J. W. (1969). Beer polyphenols and the chemical basis of haze formation. I.Phenolic constituents of beer and its raw materials. Tech. Q. Master Brew. Assoc. Am. 6,239–251.

Green, B. G. (1993). Oral astringency: A tactile component of flavor.Acta Psychol. (Amsterdam)

84, 119–125.Hagerman, A. E. and Butler, L. G. (1981). The specificity of proanthocyanidin–protein

interactions. J. Biol. Chem. 256, 4494–4497.Hagerman, A. E., Rice, M. E., and Ritchard, N. T. (1998). Mechanisms of protein precipitation

for two tannins, pentagalloyl glucose and epicatechin16 (4 ! 8) catechin (procyanidin).J. Agric. Food Chem. 46, 2590–2595.

Hahn, G. D. and Possmann, P. (1977). Colloidal silicon dioxide as a fining agent for wine.Am.

J. Enol. Vitic. 28, 108–112.Harding, D. (1979). Finings. Chem. Ind. 24, 895–896.Haslam, E. (1974). Polyphenol–protein interaction. Biochem. J. 139, 285–288.Haslam, E. and Lilley, T. H. (1988). Natural astringency in foodstuffs—A molecular inter-

pretation. CRC Crit. Rev. Food Sci. Nutr. 27, 1–40.Hickman, D., Sims, T. J., Miles, C. A., Bailey, A. J., Mari, M. D., and Koopmans, M. (2000).

Isinglass/collagen: Denaturation and functionality. J. Biotechnol. 79, 245–257.Hsu, J. C. and Heatherbell, D. A. (1987). Heat-unstable proteins in wine. I. Characterization

and removal by bentonite fining and heat treatment. Am. J. Enol. Vitic. 38, 11–16.Hsu, J. C., Heatherbell, D. A., Flores, J. H., andWatson, B. T. (1987). Heat-unstable proteins in

grape juice andwine. II. Characterization and removal by ultrafiltration.Am. J. Enol. Vitic.

38, 17–22.Hsu, J. C., Heatherbell, D. A., and Yorgey, B. M. (1989). Effects of fruit storage and processing

on clarity, proteins, and stability of Granny Smith apple juice. J. Food Sci. 54, 660–662.Hsu, J. C., Heatherbell, D. A., and Yorgey, B. M. (1990). Effects of variety, maturity and

processing on pear juice quality and protein stability. J. Food Sci. 55, 1610–1613.Jackson, R. S. (1994). ‘‘Wine Science: Principles and Applications’’. Academic Press, New

York.Jackson, G. and Bamforth, C.W. (1983). Anomalous haze readings due to beta-glucans. J. Inst.

Brew. 89, 155–156.Lewis, M. J. and Poerwantaro, W. M. (1991). Release of haze material from the cell walls of

agitated yeast. J. Am. Soc. Brew. Chem. 49, 43–46.Lopez, M. and Edens, L. (2005). Effective prevention of chill-haze in beer using an acid

proline-specific endoprotease from Aspergillus niger. J. Agric. Food Chem. 53, 7944–7949.McCarthy, S. L., Melm, G. D., and Pringle, A. T. (2005). Comparison of rapid physical

stability tests. J. Am. Soc. Brew. Chem. 63, 69–72.


McGuinness, J. D., Eastmond, R., Laws, D. R. J., and Gardner, R. J. (1975). The use of14C-labelled polyphenols to study haze formation in beer. J. Inst. Brew. 81, 287–292.

McManus, J. P., Davis, K. G., Beart, J. E., Gaffney, S. H., Lilley, T. E., and Haslam, E. (1985).Polyphenol Interactions. Part 1. Introduction; Some observations on the reversible com-plexation of polyphenols with proteins and polysaccharides. J. Chem. Soc., Perkin Trans. II

1429–1438.McMurrough, I., Kelly, R., and Byrne, J. (1992). Effect of the removal of sensitive proteins and

proanthocyanidins on the colloidal stability of lager beer. J. Am. Soc. Brew. Chem. 50,67–76.

McMurrough, I., Madigan, D., and Kelly, R. J. (1997). Evaluation of rapid colloidal stabiliza-tion with polyvinylpolypyrrolidone (PVPP). J. Am. Soc. Brew. Chem. 52, 38–43.

Mie, G. (1908). [Contributions to optics of hazy media, especially colloidal metal solutions.]Beitrage zur Optik truberMedien, speziell kolloidalerMetallosungen.Ann. Phys. (Leipzig)330, 377–445.

Moll, M., That, V., Schmitt, A., and Parisot, M. (1976). Methods for predicting the colloidalstability of beer. J. Am. Soc. Brew. Chem. 34, 187–191.

Morel du Boil, P. G. (1997). Refined sugar and floc formation. A survey of the literature. Int.Sugar J. 99, 310–312, 314.

Mulkay, P. and Jerumanis, J. (1983). [Effects of molecular weight and degree of hydroxyl-ation of proanthocyanidins on the colloidal stability of beer.] Effets du poids moleculaireet du degre d’hydroxylation des proanthocyanidines sur la stabilite colloidale de la biere.Cerevisia 8, 29–35.

Oh, H. I., Hoff, J. E., Armstrong, G. S., and Haff, L. A. (1980). Hydrophobic interaction intannin–protein complexes. J. Agric. Food Chem. 28, 394–398.

Owades, J. L. and Jakovac, J. (1966). Study of beer oxidation with O18. Proc. Annu. Meeting

Am. Soc. Brew. Chem. 180–183.Pocock, K. F. and Rankine, B. C. (1973). Heat test for detecting protein instability in wine.

Aust. Wine Brew. Spirit Rev. 91, 42–43.Rehberger, A. J. and Luther, G. E. (1999). Wort boiling. In ‘‘The Practical Brewer,’’

( J. T. McCabe, ed.), pp. 165–199. Master Brewers Association of the Americas,Wauwatosa.

Schneider, J. and Raske, W. (1997). The protein–polyphenol balance in the brewing process.Measured by nephelometric opto-electronical rapid methods. Brauwelt Int. 15, 228–234.

Senee, J., Robillard, B., and Vignes Adler, M. (1999). Films and foams of Champagne wines.Food Hydrocoll. 13, 15–26.

Siebert, K. J. (1999). Effects of protein–polyphenol interactions on beverage haze, stabiliza-tion, and analysis. J. Agric. Food Chem. 47, 353–362.

Siebert, K. J. (2000). Relationship of particle size to light scattering. J. Am. Soc. Brew. Chem. 58,97–100.

Siebert, K. J. (2006). Haze formation in beverages. LWT Food Sci. Technol. 39, 987–994.Siebert, K. J. (2008). Visual versus instrumental perception of haze – A review. Tech. Q.Master

Brew. Assoc. Am. 45, 90–98.Siebert, K. J. and Chassy, A. W. (2003). An alternate mechanism for the astringent sensation

of acids. Food Qual. Pref. 15, 13–18.Siebert, K. J. and Lynn, P. Y. (1997a). Mechanisms of adsorbent action in beverage stabiliza-

tion. J. Agric. Food Chem. 45, 4275–4280.Siebert, K. J. and Lynn, P. Y. (1997b). Mechanisms of beer colloidal stabilization. J. Am. Soc.

Brew. Chem. 55, 73–78.Siebert, K. J. and Lynn, P. Y. (1998). Comparison of polyphenol interactions with PVPP and

haze-active protein. J. Am. Soc. Brew. Chem. 56, 24–31.Siebert, K. J. and Lynn, P. Y. (2000). Effect of protein/polyphenol ratio on the size of haze

particles. J. Am. Soc. Brew. Chem. 58, 117–123.

86 Karl J. Siebert

Siebert, K. J. and Lynn, P. Y. (2008). On the mechanisms of adsorbent interactions with haze-active protein and polyphenol. J. Am. Soc. Brew. Chem. 66, 46–54.

Siebert, K. J., Stenroos, L. E., and Reid, D. S. (1981). Characterization of amorphous-particlehaze. J. Am. Soc. Brew. Chem. 39, 1–11.

Siebert, K. J., Stenroos, L. E., Reid, D. S., and Grabowski, D. (1987). Filtration difficultiesresulting from damage to yeast during centrifugation. Tech. Q. Master Brew. Assoc. Am. 24,1–8.

Siebert, K. J., Lynn, P. Y., Clark, D. F., and Hatfield, G. R. (2005). Comparison of methods forassessing colloidal stability of beer. Tech. Q. Master Brew. Assoc. Am. 42, 7–12.

Siebert, K. J., Carrasco, A., and Lynn, P. Y. (1996a). Formation of protein–polyphenol haze inbeverages. J. Agric. Food Chem. 44, 1997–2005.

Siebert, K. J., Lynn, P. Y., and Carrasco, A. (1996b). Analysis of haze-active polyphenols andproteins in grape juices and wines. In ‘‘4th International Symposium on Cool ClimateViticulture and Enology,’’ pp. VII-18–VII-21. Rochester, New York.

Siebert, K. J., Troukhanova, N. V., and Lynn, P. Y. (1996c). Nature of polyphenol–proteininteractions. J. Agric. Food Chem. 44, 80–85.

Sitters, J. H. and Rankine, B. C. (1980). Identification of hazes and deposits in wine. Aust.Grapegrower Winemaker 17.

Somers, T. C. and Ziemelis, G. (1985). Flavonol haze in white wines. Vitis 24, 43–50.Steiner, K. and Stocker, H. R. (1969). [Polyphenols and cold stability of beer.] Polyphenole

und Kaltestabilitat des Bieres. Tageszeitung fuer Brauerei 66, 550.Stoupis, T., Stewart, G. G., and Stafford, R. A. (2003). Hydrodynamic shear damage of

brewer’s yeast. J. Am. Soc. Brew. Chem. 61, 219–225.Strutt, J. W. R. (1871). On the scattering of light by small particles. Philos. Mag. 41, 447–454.Thompson, C. C. and Forward, E. (1969). Towards the chemical prediction of shelf life. J. Inst.

Brew. 75, 37–42.Thorne, R. S. W. and Svendsen, K. (1962). Particle size of beer turbidigens and its influence on

nephelometry. J. Inst. Brew. 68, 257–270.Van Buren, J. P. and Way, R. D. (1978). Tannin hazes in deproteinized apple juice. J. Food Sci.

43, 1235–1237.Waters, E. J., Peng, Z., Pocock, K. F., and Williams, P. J. (1995). Proteins in white wine,

I: Procyanidin occurrence in soluble proteins and insoluble protein hazes and its rela-tionship to protein instability. Aust. J. Grape Wine Res. 1, 86–93.

Wilson, E. L. and Burns, D. J. W. (1983). Kiwifruit juice process using heat treatmenttechniques and ultrafiltration. J. Food Sci. 48, 1101–1105.

Wu, L. C. and Lu, Y. W. (2004). Electrophoretic method for the identification of a haze-activeprotein in grape seeds. J. Agric. Food Chem. 52, 3130–3135.

Wu, L. C. and Siebert, K. J. (2002). Characterization of haze-active proteins in apple juice.J. Agric. Food Chem. 50, 3828–3834.

CHAPTER 3

Carnosine and Its Possible Rolesin Nutrition and Health

Alan R. Hipkiss

Contents I. Introduction 89

II. Carnosine Metabolism 91

III. Carnosine and Neurological Activity 91

IV. Carnosine and Other Tissues 92

V. Possible Functions of Carnosine 92

VI. Control of pH 92

VII. Carnosine and Chelation of Zinc

and Copper Ions 93

VIII. Carnosine and Aging 93

IX. Carnosine and the Causes of Aging 94

X. Proteotoxicity and Aging 95

XI. Carnosine, Oxygen Free Radicals,

and Oxidative Stress 96

XII. Carnosine and Nonenzymic Protein

Glycosylation (Glycation) 97

XIII. Carnosine and Proteolysis of Altered Proteins 99

XIV. Carnosine and Gene Expression 100

XV. Carnosine, Anticonvulsants, and Aging 102

XVI. Carnosine and Dietary Restriction-Mediated

Delay of Aging 103

XVII. Carnosine, Regulation of Protein

Synthesis, and Aging 108

XVIII. Carnosine and Corticosteroids 109

XIX. Carnosine and Age-Related Pathology 110

XX. Carnosine, Diabetes, and Secondary

Complications 110


School of Clinical and Experimental Medicine, College of Medical and Dental Sciences, The University ofBirmingham, Edgbaston, Birmingham, United Kingdom

87

XXI. Carnosine and Neurodegeneration 112

XXII. Alzheimer’s Disease 114

XXIII. Parkinson’s Disease 116

XXIV. Carnosine and Ischemia 119

XXV. Carnosine and Osteoporosis 120

XXVI. Carnosine and Cataractogenesis 120

XXVII. Carnosine and Deafness 121

XXVIII. Carnosine and Cancer 121

XXIX. Carnosine and Wound Healing 122

XXX. Carnosine and Immune Function 122

XXXI. Carnosine, Calcium, and Heart Failure 123

XXXII. Carnosine and Autistic Spectrum Disorders 123

XXXIII. Carnosine and Blood Pressure 124

XXXIV. Carnosine and Consumption

of Alcoholic Beverages 124

XXXV. Carnosine and High Fructose Foods

and Drinks 125

XXXVI. Carnosine and Dialysis Fluids 125

XXXVII. Possible Ways to Increase Tissue Carnosine

Levels: Physiological Regulation 126

XXXVIII. Possible Ways to Increase Tissue Carnosine

Levels: Dietary Supplementation 128

XXXIX. Is there any Evidence that Changes in Dietary

Carnosine Have any Effects in Humans? 129

XXXX. Would Vegetarians Benefit from Carnosine

Supplementation? 130

XXXXI. Deleterious Effects of Carnosine 131

XXXXII. Conclusions 132

References 133

88 Alan R. Hipkiss

Abstract The dipeptide carnosine has been observed to exert antiaging

activity at cellular and whole animal levels. This review discusses

the possible mechanisms by which carnosine may exert antiaging

action and considers whether the dipeptide could be beneficial to

humans. Carnosine’s possible biological activities include scavenger

of reactive oxygen species (ROS) and reactive nitrogen species

(RNS), chelator of zinc and copper ions, and antiglycating and

anticross-linking activities. Carnosine’s ability to react with delete-

rious aldehydes such as malondialdehyde, methylglyoxal, hydroxy-

nonenal, and acetaldehyde may also contribute to its protective

functions. Physiologically carnosine may help to suppress some

secondary complications of diabetes, and the deleterious conse-

quences of ischemic–reperfusion injury, most likely due to antiox-

idation and carbonyl-scavenging functions. Other, and much more

speculative, possible functions of carnosine considered include

transglutaminase inhibition, stimulation of proteolysis mediated

via effects on proteasome activity or induction of protease and

Carnosine and Its Possible Roles in Nutrition and Health 89

stress-protein gene expression, upregulation of corticosteroid syn-

thesis, stimulation of protein repair, and effects on ADP-ribose

metabolism associated with sirtuin and poly-ADP-ribose polymer-

ase (PARP) activities. Evidence for carnosine’s possible protective

action against secondary diabetic complications, neurodegenera-

tion, cancer, and other age-related pathologies is briefly discussed.

I. INTRODUCTION

Carnosine (b-alanyl-L-histidine) and related compounds, homocarnosineand anserine, together with N-acetylated forms (see Fig. 3.1 for struc-tures), are common dipeptides found in mammals, birds, and fish (Abe,2000; Bonfanti et al., 1999; de Marchis et al., 2000; Lamas et al., 2007;Tsubone et al., 2007). One remarkable feature of these compounds is thatthey are often found at relatively high concentrations (Table 3.1). Thehighest value reported for terrestrial mammals is that of themiddle glutealmuscle of the thoroughbred racehorse which contained over 100 mmol ofcarnosine per kg dry weight of muscle (Dunnet and Harris, 1997). It hasrecently been reported, however, that the carnosine plus anserine levels inturkey breast muscle can exceed 200 mM (Jones et al., 2007).

It is a valid generalization that there is more carnosine in anaerobic,glycolytic, white muscle than in red, aerobic, muscle (Table 3.2).

NHR

HO

O

O

NH2

HO

O

O N

ANS

R = H CARR = COCH3 N-CAR HCAR

BAL

NH2 NH2

HO

O

O NH

HO

O

O NN

N N

NH

NH NHN

HN

HN

FIGURE 3.1 Structures of carnosine (CAR), N-acetylcarnosine (N-CAR), homocarnosine

(HCAR), anserine (ANS), and balanine (BAL).

TABLE 3.1 Carnosine and anserine concentrations in common animal species (from

Aristoy and Toldra, 2004)

Carnosine

(mg/100 g)

Anserine

(mg/100 g)

Pig Loin 313 14.5

Ham 449 22.9

Neck 186 10.7

Beef Loin 375 59.7

Neck 201 25.4

Lamb Shoulder 39.3 31.5

Neck 94.2 119.5Chicken Pectoral 180 772

Leg 63 233

Turkey Wing 66 775

Salmon 0.53 589

Trout 1.6 344

Sardine 0.1 1.33

TABLE 3.2 Carnosine content varies according to tissue (from Purchas et al., 2004 and

Cornet and Bousset, 1999)

Animal

tissues/muscle

Carnosine

(mg/100 g)

Anserine

(mg/100 g)

Beef Cheek 42.9

Heart 32.6

Liver 77.5

Semitendinosus muscle 452Lamb Longissimus muscle 491

Semitendinosus muscle 356

Triceps brachii muscle 251

Pig Massetuer muscle 38 6

Trapezius muscle 147 6

Longissimus dorsi muscle 268 6

90 Alan R. Hipkiss

Carnosine is also associated with nervous tissues, including the brain,where it is concentrated especially in the olfactory lobe (Bonfanti et al.,1999; de Marchis et al., 2000). However, human cerebral spinal fluidcontains homocarnosine but no carnosine (Huang et al., 2005).


II. CARNOSINE METABOLISM

Carnosine is synthesized from b-alanine and L-histidine by the enzymecarnosine synthase, a reaction which also requires ATP. Studies usingprimary cell culture have indicated that the dipeptide is synthesized bymuscle cells, glial cells and oligodendrocytes (Bauer, 2005). Althoughcarnosine is found enriched in neurons, especially those of the olfactorylobe, it appears that these cells are capable of taking up the dipeptidefollowing its release from glial cells in which it is synthesized. Notunexpectedly, carnosine synthesis is subject to some form of metabolicregulation; synthesis of the dipeptide is reduced when astroglia-richcultures are treated with dibutyryl cyclic AMP and other agents whichactivate cyclic AMP-dependent protein kinases (Schulz et al., 1989). It hasbeen suggested that changes in carnosine synthesis accompany morpho-logical differentiation in muscle and astroglia (Bauer, 2005).

Carnosine can be acetylated at its amino terminus to form N-acetyl-carnosine, although the enzyme responsible has not been characterized.It has also been reported that N-acetyl-carnosine is readily de-acetylatedin the tissues (Barbizhayev, 2008). A phosphorylated form of carnosinehas also been described (Quinn et al., 1992), but again little more is knownabout the enzymes responsible or its function.

III. CARNOSINE AND NEUROLOGICAL ACTIVITY

Animal studies have shown that carnosine can affect neurological func-tion, not surprising given that fact the dipeptide is synthesized by thebrain and that specific transporters for it are present in the choroid plexus(Teuscher et al., 2004), part of the blood–brain barrier. One possible rolefor carnosine within the neuronal system is modulation in glutamatergicsensory neurons (Bonfanti et al., 1999). For a detailed discussion of carno-sine’s function within the mammalian brain, the reader is referred to thefine review by Bonfanti et al. (1999).

The kidney brush border also possesses a carnosine transport systemand there is evidence that kidney also contains an active carnosinase(Sauerhoefer et al., 2005). There is also evidence that carnosine can influ-ence sympathetic nervous activity in kidney (Tanida et al., 2005) as well asbrown (Tanida et al., 2007) and white adipose tissue (Shen et al., 2008).Other studies have shown that carnosine has antidepressant activity inrats (Tomonaga et al., 2008). In chicks, carnosine induces hyperactivity(Tsuneyoshi et al., 2007) whereas its reverse structure (L-histidinyl-b-alanine) has sedative and hypnotic effects (Tsuneyoshi et al., 2008).The mechanisms involved in remain obscure however.

92 Alan R. Hipkiss

IV. CARNOSINE AND OTHER TISSUES

Although carnosine seems to be primarily associated with the brain andinnervated tissues such as muscles (skeletal and heart) at least in the rat(Aldini et al., 2004), carnosine has been reported to be present in the eyelens (Quinn et al., 1992), which suggests that its function might not berestricted nervous tissue.

V. POSSIBLE FUNCTIONS OF CARNOSINE

Although carnosine was discovered over 100 years ago, much remains tobe revealed about its functions; indeed carnosine and homocarnosinehave been described as forgotten and enigmatic dipeptides (Bauer,2005). There are numerous examples of protective actions of carnosineagainst a variety of insults mediated by discrete entities (oxygen freeradicals, reactive nitrogen species, glycating agents, deleterious alde-hydes, toxic metal ions) as well as ameliorating conditions associatedwith aging. Carnosine has been shown to protect various cells againstischemia–reperfusion injury, for example in rat liver (Fouad et al., 2007;Fujii et al., 2003), kidney (Kurata et al., 2006), heart (Alabovsky et al., 1997;Lee et al., 1999; Zaloga and Siddiqui, 2004), and brain (Dobrota et al., 2005).Protective activity exerted by carnosine has also been observed withrespect to diabetes, osteoporosis, neurodegeneration, wound healingand loss of vision, and hearing and immune function.

Possible biochemical functions (Quinn et al., 1992) of carnosine includecontrol of pH, immunostimulant, wound healing agent, antioxidant,metal-ion chelator, carbonyl scavenger, and antiglycator (Table 3.3). Theevidence for some of these proposals is highly varied, however.

VI. CONTROL OF pH

The most convincing proposal is that carnosine plays one or more roles incontrol of intracellular hydrogen ion concentration (Abe, 2000; Vaughan-Jones et al., 2006). Carnosine is an effective physiological buffer; it ispresumed that this property explains its predominant association withwhite, glycolytic, muscles which possess relatively few mitochondria andthereby generate lactic acid. Not only may carnosine, also possible in itsacetylated form, help to directly suppress the rise in hydrogen ion con-centration but its ability to activate the enzyme carbonic anhydrase(Temperini et al., 2005) would increase bicarbonate buffer capacity.These properties may help explain carnosine’s protective action in ischae-mia, a condition associated with severe intracellular acidosis.

TABLE 3.3 Possible Homeostatic properties of carnosine

� Buffer� Hydroxyl radical scavenger� Antioxidant� Chelator of copper and zinc ions� Aldehyde/carbonyl scavenger� Antiglycator� Stimulates nitric oxide synthesis� Stimulates proteolysis� Activates carbonic anhydrase� Upregulates synthesis of oxidized protein hydrolase (OPH)� Suppresses protein cross-linking� Reacts with protein carbonyls� Suppresses AGE reactivity� May participate in protein deglycation� May participate in histone deacetylation� May participate in repair of isoaspartate residues� May stimulate synthesis of stress proteins


VII. CARNOSINE AND CHELATION OF ZINCAND COPPER IONS

Carnosine is an avid chelator of metal ions (Baran, 2000). Complexes withcalcium, copper, and zinc ions have been described (Trombley et al., 2000).It is possible, therefore, that carnosine could exert some sort of control ofcalcium metabolism in muscle tissue (heart or skeletal). It is also likelythat the dipeptide controls the availability of zinc ions in neuronal tissue,especially the olfactory lobe where both carnosine and zinc are enriched(Bakardjiev, 1997; Bonfanti et al., 1999; Sassoe-Pognetto et al., 1993).Zinc–carnosine complexes, called polaprezinc, are also effective in therepair of ulcers and other lesions in the alimentary tract (Matsukura andTanaka, 2000).

VIII. CARNOSINE AND AGING

It has been previously suggested that carnosine might possibly be anantiaging agent (Boldyrev et al., 1999a; Hipkiss, 1998; Hipkiss et al.,2001). This suggestion was based on (i) a report of observations made inAustralia around 1990, but finally published in 1994, that carnosine could

94 Alan R. Hipkiss

not only delay senescence in cultured human fibroblasts but also reversethe senescent phenotype by promoting what appeared to be rejuvenatingeffects (McFarland andHolliday, 1994, 1999), and (ii) a substantial body ofwork from Russian laboratories in the 1980s and 1990s on the dipeptide’santioxidant activity and protective functions towards heart and kidneyischaemia (see Boldyrev, 1993; Boldyrev et al., 1993, 1995, 1997; Quinnet al., 1992; and references therein). Later work showed that growingfibroblasts with carnosine also protected telomeres against shortening(Shao et al., 2004). Antiaging effects of carnosine were subsequentlyobserved in senescence-accelerated mice and fruit flies (Yuneva et al.,1999, 2002). That carnosine seems to be specifically associated with long-lived, postmitotic tissue, such as muscle and nerves, is at least consistentwith the idea that the dipeptide does not compromise cell survival andmay help ensure longevity.

IX. CARNOSINE AND THE CAUSES OF AGING

The explanation of the causes of aging remains somewhat controversial.Most, but not all, biogerontologists reject the idea that aging is a geneti-cally programed process along the lines of growth and development,because the majority of animals in the wild die from predation, starvationor disease, before they age significantly. Hence the selection of genesspecifically programing aging would not be of any evolutionary advan-tage. The consensus of opinion is that aging is the result of a breakdown ofmolecular homeostasis, due to the chronic effects of the forces of instabil-ity (endogenous and exogenous) to which all cells and organisms arecontinuously subjected. In other words, organisms have evolved to sur-vive long enough to reproduce their genes, during which time entropicevents must be either controlled or their effects eliminated. Consequently,the changes which we refer to as aging are thought to have no evolution-ary significance, but result from the eventual failure of longevity geneswhose functions ensure survival for sufficient time for the organism toreproduce successfully. That there is a frequent correlation betweenorganism longevity and the ability to resist certain external stresses,such as heat or irradiation, is consistent with this idea.

Various possible mechanisms have been proposed to explain aging.These include environmental and endogenous factors which affect anorganism’s ability to survive by causing genetic changes (e.g., DNAdamage and telomere shortening), altering gene expression, increasingoxidative stress, compromising energy provision and promoting the


accumulation of altered proteins. There is evidence which suggests thatcarnosine has at least the potential to ameliorate to some degree most ofthese possible causes of aging (Hipkiss, 1998).

X. PROTEOTOXICITY AND AGING

At a biochemical level the most common symptom of aging and itsrelated pathologies is the accumulation of altered or abnormal proteins(Dalle-Donne et al., 2003; Hipkiss, 2006a; Levine, 2002). It should bepointed out that abnormal proteins are normally formed continuously,intracellulary, and extracellularly, and they originate from biosyntheticerrors (gene expression is not 100% perfect) and postsynthetic damagedue to the actions of deleterious endogenous and exogenous agents (e.g.,oxygen and glucose). It is generally thought that the age-related accu-mulation of aberrant polypeptides is a consequence of the decline infunctional activity of the gamut of homeostatic process (e.g., DNArepair, proteolysis, antioxidant enzymes), because the molecules whichcarry out these fun ctions are themselves also subjected to the samerange of insults which they should normally prevent or eliminate.Indeed it is becoming increasingly apparent that molecular overengi-neering of certain of these homeostatic gene products can indeedincrease both stress-resistance and organism longevity. Furthermore,not only does the accumulation of aberrant polypeptides result in loss offunction of the normal gene products, but these altered molecules alsoappear to possess gain of function toxicity, mostly due to their aggregation,oligomerization, and cross-linking potential. Certain aberrant proteins alsoinduce oxygen free radical generation. Among the better understood resul-tant effects of altered protein accumulation are compromised proteolyticactivities, inflammation (or sometimes called ‘‘inflammaging’’), and induc-tion of the stress response. An obvious contentious question is: does agingcauses proteotoxocity, or does proteotoxicity causes aging? Most likely theanswer is ‘‘yes’’ to both alternatives, simply because proper control ofprotein metabolism (synthesis and degradation) is essential for viability.

Table 3.4 lists the possible areas in which carnosine could theoreticallyexert some protective, homeostatic effects which suppress cellular and/ororganism aging, by the dipeptide mostly acting at the postsynthetic levelto suppress formation of altered proteins. However, it is also possiblethat, by scavenging oxidative and glycoxidative agents, carnosine couldinhibit gene modification and thereby prevent synthesis of altered geneproducts and general DNA damage.

TABLE 3.4 Hypothetical mechanisms of carnosine’s antiaging activity

Suppresses altered protein accumulation by:� Scavenging hydroxyl radicals and hypochlorite ions� Scavenging endogenous and exogenous toxic aldehydes andglycating agents

� Reacting with protein carbonyls/AGEs� Inhibiting transglutaminase activity� Slowing protein synthesis

Activates altered protein elimination by:� Stimulating expression of oxidized protein hydrolase� Stimulating NO synthesis which upregulates proteasomeactivity

� Stimulating synthesis of stress/chaperone proteins� Participating in protein deglycation

Participation in SIRT metabolism (gene silencing) by:� Accepting acetyl groups released by SIRT-mediated histonedeacetylation

� Scavenging (acetyl)-ADP-ribose units generated following NADcleavage which accompanies SIRT-mediated histonedeacetylation

Protects telomeres against shortening

Upregulates release of corticosterone

96 Alan R. Hipkiss

XI. CARNOSINE, OXYGEN FREE RADICALS,AND OXIDATIVE STRESS

Over 50 years ago Harman (1956) proposed the so-called ‘‘oxygen freeradical theory of aging.’’ This theory proposed that much age-relateddamage to proteins, lipids, and DNA was caused by incompletelyreduced oxygen atoms, that is, oxygen free radicals. It has often beenassumed that mitochondria are the principle source of these reactiveoxygen species (ROS) because of the organelles’ intimate associationwith oxygen. However, it should be noted that ROS can be producedelsewhere within the cytosol and extracellularly too. ROS are not exclu-sively deleterious as they are also involved in cell signaling, althoughthere are enzymes such as superoxide dismutase, catalase, and variousperoxidases, in the cytosol and mitochondria, which provide defenseagainst excessive ROS generation. There is evidence that carnosine canalso exert antioxidant activity inhibiting oxidation of lipids (Bogardusand Boissonneault, 2000; Decker et al., 2001; Nagasawa et al., 2001) andproteins (Boldyrev et al., 1999a,b; Guiotto et al., 2005a; Hipkiss et al., 1998a;


Kang et al., 2002; Kim and Kang, 2007; Quinn et al., 1992). The dipeptidecan scavenge hydroxyl radicals (Tamba and Torreggiani, 1999), which arethe most damaging ROS. Hydroxyl radicals are generated from hydrogenperoxide in the presence of bivalent metal ions such as copper; hydrogenperoxide is formed by the action of superoxide dismutase on superoxideanions. Carnosine can also scavenge at least two other deleterious ROS,the hypochlorite anion (OCl�) (Hipkiss et al., 1998a; Quinn et al., 1992),which is formed from superoxide and chlorine ions by the action ofmyeloperoxidase, and peroxynitrite (ONOO�) (Fontana et al., 2002)which is formed by the reaction of superoxide with nitric oxide. Conse-quently, it is theoretically possible that carnosine could prevent damagemediated by these ROS in vivo.

XII. CARNOSINE AND NONENZYMIC PROTEINGLYCOSYLATION (GLYCATION)

Other sources of age-related macromolecular damage are metabolic alde-hydes and ketones. And the best investigated example is the chemicalprocess, originally described in cooking, called the Maillard or browningreaction. This process, originally termed nonenzymic protein glycosyla-tion, but now called glycation, involves the reaction of a reducing sugarsuch as glucose with an amino group of a protein, eventually producing ahighly complex brown product, now known as advanced glycation end-products (AGEs) (Suji and Sivakami, 2004). In fact, it turns out that glucoseis the least reactive of all the commonmetabolic sugars due to the fact thatits aldehyde group is 99.99% unavailable for reactivity because of thepredominant cyclic form of the glucose molecule. Other common sugarssuch as galactose and fructose are much more reactive than glucose;indeed diets high in galactose and fructose are frequently employedexperimental tools to induce diabetes-like symptoms in laboratory ani-mals (Wang et al., 2008). It has also been found that certain metabolicintermediates and their by-products, if present in excess, can glycateproteins (Brownlee, 1995; Thornalley, 1999), DNA (Barea and Bonatto,2008), and amino lipids (Lankin et al., 2007) very rapidly indeed to gener-ate products very similar to those found in senescent cells and organisms.

As described above, carnosine was shown in the 1990s to exert antia-ging effects in cultured cells, and the question arose about the mechanism(s) involved. When this work was initiated, carnosine was usuallyregarded as an antioxidant (Kohen et al., 1988) but as other and betterantioxidants did not exert the antiaging/rejuvenating effects on culturedfibroblasts, this suggested that additional activities were necessary toexplain its actions. It was suggested that carnosine’s structure resembledpreferred protein glycation sites and it was demonstrated that the

98 Alan R. Hipkiss

dipeptide did indeed possess antiglycating activity (Hipkiss et al., 1995a).It inhibited protein glycation, subsequent cross-linking (Hipkiss et al.,1998a), and formation of AGEs induced by a variety of reactive aldehydeand carbonyl compounds (glucose, deoxyribose, ribose, fructose, dihy-droxyacetone, malondialdehyde, acetaldehyde, formaldehyde, andmethylglyoxal) (Brownson and Hipkiss, 2000; Hipkiss and Chana, 1998;Hipkiss et al., 1998b, 2002). These observations have been confirmed andgreatly extended by other workers (Burcham and Pyke, 2006; Gugliucciet al., 2002; Seidler, 2000; Seidler et al., 2004; Yan andHarding, 2005; Ukedaet al., 2002; see also references in Hipkiss, 2005), who have detailed thechemistry of the various carnosine–carbonyl adducts generated (Aldiniet al., 2002, 2005; Carini et al., 2003; Liu et al., 2003). A carnosine–carbonyladduct, formed between the dipeptide and hydroxynonenal, has beenisolated from oxidatively stressed biological tissue (Orioli et al., 2005)and, furthermore, detected in rat urine (Orioli et al., 2007). Given thattissue levels of carnosine are generally higher in humans than in rodents(Hipkiss and Brownson, 2000), it is anticipated that similar adducts willbe detected in human tissues.

Having shown that carnosine can suppress the reactivity of lowmolec-ular weight carbonyl compounds by simply reacting with the deleteriousglycating agents, we suggested that carnosine could react with carbonylgroups generated on macromolecules such as aminolipids and proteinsfollowing oxidation or glycation. Using radiolabeled carnosine, weshowed (Brownson and Hipkiss, 2000), at least at the test-tube level,that the dipeptide could indeed react with protein-bound carbonylgroups, and the term ‘‘protein carnosinylation’’ was coined. As yet, how-ever, no evidence for the presence of ‘‘carnosinylated’’ protein has beenobtained from biological tissue. However, protein g-glutamyl-carnosineadducts have been detected (Kuroda and Harada, 2002) in animal muscle.There are a number of differing explanations that could account for theformation of these adducts. The g-glutamyl-carnosine adduct may derivefrom the reaction of carnosine with the transient carbonyl group gener-ated during the spontaneous deamidation of a glutamine residue (Kurodaand Harada, 2002). Another possibility is that adduct formation resultsfrom the action of transglutaminase on a protein glutamine residue andcarnosine producing a g-glutamyl-carnosine residue in the protein;subsequent proteolysis would release the free g-glutamyl-carnosineadduct. A third explanation is the reaction of carnosine with glutamatesemialdehyde, formed following ROS-mediated oxidation of a proteinarginine residue; subsequent hydrolysis again releasing the free adduct.As noted above, although no ‘‘carnosinylated’’ protein has been detected,a ‘‘carnosinylated’’ lipid has been detected in human muscle (Schroderet al., 2004), possibly arising from the reaction of an oxidized (amino)-lipid


with carnosine. Indeed, carnosine has been shown to inhibit MG-inducedglycation of LDL and formation of foam cells in vitro (Rashid et al., 2007).

There is one study which investigated the effects of carnosine onformation of protein carbonyls in cultured cells following exposure toglucose degradation products present in sterilized peritoneal dialysisfluids. It was shown that preexposure of cultured mesothelial cells to20 mM carnosine suppressed ROS generation and formation of proteincarbonyls induced by the glucose degradation products (Alhamdani et al.,2007a,b). The mechanism by which this protection was exerted remainsuncertain, however; possibilities are (i) the extracellular carnosine maysimply react with the deleterious carbonyl compounds extracellularly;(ii) intracellular carnosine reacts with the carbonyl compounds therebypreventing their interaction with intracellular macromolecules; and(iii) carnosine reacts with protein carbonyls forming ‘‘carnosinylated’’proteins. This problem should not be difficult to resolve experimentally.

Deleterious protein cross-linking can also be induced by reactivenitrogen species (RNS) such as peroxynitrite ONOO formed by the reac-tion of superoxide with nitric oxide (NO). The cross-links are formedbetween tyrosine residues following nitration by peroxynitrite (Sitte,2003). Carnosine appears to play roles not only in NO generation butalso in protection against excess NO production by inducible nitric oxidesynthetase (NOS), thereby preventing ONOO-mediated protein modifi-cation (Fontana et al., 2002). Evidence for a carnosine–NO adduct has alsobeen published (Nicoletti et al., 2007).

XIII. CARNOSINE AND PROTEOLYSIS OF ALTERED PROTEINS

As noted above, accumulation of altered protein forms is a commonfeature of aging, which can be explained by either, or both, increasedgeneration of the aberrant polypeptides or a decrease in cellular ability toeliminate them by selective proteolysis (Hipkiss, 2006a). In the past fewyears, much evidence has emerged showing that cell senescence is accom-panied by decreases in either, or both, proteasome- and autophagy-mediated proteolysis (Bergamini et al., 2007; Bulteau et al., 2006; Carrardet al., 2002; Donati, 2006; Martinez-Vicente and Cuevo, 2007; Ngo andDavies, 2007; Vernace et al., 2007a). Indeed, upregulation of either, orboth, proteasome and autophagic activity has been shown to delayonset of the senescent state in cultured cells (Bergamini et al., 2007;Chondrogianni and Gonos, 2007; Donati, 2006; Hansen et al., 2008;Vernace et al., 2007b). Quite why these proteolytic activities decline dur-ing aging is uncertain; possible explanations include inhibition of protea-some activity by cross-linked proteins (Carrard et al., 2002; Ding andKeller, 2001), and accumulation of lipoprotein cross-linked material

100 Alan R. Hipkiss

(lipofuscin) in the autophagosomes (Dahlmann, 2007; Tatsuta andLanger, 2008). Also it should be noted that participation of chaperoneproteins is necessary for the recognition, delivery, and degradation ofaltered proteins (Leidhold and Voos, 2007; Otto et al., 2005; Rakwalskaand Rospert, 2004). This may help explain why increased expression ofcertain chaperone proteins can extend life span in some organisms as wellas protect against heat and other stressing agents (Liao et al., 2008Morrowand Tanguay, 2003; Rattan, 2006).

Although carnosine’s ability to suppress both formation and reactivityof some of the age-associated macromolecular modifications (e.g.,protein–protein cross-links and protein AGEs) could contribute to itsapparent antisenescent effects, these prophylactic actions cannot by them-selves explain the dipeptide’s apparent rejuvenating activity towardscultured human fibroblasts observed by McFarland and Holliday (1994,1999). It is possible that carnosine may stimulate proteolysis. We obtainedpreliminary evidence that protein breakdown is increased in ‘‘old’’ fibro-blasts when cultured with carnosine (Hipkiss et al., 1998b), whileBharadwaj et al. (2002) showed that the dipeptide stimulated proteolysisof HIF-1a protein in cultured cardiomyocytes. Carnosine was also shownto stimulate neutral (nonlysosomal) protease activity in cell-free extractsfrom rat brain (Bonner et al., 1995). Intracellular elimination of aberrantpolypeptides mostly, though not exclusively, involves the proteasomes,and recent evidence suggests that nitric oxide can stimulate proteasomalactivity (Kotamraju et al., 2006; Thomas et al., 2007). It is possible thatcarnosine can upregulate NOS as it has been suggested that carnosineitself is the source of NO rather than arginine (Alaghband-Zadeh et al.,2001). Therefore, in addition to decreasing the formation of glycated andcross-linked protein which can inhibit proteasomal activity, carnosinemay actually stimulate proteasomal function to improve the eliminationaltered protein forms. This proposal should be easy to test using cellculture and in tissues of aging animals fed a carnosine-enriched diet.

XIV. CARNOSINE AND GENE EXPRESSION

Carnosine can affect gene expression. Ikeda et al. (1999) showed thatcarnosine markedly upregulates vimentin synthesis in cultured rat fibro-blasts, while an association between carnosine and vimentin, a cytoskele-tal, intermediate filament protein has been noted in glial cells and neurons(Bonfanti et al., 1999). Interestingly, it has also been shown that theprotease, oxidized protein hydrolase (OPH), is coexpressed with vimen-tin in COS cells (Shimizu et al., 2004). Thus, it is at least possible thatcarnosine could induce synthesis of OPH in the cultured human fibro-blasts and thereby increase the cellular ability to eliminate oxidized


polypeptides. While much more needs to be done to confirm or refutemany of these proposals, they could help to explain carnosine’s rejuve-nating actions of cultured human fibroblasts, particularly as increasedprotein turnover is a well-recognized antiaging strategy (Hipkiss, 2003).However, it should be noted that excessive proteolysis may contribute tothe aging phenotype as in the case of age-related muscle wastage orsarcopenia.

It has recently been reported that vimentin is very readily and specifi-cally glycated in cultured human fibroblasts (Kueper et al., 2007). Thebiological significance of this observation is at present uncertain but thefact that carnosine seems to mimic preferred glycation sites and appar-ently promotes expression of a protein (vimentin) which is itself highlysusceptible to glycation may not be entirely coincidental and should beexplored. Other studies have shown that activated macrophages secretevimentin, an intermediate filament protein, which may play a role inbacterial killing and generation of oxidative metabolites (Mor-Vakninet al., 2003). It is possible that the released vimentin helps to quench anyexcess glycoxidation species that are generated by activated leukocytes.

It has recently been shown that senescent human fibroblasts accumu-late a particular stress protein modified by glycation (Unterluggauer et al.,2008). It was found that heat cognate protein Hsc70 appears to be a targetfor selective glycation in senescent fibroblasts. One conjectures thereforewhether the generation of the highly glycated Hsc70 protein (Hsc70AGE)is part of a triggering or sensing mechanism for the induction of stress-induced senescence, and, furthermore, whether carnosine’s antisenes-cence effects might be related to its ready glycation thus sparing theHsc70 protein from modification.

A form of rejuvenation of the senescent phenotype occurs in phorbolester-treated U937 leukemia cells and that changes in proteolytic activityand vimentin expression are involved (Hass, 2005). It has been known forsome time that the enzyme poly-ADP-ribose-polymerase-1 (PARP-1)plays an important role in suppressing cellular senescence and theresponse to cellular stress. Recent studies have shown that PARP-1 canassociate with and strongly stimulate the 20S proteasomes, an activitywhich is involved in the selective degradation of oxidized proteins (Selleet al., 2007). It appears that PARP-1 catalyzes the synthesis of poly-ADP-ribose (from NADþ units) and becomes attached to various acceptorproteins located in the nucleus. One conjectures whether there is anyparallel between the metabolic changes which accompany the rejuve-nation phenomenon described above, and carnosine’s apparent abilityto control vimentin expression, its proposed effects on proteolytic activity,its potential to react with glycating agents such as ADP-ribose, and itsability to induce cellular rejuvenation in cultured human fibroblastsdescribed by McFarland and Holliday (1994, 1999).

102 Alan R. Hipkiss

XV. CARNOSINE, ANTICONVULSANTS, AND AGING

Another form of protein dysfunction which accompanies aging is thespontaneous deamidation of asparagine residues which can result in thegeneration of isoaspartate residues in proteins. The enzyme protein-isoaspartate-methyltransferase (PIMT) plays an important role in therepair of isoaspartate residues converting them into the normo-form viaformation of a cyclic succinimide intermediate (Zhu et al., 2006). A recentstudy has shown that hydroxylamine can selectively cleave this inter-mediate generating a normal C-terminal fragment plus an N-terminalfragment with either an aspartyl-N-hydroximide or an aspartyl dihydrox-amate residue at its C-terminal end (Zhu and Aswad, 2007). It is interest-ing that hydroxylamines can, like carnosine, delay senescence in culturedhuman fibroblasts (Atamna et al., 2000) and react with carbonyl groups(Hipkiss, 2001). Given its basic nature, one therefore speculates whethercarnosine could similarly cleave peptide bonds at isoaspartate residues,possible even in vivo, and this action may further contribute to thedipeptide’s antiaging activity. Such action could generate carnosineadducted to N- and C-terminal protein fragments which would be easyto detect, should they exist.

Studies in aging models using nematode worms have shown thatanticonvulsants (valproic acid, volpramide, trimethadione, and ethosux-imide) extend life span (Evason et al., 2005; Hughes et al., 2007). Someanticonvulsants also upregulate carnosine levels in mouse brain andhomocarnosine levels in human brain (Petroff et al., 1998, 2006). Bothcarnosine and homocarnosine also have anticonvulsant activity in mice,rats, and humans (Jin et al., 2005; Kozan et al., 2008; Petroff et al., 1998;Wu et al., 2006; Zhu et al., 2007). It is also thought, however, thatcarnosine’s anticonvulsant action is exerted via a carnosine–histidine–histamine pathway (Zhu et al., 2007) activating histaminergic, GABAer-gic, and glutamicergic systems (Kozan et al., 2008).

Whether there is any other connection between anticonvulsant activityand carnosine’s antiaging actions is obviously highly speculative. It maybe relevant to note that epileptic seizures and a shortened life span,together with altered protein accumulation, are consequences of PIMT-deficiency in mice, while treatment with valproic acid, an anticonvulsant,partially suppresses these symptoms including effects on life span(Yamamoto et al., 1998). Conversely, PIMT overexpression can increaselife span of Drosophila (Bennet et al., 2003). Furthermore, the chemistryof some anticonvulsants (ethosuximide) resembles quite closelythe structure of the succinimide intermediate formed during both aspara-gine deamidation and PIMT-mediated repair of isoaspartate residues.One conjectures whether there are any relationships between these


observations particularly because PIMT levels are 50% downregulated,posttranslationally, in epileptic human hippocampus (Lanthier et al.,2002), leading to the accumulation of aberrant tubulin.

Another speculation which can be offered in this context is whethercarnosine stimulates PIMT expression. It is known that small moleculesregulate expression of PIMT mRNA. For example, R-(�)-deprenyl(Huebscher et al., 1999), lithium, and valproic acid (Lamarre andDesrosiers, 2008) can upregulate PIMT, while the cyclic tripeptideargininyl–glycyl–aspartyl inhibits PIMT expression (Lanthier andDesrosiers, 2006). One wonders therefore whether carnosine is an activatorof PIMT expression functioning at either transcriptional or translationallevels; both the isoaspartyl residue in the protein to be repaired and carno-sine contain b-peptide bonds. It may only be coincidental that PIMT expres-sion decreases with tumor malignancy (Lapointe et al., 2005) and carnosinecan inhibit tumor cell growth (Holliday and McFarland, 1996).

The biological role of PIMT involves the selective methylation of iso-aspartate residues followed by a demethylation step to reform the succi-nimide intermediate. The demethylation causes the release of methanolwhich can be converted to formaldehyde and finally to formic acid, asdemonstrated in rat brain preparations. It was found that S-adenosyl-methionine (SAM), the methyl donor, caused formaldehyde levels to risein the rat brain homogenates, thus suggesting that excessive formalde-hyde may be a precipitating factor in Parkinsons’s disease (PD) (Lee et al.,2008). It is possible that carnosine could suppress formaldehyde toxicityby reacting with it to generate a carnosine–formaldehyde adduct. Thisshould be a relatively easy experiment to perform to test this prediction.

XVI. CARNOSINE AND DIETARY RESTRICTION-MEDIATEDDELAY OF AGING

There is much evidence that caloric restriction (CR) can delay aging andonset of much age-related pathology in many species, and increase maxi-mum life span (see Partridge and Brand, 2005; and references citedtherein). Recent observations suggest that fasting periods, rather than adecrease in overall caloric intake per se, may be the cause of these effects(Goodrick et al., 1990; Mager et al., 2006; Masternak et al., 2005; Mattsonand Wan, 2005). The mechanisms involved remain uncertain but they arecurrently thought to involve an interaction between gene expression andcarbohydrate metabolism. Genetic studies have indicated that histone/protein deacetylase (sirtuin) activities have important roles in the CRphenomenon (Guarente, 2000; Westphal et al., 2007). Coupled with the

104 Alan R. Hipkiss

sirtuin-mediated deacetylation reaction is the conversion of NADþ toADP-ribose and nicotinamide (Lin et al., 2000; Medvedik et al., 2007).

There are a number of theoretical locations where carnosine mighthave some influence on sirtuin-mediated protein deacetylation and ADP-ribose metabolism. First, carnosine can behave as an acetyl acceptor byforming N-acetyl-carnosine. Secondly, ADP-ribose, a product of sirtuin-mediated NAD-coupled protein deacetylation, is a potent glycating agent.Hence carnosine could be the ultimate acceptor for ADP-ribose by gen-erating as adduct, ADP-ribosyl-carnosine. A third possibility concerns thepolyADP-ribosylation of certain proteins, as a consequence of oxidativestress, carried out by a PARP, using NADþ as the ADP-ribose source.A number of studies have shown that PARP is involved in aging regula-tion and may protect cells against senescence (Burkle et al., 2005; Hass,2005). Again, given the glycating potential of ADP-ribose and carnosine’santiglycating and antiaging properties, one speculates on whether thedipeptide plays a role in either the formation or, following proteasomalactivity on the poly-ADP-ribosylated proteins (Selle et al., 2007), thesubsequent depolymerizition of the modified protein. However, nocarnosine–ADP-ribose adduct has been reported.

Another possible explanation of the effects of dietary restriction (DR)on the aging process might involve a decrease in glycolysis which inevi-tably accompanies DR-induced fasting periods (Hipkiss, 2006b). Thiscould result in a decrease in the production of the highly deleteriousglycolytic by-product, methylglyoxal (MG), whose reactivity towardsproteins carnosine can inhibit (Brownson and Hipkiss, 2000; Hipkissand Chana, 1998). It is possible that control of sirtuin activity byNADþ/NADH levels can also influence generation of altered proteinsby increasing or decreasing MG generation. NADþ is essential for theconversion of the glycolytic intermediate glyceraldehyde-3-phosphate(G3P) to 1,3-diphosphoglycerate (1,3-DPG) by glyceraldehyde-3-phos-phate dehydrogenase (GAPDH) which also yields NADH. In the adlibitum-fed (AL) condition, it is likely that NADþ levels would be lowand NADH levels high due to continuous glycolysis. This would limitGAPDH activity and promote G3P accumulation, together with its imme-diate precursor dihydroxyacetone phosphate (DHAP). It is important tonote that both G3P and DHAP are very effective glycating agents whichcan readily modify protein amino groups, etc. More importantly, how-ever, both G3P and DHAP can spontaneously decompose into MG, theexceedingly toxic glycating agent that may also be responsible for muchof the protein/lipid glycation observed during hyperglycaemic condi-tions. Indeed there are a number of observations suggesting that MGcan induce ROS production and many of the deleterious physiologicaland biochemical changes characteristic of the aged phenotype(Cantero et al., 2007; Desai and Wu, 2008; Dhar et al., 2008; Han et al., 2007;


Jia andWu, 2007; Schalkwijk et al., 2008; Vander Jagt, 2008; Yamawaki et al.,2008; also see Hipkiss, 2006b, 2008a; and references therein). Carnosine’sability to react with MG (Brownson and Hipkiss, 2000; Hipkiss and Chana,1998) couldconceivably contribute to inhibiting thedeleteriouseffectsof thishighly reactive endogenous glycating agent, especially in tissues whereglycolysis is extensive or persistent.

In DR conditions during the fasting periods, NADH would be meta-bolized by the mitochondria and NADþ regenerated, thus allowing G3Poxidation, preventing DHAP build-up and MG production, and therebydecreasing the incidence of protein and lipid glycoxidation. This model(see Fig. 3.2) would also explain the so-called oxygen paradox whereincreased mitochondrial function (aerobic metabolism) is found to bebeneficial with respect to aging and many related conditions (Hipkiss,2008a).

It is also possible that protein AGEs have a major role in affectingaging in animal models. Cai et al. (2008) have recently demonstrated thatan oral glycotoxin (protein AGE made by treatment of albumin with MG)can substantially abolish many of the beneficial effects that CR exerts onaging, at least in mice. They found that the presence of the AGE (MG-modified protein) in the diet of CR mice promoted an age-related increasein oxidative stress similar to that observed in animals fed ad libitum; thelife span of the AGE-treated mice was also decreased to that observed inthe ad libitum-fed mice. It was concluded that as normal laboratorymouse food contains large amounts of protein AGEs, any reduction offood intake will automatically decrease the AGE load. It therefore follows,somewhat controversially, that the explanation of the beneficial effects ofCR on organism life span may have little to do with decreased calorieintake, but instead reflects the effects of AGEs, exogenous, and endoge-nous, via MG generation. It has been proposed (Hipkiss, 2007b, 2008a,b,c)that dietary restriction, induced by CR or intermittent fasting, willdecrease MG formation and thereby lower endogenous AGE generation,compared to animals fed ad libitum in which MG levels are likely to beraised due to increased frequency of food intake.

Many other studies have suggested that MG is a major source ofmetabolically generated AGEs which in turn can affect organism lifespan, especially as MG can provoke many of the deleterious changesassociated with aging (see Hipkiss, 2008a,b,c; and references therein).For example, mutation in the gene coding for triose phosphate isomerasewhich provokes the accumulation of the MG precursor dihydroxyacetonephosphate, induce a shortened life span in Drosophila (Celotto et al., 2006;Gnerer et al., 2006). Defects in the enzyme responsible for the detoxifica-tion of MG, glyoxalase-1, shortens life span of the Caenorhabditis elegans(Morcos et al., 2008), while overexpression of glyoxalase-1 can extend lifespan in the nematode (Morcos et al., 2008). There is a substantial body of

Glucose

NAD+

NADH

NADH

NAD+

1, 3 DPGATP

PyruvateMitochondrialdysfunction

O2

CO2H2O

ATPCarnosine

Mitochondrion

Nicotinamide

Controls gene expression to maintain juvenile character

A speculation on how carnosine and caloric restriction may modulate ageing

DHAP+

MG

MGG3P

GlycationROS

Acetyl –CoA

Sirtuin activity MG-carnosineadduct

FIGURE 3.2 A speculation on how dietary restriction and carnosine might regulate

methylglyoxal (MG) production and consequent generation of altered proteins which

characterize the aged state. Increased frequency of glycolysis decreases NADþ avail-

ability and increases likelihood of MG generation and protein modification. Caloric

restriction and/or aerobic exercise decreases NADþ consumption and increase NADH

oxidation to facilitate NADþ regeneration to maintain juvenile character via NADþ-dependent sirtuin activity and decreased MG generation.

106 Alan R. Hipkiss

evidence suggesting that increased MG production has a causal role inmuch diabetes-associated pathology (see Ahmed and Thornalley, 2007;and references therein)

The variation in tissue susceptibility to these and other aging-inducedchanges may partly result from differing levels of those molecules(glutathione, polyamines, carnosine, creatine, pyridoxamine, glyoxalases1 and 2) which normally exert protective activity against glycoxidatingagents such as MG. It is also possible that intermittent glycolysis could behormetic by upregulating synthesis of some of these defense molecules(Hipkiss, 2007b). Interestingly, carnosine has also been described as a

Possible protection by carnosineMethylglyoxal (MG) induceddeleterious protein-AGE

Protein-NH2 CH3COCHO(MG)

+ +

+

Carnosine Carnosine-MG adduct(excreted?)

Protein-MG adduct

Protein-CO-AGE(protein carbonyl)

Protein-CO-carnosine adduct(carnosinylated-protein)(inert lipofuscin)(degraded?)

Protein-NH2

Protein-protein–AGE(cross-linked; resists proteolysis; inhibits proteasomes;induces ROS)

Carnosine

FIGURE 3.3 Schematic showing possible sites of intervention by carnosine during

formation of cross-linked methylglyoxal-modified proteins. AGE, advanced glycation

end-product.


glyoxalase mimetic (Battah et al., 2002), in addition to its other protectiveactivities. Nevertheless, carnosine’s ability to protect against MG toxicitymay be important physiologically especially with respect to the changesresponsible for aging and related pathologies (see Fig. 3.3).

Intracellular carnosine concentration may be subject to metabolic reg-ulation. Destruction of the dipeptide by carnosinase is stimulated bycitrate (Vistoli et al., 2006), thus raising the possibility that inhibitorymolecules could be created to prevent destruction of the dipeptide insera. Carnosine’s synthesis by carnosine synthetase is downregulated byraised cAMP levels (Schulz et al., 1989), at least in astrocytes. Thus, highglucose concentrations could lower cAMP levels and hence stimulatecarnosine synthesis.

It is interesting that many of the studies using model organisms tostudy how aging is delayed by genetic, physiological, and dietary meanshave a common feature, that is the upregulation of mitogenesis

108 Alan R. Hipkiss

(Anderson et al., 2008; Bonawitz et al., 2007; Cunningham et al., 2007;Guarente, 2008; Hipkiss, 2008b; Lopez-Lluch et al., 2008; Rodgers et al.,2008; Soh et al., 2007). This effect is also induced by increased aerobicexercise in mammals where many of the symptoms of aging are sup-pressed. One possible explanation is that the stimulation of synthesis ofmitochondria also increases synthesis of the necessary chaperone proteinswhich are required for maintenance of protein quality in the growingorganelle and the cytosol. Hence, as these proteins also participate in theelimination-altered proteins arising from postsynthetic damage, it islikely that the increased ability to recognize and degrade the aberrantproteins due to erroneous protein synthesis will also improve overallcellular stress-resistance and enhance longevity. Carnosine, when com-plexed with zinc ions, has been shown to stimulate expression of certainstress proteins, for example, Hsp72 (Odashima et al., 2002, 2006; Mikamiet al., 2006; Wada et al., 2006). Furthermore, carnosine may also stimulatesynthesis of the stress hormone corticosterone (discussed below) whichmay in turn upregulate expression of a number of stress proteins.Although very hypothetical these suggestions are relatively easy to test.That the stress protein Hsc70 becomes preferentially glycated in senescentcells and that synthesis of vimentin, another readily glycated protein, isinduced by carnosine is perhaps indicative of a relationship betweenthese factors; for example, could upregulation of vimentin synthesiscompete with Hsc70 glycation and help delay onset of stress-inducedpremature senescence?

XVII. CARNOSINE, REGULATION OF PROTEINSYNTHESIS, AND AGING

It has recently been shown that carnosine can also exert suppressiveeffects on mRNA translation initiation (Son et al., 2008); the dipeptideinhibited interleukin-8 mRNA translation by suppressing phosphoryla-tion of initiation factor eIF4E in peroxide-activated intestinal epithelialcells and Caco-2 cells. eIF4E Phosphorylation is required for effectivemRNA translation, which explains the observed carnosine-mediateddecreased synthesis of the proinflammatory cytokine. Carnosine alsoinhibited phosphorylation of other regulatory proteins ERK1/2 and p38MAP kinase (Son et al., 2008). This may be important as Davis et al. (2005)have shown that accelerated aging can be suppressed by inhibiting p38MAP kinase phosporylation. The mechanism by which carnosine sup-presses kinase phosphorylation is unknown, but it could be a conse-quence of decreased glycoxidative damage within the cell, due to thepresence of carnosine, rather than direct participation of the dipeptidein the signaling pathway.


Defective eIF4E limits mRNA translation initiation and results in lifespan extension in C. elegans since studies show that aging can be delayedby partial inhibition of protein synthesis. Studies in C. elegans revealedthat senescence was delayed, stress resistance enhanced, and life spanextended in eIF4E mutants (Pan et al., 2007; Syntichaki et al., 2007). Simi-larly, life span of C. elegans was extended and stress resistance enhancedwhen translation was inhibited where synthesis of eleven ribosomalproteins was suppressed using inhibiting-RNAs (RNAi) (Hansen et al.,2007). Certain ribosomal protein defects also have beneficial effects onyeast longevity (Chiocchetti et al., 2007). Explanation of these effects isuncertain (Kaeberlein and Kennedy, 2007), but it is possible that a lowerrate of bulk protein synthesis, resulting from decreased translation initia-tion frequency, also lowers synthesis of error-proteins (Hipkiss, 2007c).This lowered production of biosynthetic aberrant proteins could directlylower the load that the chaperone and proteolytic apparatus must dealwith: the chaperone/proteolytic apparatus is responsible for the elimina-tion of altered protein generated postsynthetically as well as those formedby biosynthetic errors. Compared to normal gene products, error-proteinsare more readily glycated and oxidatively damaged by ROS (Dukan et al.,2000; Fredriksson et al., 2006) than the normal gene products, thus themutant organisms would generate fewer protein carbonyls, that normallycharacterize the senescent state (Stadtman, 1992) than the wild type.Consequently, the decreased level of biosynthetic error-proteins wouldnot only decrease formation of protein carbonyls but also increase therelative availability of chaperone and proteolytic activities for the recog-nition and elimination of altered proteins arising from deleteriouspostsynthetic modification (Hipkiss, 2007a).

Interestingly, methionine restriction (40% and 80%) also delays agingin rodents (Miller et al., 2005; Naudi et al., 2007). Methionine is the initiat-ing amino acid in protein biosynthesis; therefore, this could again indicatethat decreased translation initiation is an effective antiaging strategy bydecreasing biosynthetic formation of error-proteins, similar to the effectsof the defective eIF4E initiation factor in nematodes outlined above(Hipkiss, 2008c). It is possible that because carnosine’s also has inhibitoryeffects on eIF4E activity and slows protein synthesis, the beneficial effectson fibroblast senescence and life span could be mediated via a similarmechanism in human cells.

XVIII. CARNOSINE AND CORTICOSTEROIDS

A recent study has shown that intracerebroventricular carnosine admin-istration stimulates corticosterone release in chick brain (Tsuneyoshiet al., 2007). Studies performed some 30–40 years ago showed that

110 Alan R. Hipkiss

hydrocortisone or cortisone (Cristofalo and Kabakjian, 1975; Macieira-Coelho, 1966) have positive effects on the growth and life span of culturedhuman fibroblasts. These findings have recently been reactivated wherethe beneficial effects of glucocorticoids towards cultured human fibro-blasts have again been demonstrated (Kletsas et al., 2007). Given thatcarnosine can also affect fibroblasts life span in a positive manner(McFarland and Holliday, 1994), it is at least conceivable that carnosine’saction is mediated via glucocorticoid upregulation. A recent study at thewhole animal level has revealed mixed results however, Caro et al. (2007)showed that 4 weeks chronic treatment with corticosterone decreasedmarkers of lipid peroxidation but protein glycoxidation and oxidativedamage to mitochondrial DNA were both increased in rat liver. Clearly,this is another research area which should be explored.

XIX. CARNOSINE AND AGE-RELATED PATHOLOGY

Accumulation of altered protein forms, particularly protein carbonylgroups, is not only the most common biochemical signature of aging(Levine, 2002) but such aberrant polypeptides are associated with manyage-related diseases (Dalle-Donne et al., 2003) as well. As carnosine hasthe potential to intervene in a number of processes that possibly contrib-ute to the phenomenon we call aging, particularly where generation ofaltered proteins is involved, it follows that the dipeptide may have somebeneficial effects with respect to either the causation or progression ofthose age-related conditions which also involve accumulation of aberrantprotein forms. Table 3.5 lists the possible conditions against which carno-sine might exert some therapeutic effects. It should be emphasized thatthis list is, for the most part, purely speculative and considerable amountsof work needs to carried out to verify or eliminate these suggestions.

XX. CARNOSINE, DIABETES, AND SECONDARYCOMPLICATIONS

The secondary complications of diabetes include cardiac and circulatorydisorders, peripheral neuropathy, cataractogenesis, and stroke. Over thepast decade it has become increasingly evident that much diabetes-associated pathology derives from hyperglycaemia where glucose, ormore likely its metabolites and by-products, chemically modify intracel-lular and extracellular proteins and aminolipids via the process calledglycation (Ahmed and Thornalley, 2007; Goh and Cooper, 2008;Magalhaes et al., 2008; Singh et al., 2001; Vlassara and Palace, 2002). Theprocess is termed nonenzymic glycosylation or glycation to distinguish it

TABLE 3.5 Potential age-related conditions against which carnosine could be explored

therapeutically

� Diabetes and diabetic complications� Ischemia� Neurodegeneration� Osteoporosis� Deafness� Slow wound healing� High blood pressure� Heart disease� Cataractogenesis


from the enzyme-mediated attachment of sugars to proteins or lipidsrequired for proper cell function/distribution. It should be noted thatglycation of proteins mediated by glucose is relatively slow, but othercommon sugars such as galactose or fructose are much more rapid.Furthermore, there is much evidence that certain metabolic intermediatesof glucose catabolism (via glycolytic pathway) can almost immediatelyglycate intracellular and extracellular proteins. As noted above, MG isparticularly damaging (Cantero et al., 2007; Desai and Wu, 2008; Dharet al., 2008; Gomes et al., 2008; Kalapos, 1999; Mirza et al., 2007; Nakayamaet al., 2008; Wang et al., 2007; Yander, 2008; Yao et al., 2007) and manystudies have suggested that MG is the primary source of much of thedeleterious protein glycation which is responsible for diabetic complica-tions (see Rabbani and Thornalley, 2008; Wang et al., 2008 for recentreviews). Hence there has been an extensive search for agents whichpossess antiglycating activity which may be employed to suppress AGEformation and attendant diabetic complications. It was suggested sometime ago that carnosine might be a candidate antiglycating agent for thecontrol of secondary diabetic complications (Hipkiss, 1998; Hipkiss et al.,1995a,b).

Carnosine may help suppress some features of aging at cellular andwhole organism levels, possibly by inhibiting the reactivity of ROS anddeleterious aldehydes including formation of AGEs (Hipkiss et al., 2002).It is theoretically possible that the dipeptide is beneficial towards thoseconditions where formation of protein AGEs plays important and mostlikely causal roles. Even in complication-free diabetics, the levels AGEprecursors such as MG and glyoxal are elevated in their sera (Han et al.,2007). So it is likely that carnosine and other carbonyl scavengers mightexert beneficial effects towards diabetes and its secondary complications(Hipkiss, 2005). Lee et al. (2005) have indeed demonstrated that die-tary carnosine suppresses a number of diabetic complications in mice.

112 Alan R. Hipkiss

There is also evidence indicating that low cellular carnosine levels areassociated with diabetes (Gayova et al., 1999; Nagai et al., 2003). A recentstudy has shown that carnosine can inhibit formation of glycated low-density lipoprotein which normally provokes formation of foam cellsassociated with circulatory disorders which characterize diabetic compli-cations (Rashid et al., 2007). Furthermore, carnosine can decrease bloodpressure in rats (Nagai et al., 2003; Tanida et al., 2005) and possessesvasodilatory activity (Ririe et al., 2000); hypertension is another conse-quence of diabetes. It has been shown that the toxic effects of fructose andhigh glucose levels on blood pressure might be mediated via the genera-tion of MG (Wang et al., 2006). MG not only induces peroxynitrite pro-duction in vascular smooth muscle cells (Chang et al., 2005) but also playsa causative role hypertension (Wu, 2006). Clearly, if carnosine doesindeed scavenge MG in vivo, then it could exert protective effects towardshyperglycemia-induced hypertension.

Another complication of diabetes is the loss of peripheral neuronalfunction. It has recently been shown that carnosine and its zinc complexcan ameliorated progressive diabetic neuropathy in mice (Kamel et al.,2008), although the zinc–carnosine complex was more effective than theuncomplexed dipeptide.

An interesting observation is that serum AGE levels are higher indiabetic vegetarians compared to diabetic omnivors (Krajcovicova-Kudlackova et al., 2002). This may be because of the absence of carnosinein vegetarian diets, although a raised intake of fructose by vegetarians isan alternative explanation.

More evidence that carnosine may possess therapeutic potentialcomes from studies on carnosinase in mice and humans. These studieshave shown that higher levels of this enzyme are associated with diabeticend-stage kidney disease in humans, whereas a lower activity seemed tobe protective (Freedman et al., 2007; Janssen et al., 2005). A further studyusing diabetes-prone mice, in which serum carnosine or carnosinaselevels were manipulated, showed that the onset of diabetic complicationswas enhanced when carnosinase activity was increased by transgenicmodification with the human carnosinase gene CN1; diabetes wasmilder and delayed in animals supplemented with carnosine(Sauerhofer et al., 2007).

XXI. CARNOSINE AND NEURODEGENERATION

There is evidence from animal studies that carnosine can affect brainfunction/activity (Tanida et al., 2007; Thio and Zhang, 2006; Tomonagaet al., 2004, 2005, 2008; Tsuneyoshi et al., 2007, 2008); furthermore, thedipeptide is protective against a number of neurotoxic agents, for


example, N-methyl-D-aspartate (NMDA) (Shen et al., 2007a,b), copper(Hornung et al., 2000), zinc (Kawahara et al., 2007), ROS (Kim and Kang,2007; Kim et al., 2004), and RNS (Calabrese et al., 2005). There isalso substantial evidence suggesting that carnosine is protective againstischemia in the brain and induced seizures; both of these are discussedseparately below.

Many neurodegenerative diseases are age-related, consequently it ispossible that factors which delay general aging could also delay onsetof neurodegeneration. Furthermore, many neurodegenerative diseaseshave at least one feature in common, that is the accumulation of alteredproteins, a feature characteristic of aged cells generally. These aberrantprotein forms are found as tangles and amyloid plaque in Alzheimer’sdisease (AD), as Lewy bodies in PD and inclusion bodies in Huntington’sdisease (HD) (Bossy-Wetzel et al., 2004). From first principles, generalexplanations for their occurrence are either increased production ofaltered proteins or their decreased clearance from the tissue. The likeli-hood that neurodegeneration is accompanied by dysfunction of either theubiquitin/proteasome system (Paul, 2008) and/or the autophagic appa-ratus (Bandhyopadhyay and Cuervo, 2007) has been recognized for sometime. The possibility that carnosine might stimulate proteolytic activityby, for example, activating proteasome function by increasing nitricoxide synthesis (Thomas et al., 2007), or whether it affects autophagy orchaperone formation/activity, should be explored.

It has long been suggested that ROS may play causal roles in thesediseases and in the production of the aberrant protein molecules(de Arriba et al., 2006). In addition, protein damage inflicted by RNS hasalso been suggested. As discussed above, carnosine has been shown topossess antioxidant activity and also to react with RNS as well, hence themolecule has the potential to be considered as a theraupeutic agent(Calabrese et al., 2008).

There are a number of findings suggesting that agents that facilitateelimination of protein carbonyls (by either proteolytic elimination or byenzymically mediated chemical reduction) may suppress neurodegenera-tive conditions in model systems (Botella et al., 2004). Consequently, ascarnosine may also react with protein carbonyls, it is theoretically possiblethat it could suppress formation and/the reactivity of protein carbonyls inthe brain. Whether carnosine participates in carbonyl reductase activityhas not been investigated but it is also a reasonable speculation.

There are numerous examples where carnosine has been demon-strated to be protective activity against neurotoxic agents. For example,the dipeptide was shown to protect the mitochondria of cultured astro-glial cells against nitric oxide-induced damage (Calabrese et al., 2005).Carnosine was also shown to protect neurofilament-L against oxidativedamage, aggregation, and formation of dityrosine induced by hydrogen

114 Alan R. Hipkiss

peroxide and cytochrome c (Kim and Kang, 2007; Kim et al., 2004). Car-nosine was shown to modulate the neurotoxic effects of copper and zinc(Hornung et al., 2000). In the following sections, the possible roles forcarnosine as protective agents in specific neurodegenerative conditionsare discussed.

XXII. ALZHEIMER’S DISEASE

The causal events in AD are much discussed but increased oxidative/glycoxidative damage is acknowledged to play a role. One commonaltered protein form which accompanies AD is a small peptide fragmentcalled amyloid-b-peptide (Ab-peptide). Ab-peptide is generated from alarger protein called amyloid precursor protein (APP) via the action oftwo proteases. It appears that an enzyme (insulin degrading enzyme orIDE) which normally cleaves the Ab-peptide in the middle declines withage (Caccamo et al., 2005; Farris et al., 2005; Qiu and Folstein, 2006). Thereason for IDE’s decline in activity is unknown; possibilities includeinactivation by ROS, RNS, or glycating agents, decreased gene expressionand preferential usage in insulin metabolism.

Glycoxidation events have a role in neurodegenerative disorders, andsome recent papers have proposed that MGmay be directly involved (Baret al., 2002; Luth et al., 2005; Munch et al., 1997, 2003; Pamplona et al., 2005,2008; Reddy et al., 2004; Yan et al., 1994). Other reactive aldehydes such aslipid oxidation products, hydroxynonenal, malondialdehyde, and acro-lein are additional sources of protein damage. It is theoretically possiblethat carnosine or related structures could react with these deleteriousaldehydes and thereby suppress their damaging effects towards proteins(Hipkiss, 2007a). It may be significant that a low serum carnosine level hasbeen reported to be associated with AD (Fonteh et al., 2007). Furthermore,a raised level of protein AGEs in cerebral spinal fluid (CSF) (Ahmed et al.,2005; Luth et al., 2005; Shuvaev et al., 2001; Yamagishi et al., 2005) isassociated with AD, while homocarnosine levels in CSF generally declinemarkedly with age (Huang et al., 2005; Janssen et al., 2005). It may also besignificant that carnosine is enriched in the olfactory lobe (Barkardjiev,1997; Bonfanti et al., 1999; Sassoe-Pognetto et al., 1993) and a loss of a senseof smell may be an early symptom of neurodegeneration (Ghanbari et al.,2004; Kovaks, 2004). Given carnosine’s homeostatic properties outlinedabove, it is at leastworth consideringwhether carnosine or homocarnosinepossess therapeutic potential towardAD (Hipkiss, 2007a), especially as thechoroid plexus possesses a carnosine (homocarnosine) transport proteinwhich may control CSF homocarnosine levels (Teuscher et al., 2004).While carnosine is absent from human CSF, one speculates that homo-carnosine might act as an antiglycating agent in CSF (Hipkiss, 2007a). The


age-related decline in CSF homocarnosine levels (Jansen et al., 2006) couldat least partially explain the observed increase in glycated proteins CSF ofAD patients, as well as the strong relationship between aging and AD.

Many of the proteins which accumulate during neurodegenerativeconditions may also become cross-linked by transglutaminase(Andringa et al., 2004; Junn et al., 2003; Karpuj and Steinman, 2004;Selkoe et al., 1982), an enzyme which cross-links the glutamine sidechain to a lysine e-amino group. Transglutaminase protein is also asso-ciated with many of the inclusion bodies characteristics of AD, PD,etc. (Junn et al., 2003). It is theoretically possible that carnosine couldsubstitute for the lysine residue e-amino group in the transglutaminasereaction (Hipkiss, 2007a) to generate g-glutamyl-carnosine as a hypothet-ical reaction product. While no such linkage has been reported (or sought)in neuronal tissue from neurodegenerative brain, the predictedg-glutamyl-b-alanyl-histidine products have been isolated from animalmuscle tissues (Kuroda et al., 2000). However, as discussed above, it ispossible that such structures might be derived from the spontaneousdeamidation of glutamine residues in close proximity to carnosine orfrom the reaction of the dipeptide with oxidatively induced glutamicsemialdehyde.

Carnosine’s copper and zinc ion-chelating activity may also contributeto suppression of neurodegenerative conditions (Hipkiss, 2005). Zinc hasbeen reported to be associated with the amyloid which accumulates inAD brain (Bush and Tanzi, 2002; Danscher et al., 1997; Religa et al., 2006),while copper ion-mediated oxidation of neuronal proteins may accom-pany both AD and PD (Smith et al., 2006). Carnosine has been found toprotect cultured neurons against zinc-induced death (Kawahara et al.,2007).

Cell culture studies have shown that carnosine is protective against thetoxicity of the Ab-peptide which accumulates in the AD brain. The dipep-tide prevented Ab-peptide (1–42)-induced glutamate release, butincreased expression of the NMDA receptor. It was proposed that carno-sine’s protective activity was exerted via regulation of glutamate releaseand independent of the carnosine–histidine–histamine axis (Fu et al.,2008). Similarly, Boldyrev et al. (2004c) found that carnosine suppressedthe cytotoxicity of Ab42 in cerebellar granule cells independently of thedipeptides’ effects on calcium metabolism and ROS generation. Prestonet al. (1998) also showed that the toxicity of the Ab42-related peptidefragment (25–35) towards rat brain endothelial cells was inhibited bycarnosine and related structures, although the mechanism involved wasnot investigated.

There is one explorative study investigating whether there is anycorrelation between serum carnosinase levels and dementia. The findings,using a small sample size, indicate that while there was no significant

116 Alan R. Hipkiss

difference between control patients and those suffering from mixeddementia or AD, carnosinase activity was higher in patients who regu-larly exercised (Balion et al., 2007).

XXIII. PARKINSON’S DISEASE

Parkinson’s disease is a neurodegenerative condition associated with theloss of dopaminergic neurons in a region of the brain called the substantianigra pars compacta. The cause of PD is unknown but it seems that thesubstantia nigra is particularly susceptible to oxidative damage which inturn induces mitochondrial dysfunction and increased production ofROS, accompanied by the accumulation altered protein species whichform aggregates called Lewy bodies. A major component of Lewy bodiesis a protein called a-synuclein, an abundant presynaptic protein, but otherproteins are present in these structures including ubiquitin, transgluta-minase, and a number of heat-shock proteins.

It appears likely that oxidative events including mitochondrial dys-function play a major role in PD. Among the deleterious agents thought tobe involved are peroxynitrite and hydroxyl radicals (Yokoyama et al.,2008). As noted above, carnosine has been shown to inhibit proteindamage mediated by peroxynitrite and hydroxyl radicals in astroglialcells (Nicoletti et al., 2007). There is some evidence that carnosine cansuppress some of the oxidative damage associated with PD using a modelsystem, and possibly inhibit fibrillization of a-synuclein (Herrera et al.,2008).

In order to investigate PD in animal models, one approach is to use achemical called 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP), aknown neurotoxin which induces symptoms similar to PD in animals andhumans. When this compound was injected into senescence-acceleratedmice (SAMP1), the animals demonstrated short-term tremor, weight lossand pronounced rigidity. Changes in the brain were observed includingincreased levels of protein carbonyls, lipid hydroperoxides, and mono-amine oxidase-B activity. However, if these MPTP-treated animals werealso treated with carnosine (100 mg/kg) for 14 days, the weight loss andrigidity were decreased, as were the levels of protein carbonyls, lipidhydroperoxides, and monoamine oxidase-B activity in their brains(Boldyrev et al., 2004a,b,c). These observations seem to suggest that car-nosine suppresses some of PD-like changes induced by MPTP. At presentit is unknown whether carnosine is similarly beneficial in humans.

It has recently been shown that oxidative damage to glyceraldehydedehydrogenase (GDH), an important glycolytic enzyme, occurs in thefrontal cortex in PD patients (Gomez and Ferrer, 2009). Not only wouldthis limit ATP synthesis and generation of many necessary metabolic


intermediates, but would also increase formation of MG. As noted above,MG is a highly toxic agent which rapidly reacts with available protein andlipid amino groups, and can provoke mitochondrial dysfunction and caninduce many of the biochemical symptoms of aging. As also noted above,carnosine has been shown to protect proteins against MG-inducedmodification.

a-Synuclein is a major component of the Lewy bodies which accom-pany PD. There is an extensive literature about the roles of a-synucleinmutations and its metabolism, modification, and modes of aggregation(Bisaglia et al., 2009). There is some evidence that carnosine can inhibita-synuclein oligomerization in a model system (Kang and Kim, 2003; Kimet al., 2002). Another paper has shown that an early event in Lewy bodydiseases is formation of adducts between a-synuclein lysine aminogroups and malondialdehyde (MDA), a lipid peroxidation product, inthe substantia nigra and frontal cortex (Dalfo and Ferrer, 2008). Manyyears ago it was shown that carnosine inhibited both MDA-mediatedtoxicity in cultured neuronal cells and formation of protein carbonylsand protein cross-linking (Hipkiss et al., 1997).

It appears that the a-synuclein contains intramolecular cross-linkspossibly mediated by tissue transglutaminase (Andringa et al., 2004;Muma, 2007; Ruan and Johnson, 2007). As noted above there is a theoreti-cal possibility that carnosine may be a competitive inhibitor for tissuetransglutaminase which could prevent formation of the a-synuclein cross-links which may prevent Lewy body formation and proteasomeinhibition.

There is evidence that proteasomal function is compromised in PD(McNaught et al., 2003), which could promote accumulation ofa-synuclein, etc. and consequent inclusion body formation. Mutations inthe genes for at least two proteins, Parkin and ubiquitin carboxyterminalhydrolase L1, which are components of the ubiquitin–proteasome system(Pallancke and Greenamyre, 2006) are associated with PD. Additionally,mutations in the gene for a protein, termed Pink1, which may have a rolein mitochondrial function and appears to associate functionally withParkin (Clark et al., 2006), are also associated with PD. One possibility isthat Pink1 regulates a mitochondrial protease HtrA2 also called Omi(Plun-Favreau et al., 2007) which may be involved in protein qualitycontrol. Whether stimulation of the proteasomal system and/orchaperone-mediated autophagy by carnosine-induced increased stressprotein synthesis is beneficial to the compromised proteolytic apparatusis unknown.

A recently proposed explanation of PD involves the formation ofadducts between dopamine and the products of the peroxidation ofarachidonic acid and docosahexanoic acid (Liu et al., 2008b). At leasttwo compounds, hexanoyl-dopamine and propanoyl-dopamine, derived

118 Alan R. Hipkiss

from arachidonic acid and docosahexanoic acids respectively, are cyto-toxic to cultured neuronal cells via ROS production and mitochondrialdysfunction. It is conceivable that carnosine could substitute for dopa-mine as the dipeptide does exhibit antioxidant activity. This suggestionpredicts that hexanoyl-carnosine and propanoyl-carnosine might be gen-erated. Interestingly it has recently been reported that dietary supplemen-tation of young pigs with docosahexanoic acid provokes a decrease inmuscle carnosine levels (Li et al., 2008), which may indicate some sort ofrelationship between the dipeptide and unsaturated fatty acids and isconsistent with the idea that carnosine and docosahexanoic acid peroxi-dation products form adducts together.

Although there is no cure for PD, a common treatment to maintaindopamine supply, and hopefully slow down the degeneration, involvesusing L-dopa as a source of dopamine. It has been reported, however, thatsome of the monoamine oxidase-generated oxidative metabolites ofL-dopa are neurotoxic, most likely due to the generation of aldehydegroups on them (Burke et al., 2004). Given carnosine’s avidity for a varietyof metabolic aldehydes (discussed above), it is theoretically possible thatthe dipeptide could react with 3,4-dihydroxyphenylglycolaldehyde(dopal), the product of monomine oxidase activity on L-dopa, especiallyas L-dopa treatment in rats elevates dopal levels in their brains 18-fold(Fornai et al., 2000).

Combining carnosine treatment with L-dopa therapy in PD subjectshas recently been examined in a study using 36 patients (Boldyrev et al.,2008). It was found that carnosine treatment (1.5 g/day) significantlyimproved a number of neurological symptoms including decreased rigid-ity of the hands and legs, and increased hand movement and leg agility.At a biochemical level, it was found that the level of protein carbonyls inblood plasma was decreased after 30 days carnosine treatment; further-more, the levels of red cell Cu/Zn-superoxide dismutase were increasedin the carnosine-treated PD patients. While these results are very encour-aging with respect to the efficacy of carnosine with respect to PD andpossible other neurological conditions involving aldehydes and ROSspecies, they require much larger and extensive trials to confirm thesefindings. Nevertheless, they do seem to indicate that carnosine can exertsome therapeutic benefit despite the presence of serum carnosinases.

There has been another study where carnosine has been employed incombination with a source of dopamine. Sozio et al. (2008) chemicallylinked the b-amino group of carnosine to an L-dopa precursor and thenmeasured the release of dopamine over a 12-h period in rats. It was foundthat when presented as a co-drug the level of tissue dopamine wasretained at higher levels compared to when the animals were given freeL-dopa. There were no major effects of Parkinsonian behavior noted,


although given the very brief time over which these experiment wascarried out, this is not unexpected.

XXIV. CARNOSINE AND ISCHEMIA

Evidence that carnosine possesses anti-ischemic activity emerged fromRussian studies some years ago (see Stvolinsky and Dobrota, 2000 andreferences therein). Since then even more encouraging evidence has beenobtained; the dipeptide has therapeutic potential against a number ofischemic conditions in brain, liver, heart, and kidney. Studies of brainischemia or strokes using animal models (Gallant et al., 2000; Rajanikantet al., 2007; Tang et al., 2007; Yasuhara et al., 2008) have shown thatcarnosine is protective, even when added after the ischemic injury(Dobrota et al., 2005; Rajanikant et al., 2007; Tang et al., 2007). As part ofthe possible explanations for its protective action, it has been suggestedthat carnosine’s ability to scavenge the lipid peroxidation products hydro-xynonenal (Tang et al., 2007) and malondialdehyde (Dobrota et al., 2005)help to compensate for any ischemia-induced deficit in antioxidant activ-ity. It has also been shown in the postischemic mouse brain that carnosinetreatment, 2 h following the experimental stroke, caused a decrease in ROSlevels and matrix metalloproteinase protein levels and activity, whereasglutathione levels were preserved (Rajanikant et al., 2007). It appearedthat carnosine treatment 2 h following the experimental stroke was alsoeffective in decreasing infarct area, but when added after 4 h carnosinewas ineffective, indicating a therapeutic window if carnosine is to beconsidered for treatment of strokes in humans. Interestingly, a recentstudy has shown that the presence of bestatin, an inhibitor of the enzymecarnosinase which cleaves the dipeptide into its constituent amino acids,histidine, and b-alanine, suppressed the efficacy of carnosine in themouse brain stroke model (Min et al., 2008). Indeed the presence ofbestatin increased stroke severity but did not raise cerebral carnosinelevels. This may indicate that conversion of the dipeptide into histidineand b-alanine is required for efficacy; alternatively bestatin mayhave been exerting other unidentified effects. The carnosine analoguesanserine and N-acetyl-carnosine were much less effective than carnosinein decreasing stroke infarct size. Homocarnosine, however, does protectcultured neuronal cells against ischemia (Tabakman et al., 2004).

Protective effects have also been observed against ischemia in liver(Fouad et al., 2007), heart (Alabovsky et al., 1997), and kidney (Fujii et al.,2005, 2003; Kurata et al., 2006). Despite these clear observations of efficacy,the underlying mechanisms responsible for carnosine’s effects remainuncertain but presumably include its antioxidant and carbonyl-scavenging

120 Alan R. Hipkiss

activities and possible actions on matrix metalloproteinases (Rajanikantet al., 2007) and histamine receptors (Kurata et al., 2006).

XXV. CARNOSINE AND OSTEOPOROSIS

It is possible that regulation of protein glycation, and formation of proteinAGEs, can affect osteoporosis (Hein, 2006). Some recent studies suggestthat glycation can affect bone’s mechanical property (Shiraki et al., 2008;Tang et al., 2009) possibly by provoking deleterious changes in osteoblastfunction (Franke et al., 2007). Whether dietary carnosine would affectglycation of bone proteins is an obvious question which has beenaddressed by a Japanese group who have produced evidence suggestingthe carnosine–zinc complexes are therapeutic in terms of bone loss inanimal models and humans (Kishi et al., 1994; Sugiyama et al., 2000;Yamaguchi and Kishi, 1993; Yamaguchi and Matsui, 1996). It is suggestedthat the carnosine–zinc complex both stimulates bone formation by osteo-blasts and decreases bone resorption by the osteoclasts (Yamaguchi, 1995;Yamaguchi and Kishi, 1995a). The mechanisms involved remain obscure,but it appears that in cultured mouse marrow cells the carnosine–zinccomplex inhibits osteoclast cell formation, when present at between 10�6

and 10�4 M, by inhibiting the action of transforming growth factor-b(Yamaguchi and Kishi, 1995a) and parathyroid hormone, possibly byinterfering with calcium signaling (Yamaguchi and Kishi, 1995b). It isalso possible that the zinc–carnosine complex enhances the anaboliceffects of estrogen on osteoblasts (Yamaguchi and Matsui, 1997). It isclear from these observations that carnosine, when complexed withzinc, may have beneficial effects towards control of osteoporosis butmanymore studies, including double-blind trials in humans, are requiredbefore any unequivocal statement of its efficacy can be made.

XXVI. CARNOSINE AND CATARACTOGENESIS

A Russian group headed by Barbizhayev have produced a substantialbody of work emphasizing carnosine’s potential for the treatment oflenticluar cataracts in humans (Barbizhayev, 2008; Barbizhayev et al.,2004). In particular, Barbizhayev has suggested (Barbizhayev et al., 2001)that N-acetylcarnosine, which is unsusceptible to the action of serumcarnosinase, might be useful as a prodrug as the acetyl group is appar-ently readily cleaved intracellularly to release carnosine which thenexerts its anticataractogenic effects, most probably via a combinationof antioxidant and antiglycating activities. It is thought that there islittle carnosinase activity in the eye lens. The use eyedrops containing


N-acetyl-carnosine over trial periods of 2 and 6 months were reported toalleviate vision deficiency (lens opacity, visual acuity) associated withcataractogenesis, compared to placebo group. The improvements weresustained for 24 months (Barbizhayev, 2004, 2005).

XXVII. CARNOSINE AND DEAFNESS

Production of ROS is associated with deafness in animals and humans.It appears that carnosine can suppress loss of hearing induced by antibioticsand other agents, although it is uncertain as to the precise mechanismsinvolved (Zhuravskii et al., 2004a,b). Early studies had shown, however,that carnosine exhibited excitatory activity to the afferent fibers in the lateralline organof frogs (Mroz andSewell, 1989; Panzanelli et al., 1994)whichmayindicate an evolutionary role of the dipeptide in sound detection.

XXVIII. CARNOSINE AND CANCER

Antineoplastic activity of carnosine was first reported more than twodecades ago (Nagai and Suda, 1986). L-Carnosine’s ability to kill culturedtransformed cells (3T3 cells and HeLa cells), selectively, was found to bedependent on the absence of pyruvate in the growth medium (Hollidayand McFarland, 1996, 2000); D-carnosine was nontoxic to HeLa cells.When pyruvate or certain other metabolic intermediates (oxaloacetateand a-ketoglutarate) were present in the growth medium, the toxic effectsof carnosine towards the transformed cells were inhibited, but citrate,isocitrate, succinate, fumarate, andmalate had no effects upon carnosine’sability to kill the cells. The explanation of carnosine’s toxicity towardstransformed cells is very uncertain. It is possible that carnosine may beinhibiting glycolysis by reacting with glyceraldhyde-3-phosphate, andthereby limiting the supply of metabolic precursors and possibly ATP,whereas the addition of pyruvate, oxaloacetate, and a-ketoglutarateenables these limitations to be overcome. It should be pointed out thatmany transformed cells are highly dependent on glycolysis for their ATPsupply and are more sensitive to agents that interfere with this pathway.

The presence of dietary carnosine in vitamin E-deficient rats wasfound to increase mammary tumor latency, while not affecting tumorincidence (Boissoneault et al., 1998). Another beneficial effect of carnosinein relation to cancer has recently been reported: carnosine was shownto inhibit metastasis of hepatocarcinoma SK-Hep-1 cells (Chung and Hu,2008). Unlike the effects reported above, carnosine did not affect theviability of these cells but instead the dipeptide inhibited cell migrationand invasion. The mechanism responsible apparently involves a decrease

122 Alan R. Hipkiss

in extracellular matrix metalloproteinase-9 (MMP-9), an activity neces-sary for tumor invasion and angiogenesis. However, carnosine did notdirectly affect MMP-9 activity. Instead the dipeptide appears upregulateexpression of the antimetastatic gene nm23-H1, whose gene productinhibits MMP-9 gene expression, and thereby suppresses synthesis ofthis activity necessary for tumor invasion and metastasis.

XXIX. CARNOSINE AND WOUND HEALING

One problematic aspect of the aged organism is slower woundhealing. There is evidence that carnosine can have beneficial effects here(Roberts et al., 1998). When carnosine is complexed with zinc to form‘‘polaprezinc,’’ it behaves as an antiulcer drug which also possesseswound-healing activity (Nagai et al., 1986). Investigation of the possiblemechanisms involved has revealed that the zinc–carnosine complex maystimulate synthesis of insulin-like growth factor-1 (Watanabe et al., 1998)and decreased secretion of interleukin-8 (IL-8), due to suppression of IL-8mRNA expression in gastric epithelial cells. Polaprezinc also downregu-lated NF-kB activation by a number of activators suggesting overallanti-inflammatory action (Shimada et al., 1999), but did not involvemodification of prostaglandin E2 production in gastric epithelial cells(Arakawa et al., 1990). Further investigation revealed that zinc–carnosineinduced expression of the stress protein Hsp72 while inhibiting NF-kBactivation in colonic mucosa (Odashima et al., 2002, 2006). These observa-tions may help explain the beneficial effects of polaprezinc in rodent andhuman gut (Mahmood et al., 2007). Polaprezinc also ameliorated aspirin-induced mucosal injury in rats (Naito et al., 2001) most probably byinhibiting the increase in neutrophil myeloperoxidase via inhibition ofTNF-a expression.

Vimentin is thought to play a role in wound healing (Mor-Vaknin et al.,2003), and carnosine has been shown to stimulate vimentin expression inrat fibroblasts (Ikeda et al., 1999). Therefore, it is possible that thisprovides an additional mechanism for carnosine’s beneficial effects onwound healing.

XXX. CARNOSINE AND IMMUNE FUNCTION

There is some evidence suggesting that carnosine can upregulateimmune function. Carnosine’s ability to react with hypochlorite anions(Formazyuk et al., 1992; Quinn et al., 1992) generated in activated leuko-cytes via the myeloperoxidase reaction, suggests that the dipeptide maylimit hypochlorite-mediated oxidation in vivo (Pattison and Davies, 2006)


and moderate neutrophil function (Tan and Candish, 1998). There is alsosome evidence that carnosine can suppress contact hypersensitivity inmice, but the mechanisms involved have not been studied in detail (Reeveet al., 1993a,b).

XXXI. CARNOSINE, CALCIUM, AND HEART FAILURE

Carnosine occurs in cardiac muscle at concentrations between 2 and10 mM (Roberts and Zaloga, 2000). Heart failure is thought to be asso-ciated with dysregulation of myocardial calcium metabolism resulting incontractile failure. There is evidence that carnosine can improve cardiaccontractility, possibly via its effects on regulation of intracellular calciumlevels, in a concentration-dependent manner (Zaloga et al., 1996). Studiesin rats have shown that carnosine increases the levels of free calcium ionswhile also increasing the sensitivity of the contractile proteins to calcium(Batrukova and Rubstov, 1997; Roberts and Zaloga, 2000; Zaloga et al.,1997). It is unknown if there is any relationship between heart failure andmyocardial carnosine levels in humans patients, although it is known thattissue carnosine levels are decreased in animals suffering from traumaand chronic infection which are associated with impaired cardiac contrac-tility (Roberts and Zaloga, 2000).

XXXII. CARNOSINE AND AUTISTIC SPECTRUM DISORDERS

Autism and Asperger’s syndrome are regarded as pervasive develop-mental disorders. Autism is a neurological disorder associated withimpairment of language, cognition, and socialization, whereas Asperger’ssyndrome is an autistic condition not associated with language delay orintellectual impairment. The causes of these conditions are unknown. In adouble-blind, placebo-controlled, trial it was found that that carnosinesupplementation improved the behavior, communication, and socializa-tion in children with autistic spectrum disorders (Chez et al., 2002). Themechanism responsible for these effects is very uncertain, but it has beenhypothesized that increased oxidative stress may be associated withautism (Chauhan and Chauhan, 2006; Chauhan et al., 2004; Yorbik et al.,2002), and that polymorphisms in the gene coding for the aldehyde-scavenging enzyme glyoxalase 1 could be a susceptibility factor (Junaidet al., 2004). Other workers have questioned this conclusion (Rehnstromet al., 2008; Sacco et al., 2007; Wu et al., 2008). However, a recent finding byFujimoto et al (2008) has shown that expression of glyoxalase mRNA inwhite blood cells correlated inversely with the onset of depression inbipolar disorder patients, compared to controls. It is possible that changes

124 Alan R. Hipkiss

in glyoxalase expression in neocortex tissue may play a role in autism(Sacco et al., 2007) as increased protein AGEs have been detected inpostmortem autistic brain ( Junaid et al., 2004). These observations mayunderline carnosine’s effects on the autistic children as the dipeptidepossesses antioxidant activity and is protective against methylglyoxal-mediated protein modification. However, it should be emphasized thatthere have been no other published reports of the beneficial effects ofcarnosine towards autistic spectrum disorders.

XXXIII. CARNOSINE AND BLOOD PRESSURE

There is evidence that carnosine is a vasodilator (Ririe et al., 2000) andthus can lower blood pressure (Niijima et al., 2002; Tanida et al., 2005).It has been shown that carnosine promotes synthesis of nitric oxide(Nicoletti et al., 2007; Tomonaga et al., 2005), a well-known dilator ofblood vessel walls. Also carnosine can inhibit angiotensin-convertingenzyme (ACE) activity (Hou et al., 2003; Nakagawa et al., 2006) possiblyvia effects on cGMP and nitric oxide, which again points to the possibilitythat the dipeptide or carnosine-enriched foods could be explored tocombat raised blood pressure in humans.

XXXIV. CARNOSINE AND CONSUMPTIONOF ALCOHOLIC BEVERAGES

Consumption of alcoholic drinks leads to the generation of acetaldehydein the tissues, predominantly the liver but also in the brain. Acetaldehydecan react with protein amino groups to generate carbonyls with the poten-tial for cross-linking to other macromolecules. It is thought that acetalde-hyde generation is a major source of ‘‘hangovers’’ experienced followingexcessive alcohol consumption. Given carnosine’s ability to react withacetaldehyde and protect cultured human fibroblasts and lymphocytesagainst its toxicity (Hipkiss et al., 1998a) as well as prevent cross-linkingbetween protein and DNA (Hipkiss et al., 1995b), it has been suggestedthat ingestion of carnosine, either as a supplement or as a high-carnosinefood (meat), could be an effective way to prevent ‘‘hangovers’’(Hipkiss, 1998), as well as protecting the brain and other tissues againstalcohol-induced glycoxidative damage. This should be relative easyto test.

Beneficial effects of carnosine have been described with respect toethanol-induced liver injury in mice (Liu et al., 2008a,b). It was foundthat following 3 weeks of ethanol treatment (present in drinking water),subsequent exposure to carnosine decreased liver malondialdehyde


levels by around 40% compared to ethanol-treated animals. The dipeptidealso promoted a decline in indices of cell damage (release of liverenzymes), increased glutathione content and catalase and glutathioneperoxidase activities, and downregulated expression of inflammatory-associated cytokines (IL-6 and TNF-a).

XXXV. CARNOSINE AND HIGH FRUCTOSE FOODSAND DRINKS

There has been much interest in the metabolic effects of fructose andwhether its consumption should be restricted due to the sugar’s potentialdeleterious effects with respect to diabetes-associated phenomena(Abdel-Sayred et al., 2008; Brown et al., 2008; Le and Tappy, 2006; Millerand Adeli, 2008). In particular, fructose glycates proteins far more readilyglucose to generate protein AGEs, and as a consequence, could possiblybe responsible for increasing the incidence of type-2 diabetic complica-tions ( Jia and Wu, 2007; Lo et al., 2008). Additionally, fructose is also aready metabolic source of methylglyoxal which, as already described, is ahighly deleterious agent which is thought to be a major causal agent ofAGE formation and therefore much of the secondary complications oftype-2 diabetes. As carnosine can react with methylglyoxal and fructoseto prevent their damaging effects on proteins at the test-tube level, onemight consider whether increasing tissue carnosine levels might be bene-ficial in high fructose diets. However, no such study has yet been carriedout to test whether any of these ideas are justified. A further observationhas recently been revealed in that high fructose consumption by men canincrease the risk of gout due to an increased production of uric acid (Choiand Curhan, 2008; Gao et al., 2008). Consequently it would be interestingto determine whether carnosine ameliorates uric acid production inhumans.

XXXVI. CARNOSINE AND DIALYSIS FLUIDS

Treatment of kidney failure involves dialysis using heat-sterilized dialysisfluids. Because the dialysis fluid contains glucose, the heating inevitablygenerates glucose degradation products such as methylglyoxal, glyoxal,and acetaldehyde, which are well recognized for their ability to induceAGEs on protein targets. Hence dialysis with aldehyde-containing dialy-sis fluid will not be expected to improve kidney health, but exacerbate thekidney dysfunction. Because of carnosine’s ability to protect proteinsagainst aldehydic glycating agents, the possibility that the dipeptidemay decrease reactivity of the AGE precursors was explored

126 Alan R. Hipkiss

(Alhamdani et al., 2007a,b). It was shown that heat-treated peritonealdialysis fluid compromised the viability of cultured human peritonealmesothelial cells, whereas the additional presence of carnosine in theincubation medium considerably enhanced cell viability, and markedlydecreased cell-associated protein carbonyl groups and ROS generation.These observations obviously suggest that carnosine could be employedto either remove deleterious glucose degradation products fromdialysis fluid prior to use, or that the dipeptide could be added to dialysisfluid to suppress the reactivity of the protein damaging agents. However,in the latter case it is uncertain whether patients’ kidneys would be able todeal with (i.e., selectively excrete) the putative aldehyde–carnosineadducts.

XXXVII. POSSIBLE WAYS TO INCREASE TISSUE CARNOSINELEVELS: PHYSIOLOGICAL REGULATION

There have been relatively few studies of age-related changes in tissuecarnosine levels despite the fact that the initial observation of the dipep-tide’s ability to suppress some features of senescence were made morethan 15 years ago. Carnosine levels have been reported to decline with agein the rats ( Johnson and Hammer, 1992; Stuerenburg and Kunze, 1999)and human muscle (Stuerenburg and Kunze, 1999). More recently, Tallonet al. (2007) found evidence of carnosine’s age-related decline in humanmuscle fibers.

On the assumption that increasing tissue levels of carnosine might bebeneficial in terms of aging and some of its related conditions, this can beachieved either by physiological regulation or by dietary supplementa-tion. Muscle carnosine levels are generally higher when accompanied byintense exercise in fast-twitch type II fibers compared to slow-twitch typeI fibers. It has been found that muscle levels of the dipeptide can beincreased following resistance exercise in humans (Hill et al., 2006) andthere are also reports that very high levels of carnosine are present inhighly trained race-horses (Harris et al., 1990). In humans, the carnosinecontent of vastus lateralis muscle is generally high in sprinters and bodybuilders (Tallon et al., 2005); 8 weeks intensive training resulted in adoubling of the carnosine content of the vastus lateralis muscle (Kimet al., 2005).

In athletes involved in explosive/intense muscle exercise, it is likelythat the raised carnosine levels are required as physiological buffers.Hence it is possible that raising carnosine levels may improve muscleperformance by increasing buffer capacity. In an attempt to increasecarnosine levels by dietary means, but circumventing the effects ofserum carnosinase, increasing b-alanine intake has been investigated


(Harris et al., 2006). Studies have shown that the availability of b-alaninemay limit carnosine synthesis, histidine being generally available meta-bolically. Therefore, it was suggested that dietary supplementation withb-alanine could raise carnosine synthesis in the tissues and studies onhuman subjects have shown that b-alanine is effective in raising musclecarnosine levels (Harris et al, 2006). Furthermore, b-alanine supplementa-tion for 4 and 10 weeks increased vastus lateralis carnosine content by58% and 80%, respectively, in subjects subjected to high-intensity cycling.A similar study using Vietnamese sports-science students showed thatb-alanine supplementation promoted an increase in muscle carnosineconcentration (Kendrick et al., 2008), although there were no improve-ments in any of the exercise parameters measured. In a double-blindrandomized study of 26 elderly subjects, aged between 55 and 92 years,it was found that b-alanine supplementation for 90 days improvedmuscleendurance (physical working capacity) by 28%, presumably due to theincreased synthesis of carnosine (Stout et al., 2008). It is interesting thatwhile no beneficial effects in terms of muscle performance were observedin young subjects (Kendrick et al., 2008), whereas in the elderly, improve-ment was detected (Stout et al., 2008), presumably due to the lower tissuecarnosine levels which limit performance in old muscles.

It should be pointed out that subjects consuming b-alanine as a sup-plement (40 mg/kg body weight) experienced symptoms of flushing, skinirritation and prickly sensations for up to 1 h, first of the ears, foreheadand scalp and then the trunk, arms, hands, spine, and buttocks. Loweringthe b-alanine dose to 10 mg/kg body weight effectively eliminated thesesymptoms (Harris et al., 2006). Interestingly, consumption of chickenbroth (enriched in carnosine) containing the equivalent of 40 mg/kgbody weight b-alanine did not induce any of the unpleasant symptoms,but carnosine was not detected in the plasma of these subjects.

Whether it is possible to raise carnosine levels in human brain isunknown. One study in rats has shown that oral administration of chickenextract (a major source of carnosine in humans too) did provoke an increasein brain carnosine levels: a single dose of the chicken extract led to anincrease in carnosine levels within 30 min in plasma, but 1 or 2 h durationwere required for increased levels of carnosine to be observed in the cerebralcortex, hypthalamus, and hippocampus (Tomonaga et al., 2007). It is uncer-tain whether these effects result from direct uptake of the carnosine fromplasma or a consequence of de novo synthesis. In a study using senescence-accelerated mice (SAMP8), it was found that oral supplementation withcreatine provoked, at 25 weeks of age, a transient 88% increase in musclecarnosine content, accompanied by a 40% increase in anserine content,which coincided with an improvement in resistance to contractile fatigue(Derave et al., 2008). At 60 weeks, no differences were detectable betweenthe creatine-supplemented and control animals in terms of their muscle

128 Alan R. Hipkiss

carnosine and anserine levels. The mechanism responsible for this effectis uncertain but could involve an upregulation of carnosine synthesis orits decreased catabolism, but why either of these is influenced by creatinesupply remains unclear. This study also showed that muscle carnosinedeclines by 45%with age (from 10 to 60 weeks) in the control SAMP8 mice.

Relatively little is known of the factors that control carnosine synthe-sis, although as mentioned above, the enzyme involved, carnosine syn-thetase, does appear to be regulated by cyclic AMP (Schulz et al., 1989)which suggests the possibility that conditions which lower cyclic AMPlevels may increase carnosine synthesis. Thus, increased glucose metabo-lism via glycolysis might be accompanied by an upregulation of carnosinesynthesis. It is possible that this suggested relationship is beneficial due tocarnosine’s ability to suppress the reactivity of a major deleterious bypro-duct of the glycolytic pathway, methylglyoxal (MG) (discussed above).The primary enzyme responsible for carnosine’s hydrolysis into b-alanineand histidine is carnosinase, an activity which is stimulated by citrate(Vistoli et al., 2006). These observations raise the possibility that intracel-lular carnosine levels may be subject to metabolic regulation due to theeffects of major metabolic intermediates and effectors on the enzymesresponsible for its synthesis and degradation.

Carnosine levels in the tissues have been seen to decline followingtrauma and during chronic infection (Fitzpatrick et al., 1980). It is interest-ing that pathological states are associated with decreased cardiac functionpossible due to problems with cardiac muscle contraction. It has beensuggested that decreased carnosine levels may play a role in decreasedmuscle contractivity in a number of disease states including congestiveheart failure (Roberts and Zaloga, 2000).

XXXVIII. POSSIBLE WAYS TO INCREASE TISSUE CARNOSINELEVELS: DIETARY SUPPLEMENTATION

While carnosine is absorbed intact from the gut, the presence of serumcarnosinase is frequently cited as an impediment to the dipeptide’s poten-tial efficacy. However, studies have shown that serum carnosine levelsare raised at least temporarily, up to 4–5 h, following a carnosine-containing meal (Antonini et al., 2002; Park et al., 2005). Such studiesindicate a window of opportunity for carnosine administration. Oneapproach to overcoming the carnosinase effect would be to employ acarnosinase inhibitor such as bestatin, although undoubtedly therewould be some side effects. The fact that carnosinase has been shown tobe upregulated by citrate (Vistoli et al., 2006) may permit the design ofspecific inhibitory molecules. Another approach would be to employ aform of carnosine which is resistant to carnosinase attack such as


N-acetyl-carnosine or the decarboxylated form carcinine. In fact N-acetyl-carnosine has been proposed as a prodrug to treat cataracts in the eyelens, as the acetyl group is readily removed intracellularly (Barbizhayevet al., 2004) (see section on cataracts for more details). An alternative wayto evade serum carnosinase activity would be to introduce carnosine vianasal administration (Hipkiss, 2005). This route may be particularlyappropriate for raising carnosine levels in the brain as the olfactory lobeis normally enriched in the dipeptide.

Synthesis of carnosine analogues resistant to carnosinase attack isanother method which has been employed in an attempt to circumventthe problem of serum carnosinase (Bellia et al., 2008; Cacciatore et al., 2005;Calcagni et al., 1999; Guiotto et al., 2005b). Some of the resultant structureswith the b-alanine replaced by 2,3-diaminoproprionic acid residue andwith acetyl groups on either of its amino groups were shown to not onlyresist attack by carnosinase but also inhibit the enzyme’s ability to cleavecarnosine, while still retaining hydroxyl radical-scavenging activity andpreventing peroxynitrite-mediated tyrosine nitration (Cacciatore et al.,2005). Other structures synthesized have included sulfonamido-pseudopeptides, tauryl-histidine, tauryl-1-methylhistidine, and tauryl-3-methylhistidine (Calcagni et al., 1999), while another Italian laboratorygenerated cyclodetrin conjugates of carnosine, attached via the dipep-tide’s b-amino group to either carbon-3 or carbon-6 of the glucose moiety.These glycosidic derivatives were resistant to carnosinase attack, andseemed to be better inhibitors of copper-induced lipid peroxidation thanthe parent dipeptides, carnosine, and anserine (Bellia et al., 2008). Therewere no reports on the efficacy of the cyclodextrin conjugates with respectto carnosine’s antiglycating activity, however.

XXXIX. IS THERE ANY EVIDENCE THAT CHANGES INDIETARY CARNOSINE HAVE ANY EFFECTSIN HUMANS?

There have been few studies on carnosine consumption in humans. Dueto the documented presence of carnosinase in blood, many scientists haveassumed that the dipeptide’ survival would be relatively short due to itsrapid hydrolysis by the enzyme. Nevertheless, a study by Gardner et al.(1991) showed that plasma carnosine levels peaked at over 180 mg/ml,0.5 h after intake of a beverage containing 3 g of carnosine. Maximalcarnosine levels in urine occurred within 2 h. In another study, Parket al. (2005) showed that ingestion of cooked ground beef containing248 mg of carnosine led to plasma carnosine concentration rising fromessentially 0 to around 30 mg/ml within 3.5 h and thereafter rapidlydeclining such that none was detectable 2 h later. These studies indicate

130 Alan R. Hipkiss

that, despite the presence of serum carnosinase, ingestion of carnosine canlead to raised levels of the dipeptide in blood which could then lead toeffects on tissue carnosine.

Studies of the effects of carnosine consumption on humans have beenrare. One study showed that, following intake of a bolus of carnosine(450 mg), serum total antioxidant activity was increased by 11% 1 h afteringestion (Antonini et al., 2002). Such an increase in antioxidant function isconsistent with the dipeptide’s recognized antioxidant activity. It is pos-sible that human brain function can be affected by dietary carnosine as ithas been shown that the dipeptide (two 400 mg doses per day) canmodulate the behavior, socialization, and communication skills of autisticchildren (Chez et al., 2002 and discussed above). Homocarnosine, whichalso possesses anticonvulsant activity in humans, is found in human CSF.It is unknown whether the sevenfold age-related decline in homocarnsinelevels in human CSF (Huang et al., 2005; Janssen et al., 2005) contributesto the onset or progression of any age-related pathology. However, asprotein AGEs accumulate in CSF of AD patients (Ahmed et al., 2005;Shuvaev et al., 2001), one has to at least consider whether there is a causalrelationship between these observations (Hipkiss, 2007a).

Carnosine (400 mg/day) together with omega-3-fatty acids (eicosa-pentaenonic acid) was employed in a study of dietary effects on dyslexicchildren. In this study there were no significant effects of the dietarysupplements on a range of language skills and behavior problems(Kairaluoma et al., 2008).

There is some evidence that carnosine supplementation can restoresense of taste in humans. A report from Japan states that polaprezinc(a zinc–carnosine complex) is frequently effective in treating patientsexperiencing taste disorders (Ikeda et al., 2005). Experiments with zinc-deficient rats showed that polaprezinc was effective in restoration of tastebud proliferation (Hamano et al., 2006).

Polaprezinc has also been shown to be beneficial in treatment of ulcersand other gut lesions (discussed above) and in inhibiting some of thechanges surrounding osteoporosis (also discussed above).

XXXX. WOULD VEGETARIANS BENEFIT FROM CARNOSINESUPPLEMENTATION?

The possibility that vegetarian diets, deficient in carnosine, could besomewhat deleterious in the ability to suppress aldehyde-inducedprotein modification and AGE has been discussed (Hipkiss, 2005,2006c). Indeed, Harris et al. (2007) showed that muscle carnosine levelswere reduced by up 50% in vegetarian subjects. There is one report of


increased AGEs in vegetarian type-2 diabetics’ sera (Krajcovicova-Kudlackova et al., 2002). This may be due to the absence of carnosine inthe vegetarian diet, but an alternative explanation could be that theincreased fructose in the vegetarian diet increases AGE formation. Itshould be pointed out that many plants contain high levels of aldehydescavengers which are probably present to protect plant proteins againstglycating sugars such as fructose, and which could be exploited as dietaryantiglycating agents.

XXXXI. DELETERIOUS EFFECTS OF CARNOSINE

Carnosine is usually regarded as being almost nontoxic (Sato et al., 2008).However, there are some indications that the dipeptide can have deleteri-ous effects. It has been known for a long time that humans with mutationsin the gene coding for serum carnosinase show high levels of the dipep-tide in their blood which is accompanied by neurological dysfunction(Gjessing et al., 1990; Wassif et al., 1994; Willi et al., 1997). This may suggestthat elevated serum carnosine or a failure to cleave the dipeptide else-where has deleterious effects, although there have been claims made thatelevated levels of serum carnosine is not in itself a problem. There is onerather obvious possible way in which carnosine could be deleteriousshould the dipeptide prove to be an inhibitor of serum transglutaminaseactivity (Hipkiss, 2007a). Inhibition of this enzyme would suppress thedevelopment of cross-linking between fibrin molecules and could there-fore compromise blood clotting. However, this remains a speculation atthis stage.

There is one report suggesting that carnosine affects spermatogenesisin senescence-accelerated mice (SAMP1), reducing cell yield and increas-ing destructive changes in spermatogenic epithelium in the testiculartubules (Gopko et al., 2005). However, the same research group had earlierstated (Zakhidov et al., 2002) that carnosine did not modify the incidenceof chromosome mutations in spermatogenic cells in these animals. Thereis also a paradoxical situation in the aging-resistant SAMR1 mice. In 2002it was stated that carnosine increased the count of aberrant spermatogo-nia in the SAMR1 animals (Zakhidov et al., 2002), whereas it was laterreported that in these animals, carnosine treatment resulted in no increasein the incidence of aberrant spermatogia (Gopko et al., 2005). Hence thesignificance of these observations is uncertain, but there have been noother reports of deleterious effects of carnosine. Nevertheless, it cannot bearbitrarily assumed that carnosine may not be without toxicity in somesystems or organs.

132 Alan R. Hipkiss

XXXXII. CONCLUSIONS

Studies using model systems, cell culture, and animals have indicatedthat carnosine possesses a range of potential homeostatic functions whichtogether may help to suppress many of the biochemical changes to macro-molecules which accompany aging and a number of related pathologicalconditions. Especially relevant is carnosine’s carbonyl-scavenging abilitywhich may prove to be particularly important in suppressing formationof protein carbonyls and those cross-linked protein species which inhibitproteasomal elimination of altered proteins. Carnosinemay also stimulatenitric oxide synthesis and thereby increase proteasome activity, as well asupregulate synthesis of another protease, OPH. There is also some evi-dence that carnosine’s other properties could also contribute to its antia-ging activities include antioxidant, wound healing agent, aldehydescavenger (including methylglyoxal), copper and zinc chelator, heat-shock protein inducer, anti-inflammatory agent, and antiepileptic agent.Much more work is required to explore all these proposals and specula-tions. As aging does appear to be multifactorially controlled, it is perhapsnot unsurprising that a pluripotent agent such as carnosine might exertantiaging effects via more than one mode of action.

Evidence of carnosine’s efficacy towards human health is relativelysparse in comparison with the range of effects observed in model systemsand in animals. The most intensely investigated is cataractogenesis,although predominantly undertaken by a single research group. Thereis also relatively strong evidence that, in its zinc complex form (polapre-zinc), carnosine has positive effects on in repair of gut lesions such asulcers. While a good case can be made for the use of carnosine in alleviat-ing the deleterious effects of ischemic conditions, especially stroke-relatedevents, there is little or no direct evidence as yet that the dipeptide isefficacious in human patients. This is also the position for the neurode-generative diseases, AD and PD, although there is evidence from oneresearch group showing that carnosine does have beneficial effects onParkinson’s patients undergoing treatment with L-dopa. The beneficialeffects of carnosine on children with autistic spectrum disorders havebeen described by only one research group, so this needs verification.While a case can be made that carnosine may be useful in controlling thesecondary complications associated with type-2 diabetes, there is nodirect evidence that increased carnosine consumption suppresses theirdevelopment in humans.

Clearly, much more work is required to verify or refute the manyproposals made of carnosine’s efficacy towards human health. Whethersuch studies will be undertaken is in doubt simply because of the non-patentability of the molecule and therefore it is unlikely to generate large


monetary benefit to any company or institution. However, if one were totry to determine whether carnosine does prevent many of the unpleasanteffects of ‘‘hangovers’’ following excess ethanol consumption, one wouldanticipate that there would be no difficulty in finding volunteers. Sciencecan sometimes be fun as well as intellectually challenging.

REFERENCES

Abdel-Sayred, A., Binnert, C., Le, K. A., Bortolotti, M., Schneiter, P., and Tappy, L. (2008).A high fructose diet impars basal and stress-mediated lipid metabolism in healthy malesubjects. Br. J. Nutr. 21, 1–7.

Abe, H. (2000). Role of histidine-related compounds as intracellular proton buffering con-stituents in vertebrate muscle. Biochemistry (Moscow) 65, 757–765.

Ahmed, N. and Thornalley, P. J. (2007). Advanced glycation endproducts: What is theirrelevance to diabetic complications? Diabetes Obes. Metab. 9, 233–246.

Ahmed, N., Ahmed, U., Thornalley, P. J., Hager, K., Fleischer, G., and Munch, G. (2005).Protein glycation, oxidation and nitration adduct residues and free adducts of cerebro-spinal fluid in Alzheimer’s disease and link to cognitive impairment. J. Neurochem. 92,255–263.

Alabovsky, V. V., Boldyfer, A. A., Vinokurov, A. A., and Shchavratsky, V. Kh. (1997). Effectof histidine-containing dipeptides on isolated heart under ischemia/reperfusion. Bio-chemistry (Moscow) 62, 77–87.

Alaghband-Zadeh, J., Mehdizadeh, S., Khan, N. S., O’Farrell, A., Bitensky, L., and Chayen, J.(2001). The natural substrate for nitric oxide synthase activity. Cell Biochem. Funct. 19,277–280.

Aldini, G., Carini, M., Beretta, G., Bradamante, S., and Facino, R. B. (2002). Carnosine is aquencher of 4-hydroxy-nonenal: Though what is the mechanism of reaction? Biochem.

Biophys. Res. Commun. 298, 699–706.Aldini, G., Orioli, M., Carini, M., and Facino, M. (2004). Profiling histidine-containing

dipeptides in rat tissues by liquid chromatography/electrospray ionization tandemmass spectrometry. J. Mass Spectrom. 39, 1417–1428.

Aldini, G., Maffei-Fracino, R., Beretta, G., and Carini, M. (2005). Carnosine and relateddipeptides as quenchers of reactive carbonyl species: From structural studies to thera-peutic perspectives. BioFactors 24, 77–87.

Alhamdani, M. S. S., Al-Kassir, A. H. A. M., Abbas, F. K. H., Jaleel, N. A., and Al-Taee, M. F.(2007a). Antiglycating and antioxidant effect of carnosine against glucose degradationproducts in peritoneal mesothelial cells. Nephron Clin. Pract. 107, 26–34.

Alhamdani, M. S. S., Al-Azzawie, H. F., and Abbas, F. K. (2007b). Decreased formation ofadvanced glycation end-products in peritoneal fluid by carnosine and related peptides.Perit. Dial. Int. 27, 86–89.

Anderson, R. M., Barger, J. L., Edwards, M. G., Braun, K. H., O’Conner, C. E., Prolla, T. A.,and Weindruch, R. (2008). Dynamic regulation of PGC-1alpha localization and turnoverimplicates mitochondrial adaptation in caloric restriction and the stress response. AgingCell 7, 101–111.

Andringa, G., Lam, K. Y., Chegary, M., Wang, X., Chase, T. N., and Bennett, M. C. (2004).Tissue transglutaminase catalyzes the formation of alpha-synuclein crosslinks inParkinson’s disease. FASEB J. 18, 932–934.

Antonini, F. M., Petruzzi, E., Pinzani, P., Orlando, C., Poggesi, M., Serio, M., Pazzagli, M.,andMasotti, G. (2002). The meat in the diet of aged subjects and the antioxidant effects ofcarnosine. Arch. Gerontol. Geriatr. Suppl. 8, 7–14.

134 Alan R. Hipkiss

Arakawa, T., Satoh, H., Nakamura, A., Nebiki, H., Fukuda, T., Sakuma, H., Nakamura, H.,Ishikawa, M., Seiki, M., and Kobayashi, K. (1990). Effects of zinc L-carnosine on gastricmucosal and cell damage caused by ethanol inrats. Correlation with endogenousprostaglandin E2. Dig. Dis. Sci. 35, 559–566.

Aristoy, M. C. and Toldra, F. (2004). Histidine dipeptides HPLC-based test for the detectionof mammalian origin proteins in feeds for ruminants. Meat Sci. 67, 211–217.

de Arriba, S. G., Stuchbury, G., Yarin, J., Burnell, J., Loske, C., and Munch, G. (2006).Methylglyoxal impairs glucose metabolism and leads to energy depletion in neuronalcells—Protection by carbonyl scavengers. Neurobiol. Aging 28, 1044–1050.

Atamna, H., PalerMartinez, A., and Ames, B. N. (2000). N-t-Butylhydroxylamine, a hydroly-sis product of a-phenyl-N-t-butylnitrone, is more potent in delaying senescence in humanlung fibroblasts. J. Biol. Chem. 275, 6741–6748.

Bakardjiev, A. (1997). Biosynthesis of carnosine in primary cultures of rat olfactory bulb.Neurosci. Lett. 227, 115–118.

Balion, C. M., Benson, C., Raina, P. S., Papaioannou, A., Patterson, C., and Ismaila, A. S.(2007). Brain type carnosinase in dementia: A pilot study. BMC Neurol. 7, 38.

Bandhyopadhyay, U. and Cuervo, A. M. (2007). Chaperone-mediated autophagy in agingand neurodegeneration: Lessons from a-synuclein. Exp. Gerontol. 42, 120–128.

Bar, K. J., Franke, S., Wenda, B., Muller, S., Kientsch-Engel, R., Stein, G., and Sauer, H. (2002).Pentosidine and N(epsilon)-(carboxymethyl)-lysine in Alzheimer’s disease and vasculardementia. Neurobiol. Aging 24, 333–338.

Baran, E. J. (2000). Metal complexes of carnosine. Biochemistry (Moscow) 65, 789–797.Barea, F. and Bonatto, D. (2008). Relationships among carbohydrate intermediate metabolites

andDNAdamage and repair in yeast from a systems biology perspective.Mutat. Res. 642,43–56.

Bauer, K. (2005). Carnosine and homocarnosine, the forgotten, enigmatic peptides of thebrain. Neurochem. Res. 30, 1339–1345.

Barbizhayev, M. A. (2004). Rejuvenation of visual function in older adult drivers and driverswith cataract during a short-term administration of N-acetylcarnosine lubricant eyedrops. Rejuvenation Res. 7, 186–198.

Barbizhayev, M. A. (2005). Analysis of lipid peroxidation and electron microscopicsurvey of maturation stages during human cataractogenesis: Pharmacokinetic assay ofCan-C N-acetylcarnosine prodrug lubricant eye drops for cataract prevention. Drugs 6,345–369.

Barbizhayev, M. A. (2008). Ocular drug metabolism of the bioactivating antioxidantN-acetylcarnosine for vision in ophthalmic prodrug and codrug design and delivery.Drug Dev. Ind. Pharm. 34, 1071–1089.

Barbizhayev, M. A., Deyev, A. I., Yermanova, V. N., Semiletov, Y. A., Davydova, N. G.,Kurysheva, N. I., Zhukotskii, A. V., and Goldman, I. M. (2001). N-Acetylcarnosine, anatural histidine-containing dipeptide, as a potent ophthalmic drug in treatment ofhuman cataracts. Peptides 22, 979–994.

Barbizhayev, M. A., Deyev, A. I., Yermanova, V. N., Brikman, I. V., and Bours, J. (2004). Lipidperoxidation and cataracts:N-acetylcarnosine as a therapeutic tool to manage age-relatedcataracts in human and canine eyes. Drugs 5, 125–139.

Batrukova, M. A. and Rubstov, A. M. (1997). Histidine-containing dipeptides as endogenousregulators of the activity of sarcoplasmic reticulum Ca-releasing channels. Biochim.Biophys. Acta 1324, 142–150.

Battah, S., Ahmed, N., and Thornalley, P. J. (2002). Novel anti-glycation therapeutic agents:Glyoxalase-1-mimetics. Int. Congr. Ser. 1245, 107–111.

Bellia, F., Amorini, A. M., La Mendola, D., Vacchio, G., Tavazzi, B., Giardini, B., Di Pietro, V.,Lazzzarino, G., and Rizzarelli, E. (2008). New glycosidic derivaties of histidine-contining


dipeptides with antioxidant properties and resistant to carnosinase activity. Eur. J. Med.

Chem. 43, 373–380.Bennet, E. J., Bjerregaard, J., Knapp, J. E., Chavous, D. A., Friedman, A. M., Royer, W. E.,

and O’Connor, C. M. (2003). Catalytic implications from the Drosophila proteinL-isoaspartyl methyltransferase structure and site-directed mutagenesis. Biochemistry

42, 12844–12853.Bergamini, E., Cavallini, G., Donati, A., and Gori, Z. (2007). The role of autophagy in aging.

Its essential part in the anti-aging mechanism of caloric restriction. Ann. N. Y. Acad. Sci.

1114, 69–78.Bharadwaj, L. A., Davies, G. F., Xavier, I. J., and Ovsenek, N. (2002). L-Carnosine and

verapamil inhibit hypoxia-induced expression of hypoxia inducible factor (HIF1-alpha)in H9c2 cardiomyoblasts. Pharmacol. Res. 45, 175–181.

Bisaglia, M., Mammi, S., and Bubacco, L. (2009). Structural insights on physiological func-tions and pathological effects of a-synuclein. FASEB J. 23, 329–340.

Bogardus, S. L. and Boissonneault, G. A. (2000). Carnosine inhibits in vitro low-densitylipoprotein oxidation. Nutr. Res. 20, 967–976.

Boissoneault, G. A., Hardwick, T. A., Bogardus, S. L., Chan, W. K. M., Tatum, V.,Glauert, H. P., Chow, C. K., and Decker, E. A. (1998). Interactions between carnosineand vitamin E in mammary cancer risk determination. Nutr. Res. 18, 723–733.

Boldyrev, A. A. (1993). Does carnosine possess direct antioxidant activity? Int. J. Biochem. 25,1101–1107.

Boldyrev, A.A., Koldo, A., Kurella, E., Maltseva, V., and Stvolinski, S. (1993). Natural histidine-containing dipeptide carnosine as a potent hydrophilic antioxidant with membranestabilizing function. A biomedical report. Mol. Chem. Neuropathol. 19, 185–192.

Boldyrev, A. A., Abe, H., and Stvolinsky, S. (1995). Effects of carnosine and related com-pounds on generation of free oxygen species: A comparative study. Comp. Biochem.

Physiol. B Biochem. Mol. Biol. 112, 481–485.Boldyrev, A., Stvolisky, S. L., Tyulina, O. V., Koshelev, V. B., Hori, N., and Carpenter, O.

(1997). Biochemical and physiological evidence that carnosine is an endogenous neuro-protector against free radicals. Cell. Mol. Neurobiol. 17, 259–271.

Boldyrev, A. A., Gallant, S. C., and Sukhich, G. T. (1999a). Carnosine, the protective, anti-ageing peptide. Biosci. Rep. 19, 581–587.

Boldyrev, A., Song, R., Lawrence, D., and Carpenter, D. O. (1999b). Carnosine protectsagainst excitotoxic cell death independently of effects on reactive oxygen species. Neuro-

science 94, 571–577.Boldyrev, A., Bulygina, E., Leinsoo, T., Petrshanko, I., Tsebone, S., and Abe, H. (2004a).

Protection of neuronal cells against reactive oxygen species by carnosine and relatedcompounds. Comp. Biochem. Physiol. B Biochem. Mol. Biol. 137, 81–88.

Boldyrev, A. A., Fedorova, T., Stvolinsky, S., Borras, C., Sastre, J., and Vina, J. (2004b).Chemical intervention in senescence-accelatered mice metabolism for modeling neuro-degenerative diseases: An overview. Int. Cong. Ser. 1280, 109–115.

Boldyrev, A. A., Koudinov, A., Berezov, T., and Carpenter, D. O. (2004c). Amyloid-betainduced cell death is independent of free radicals. J. Alzheimers Dis. 6, 633–638.

Boldyrev, A. A., Fedorova, T., Stepanova, M., Dobrolvorskaya, I., Kozlova, E., Boldanova, N.,Bagyeva, G., Ivanova-Smolenskaya, I., and Illarioshkin, S. (2008). Carnosine increasesefficiency of DOPA therapy of Parkinsons’s disease: A pilot study. Rejuvenation Res. 11,821–827.

Bonawitz, N. D., Chatenay-Lapointe, M., Pan, Y., and Shadel, G. S. (2007). Reduced TORsignalling extends chronological life span via increased respiration and upregulatedmitochondrial gene expression. Cell Metab. 5, 265–277.

Bonfanti, L., Peretto, P., deMarchis, S., and Fasolo, A. (1999). Carnosine-related dipeptides inthe mammalian brain. Prog. Neurobiol. 59, 333–353.

136 Alan R. Hipkiss

Bonner, A. B., Swann, M. E., Marway, J. S., Heap, L. C., and Preedy, V. R. (1995). Lysosomaland non-lysosomal protease activities of the brain in response to ethanol feeding. Alcohol12, 505–509.

Bossy-Wetzel, E., Schwarzenbacher, R., and Lipton, S. A. (2004). Moelcular pathways toneurodegeneration. Nat. Med. 10, S2–S9.

Botella, J. A., Ulschmid, J. K., Gruenewald, C., Moehle, C., Kretzschmar, D., Becker, K., andScheuwly, S. (2004). The Drosophila carbonyl reductase sniffer prevents oxidative stress-induced neurodegeneration. Curr. Biol. 14, 782–786.

Brown, C. M., Dulloo, A. G., Yepuri, G., and Montani, J. P. (2008). Fructose ingestion acutelyelevates blood pressure in healthy young humans. Am. J. Physiol. Regul. Integr. Comp.

Physiol. 294, R730–R737.Brownlee, M. (1995). Advanced protein glycation in diabetes and aging. Annu. Rev. Med. 46,

223–234.Brownson, C. and Hipkiss, A. R. (2000). Carnosine reacts with a glycated protein. Free Radic.

Biol. Med. 28, 1564–1570.Bulteau, A. L., Szweda, L. I., and Friguet, B. (2006). Mitochondrial protein oxidation

and degradation in response to oxidative stress and aging. Exp. Gerontol. 41, 653–657.Burcham, P. C. and Pyke, S. M. (2006). Hydralazine inhibits rapid acrolein-induced protein

oligomerization: Role of aldehyde scavengers and adduct trapping in cross-link blockingand cytoprotection. Mol. Pharm. 69, 1056–1065.

Burke, W. J., Li, S. W., Chung, H. D., Ruggiero, D. A., Kristal, B. S., Johnson, E. M., Lampe, P.,Kumar, V. B., Franko, M., Williams, E. A., and Zahm, D. S. (2004). Neurotoxicity of MAOmetabolites of catecholamine neurotransmitters: Role in neurodegenerative disease.Neurotoxicology 25, 101–115.

Burkle, A., Diefenbach, J., Brabeck, C., and Beneke, S. (2005). Ageing and PARP. Pharmacol.

Res. 52, 1599–1601.Bush, A. I. and Tanzi, R. E. (2002). The galvanization of beta-amyloid in Alzheimer’s disease.

Proc. Natl. Acad. Sci. USA 99, 7317–7319.Caccamo, A., Oddo, S., Sugarman, M. C., Akbari, Y., and LaFerla, F. M. (2005). Age and

region-dependent alterations in Abeta-degrading enzymes: Implications for Abeta-induced disorders. Neurobiol. Aging 26, 645–654.

Cacciatore, I., Cocco, A., Costa, M., Fontana, M., Lucente, G., Pecci, L., and Pinnen, F. (2005).Biochemical properties of new synthetic carnosine analogues containing the residue of2,3-diaminoproprionic acid: The effect of N-acetylation. Amino Acids 28, 77–83.

Cai, W., He, J. C., Zhu, L., Chen, X., Zheng, F., Striker, G. E., and Vlassara, H. (2008). Oralglycotoxins determine the effects of calorie restriction on oxidant stress, age-relateddiseases and lifespan. Am. J. Path. 173, 327–336.

Calcagni, A., Ciattini, P. G., Di Stefano, A., Dupre, S., Luisi, G., Pinnen, F., Rossi, D., andSpirito, A. (1999). psi(SO2NH) transition state isosters of peptides. Synthesis and bioac-tivity of sulfonamido pseudopeptides related to carnosine. Farmaco 54, 673–677.

Calabrese, V., Coombrita, C., Guagliano, E., Sapoenza, M., Ravagna, A., Cardile, V.,Scapagnini, G., Santoro, A. M., Mangiameli, A., Butterfield, D. A., GiuffirdaStella, A. M., and Rizzarelli, E. (2005). Protective effect of carnosine during nitrosactivestress in astroglial cell cultures. Neurochem. Res. 30, 797–807.

Calabrese, V., Cornelius, C., Mancuso, C., Pennisi, G., Calafato, S., Bellia, F., Bates, T. E.,Stella, A. M. G., Schapira, T., Kostova, A. T. D., and Rizzarelli, E. (2008). Cellular stressreponse: A novel target for chemoprevention and nutritional neuroprotection in aging,neurodegenerative disorders and longevity. Neurochem. Res. 33, 2444–2471.

Cantero, A. V., Portero-Otin, M., Ayala, V., Auge, N., Sanson, M., Elbaz, M., Thiers, J. C.,Pamplona, R., Salvayre, R., and Negre-Salvayre, A. (2007). Methylglyoxal inducesadvanced glycation end product (ages) formation and dysfunction of PDGF receptor-beta: Implications for diabetic atherosclerosis. FASEB J. 21, 3096–3106.


Carini, M., Aldini, G., Beretta, G., Arlandini, E., and Facino, R.M. (2003). Acrolein-sequesteringability of endogenous dipeptides: Characterization of carnosine and homocarnosine/acrolein adducts by electrospray ionization tandem mass spectrometry. J. Mass Spectrom.

38, 996–1006.Caro, P., Gomez, J., Sanz, A., Portero-Otin, M., Pamplona, R., and Barja, G. (2007). Effect of

graded corticosterone treatment on aging-related markers of oxidative stress in rat livermitochondria. Biogerontology 8, 1–11.

Carrard, G., Bulteau, A. L., Petropoulos, I., and Friguet, B. (2002). Impairment of proteasomestructure and function in aging. Int. J. Biochem. Cell Biol. 34, 1461–1474.

Celotto, A. M., Frank, A. C., Seigle, J. L., and Palladino, M. J. (2006). Drosophila model ofhuman inherited triosephosphate isomerase deficiency glycolytic enzymopathy. Genetics174, 1237–1246.

Chang, T., Wang, R., and Wu, L. (2005). Methylglyoxal-induced nitric oxide and peroxyni-trite production in vascular smooth muscle cells. Free Radic. Biol. Med. 38, 286–293.

Chauhan, A. and Chauhan, V. (2006). Oxidative stress in autism. Pathophysiology 13, 171–181.Chauhan, A., Chauhan, V., Brown, W. T., and Cohen, I. (2004). Oxidative stress in autism:

Increased peroxidationand reduced serum levels of ceruloplasmin and transferring—Theantioxidant proteins. Life Sci. 75, 2539–2549.

Chez, M. G., Buchanan, C. P., and Aimonovitch, M. C. (2002). Double-blind, placebo-controlled study of L-carnosine supplementation in children with autistic spectrumdisorders. J. Child Neurol. 17, 833–837.

Chiocchetti, A., Zhou, J., Zhu, H., Karl, T., Haubenreisser, O., Rinnerthaler, M., Heeren, G.,Oender, K., Bauer, J., Hintner, H., Breitenbach, M., and Breitenbach-Koller, L. (2007).Ribosomal proteins Rp110 and Rps6 are potent regulators of yeast replicative life span.Exp. Gerontol. 42, 275–286.

Choi, H. K. and Curhan, G. T. (2008). Soft drinks, fructose consumption, and the risk of goutin men: Prospective cohort study. BMJ 336, 309–312.

Chondrogianni, N. and Gonos, E. S. (2007). Overexpression of hump11/POMP proteasomeaccessory protein enhances proteasome-mediated antioxidant defence. Exp. Gerontol. 42,899–903.

Chung, C. H. and Hu, M. L. (2008). L-Carnosine inhibits metastasis of SK-Hep-1 cells byinhibition of matrix metalloproteinase-9 expression and induction of an antimetastaticgene, nm23-H1. Nutr. Cancer 60, 526–533.

Clark, I. E., Dodson, M. W., Jiang, C., Cao, J. H., Huh, J. R., Seol, J. H., Yoo, S. J., Hay, B. A.,and Guo, M. (2006).Drosophila pink1 is required for mitochondrial function and interactsgenetically with parkin. Nature 441, 1162–1166.

Cornet, M. and Bousset, J. (1999). Free amino acids and dipeptides in porcine muscles:Differences between ‘‘red’’ and ‘‘white’’ muscles. Meat Sci. 51, 215–219.

Cristofalo, V. J. and Kabakjian, J. (1975). Lysosomal enzymes and aging in vitro: Subcellularenzyme distribution and effect of hydrocortisone on cell lifespan. Mech. Ageing Dev. 4,19–28.

Cunningham, J. T., Rodgers, J. T., Arlow, D. H., Vazquez, F., Mootha, V. K., andPuigserver, P. (2007). mTOR controls mitochondrial oxidative function through a YY1-PGC-1 alpha transcription complex. Nature 450, 736–740.

Dalfo, E. and Ferrer, I. (2008). Early a-synuclein lipoxidation in neocortex in Lewy bodydiseases. Neurobiol. Aging 29, 408–417.

Dalle-Donne, I., Giustarini, D., Clombo, R., Rossi, R., and Milzani, A. (2003). Protein carbon-ylation in human diseases. Trends Mol. Med. 9, 169–176.

Dahlmann, B. (2007). Role of proteasomes in disease. BMC Biochem. 8, S3–S15.Danscher, G., Jensen, K. B., Fredericksonm, C. J., Kemp, K., Andreason, A., Juhl, S.,

Stoltenberg, M., and Ravid, R. (1997). Increased amount of zinc in the hippocampus

138 Alan R. Hipkiss

and amygdale of Alzheimer’s disease brains: A proton-induced X-ray emission spectro-scopic analysis of cryostat sections from autopsy material. J. Neurosci. Methods 76, 53–59.

Davis, T., Baird, D. M., Haughton, M. F., Jones, C. J., and Kipling, D. (2005). Prevention ofaccelerated cell aging in Werner syndrome using a p38 mitogen-activated protein kinaseinhibitor. J. Gerontol. A Biol. Sci. Med. Sci. 60, 1386–1393.

Decker, E. A., Ivanov, V., Zhu, B. Z., and Frei, B. (2001). Inhibition of low-density lipoproteinoxidation by carnosine and histidine. J. Agric. Food Chem. 49, 511–516.

Derave, W., Jones, G., Hespel, P., and Harris, R. C. (2008). Creatine supplementation aug-ments skeletal muscle carnosine content in senescence-accelerated mice (SAMP8). Reju-venation Res. 11, 641–647.

Desai, K. M. and Wu, L. (2008). Free radical generation by methylglyoxal in tissues. Drug

Metabol. Drug Interact. 23, 151–173.Dhar, A., Desai, K., Kasachmov, M., Yu, P., and Wu, L. (2008). Methylglyoxal production in

vascular smooth muscle cells from different metabolic precursors. Metabolism 57,1211–1220.

Ding, Q. and Keller, J. N. (2001). Proteasome inhibition in oxidative stress neurotoxicity:Implications for heat shock proteins. J. Neurochem. 77, 1010–1017.

Dobrota, D., Fedorova, T., Stvolinsky, S. L., Babusikova, E., Likavcanova, K., Drgova, A.,Strapkova, A., and Boldyrev, A. A. (2005). Carnosine protects the brain of rats and mongo-lian gerbils against ischemic injury: After-stroke-effect.Neurochem. Res. 30, 1283–1288.

Donati, A. (2006). The involvement of macroautophagy in aging and anti-aging interven-tions. Mol. Aspects Med. 27, 455–470.

Dukan, S., Farewell, A., Ballesteros, M., Taddei, F., and Nystrom, T. (2000). Proteins areoxidatively carbonylated in response to reduced transcriptional and translation. Proc.Natl. Acad. Sci. USA 97, 5746–5749.

Dunnet, M. and Harris, R. C. (1997). High-performance liquid chromatographic determina-tion of imidazole dipeptides, histidine, 1-methylhistidine and 3-methylhistidine inequine and camel muscles and individual fibres. J. Chromatogr. B 688, 47–55.

Evason, K., Huang, C., Yamben, I., Covey, D. F., and Kornfield, K. (2005). Anticonvulsantmedications extend worm life-span. Science 307, 258–262.

Farris, W., Mansourian, S., Chang, Y., Linsley, L., Eckman, E. A., Frosch, M. P., Eckman, C. B.,Tanzi, R. E., Selkoe, D. J., and Guennette, S. (2005). Insulin-degrading enzyme regulatesthe levels of insulin, amyloid beta-protein, and the beta-amyloid precursor proteinintracellular domain in vivo. Proc. Natl. Acad. Sci. USA 100, 4162–4167.

Fitzpatrick, D., Amend, J. F., Squibb, R. L., and Fisher, H. (1980). Effects of chronic and acuteinfections on tissue levels of carnosine, anserine and free histidine in rats and chickens.Proc. Soc. Exp. Biol. Med. 165, 404–408.

Fontana, M., Pinnen, F., Lucente, G., and Pecci, L. (2002). Prevention of peroxynitrate-dependent damage by carnosine and related sulphonamide pseudopeptides. Cell. Mol.Life Sci. 59, 546–555.

Fonteh, A. N., Harrington, R. J., Tsai, A., Liao, P., and Harrington, M. G. (2007). Free aminoacid and dipeptide changes in the body fluids of Alzheimer’s disease subjects. Amino

Acids 32, 213–224.Formazyuk, V. E., Gorshkova, T. Y., Boldyrev, A. A., and Sergienko, V. T. (1992). Character-

istics of the chloramines complexes of carnosine and hypochlorite. Biochemistry (Moscow)

57, 915–918.Fornai, F., Giorgi, F. S., Bassi, L., Ferrucci, M., Alessandri, M. G., and Corsini, G. U. (2000).

Modulation of dihydrophenylacetaldehyde extracellular levels in vivo in the rat striatumafter different kinds of pharmacological treatment. Brain Res. 861, 126–134.

Fouad, A. A., El-Rehany, M. A., and Maghraby, H. K. (2007). The hepatoprotective effectof carnosine against ischemia/reperfusion liver injury in rats. Eur. J. Pharmacol. 572,61–68.


Franke, S., Siggelkow, H., Wolf, G., and Hein, G. (2007). Advanced glycation endproductsunfluence the mRNA expression of RAGE, RANKL and various osteoblastic genes inhuman osteoblasts. Arch. Physiol. Biochem. 113, 154–161.

Fredriksson, A., Ballasteros, M., Dukan, S., and Nystrom, T. (2006). Induction of the heatshock regulaon in response to increasedmistranslation requires oxidative modification ofthe malformed proteins. Mol. Microbiol. 59, 350–359.

Freedman, B. I., Hicks, P. J., Sale, M. M., Pierson, E. D., Langefeld, C. D., Rich, S. S.,McDonough, C., Janssen, B., Yard, B. A., van der Woude, F. J., and Bowden, D. W.(2007). A leucine repeat in the carnosinase gene CNDP1 is accociated with diabetic end-stage renal disease in European Americans. Nephrol. Dial. Transplant. 22, 1131–1135.

Fu, Q., Dai, H., Hu, W., Fan, Y., Shen, Y., Zhang, W., and Chen, Z. (2008). Carnosine protectsagainst AB42-induced neurotoxicity in differentiated rat PC12 cells. Cell. Mol. Neurobiol.

28, 307–316.Fujii, T., Takaoka, M., Muraoka, T., Kurata, H., Tsuruoka, N., Ono, H., Kiso, Y., Tanaka, T.,

and Matsumura, Y. (2003). Preventive effect of L-carnosine on ischemia/reperfusion-induced acute renal failure in rats. Eur. J. Pharmacol. 474, 261–267.

Fujii, T., Takaoka, M., Tsuruoka, N., Kiso, Y., and Matsumura, Y. (2005). Dietary supplemen-tion with L-carnosine prevents ischemia/reperfusion renal injury in rats. Biol. Pharm.

Bull. 28, 361–363.Fujimoto, M., Uchida, S., Watanuki, T., Wakabayashi, Y., Otsuki, K., Matsugara, T.,

Suetsugi, M., Funato, H., and Watanabe, Y. (2008). Reduced expression of glyoxalase-1mRNA in mood disorder patients. Neurosci. Lett. 438, 196–199.

Gallant, S., Kukley, M., Stvolinsky, S., Bulgina, E., and Boldyrev, A. A. (2000). Effect ofcarnosine on rats under experimental ischemia. Tohoku J. Exp. Med. 191, 85–99.

Gao, X., Qi, L., Qiao, N., Choi, H. K., Curham, G., Tucker, K. L., and Ascherio, A. (2008).Intake of added sugar and sugar-sweetened drink and serum uric acid concentration inUS men and women. Hypertension 50, 306–313.

Gardner, M. L. G., Illingworth, K. M., Kelleher, J., andWood, D. (1991). Intestinal absorptionof the intact peptide carnosine in man, and comparison with intestinal permeability tolactulose. J. Physiol. 439, 411–422.

Gayova, E., Kron, I., Suchozova, K., Pavlisak, V., Fedurco, M., and Novakova, I. (1999).Carnosine in patients with type 1 diabetes mellitus. Bratisl. Lek. Lesty 100, 500–502.

Ghanbari, H. A., Ghanbari, K., Harris, P. L. R., Jones, P. K., Kubat, Z., Castellani, R. J.,Wolozin, B. L., Smith, M. A., and Perry, G. (2004). Oxidative damage in culturedhuman olfactory neurons from Alzheimer’s disease patients. Aging Cell 3, 41–44.

Gjessing, L. R., Lunde, H. A., Morkrid, L., Lenney, J. F., and Sjaastad, O. (1990). Inborn errors ofcarnosine and homocarnosine metabolism. J. Neural Transm. Suppl. 29, 91–106.

Gnerer, J. P., Kreber, R. A., and Ganetsky, B. (2006). Wasted away, a drosophila mutation intriosephosphate isomerase, causes paralysis, neurodegeneration, and early death. Proc.Natl. Acad. Sci. USA 103, 14987–14993.

Goh, S. Y. and Cooper, M. E. (2008). Review: The role of advanced glycation end productsin progression and complications of diabetes. J. Clin. Endocrinol. Metab. 93, 1143–1152.

Gomes, R. A., Oliveira, L. M. A., Silva, M., Ascenso, C., Quintas, A., Costa, G., Coelho, A. V.,Silva, M. S., Ferreira, A. E. N., Freire, P., and Cordeiro, C. (2008). Protein glycation in vivo:Functional and structural effects on yeast enolase. Biochem. J. 416, 317–326.

Gomez,A. and Ferrer, I. (2009). Increased oxidationof certain glycolytic and energymetabolismenzymes in the frontal cortex in Lewy body diseases. J. Neurosci. Res. 87, 1002–1013.

Gopko, A. V., Kulibib, A. Y., Semenova, M. L., and Zakhidov, S. T. (2005). Carnosine-inducedchanges in the development of spermatogenic epithelial cells in senescence acceleratedSAM mice. Bull. Exp. Biol. Med. 140, 235–237.

140 Alan R. Hipkiss

Goodrick, C. L., Ingram, D. K., Reynolds, M. A., Freeman, J. R., and Cider, N. (1990). Effect ofintermittent feeding upon body weight and lifespan in inbred mice: Interaction ofgenotype and age. Mech. Ageing Dev. 55, 69–87.

Guarente, L. (2000). Sir2 links chromatin silencing, metabolism, and aging. Genes Dev. 14,1021–1026.

Guarente, L. (2008). Mitochondria – a nexus of aging, calorie restriction and sirtuins? Cell 132,171–176.

Gugliucci, A., Smith, M. E., and Menini, T. (2002). Carnosine in physiological concentrationsinhibits advanced glycation of histones by reactive trioses methylglyoxal and glyceralde-hyde. Diabetes 51, 2111.

Guiotto, A., Calderan, A., Ruzzo, P., and Borin, G. (2005a). Carnosine and carnosine-relatedantioxidants: A review. Curr. Med. Chem. 12, 2293–2315.

Guiotto, A., Calderan, A., Ruzzo, P., Osler, A., Rubini, C., Jo, D. G., Mattson, M. P., andBorin, G. (2005b). Synthesis and evaluation of neuroprotective a,b-unsaturated aldehydescavenger histidyl-containing analogues of carnosine. J. Med. Chem. 48, 6156–6161.

Hamano, H., Yoshinaga, K., Eta, R., Emori, Y., Kawasaki, D., Ilno, Y., Sawada, M.,Kuroda, H., and Takei, M. (2006). Effect of polaprezinc on taste disorders in zinc-deficientrats. Biofactors 28, 185–193.

Han, Y., Randell, E., Vasdev, S., Gill, A., Gadag, V., Newhook, L. A., Grant, M., andHagerty, D. (2007). Plasma methylglyoxal and glyoxal are elevated and related to earlymembrane alteration in young, complication-free patients with Type 1 diabetes.Mol. Cell.

Biochem. 305, 123–131.Hansen, M., Taubert, T., Crawford, D., Libina, N., Lee, S. J., and Kenyon, C. (2007). Lifespan

extension by conditions that inhibit translation in Caenorhabditis elegans. Aging Cell 6,95–110.

Hansen, M., Chandos, A., Mitic, L. L., Onken, B., Driscoll, M., and Kenyon, C. (2008). A rolefor autophagy in the extension of lifespan by dietary restriction in C. Elegans. PLoS Genet.

4, e24.Harman, D. (1956). Aging: A theory based on free radical and radiation chemistry. J. Gerontol.

2, 298–300.Harris, R. C., Marlin, D. J., Dunnett, M., Snow, D. H., and Hultman, E. (1990). Muscle

buffering capacity and dipeptide content in the thoroughbred horse, greyhound dogand man. Comp. Biochem. Physiol. 97A, 249–251.

Harris, R. C., Tallon, M. J., Dunnett, M., Boobis, L., Coakley, J., Kim, H. J., Fallowfield, J. L.,Hill, C. A., and Wise, J. A. (2006). The absorption of orally supplied beta-alanine andits effects on muscle carnosine synthesis in human vastus lateralis. Amino Acids 30,279–289.

Harris, R. C., Jones, G., Hill, C. H., Kendrick, I. P., Boobis, L., Kim, C. K., Kim, H. J.,Dang, V. H., Edge, J., and Wise, J. A. (2007). The carnosine content of vastus lateralis invegetarians and omnivores. FASEB J. 21, 76.

Hass, R. (2005). Retrodifferentiation and reversibility of aging: Forever young? Signal

Transduct. 5, 93–102.Hein, G. E. (2006). Glycation endproducts in osteoporosis—Is there a pathophysiologic

importance? Clin. Chim. Acta 371, 32–36.Herrera, F. E., Chesi, A., Paleologou, K. E., Schmid, A., Munoz, A., Vendruscolo, M.,

Gustincich, S., Lashuel, H. A., and Carloni, P. (2008). Inhibition of a-synuclein fibrilliza-tion by dopamine is mediated by interactions with five C-terminal residues and with E83in the NAC region. PLoS ONE 3, e3394.

Hill, C. A., Harris, R. C., Kim, H. J., Harris, B. D., Sale, C., Boobis, L. H., Kim, C. K., andWise, J. A. (2006). Influence of beta-alanine supplementation on skeletal muscle carnosineconcentrations and high intensity cycling capacity. Amino Acids 32, 225–233.


Hipkiss, A. R. (1998). Carnosine, a protective anti-ageing peptide? Int. J. Biochem. Cell Biol. 30,863–868.

Hipkiss, A. R. (2001). On the anti-aging activities of aminoguanidine and N-t-butylhydrox-ylamine. Mech. Ageing Dev. 122, 169–171.

Hipkiss, A. R. (2003). Do developmental-related changes in constitutive proteolysis affectaberrant protein accumulation and generation of the aged phenotype? Mech. Ageing Dev.124, 575–579.

Hipkiss, A. R. (2005). Glycation, ageing and carnosine: Are carnivorous diets beneficial?Mech. Ageing Dev. 126, 1034–1039.

Hipkiss, A. R. (2006a). Accumulation of altered proteins and ageing: Causes and effects. Exp.Gerontol. 41, 464–473.

Hipkiss, A. R. (2006b). On the mechanisms of ageing suppression by dietary restriction—Ispersistent glycolysis the problem? Mech. Ageing Dev. 127, 8–15.

Hipkiss, A. R. (2006c). Would carnosine or a carnivorous diet help suppress aging andassociated pathologies? Ann. N. Y. Acad. Sci. 1067, 369–374.

Hipkiss, A. R. (2007a). Could carnosine or related structures suppress Alzheimer’s disease?J. Alzheimers Dis. 11, 229–240.

Hipkiss, A. R. (2007b). Dietary restriction, glycolysis, hormesis and ageing. Biogerontology8, 221–224.

Hipkiss, A. R. (2007c). On why decreasing protein synthesis can increase lifespan. Mech.Ageing Dev. 128, 412–414.

Hipkiss, A. R. (2008a). Energy metabolism, altered proteins, sirtuins and ageing: Convergingmechanisms? Biogerontology 9, 49–55.

Hipkiss, A. R. (2008b). Error-protein metabolism and ageing. Biogerontology (in press).Hipkiss, A. R. (2008c). On methionine restriction, suppression and mitochondrial dysfunc-

tion and aging. Rejuvenation Res. 11, 685–688.Hipkiss, A. R. and Brownson, C. (2000). Carnosine reacts with protein carbonyl groups:

Another possible role for the anti-ageing peptide? Biogerontology 1, 217–223.Hipkiss, A. R. and Chana, H. (1998). Carnosine protects proteins against methylglyoxal-

mediated modifications. Biochem. Biophys. Res. Commun. 248, 28–32.Hipkiss, A. R., Michaelis, J., and Syrris, P. (1995a). Non-enzymic glycosylation of the dipep-

tide L-carnosine, a potential anti-protein-cross-linking agent. FEBS Lett. 371, 81–85.Hipkiss, A. R., Michaelis, J., Syrris, P., and Dreimanis, M. (1995b). Strategies for the extension

of human lifespan. Perspect. Hum. Biol. 1, 59–70.Hipkiss, A. R., Preston, J. E., Himsworth, D. T., Worthington, V. C., and Abbott, N. J. (1997).

Protective effects of carnosine against malondialdehyde-induced toxicity towardscultured rat brain endothelial cells. Neurosci. Lett. 238, 135–138.

Hipkiss, A. R., Worthington, V. C., Himsworth, D. T. J., and Herwig, W. (1998a). Protectiveeffects of carnosine against protein modification mediated by malondialdehyde andhypochlorite. Biochim. Biophys. Acta 1380, 46–54.

Hipkiss, A. R., Preston, J. E., Himsworth, D. T. M., Worthington, V. C., Keown, M.,Michaelis, J., Lawrence, J., Mateen, A., Allende, L., Eagles, P. A. M., and Abbott, N. J.(1998b). Pluripotent protective effects of carnosine, a naturally-occurring dipeptide. Ann.N. Y. Acad. Sci. 854, 37–53.

Hipkiss, A. R., Brownson, C., and Carrier, M. J. (2001). Carnosine, the anti-ageing, anti-oxidant dipeptide, may react with protein carbonyl groups. Mech. Ageing Dev. 122,1431–1445.

Hipkiss, A. R., Brownson, C., Bertani, M. F., Ruiz, E., and Ferro, A. (2002). Reaction ofcarnosine with aged proteins. Another protective process. Ann. N. Y. Acad. Sci. 959,285–294.

Holliday, R. and McFarland, G. A. (1996). Inhibition of the growth of transformed andneoplastic cells by the dipepitde carnosine. Br. J. Cancer 73, 966–971.

142 Alan R. Hipkiss

Holliday, R. and McFarland, G. A. (2000). A role for carnosine in cellular maintenance.Biochemistry (Moscow) 65, 843–848.

Hornung, M. S., Blakesmore, L. J., and Trombley, P. Q. (2000). Endogenous mechanisms ofneuroprotection: Role of zinc, copper and carnosine. Brain Res. 852, 56–61.

Hou, W. C., Chen, H. J., and Lin, Y. H. (2003). Antioxidant peptides with angiotensinconverting enzyme inhibitory activities and applications for angiotensin convertingenzyme purification. J. Agric. Food Chem. 51, 1706–1709.

Huang, Y., Duan, J., Chen, H., Chen, M., and Chen, G. (2005). Separation and determinationof carnosine-related peptides using capillary electrophoresis with laser-induced fluores-cence detection. Electrophoresis 26, 593–599.

Huebscher, K. J., Lee, J., Rovelli, G., Ludin, B., Matus, A., Stauffer, D., and Furst, P. (1999).Protein isoaspartyl methyltransferase protects from Bax-induced apoptosis. Gene 240,333–341.

Hughes, S. E., Evason, K., Xiong, C., and Kornfeld, K. (2007). Genetic and pharmacologicalfactors that influence reproductive aging in nematodes. PLoS Genet. 3, 254–265.

Ikeda, D., Wada, C., Yoneda, C., Abe, H., and Watabe, S. (1999). Carnosine stimulatesvimentin expression in cultured rat fibroblasts. Cell Struct. Funct. 24, 79–87.

Ikeda, M., Aiba, T., Ikui, A., Kurono, Y., Sakagami, M., Takeda, N., and Tomita, H. (2005).Taste disorders: A survey of the examination methods and treatments used in Japan.ActaOtolaryngol. 125, 1203–1210.

Janssen, B., Hohenadel, D., Brinkkoeter, P., Peters, V., Rind, N., Fischer, C., Rychlik, I.,Cerna, M., Rondeau, E., Mathieson, P., Saleem, M. A., Meyer, J., et al. (2005). Carnosineas a protective factor in diabetic nephropathy: Association with a leucine repeat of thecarnosinase gene CNDP1. Diabetes 54, 2320–2327.

Jansen, E. E. W., Gibson, K. M., Shigematsu, Y., Jakobs, C., and Verhoeven, N. M. (2006).A novel, quantitative assay for homocarnosine in cerebrospinal fluid using stable-isotopedilution liquid chromatography-tandem mass spectrometry. J. Chromatogr. B Analyt.

Technol. Biomed. Life Sci. 830, 196–200.Jia, X. and Wu, L. (2007). Accumulation of endogenous methylglyoxal impaired insulin

signalling in adipose tissue of fructose-fed rats. Mol. Cell. Biochem. 306, 133–139.Jin, C. L., Yang, L. X., Wu, X. H., Li, Q., Ding, M. P., Fan, Y. Y., Zhang, W. P., Luo, J. H., and

Chen, Z. (2005). Effects of carnosine on amygdaloid-kindled seizures in Sprague–Dawleyrats. Neuroscience 135, 939–947.

Johnson, P. and Hammer, J. L. (1992). Histidine dipeptide levels in aging and hypertensiverat skeletal and cardiacmuscles.Comp. Biochem. Physiol. B Biochem.Mol. Biol. 103, 981–984.

Jones, G. A., Harris, R. C., and Nixey, C. (2007). Distribution of the histidine-containingdipeptides, anserine and carnosine, within turkey muscle. In ‘‘10th InternationalCongress on Amino Acids and Proteins (ICAAP)’’, Kallithea, Greece, Vol. 33, pLIIabstract.

Junaid,M. A., Kowal, D., Barua,M., Pullartkat, P. S., Brooks, S. S., and Pullartkat, R. K. (2004).Proteomic studies identified a sungle nucleotide polymorphism in glyoxalase 1 as autismsusceptibility factor. Am. J. Med. Genet. 131A, 11–17.

Junn, E., Ronchetti, R. D., Quezado, M. M., Kim, S. Y., and Mouradian, M. M. (2003). Tissuetransglutaminase-induced aggregation of alpha-synuclein: Implications for Lewy bodyformation in Parkinson’s disease and dementia with Lewy bodies. Proc. Natl. Acad. Sci.

USA 100, 2047–2052.Kaeberlein, M. and Kennedy, B. K. (2007). Protein translation, 2007. Aging Cell 6,

731–734.Kairaluoma, L., Narhl, V., Ahonen, T., Westerholm, J., and Aro, M. (2008). Do fatty acids help

in overcoming reading difficultires? A double-blind, placebo-controlled study of theeffects of eicosapentaenoic acid and carnosine supplementation on children with dys-lexia. Child Care Health Dev. (in press).


Kalapos, M. P. (1999). Methylglyoxal in living organisms: Chemistry, biochemistry, toxicol-ogy and biological implications. Toxicol. Lett. 110, 145–175.

Kamel, J., Ohsawa,M., Miyata, S., and Tanaka, S. I. (2008). Preventive effect of L-carnosine onchanges in the thermal nociceptive threshold in streptozotocin-induced diabetic mice.Eur. J. Pharmacol. 600, 83–86.

Kang, J. H. and Kim, K. S. (2003). Enhanced oligomerization of alpha-synuclein mutant byCu,Zn-superoxide dismutase and hydrogen peroxide system. Mol. Cells 28, 87–93.

Kang, J. H., Kim, K. S., Choi, S. Y., Kwon, H. Y., Won, M. H., and Kang, T. C. (2002).Protective effects of carnosine, homocarnosine and anserine against peroxylradical-mediated Cu,Zn-superoxide dismutase modification. Biochim. Biophys. Acta

1570, 89–96.Karpuj, M. and Steinman, L. (2004). The multifaceted role of transglutaminase in neurode-

generation: Review article. Amino Acids 26, 373–379.Kawahara,M., Konoha, K., Nagata, T., and Sadakane, Y. (2007). Protective substances against

zinc-induced neuronal death after ischemia: Carnosine as a target for drug vascular typedementia. Recent Pat. CNS Drug Discov. 2, 145–149.

Kendrick, I. P., Harris, R. C., Kim, H. J., Kim, C. K., Dang, V. H., Lam, T. Q., Bui, T. T.,Smith, M., and Wise, J. A. (2008). The effects of 10 weeks of resistance training combinedwith beta-alanine supplementation on whole body strength, force production, muscleendurance and body composition. Amino Acids 34, 547–554.

Kim, N. H. and Kang, J. H. (2007). Protective effects of histidine dipeptides on the modifica-tion of neurofilament-L by the cytochrome c/hydrogen peroxide system. J. Biochem. Mol.

Biol. 40, 125–129.Kim, K. S., Choi, S. Y., Kwon, H. Y., Won, M. H., Kang, T. C., and Kang, J. H. (2002). The

ceruplasmin and hydrogen peroxide system induces alpha-synuclein aggregationin vitro. Biochimie 84, 625–631.

Kim, N. H., Jeong, M. S., Choi, S. Y., and Kang, J. H. (2004). Oxidative modification ofneurofilament-L by the Cu,Zn-superoxide dismutase and hydrogen peroxide system.Biochimie 86, 553–559.

Kim, H. J., Cho, J., Kim, C. K., Harris, R. C., Harris, D. B., Sale, C., andWise, J. A. (2005). Effectof muscle fibre morphology and carnosine content after 12 days training of Korean speedskaters. Med. Sci. Sports Exerc. 37, S192.

Kishi, S., Segawa, Y., and Yamaguchi, M. (1994). Histomorphological confirmation of thepreventative effect of beta-alanyl-L-histidinato zinc on bone loss in ovariectomized rats.Biol. Pharm. Bull. 63, 203–207.

Kletsas, D., Pratsinis, H., Gioni, V., Pilichos, K., Yiacoumettis, A. M., and Tsagarakis, S.(2007). Prior chronic in vivo glucocorticoid excess leads to an anabolic phenotype andan extension of cellular life span of skin fibroblasts in vitro. Ann. N. Y. Acad. Sci. 1100,449–454.

Kohen, R., Yamamoto, Y., Cundy, K. C., and Ames, B. N. (1988). Antioxidant activity ofcarnosine, homocarnosine and anserine present in muscle and brain. Proc. Natl. Acad. Sci.

USA 95, 2175–2179.Kotamraju, S., Matalon, S., Matsunaga, T., Shang, T., Hickman-Davis, J. M., and

Kalyanaraman, B. (2006). Upregulation of immunoproteasomes by nitric oxide: Potentialantioxidative mechanism in endothelial cells. Free Radic. Biol. Med. 40, 1034–1044.

Kovaks, T. (2004). Mechanisms of olfactory dysfunction in aging and neurodegenerativedisorders. Ageing Res. Rev. 3, 215–232.

Kozan, R., Sefil, F., and Bagirici, R. (2008). Anticonvulsant effect of carnosine on penicillin-induced epileptiform activity in rats. Brain Res. (in press).

Krajcovicova-Kudlackova, M., Sebekova, K., Schinzel, R., and Klvanova, J. (2002). Advancedglycation end products and nutrition. Physiol. Res. 51, 313–316.

144 Alan R. Hipkiss

Kueper, T., Grune, T., Prahl, S., Lenz, H., Welge, V., Biernoth, T., Vogt, Y., Muhr, G. M.,Gaemlich, A., Jung, T., Boemke, G., Elsasser, H. P., et al. (2007). Vimentin is the specifictarget in skin glycation. Structural prerequisites, functional consequences, and role inskin aging. J. Biol. Chem. 282, 23427–23436.

Kurata, H., Fujii, T., Tsutsui, H., Katayama, T., Ohkita, M., Takaoka, M., Tsuruoka, N.,Kiso, Y., Ohno, Y., Fujisawa, Y., Shokoji, T., Nishiyama, A., et al. (2006). Renoprotectiveeffects of L-carnosine on ischemia/reperfusion-induced renal injury in rats. J. Pharmacol.

Exp. Ther. 319, 640–647.Kuroda, M. and Harada, T. (2002). Distribution of gamma-glutamyl-beta-alanylhistidine

isopeptide in the macromolecular fractions of commercial meat extracts and correlationwith the color of the macromolecular fractions. J. Agric. Food Chem. 50, 2088–2093.

Kuroda, M., Ohtake, R., Suzuki, E., and Harada, T. (2000). Investigation on the formation andthe determination of gamma-glutamyl-beta-alanylhistidine and related isopeptide in themacromolecula fraction of beef soup. J. Agric. Food Chem. 48, 6317–6324.

Lamarre, M. and Desrosiers, R. R. (2008). Up-regulation of protein L-isoaspartyl methyl-transferase expression by lithium is mediated by glycogen synthase kinase-3 inactivationand beta-catenin stabilization. Neuropharmacology 55, 669–676.

Lamas, I., Anadon, R., and Diaz-Requeira, S. (2007). Carnosine-like immunoreactivity inneurons of the brain of an advanced teleost, the gray mullet (Chelon labrosus, Risso). BrainRes. 1149, 87–100.

Lankin, V. Z., Tikhaze, A. K., Kapel’ko, V. I., Shepel’kova, G. S., Shumaev, K. B.,Panasenko, O. M., Konovalova, G. G., and Belenkov, Y. N. (2007). Mechanisms of oxida-tive modification of low density lipoproteins under conditions of oxidative and carbonylstress. Biochemistry (Moscow) 72, 1081–1090.

Lanthier, J. and Desrosiers, R. R. (2006). Regulation of protein L-isoaspartyl methyltrans-fearse by cell and matrix interactions: Involvement of integrin PI 3-kinase and theproteasome. Biochem. Cell Biol. 84, 684–694.

Lanthier, J., Bouthillier, A., Lapointe, M., Demeule, M., Beliveau, R., and Desrosiers, R. R.(2002). Down-regulation of protein L-isoaspartyl methyltransferase in human epileptichippocampus contributes to generation of damaged tubulin. J. Neurochem. 83, 581–591.

Lapointe, M., Lanthier, J., Moumdjian, R., Regina, A., and Desrosiers, R. R. (2005). Expressionand activity of L-isoaspartyl methyltransferase decrease in stage progression of humanastrocyte tumors. Mol. Brain Res. 135, 93–103.

Le, K. A. and Tappy, L. (2006). Metabolic effects of fructose. Curr. Opin. Clin. Nutr. Metab.

Care 9, 469–475.Lee, J. W., Miyawski, H., Bobst, E. V., Hester, J. D., Ashraf, M., and Bobst, A. M. (1999).

Improved functional recovery of ischemic rat hearts due to singlet oxygen scavengershistidine and carnosine. J. Mol. Cell. Cardiol. 31, 13–21.

Lee, Y., Hsu, C., Lin, K., and Yin, M. (2005). Histidine and carnosine delay diabetic deterio-ration in mice and protect human low density lipoprotein against oxidation and glyca-tion. Eur. J. Pharmacol. 512, 145–150.

Lee, E. S., Chen, H., Hardman, C., Simm, A., and Charlton, C. (2008). Excessive S-adenosyl-l-methionine-dependent methylation increases levels of methanol, formaldehyde and for-mic acid in rat brain striatal homogenates: Possible role in S-adenosyl-l-methionone-induced Parkinsons’s disease-like disorders. Life Sci. 83, 821–827.

Leidhold, C. and Voos,W. (2007). Chaperones and proteases—Guardians of protein integrityin eukaryotic organelles. Ann. N. Y. Acad. Sci. 1113, 72–86.

Levine, R. L. (2002). Carbonyl modified proteins in cellular regulation, aging and disease.Free Radic. Biol. Med. 32, 790–796.

Li, P., Kim, S. W., Datta, S., Pond, W. G., and Wu, G. (2008). Dietary supplementation withcholesterol and docosahexanoic acid affects concentrations of amino acids in tissues ofyoung pigs. Nitric Oxide 19, 259–265.


Liao, P. C., Lin, H. Y., Yuh, C. H., Yu, L. K., and Wang, H. D. (2008). The effect of neuronalexpression of heat shock proteins 26 and 27 on lifespan, neurodegeneration and apoptosisin Drosophila. Biochem. Biophys. Res. Commun. 376, 637–641.

Lin, S. J., Defossez, P. A., andGuarente, L. (2000). Requirement of NAD and SIR2 for life-spanextension by caloric restriction in Saccharomyces cerevisiae. Science 289, 2126–2128.

Liu, Y., Xu, G., and Sayre, L. M. (2003). Carnosine inhibits (E)-4-hydroxy-2-nonenal-inducedprotein cross-linking: Structural characterization of the carnosine-HNE adducts. Chem.

Res. Toxicol. 16, 1589–1597.Liu, W. H., Liu, T. C., and Yin, M. C. (2008a). Beneficial effects of histidine and carnosine on

ethanol-induced chronic liver injury. Food Chem. Toxicol. 46, 1503–1509.Liu, X., Yamada, N., Maruyama, W., and Osawa, Y. (2008b). Formation of dopamine adducts

derived from brain polyunsaturated fatty acids: Mechanism for Parkinson’s disease.J. Biol. Chem. 283, 34887–34895.

Lo, C. Y., Li, S., Wang, Y., Tan, Y., Pan, M. H., Sang, S., and Ho, C. T. (2008). Reactivedicarbonyl compounds and 5-(hydroxymethyl)-2-furfural in carbonated beverages con-taining high fructose corn syrup. Food Chem. 107, 1099–1105.

Lopez-Lluch, G., Irusta, P. M., Navas, P., and de Cabo, R. (2008). Mitochondrial biogenesisand healthy aging. Exp. Gerontol. 43, 813–819.

Luth, H. J., Ogunlade, V., Kuhla, B., Kientsch-Engel, R., Stahl, P., Webster, J., Arendt, T., andMunch, G. (2005). Age- and stage-dependent accumulation of advanced glycation endproducts in intracellular deposits in normal and Alzheimer’s disease brains. Cereb. Cortex15, 211–220.

Macieira-Coelho, A. (1966). Action of cortisone on human fibroblasts in vitro. Experientia 22,390–391.

Magalhaes, P. M., Appel, H. J., and Duarte, J. A. (2008). Involvement of advanced glycationend products in the pathogenesis of diabetic complications: The protective role of regularphysical activity. Eur. Rev. Aging Phys. Act. 5, 17–29.

Mager, D. E., Wan, R., Brown, M., Cheng, A., Wareski, P., Abernethy, D. R., andMattson, M. P. (2006). Caloric restriction and intermittent fasting alter spectral measuresof heart rate and blood pressure variability in rats. FASEB J. 20, 631–637.

Mahmood, A., Fitzgerald, A. J., Marchbank, T., Ntatsaki, E., Murray, D., Ghosh, S., andPlayford, R. J. (2007). Zinc carnosine, a health food supplement that stabilises small bowelintegrity and stimulates gut repair processes. Gut 56, 168–175.

de Marchis, S., Modena, C., Peretto, P., Migheli, A., Margolis, F. L., and Fasalo, A. (2000).Carnosine-related dipeptides in neurons and glia. Biochemistry (Moscow) 65, 824–833.

Martinez-Vicente, M. and Cuevo, A. M. (2007). Autophagy and neurodegeneration: Whenthe cleaning crew goes on strike. Lancet Neurol. 6, 352–362.

Masternak, M. M., Al-Regaiey, K. A., Bonkowski, M. S., Panici, J. A., and Bartke, A. (2005).Effect of every other day feeding diet on gene expression in normal and in long-livedAmes dwarf mice. Exp. Gerontol. 40, 491–497.

Matsukura, T. and Tanaka, H. (2000). Applicability of zinc complex of L-carnosine formedical use. Biochemistry (Moscow) 65, 817–823.

Mattson, M. P. and Wan, R. (2005). Beneficial effects of intermittent fasting and caloricrestriction on the cardiovascular and cerebrovascular systems. J. Nutr. Biochem. 16,129–137.

McFarland, G. A. and Holliday, R. (1994). Retardation of senescence in cultured humandiploid fibroblasts by carnosine. Exp. Cell Res. 212, 167–175.

McFarland, G. A. and Holliday, R. (1999). Further evidence for the rejuvenating effects of thedipeptide L-carnosine on cultured diploid fibroblasts. Exp. Gerontol. 34, 35–45.

McNaught, K. St. P., Belizaire, R., Isacson, O., Jenner, P., and Olanow, C. W. (2003). Alteredproteasome function in sporadic Parkinson’s disease. Exp. Neurol. 179, 38–46.

146 Alan R. Hipkiss

Medvedik, O., Lamming, D. W., Kim, K. D., and Sinclair, D. A. (2007). MSN2 and MSN4 linkcalorie restriction and TOR to sirtuin-mediated lifespn extension in Saccharomycescerevisiae. PLoS Biol. 5, 2330–2341.

Mikami, K., Otaka, M., Watanabe, D., Goto, T., Endoh, A., Miura, K., Ohshima, S.,Yoneyama, K., Sato, M., Shibuya, T., Segawa, D., Kataoka, E., et al. (2006). ZincL-carnosine protects against mucosal injury in portal hypertensive gastropathy throughinduction of heat shock protein 72. J. Gastroenterol. Hepatol. 21, 1669–1674.

Miller, A. and Adeli, K. (2008). Dierty fructose and metabolic syndrome. Curr. Opin.

Gastroenterol. 24, 204–209.Miller, R. A., Buehner, G., Chang, Y., Harper, J. M., Sigler, R., and Smith-Wheelock, M. (2005).

Methionine-deficient diet extends mouse lifespan, slow immune and lens aging, altersglucoase, T4, IGF-1 and insulin levels, and increases hepatocyte MIF levels and stressresistance. Aging Cell 4, 119–125.

Min, J., Senut, M. C., Rajanikant, K., Greengerg, E., Banagi, R., Zemke, D., Mousa, A.,Kassab, M., Farooq, M. U., Gupta, R., and Majid, A. (2008). Differential neuroprotectiveeffects of carnosine, anserine and N-acetyl carnosine against permanent focal ischemia.J. Neurosci. Res. 86, 2984–2991.

Mirza, M. A., Kandhro, A. J., Memon, S. Q., Khuhawar, M. Y., and Arain, R. (2007).Determination of glyoxal and methylglyoxal in the serum of diabetic patients by MEKCusing stilbenediamine as derivatizing agent. Electrophoresis 28, 3940–3947.

Morcos, M., Du, X., Pfisterer, F., Hunter, H., Sayed, A. A., Thornalley, P. J., Ahmed, N.,Baynes, J., Thorpe, S., Kukudov, G., Schlotterer, A., Bozorgmehr, F., et al. (2008).Glyoxalase-1 prevents mitochondrial protein modification and enhances lifespan inC. Elegans. Aging Cell 7, 260–269.

Mor-Vaknin, N., Puntutieri, A., Sitwala, K., and Markovitz, D. M. (2003). Vimentin issecreted by activated macrophages. Nat. Cell Biol. 5, 59–63.

Morrow, G. and Tanguay, R. M. (2003). Heat shock proteins and aging in Drosophila melano-

gaster. Semin. Cell Dev. Biol. 14, 291–299.Mroz, E. A. and Sewell, W. F. (1989). Pharmacological altereations of the activity of afferent

fibres innervating hair cells. Hear. Res. 38, 141–162.Muma, N. A. (2007). Transglutaminse is linked to neurodegenerative diseases. J. Neuropathol.

Exp. Neurol. 66, 258–263.Munch, G., Thome, J., Foley, P., Schinzel, R., and Riederer, P. (1997). Advanced glycation

endproducts in ageing and Alzheimer’s disease. Brain Res. Rev. 23, 134–143.Munch, G., Kuhla, B., Luth, H. J., Arendt, T., and Robinson, S. R. (2003). Anti-ageing defences

against Alzheimer’s disease. Biochem. Soc. Trans. 31, 1397–1399.Nagai, K. and Suda, T. (1986). Antineoplastic effects of carnosine and beta-alanine—Physio-

logical considerations of its antineoplastic effects. Nippon Seirigaku Zasshi 48, 741–747.Nagai, K., Suda, T., Kawasaki, K., and Mathuura, S. (1986). Action of carnosine and beta-

alanine on wound healing. Surgery 100, 815–821.Nagai, K., Niijima, A., Yamano, T., Otani, H., Okumra, N., Tsuruoka, N., Nakai, M., and

Kiso, Y. (2003). Possible role of L-carnosine in the regulation of blood glucose throughcontrolling autonomic nerves. Exp. Biol. Med. (Maywood) 228, 1138–1145.

Nagasawa, T., Yonekura, T.,Nishizawa,N., andKitts, D.D. (2001). In vitro and in vivo inhibitionof muscle lipid and protein oxidation by carnosine.Mol. Cell Biochem. 225, 29–34.

Naito, Y., Yoshikawa, T., Yagi, N., Matsuyama, K., Yoshida, N., Seto, K., and Yoneta, T.(2001). Effects of polaprezinc on lipid peroxidation, neutrophil accumulation, and TNF-alpha expression in rats with aspirin-induced gastric mucosal injury. Dig. Dis. Sci. 46,845–851.

Nakagawa, K., Ueno, A., and Nishikawa, Y. (2006). Interactions between carnosine andcaptopril on free radical scavenging activity and angiotensin-converting enzyme activityin vitro. Yakugaku Zasshi 125, 37–42.


Nakayama, K., Nakayama, M., Iwabuchi, M., Terawaki, H., Sato, T., Kohno, M., and Ito, S.(2008). Plasma alpha-oxoaldehyde levels in diabetic and ondiabetic chronic kidney dis-ease patients. Am. J. Nephrol. 28, 871–878.

Naudi, A., Caro, P., Jove, M., Gomez, J., Boada, J., Ayala, V., Portero-Otin, M., Barja, G., andPamplona, R. (2007). Methionine restriction decreases endogenous oxidative damage andincreases mitochondrial biogenesis and uncoupling protein 4 in rat brain. RejuvenationRes. 10, 473–484.

Ngo, J. K. and Davies, K. J. (2007). Importance of the lon protease in mitochondrialmaintenance and the significance of declining lon in aging. Ann. N. Y. Acad. Sci. 1119,78–87.

Nicoletti, V. G., Santoro, A. M., Grasso, G., Vagliasindi, L. I., Giuffrida, M. L., Cuppari, C.,Purrello, V. S., Giuffrida Stella, A. M., and Rizzarelli, E. (2007). Carnosine interaction withnitric oxide and astroglial protection. J. Neurosci. Res. 85, 2239–2245.

Niijima, A., Okui, T., Matsumura, Y., Yamano, T., Tsuroka, N., Kiso, Y., and Nagai, K. (2002).Effects of L-carnosine on renal sympathetic nerve activity and DOCA-salt hypertensionin rats. Auton. Neurosci. 97, 99–102.

Odashima, M., Otaka, M., Jin, M., Konishi, N., Kato, S., Matsuhashi, T., Nakamura, C., andWatanabe, S. (2002). Induction of a 72-kDa heat-shock protein in cultured rat gastricmucosal cells and rat gastric mucosa by zinc L-carnosine. Dig. Dis. Sci. 47, 2799–2804.

Odashima, M., Otaka, M., Jin, M., Wada, I., Horikawa, Y., Matsuhashi, T., Ohba, R.,Hatakeyama, N., Oyake, J., and Watanabe, S. (2006). Zinc L-carnosine protects colonicmucosal injury through induction of heat shock protein 72 and suppression of NF-kbactivation. Life Sci. 79, 2245–2250.

Orioli, M., Aldini, G., Beretta, G., Mattei Facino, R., and Carini, M. (2005). LC-ESI-MS/MSdetermination of 4-hydroxy-trans-2-nonenal Michael adducts with cysteine andhistidine-containing peptides as early markers of oxidative stress inexcitable tissues.J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 827, 109–118.

Orioli, M., Aldini, G., Benfatto, M. C., Facini, R. M., and Carini, M. (2007). HNE Michaeladducts to histidine and histidine-containing peptides as biomarkers of lipid-derivedcarbonyl stress in urines: LC-MS/MS profiling in Zucker obese rats. Anal. Biochem. 79,9174–9184.

Otto, H., Conz, C., Maier, P., Suzuki, C. K., Jeno, P., Rucknagel, P., Stahl, J., and Rospert, S.(2005). The chaperones MPP11 and Hsp70L1 form the mammalian ribosome-associatedcomplex. Proc. Natl. Acad. Sci. USA 102, 10064–10069.

Pallancke, L. and Greenamyre, J. T. (2006). Pink, parkin and the brain.Nature 441, 1058–1064.Pamplona, R., Dalfo, E., Ayala, V., Bellmunt, M. J., Prat, J., Ferrer, I., and Portero-Otin, M.

(2005). Proteins in human brain cortex are modified by oxidation, glycoxidation andlipoxidation: Effects of Alzheimer’s disease and identification of lipoxidation targets.J. Biol. Chem. 280, 21522–21530.

Pamplona, R., Naudi, A., Gavin, R., Pastrana, M. A., Sajnani, G., Ilieva, E. V., Del Rio, J. A.,Portero-Otin, M., Ferrer, I., and Requena, J. R. (2008). Increased oxidation, glycoxidation,and lipoxidation in prion disease. Free Radic. Biol. Med. 45, 1159–1166.

Pan, K. Z., Palter, J. E., Rogers, A. N., Olsen, A., Chen, D., Lithgow, G. J., and Kapahi, P.(2007). Inhibition of mRNA translation extends lifespan in Caenorhabditis elegans. AgingCell 6, 111–119.

Panzanelli, P., Valli, P., Cantino, D., and Fasolo, A. (1994). Glutamate and carnosine in thenvestibular system of the frog. Brain Res. 662, 293–296.

Park, Y. J., Volpe, S. L., and Decker, E. A. (2005). Quantitation of carnosine in human plasmaafter dietary consumption of beef. J. Agric. Food Chem. 53, 4735–4739.

Partridge, L. and Brand, M. D. (2005). Preface. Special issue on dietary restriction. Longevityand ageing – the current state of our knowledge and ignorance. Mech. Ageing Devel. 126,911–912.

148 Alan R. Hipkiss

Pattison, D. I. and Davies, M. J. (2006). Evidence for rapid inter- and intramolecular chlorinetransfer reactions of histamine and carnosine chloramines: Implications for the preven-tion of hypochlorous-acid-mediated damage. Biochemistry 45, 8152–8162.

Paul, S. (2008). Dysfunction of the ubiquitin-proteasome system in multiple disease condi-tions: Therapeutic approaches. Bioessays 30, 1172–1184.

Petroff, O. A., Mattson, R. H., Behar, K. L., Hyder, F., and Rothman, D. L. (1998). Vigabatrinincreases human brain homocarnosine and improves seizure control. Ann. Neurol. 44,948–952.

Petroff, O. A., Hyder, F., Rothman, D. L., andMattson, F. H. (2006). Brain homocarnosine andseizure control of patients taking gabapentin or topiramate. Epilepsia 47, 495–498.

Plun-Favreau, H., Klupsch, K., Moisoi, N., Gandhi, S., Kjaer, S., Frith, D., Harvey, K., Deas, E.,Harvey, R. J., McDonald, N., Wood, N. W., Martins, L. M., et al. (2007). The mitochondrialprotease htrA2 is regulated by Parkinson’s disease-associated kinase PINK1. Nat. CellBiol. 9, 1243–1252.

Preston, J. P., Hipkiss, A. R., Himsworth, D. J. T., Romero, I. A., and Abbott, J. N. (1998). Toxiceffects of beta-amyloid (25–35) on immortalised rat brain endothelial cells: Protection bycarnosine, homocarnosine and beta-alanine. Neurosci. Lett. 242, 105–108.

Purchas, R. W., Rutherford, S. M., Pearce, P. D., Vather, R., and Wilkinson, B. H. P. (2004).Concentrations in beef and lamb of taurine, carnosine, coenzyme Q10 and creatine. Meat

Sci. 66, 629–637.Qiu, W. Q. and Folstein, M. F. (2006). Insulin, insulin-degrading enzyme and amyloid-beta

peptide inn Alzheimer’s disease: Review and hypothesis. Neurobiol. Aging 27, 190–198.Quinn, P. J., Boldyrev, A. A., and Formazuyk, V. E. (1992). Carnosine: Its properties, func-

tions and potential therapeutic applications. Mol. Aspects Med. 13, 379–444.Rabbani, N. and Thornalley, P. J. (2008). The dicarbonyl proteome: Proteins susceptible to

dicarbonyl glycation at functional sites in health, aging and disease. Ann. N. Y. Acad. Sci.

1126, 124–127.Rajanikant, G. K., Zemke, D., Senut, M. C., Frenzel, M. B., Chen, A. F., Gupta, R., and

Majid, A. (2007). Carnosine is neuroprotective against permanent focal cerebral ischemiain mice. Stroke 38, 3023–3031.

Rakwalska, M. and Rospert, S. (2004). The ribosome-bound chaperones RAC and Ssb1/2pare required for accurate translation in Saccharomyces cerevisiae. Mol. Cell Biol. 24,9186–9197.

Rashid, I., van Reyk, D. M., and Davies, M. J. (2007). Carnosine and its constituents inhibitglycation of low-density lipoproteins that promote foam cell formation in vitro. FEBS Lett.581, 1067–1070.

Rattan, S. I. (2006). Hormetic modulation of aging and longevity by mild heat stress. Dose

Response 3, 533–646.Reddy, V. P., Obrenovich, M. E., Atwood, C. S., Perry, G., and Smith, M. A. (2004). Involve-

ment of Maillard reactions in Alzheimer’s disease. Neurotox. Res. 4, 191–209.Reeve, V. E., Boehm-Wilcox, C., Bosnic, M., and Rozinova, E. (1993a). Suppressive effect of

2-acetyl-4-tetrahydroxybutylimidazole on contact hypersensitivity in Skh:HR hairlessmouse. Int. Arch. Allergy Immunol. 102, 101–106.

Reeve, V. E., Bosnic, M., and Rozinova, E. (1993b). Carnosine (beta-alanylhistidine) protectsfrom the suppression of contact hypersensitivity by ultraviolet B (280–320nm) radiationor by cis urocanic acid. Immumnology 78, 99–104.

Rehnstrom, K., Ylisaukko-Oja, T., Vanhala, R., von Wendt, L., Peltonen, L., and Hovatta, I.(2008). No association between common variants in glyoxalase 1 and autistic spectrumdisorders. Am. J. Med. Genet. B Neuropsychiatr. Genet. 1478, 124–127.

Religa, D., Strozyk, D., Cherny, R. A., Volotakis, I., Haroutunian, V., Winblad, B., Naslund, J.,and Bush, A. I. (2006). Elevated cortical zinc in Alzheimer’s disease. Neurology 67,69–75.


Ririe, D. G., Roberts, P. R., Shouse,M. N., and Zaloga, G. P. (2000). Vasodilatory actions of thedietary peptide carnosine. Nutrition 16, 168–172.

Roberts, P. R. and Zaloga, G. P. (2000). Cardiovascular effects of carnosine. Biochemistry

(Moscow) 65, 856–861.Roberts, P. R., Black, K. W., Santamauro, J. T., and Zaloga, G. P. (1998). Dietary peptides

improve wound healing following surgery. Nutrition 14, 266–269.Rodgers, J. T., Lerin, C., Gerhart-Hines, Z., and Puigserver, P. (2008). Metabolic adaptations

through PGC-1 alpha and SIRT1 pathways. FEBS Lett. 582, 46–53.Ruan, Q. and Johnson, G. V. (2007). Transglutaminase 2 in neurodegenerative disorders.

Front. Biosci. 12, 891–904.Sacco, R., Papaleo, V., Hager, J., Rousseau, F., Moessner, R., Militerni, R., Bravaccio, P.,

Trillo, S., Schneider, C., Melmed, R., Elia, M., Curatolo, P., et al. (2007). Case-controland family-based association studies of candidate genes in autistic disorder and itsendophenotypes: TPH2 and GLO 1. BMC Med. Genet. 8, 11.

Sassoe-Pognetto, M. M., Cantino, D., Panzanelli, P., di Cantagno, L. V., Giustetto, M.,Margolis, F., de Biasi, L., and Fasolo, A. (1993). Presynaptic colocalization of carnosineand glutamate in olfactory neurones. Neuroreport 5, 7–10.

Sato, M., Karasawa, N., Shimizu,M., Morimatsu, F., and Yamada, R. (2008). Safety evaluationof chicken breast extract containing carnosine and anserine. Food Chem. Toxicol. 46,480–489.

Sauerhoefer, S., Bartram, C. R., Nawroth, P., Hammes, H. P., Yard, B. A., Zschocke, J., andvan der Woude, F. J. (2005). Carnosine as a protective factor in diabetic nephropathy:Association with a leucine repeat of the carnosinase gene CNDP1.Diabetes 54, 2320–2327.

Sauerhofer, S., Yuan, G., Braun, G. S., Deinzer, M., Neumaier, M., Gretz, N., Floege, J.,Kriz, W., van der Woude, F., and Moeller, M. J. (2007). L-Carnosine, a substrate ofcarnosinase-1, influences glucose metabolism. Diabetes 56, 2425–2432.

Schalkwijk, C. G., Brouwers, O., and Stehouer, C. D. (2008). Modulation of insulin action byadavanced glycation endproducts: A new player in the field. Horm. Metab. Res. 40,614–619.

Schroder, L., Schmitz, C., and Bachert, P. (2004). Molecular dynamics and information onpossible sites of interaction of intramyocellular metabolites in vivo from resolved dipolarcouplings in localized 1H NMR spectra. J. Magn. Reson. 171, 213–224.

Schulz, M., Hamprecht, B., Kleinkauf, H., and Bauer, K. (1989). Regulation by dibutyrylcyclic AMP of carnosine synthesis in astroglial-rich primary cultures kept in serum-freemedium. J. Neurochem. 52, 229–234.

Seidler, N. W. (2000). Carnosine prevents glycation-induced changes in electrophoreticmobility of aspartate aminotransferase. J. Biochem. Mol. Toxicol. 14, 215–220.

Seidler, N. W., Yeargans, G. S., and Morgan, T. G. (2004). Carnosine disaggregates glycatedalpha-crystallin: An in vitro study. Arch. Biochem. Biophys. 427, 110–115.

Selkoe, D. J., Abraham, C., and Ihara, Y. (1982). Brain transglutaminase: In vitro crosslinkingof human neurofilament proteins into insoluble polymers. Proc. Natl. Acad. Sci. USA 79,6070–6074.

Selle, A., Ullrich, O., Harnacke, K., and Hass, R. (2007). Retrodifferentaion and rejuvenationof senescent monocyte cells requires PARP-1. Exp. Gerontol. 42, 554–562.

Shao, L., Li, Q. H., and Tan, Z. M. (2004). L-carnosine reduces telomere damage andshortening rate in cultured normal fibroblasts. Biochem. Biophys. Res. Commun. 324,931–936.

Shen, Y., Fan, Y., Dai, H., Fu, Q., Hu, W., and Chen, Z. (2007a). Neuroportective effect ofcarnosine on necrotic cell death in PC12 cells. Neurosci. Lett. 414, 145–149.

Shen, Y., Hu, W. W., Fan, Y. Y., Dai, H. B., Fu, Q. L., Wei, E. Q., Luo, J. H., and Chen, Z.(2007b). Carnosine protects against NMDA-induced neurotoxicity in differentiated PC12

150 Alan R. Hipkiss

cells through carnosine-histidine-histamine pathway and H1/H3 receptors. Biochem.

Pharmacol. 73, 709–717.Shen, J., Yao, J. F., Tanida, M., and Nagai, K. (2008). Regulation of sympathetic nerve activity

by L-carnosine in mammalian white adipose tissue. Neurosci. Lett. 44, 100–104.Shimada, T., Waranabe, N., Ohtsuka, Y., Endoh, M., Kojima, K., Hiraishi, H., and Terano, A.

(1999). Polaprezinc down-regulates proinflammatory cytokine-induced nuclear factor-kappa B activation and interleukin-8 expression in gastric epithelial cells. J. Pharmacol.

Exp. Ther. 291, 345–352.Shimizu, K., Kiuchi, Y., Ando, K., Hayakawa, M., and Kikugawa, K. (2004). Coordination of

oxidized protein hydrolase and the proteasome in the clearance of cytotoxic denaturedproteins. Biochem. Biophys. Acta 324, 140–146.

Shiraki, M., Kuroda, Y., Tanaka, S., Salto, M., Fukunaga, M., and Nakamura, T. (2008).Nonenzymic collagen cross-links induced by glycoxidation (pentosidine) predicts verte-bral fractures. J. Bone Miner. Metab. 26, 93–100.

Shuvaev, V. V., Laffont, I., Serot, M., Fujii, J., Taniguchi, N., and Siest, G. (2001). Increasedprotein glycation in cerebrospinal fluid of Alzheimer’s disease. Neurobiol. Aging 22,397–402.

Singh, R., Barden, A., Mori, T., and Beilin, L. (2001). Advanced glycation end-products: Areview. Diabetologia 44, 129–148.

Sitte, N. (2003). Oxidative damage to proteins. In ‘‘Aging at the Molecular Level’’ (T. vonZglinicki, ed.), pp. 27–45. Kluwer Academic Publishers, The Netherlands.

Smith, D. P., Smith, D. G., Curtain, C. C., Boas, J. F., Pilbrow, J. R., Ciccotosto, G. D., Lau, T. L.,Tew, D. J., Perez, K., Wade, J. D., Bush, A. I., Drew, S. C., et al. (2006). Copper-mediatedamyloid-beta toxicity is associated with an intermolecular bridge. J. Biol. Chem. 281,15145–15154.

Soh, J. W., Hotic, S., and Arking, R. (2007). Dietary restriction in Drosophila is dependent onmitochondrial efficiency and constrained by pre-existing extended longevity. Mech.

Ageing Dev. 128, 581–593.Son, D. O., Satsu, H., Kiso, Y., Totsuka, M., and Shimizu, M. (2008). Inhibitory effect of

carnosine on interleukin-8 in intestinal epithelial cells through translational regulation.Cytokine 43, 265–276.

Sozio, P., Iannitelli, A., Cerasa, L. S., Cacciatre, I., Cornacchia, C., Giorgioni, G.,Ricciutelli, M., Nasuti, C., Cantalamessa, F., and Stefano, A. (2008). New L-dopa codrugsas potential antiparkinson agents. Arch. Pharm. (Weinheim) 341, 412–417.

Stadtman, E. R. (1992). Protein oxidation and aging. Science 257, 1220–1224.Stout, J. R., Graves, B. S., Smith, A. E., Hartman, M. J., Cramer, J. T., Beck, T. W., and

Harris, R. C. (2008). The effect of beta-alanine supplementation on neuromuscular fatiguein elderly (55–92 years): A double-blind randomized study. J. Int. Soc. Sports Nutr. 5, 21.

Stuerenburg, H. J. and Kunze, K. (1999). Concentrations of free carnosine (a putativemembrane-protective antioxidant) in human muscle and rat muscles. Arch. Gerontol.Geriatr. 29, 107–113.

Stvolinsky, S. L. and Dobrota, D. (2000). Anti-ischemic activity of carnosine. Biochemistry

(Moscow) 65, 849–855.Sugiyama, T., Tanaka, H., and Kawai, S. (2000). Improvement of periarticular osteoporosis

in postmenopausal women with rheumatoid arthritis by beta-alanyl-L-histidinatozinc:A pilot study. J. Bone Miner. Metab. 18, 335–338.

Suji, G. and Sivakami, S. (2004). Glucose, glycation and aging. Biogerontology 5, 365–373.Syntichaki, P., Troulinaki, K., and Tavernarakis, N. (2007). eIF4E function in somatic cells

modulates ageing in Caenorhabditis elegans. Nature 445, 922–926.Tabakman, R., Jiang, H., Levine, R. A., Kohen, R., and Lazarovici, P. (2004). Apoptotic

characteristics of cell death and neuroprotective effect of homocarnosine on pheochro-mocytoma PC12v cells exposed to ischemia. J. Neurosci. Res. 75, 499–507.


Tallon, M. J., Harris, R. C., Boobis, L., Fallowfield, J., and Wise, J. A. (2005). The carnosinecontent of vastus lateralis is elevated in resistance-trained bodybuilders. J. Strength. Cond.Res. 19, 725–729.

Tallon, M. J., Harris, R. C., Maffulli, N., and Tarnopolsky, M. A. (2007). Carnosine, taurineand enzyme activities of human skeletal muscle fibres from elderly subjects withosteoarthritis and young moderately active subjects. Biogerontology 8, 129–137.

Tamba, M. and Torreggiani, A. (1999). Hydroxyl radical scavenging by carnosine and Cu(11)-carnosine complexes: A pulse-radiolysis and spectroscopic study. Int. J. Radiat. Biol.75, 1177–1188.

Tan, K. M. and Candish, J. K. (1998). Carnosine and anserine as modulators of neutrophilfunction. Clin. Lab. Haematol. 20, 239–244.

Tang, S. C., Arumugam, T. V., Cutler, R. G., Jo, D. G., Magnus, T., Chan, S. L., Mughal, M. R.,Telliohann, R. S., Nassar, M., Ouyang, X., Calderan, A., Ruzza, P., et al. (2007). Neuro-protective actions of a histidine analogue in models of ischemic stroke. J. Neurochem. 101,729–736.

Tang, S. Y., Allen, M. R., Phipps, R., Burr, D. B., and Vashishth, D. (2009). Changes in non-enzymic glycation and its association with altered mechanical properties following1-year treatment with risedronate or alendronate. Osteoporos. Int. 20, 887–894.

Tanida, M., Niijima, A., Fukuda, Y., Sawai, H., Tsuruoka, N., Shen, J., Yamada, S., Kiso, Y.,and Nagai, K. (2005). Dose-dependent effects of L-carnosine on the renal sympatheticnerve and blood pressure in irethane-anesthetized rats. Am J Physiol. Regul. Integr. Comp.

Physiol. 288, R447–455.Tanida, M., Gotoh, H., Taniguchi, H., Otani, H., Shen, J., Nakamura, T., Tsuroka, N., Kiso, Y.,

Okumura, N., and Nagai, K. (2007). Effects of central injection of L-carnosine on sympa-thetic nerve activity innervating brown adipose tissue and body temperature in rats.Regul. Pept. 144, 62–71.

Tatsuta, T. and Langer, T. (2008). Quality control of mitochondria: Protection against neuro-degeneration and ageing. EMBO J. 27, 306–314.

Temperini, C., Scozzafava, A., Puccetti, L., and Supuran, C. (2005). Carbonic anhydrase activa-tors: X-ray crystal structure of the adduct of human isozyme II with histidine as a platformfor the design of stronger activators. Bioorg. Med. Chem. Lett. 15, 5136–5141.

Teuscher, N. S., Shen, H., Shu, C., Xiang, J., Keep, R. F., and Smith, D. E. (2004). Carnosineuptake in rat choroids plexus primary cell cultures and choroids plexus whole tissuefrom PEPT2 null mice. J. Neurochem. 89, 375–382.

Thio, L. L. and Zhang, H. X. (2006). Modulation of inhibitory glycine receptors in culturedembryonic mouse hippcampal neurons by zinc, thiol containing redox agents and carno-sine. Neuroscience 139, 1315–1327.

Thomas, S., Kotamraju, S., Zielonka, J., Harder, D. R., and Kalyanaraman, B. (2007). Hydro-gen peroxide induces nitric oxide and proteasome activity in endothelial cells: A bell-shaped signaling response. Free Radic. Biol. Med. 42, 1049–1061.

Thornalley, P. J. (1999). Clinical significance of glycation. Clin. Lab. 45, 263–273.Tomonaga, S., Tachibana, T., Takagi, T., Saito, E. S., Zhang, R., Denbow, D. M., and

Furuse, M. (2004). Effect of central administration of carnosine and its constituents onbehaviours in chicks. Brain Res. Bull. 63, 75–82.

Tomonaga, S., Tachibana, T., Takahashi, H., Sato, M., Denbow, D. M., and Furuse, M. (2005).Nitric oxide involves in carnosine-induced hyperactivity in chicks. Eur. J. Pharmacol. 524,84–88.

Tomonaga, S., Hayakawa, T., Yamane, H., Maemura, H., Sato, M., Takahata, Y.,Morimatsu, F., and Furuse, M. (2007). Oral administration of chicken breast extractincreases brain carnosine and anserine in rats. Nutr. Neurosci. 10, 181–186.

152 Alan R. Hipkiss

Tomonaga, S., Yamane, H., Onitsuka, E., Yamada, S., Sato, M., Takahata, Y., Morimatsu, F.,and Furuse, M. (2008). Carnosine-induced antidepressant-like activity in rats. Pharmacol.Biochem. Behav. 89, 627–632.

Trombley, P. Q., Hornung, M. S., and Blakemore, L. J. (2000). Interactions between carnosineand zinc and copper: Implications for neuromodulations and neuroprotection. Biochem-

istry (Moscow) 65, 949–960.Tsubone, S., Yashikawa, N., Okada, S., and Abe, U. (2007). Purification and characterization

of a novel imidazole dipeptide synthetase from the muscle of the Japanese eel Anguillajaponica. Comp. Biochem. Physiol. B Biochem. Mol. Biol. 146, 560–567.

Tsuneyoshi, Y., Tomonga, S., Asechi, M., Morishita, K., Denbow, D. M., and Furuse, M.(2007). Central administration of dipeptides, beta-alanyl-BCAAs, induces hyperactivityin chicks. BMC Neurosci. 8, 37–42.

Tsuneyoshi, Y., Yamane, H., Tomonga, S., Morishita, K., Denbow, D. M., and Furuse, M.(2008). Reverse structure of carnosine-induced sedative and hypnotic effects in chickunder stress. Life Sci. 82, 1065–1069.

Ukeda, H., Hasegawa, Y., Harada, Y., and Sawamura, M. (2002). Effect of carnosine andrelated compounds on the inactivation of human Cu,Zn-superoxide dismutase by modi-fication of fructose and glycolaldehyde. Biosci. Biotechnol. Biochem. 66, 36–43.

Unterluggauer, H., Micutkova, L., Lindner, H., Sarg, B., Hernebring, M., Nystrom, T., andJansen-Durr, P. (2008). Identification of Hsc70 as target for AGEmodification in senescenthuman fibroblasts. Biogerontology (in press); DOI.10.10007/810522-0089193-z.

Vander Jagt, D. L. (2008). Methylglyoxal, diabetes mellitus and diabetic complications. Drug

Metabol. Drug Interact. 23, 93–124.Vaughan-Jones, R. D., Spitzer, K. W., and Swietach, P. (2006). Spatial aspects of intracellular

pH regulation in heart muscle. Prog. Biophys. Mol. Biol. 90, 207–224.Vernace, V. A., Arnaud, L., Schmidt-Glenewinkel, T., and Figueiredo-Pereira, M. E. (2007a).

Agingperturbs 26Sproteasomeassembly inDrosophilamelanogaster. FASEB J. 21, 2672–2682.Vernace, V. A., Schmidt-Glenewinkel, T., and Figueiredo-Pereira, M. E. (2007b). Aging and

regulated protein degradation: Who has the upper hand? Aging Cell 6, 599–606.Vistoli, G., Pedretti, A., Cattaneo,M., Aldini, G., and Testa, B. (2006). Homologymodelling of

human serum carnosinase, a potential medicinal target, and MD simulation of its alloste-ric activation by citrate. J. Med. Chem. 49, 3269–3277.

Vlassara, H. and Palace, M. R. (2002). Diabetes and advanced glycation endproducts: Areview. J. Intern. Med. 251, 87–101.

Wada, I., Otaka, M., Jin, M., Odashima, M., Komatsu, K., Konishi, N., Matsuhashi, T.,Horikawa, Y., Ohba, R., Itoh, H., and Watanabe, S. (2006). Expression of HSP72 in thegastric mucosa is regulated by gastric acid in rats—Correlation of HSP72 expression withmucosal protection. Biochem. Biophys. Res. Commun. 349, 611–618.

Wang, H., Meng, Q. H., Chang, T., and Wu, L. (2006). Fructose-induced peroxynitriteproduction is mediated by methylgyoxal in vascular smooth muscle cells. Life Sci. 79,2448–2454.

Wang, X., Chang, T., Jiang, B., Desai, K., and Wu, L. (2007). Attenuation of hypertensiondevelopment by aminoguanidine in spontaneously hypertensive rats: Role of methyl-glyoxal. Am. J. Hypertens. 20, 629–636.

Wang, X., Jia, X., Chang, T., Desai, K., and Wu, L. (2008). Attenuation of hypertensiondevelopment by scavenging methylglyoxal in fructose-fed rats. J. Hypertens. 26, 765–772.

Wassif, W. S., Sherwood, R. A., Amir, A., Idowu, B., Summers, B., Leigh, N., and Peters, T. J.(1994). Serum carnosinase activities in central nervous system disorders. Clin. Chim. Acta

225, 57–64.Watanabe, S., Wang, X. E., Hirose, M., Kivilioto, T., Osada, T., Miwa, H., Oide, H.,

Kitamura, T., Yoneta, T., Seto, K., and Sato, N. (1998). Insulin-like growth factor 1 plays


a role in gastric wound healing: Evidence using a zinc derivative, polaprezinc, and anin vitro rabbit wound repair model. Aliment. Pharmacol. Ther. 12, 1131–1138.

Westphal, C. H., Dipp, M. A., and Guarente, L. (2007). A therapeutic role for sirtuins indiseases of aging? Trends Biochem. Sci. 32, 555–560.

Willi, S. M., Zhang, Y., Hill, J. B., Phelan, M. C., Michaelis, R. C., and Holden, K. R. (1997).A deletion in the long arm of chromosome 18 in a child with serum carnosinasedeficiency. Pediatr. Res. 41, 210–213.

Wu, L. (2006). Is methylgyoxal a causative factor for hypertension development? Can. J.

Physiol. Pharmacol. 84, 129–139.Wu, X. H., Ding, M. P., Zhu-Ge, Z. J., Zhu, Y. Y., Jin, C. L., and Chen, Z. (2006). Carnosine, a

precursor of histidine, ameliorates pentylenetetrazole-induced kindled seizures in rat.Neurosci. Lett. 400, 146–149.

Wu, Y. Y., Chlen, W. H., Huang, Y. S., Gau, S. S., and Chen, C. H. (2008). Lack of evidence tosupport the glyoxalase 1 gene (GLO1) as a risk gene of autism in Han Chinese patientfrom Taiwan. Prog. Neuropsychpharmacol. Biol. Psychiatry 32, 1740–1744.

Yamagishi, S., Nakamura, K., Inoue, H., Kikushi, S., and Takeuchi, M. (2005). Serum andcerebrospinal fluid levels of glyceraldehyde-derived advanced glycation end products(ages) may be a promising biomarker for early detection of Alzheimer’s disease. Med.

Hypotheses 64, 1205–1207.Yamaguchi, M. (1995). b-Alanyl-L-histidinato zinc and bone resorption. Gen. Pharmacol. 26,

1179–1183.Yamaguchi, M. and Kishi, S. (1993). Prolonged administration of beta-alanyl-L-histidine zinc

prevents bone loss in ovariectomized rats. Jpn. J. Pharmacol. 63, 203–207.Yamaguchi, M. and Kishi, S. (1995a). Differential effects of transforming growth factor-beta

on osteoblast-like cell formatiom in mouse marrow culture: Relation to the effect of zinc-chelating dipeptides. Peptides 16, 1483–1488.

Yamaguchi, M. and Kishi, S. (1995b). Inhibitory effect of zinc-chelating dipeptide on para-thyroid hormone-stimulated osteoclast-like cell formation in mouse marrow cultures:Involvement of calcium signalling. Peptides 16, 629–633.

Yamaguchi, M. and Matsui, T. (1996). Stimulatory effect of zinc-chelating dipeptide ondeoxyribonucleic acid synthesis in osteoblastic MC3T3-E1 cells. Peptides 17, 1207–1211.

Yamaguchi, M. and Matsui, T. (1997). Zinc enhancement of 17beta-estradiol’s anabolic effectin osteoblastic MC3T3-E1 cells. Calcif. Tissue Int. 60, 527–532.

Yamamoto, A., Takagi, H., Kitamura, H., Nakano, H., Kawano, H., Kuroyanagi, H.,Yahagi, Y., Kobayashi, S., Koizumi, K., Sakai, T., Saito, K., Chibo, T., et al. (1998).Deficiency in protein L-isoaspartyl methyltransferase results in a fatal progressive epi-lepsy. J. Neurosci. 18, 2063–2074.

Yamawaki, H., Saito, K., Okada, M., and Hara, Y. (2008). Methylglyoxal mediates vascularinflammation via JNK and p38 in human endothelial cells. Am. J. Physiol. Cell Physiol. 295,C1510–C1517.

Yan, H. and Harding, J. J. (2005). Carnosine protects against the inactivation of esteraseinduced by glycation and a steroid. Biochim. Biophys. Acta 1741, 120–126.

Yan, S. D., Chen, X., Schmidt, A. M., Brett, J., Godman, G., Zou, Y. S., Scott, C. W., Caputo, C.,Frappier, T., Smith, M. A., Perry, G., Yen, S. H., and Stern, D. (1994). Glycated tau proteinin Alzheimer’s disease: A mechanism for induction of oxidant stress. Proc. Natl. Acad. Sci.

USA 91, 7787–7791.Yander Jagt, D. L. (2008). Methylglyoxal, diabetes mellitus and diabetic complications. Drug

Metabol. Drug Interact. 23, 93–124.Yao, D., Taguchi, T., Matsumura, T., Pestell, R., Edelstein, D., Giardino, I., Suske, G.,

Rabbani, N., Thornalley, P. J., Sarthy, V. P., Hammes, H. P., and Brownlee, M. (2007).High glucose increases angiopoietin-2 transcription in microvascular endothelial cellsthrough methylglyoxal modification of mSin3A. J. Biol. Chem. 282, 31038–31045.

154 Alan R. Hipkiss

Yasuhara, T., Hara, K., Maki, M., Masuda, T., Sanberg, C. D., Sanberg, P. R., Bickford, P. C.,and Borlongan, C. V. (2008). Dietary supplementation exerts neuroprotective effects inischemic stroke model. Rejuvenation Res. 11, 201–214.

Yokoyama, H., Kurolwa, H., and Yano R Araki, T. (2008). Targeting reactive oxygen species,reactive nitrogen species and inflammation in MPTP neurotoxicity and Parkinson’sdisease. Neurol. Sci. 29, 292–301.

Yorbik, O., Sayal, A., Akay, C., Akbiyik, D. I., and Sohmen, T. (2002). Investigation ofantioxidant enzymes in children with autistic disorder. Prostglandins Leukot. Essent.

Fatty Acids 67, 341–343.Yuneva, M. O., Bulygina, E. R., Gallant, S. C., Kramarenko, G. G., Stvolinsky, S. L.,

Semyonova, M. L., and Boldyrev, A. A. (1999). Effect of carnosine on age-inducedchanges in senescence-accelerated mice. J. Anti Aging Med. 2, 337–342.

Yuneva, A. O., Kramarenko, G. G., Vetreshchak, T. V., Gallant, S., and Boldyrev, A. A.(2002). Effect of carnosine on Drosophila melanogaster lifespan. Bull. Exp. Biol. Med. 133,559–561.

Zakhidov, S. T., Gopko, A. V., Semenova, M. L., Mikhaleva, Y. Y., Makarov, A. A., andKulibin, A. Y. (2002). Carnosine modifies the incidence of genetically abnormal sex cellsin the testes of senescence-accelerated mice. Bull. Exp. Biol. Med. 134, 78–80.

Zaloga, G. P. and Siddiqui, R. A. (2004). Biologically active dietary peptides. Mini Rev. Med.

Chem. 4, 815–821.Zaloga, G. P., Roberts, P. R., and Nelson, T. E. (1996). Carnosine: A novel peptide regulator of

intreacellular calcium and contractility in cardiac muscle. New Horiz. 4, 26–35.Zaloga, G. P., Roberts, P. R., Black, K. W., Lin, M., Zapata-Sudo, G., and Sudo, R. T. (1997).

Carnosine is a novel peptide modulator of intracellular calcium contractility in cardiaccells. Am. J. Physiol. 272, H462–H468.

Zhu, J. X. and Aswad, D. W. (2007). Selective cleavage of isoaspartyl peptide bonds byhydroxylamine after methyltransferase priming. Anal. Biochem. 364, 1–17.

Zhu, J. X., Doyle, H. A., Mamula, M. J., and Aswad, D. W. (2006). Protein repair in the brain,proteomic analysis of endogenous substrates for protein L-isoaspartyl methyltransferasein mouse brain. J. Biol. Chem. 28, 33802–33813.

Zhu, Y. Y., Zhu-Ge, Z. B., Wu, D. C., Wang, S., Liu, L. Y., Ohtsu, H., and Chen, Z. (2007).Carnosine inhibits pentylenetetrazol-induced seizures by histaminergic mechanisms inhistidine decarboxylase knock-out mice. Neurosci. Lett. 416, 211–216.

Zhuravskii, S. G., Aleksandrova, L. A., Sirot, V. S., and Ivanov, S. A. (2004a). Naturalantioxidant L-carnosine inhibits LPO intensification in structures of the auditory analyzerunder conditions of chronic exposure to aminoglycoside antibiotics. Bull. Exp. Biol. Med.

138, 361–364.Zhuravskii, S. G., Aleksandrova, L. A., Ivanov, S. A., Sirot, V. S., Lopotko, A. I., and

Zhloba, A. A. (2004b). Protective effect of carnosine on excitable structures of the auditoryapparatus in albino rats with acute acoustic trauma. Bull. Exp. Biol. Med. 137, 98–102.

CHAPTER 4


* Department of Food Scie{ Soil Resource Group, Gu{ Department of Land Res

Recent Advances in theMicrobial Safety of FreshFruits and Vegetables

Keith Warriner,* Ann Huber,† Azadeh Namvar,*

Wei Fan,* and Kari Dunfield‡


rition R016/S1

nce, Uelph, Oources

esearch, Volume 57 # 2009043-4526(09)57004-0 All righ

niversity of Guelph, Guelph, Ontario, Canadantario, Canada

, University of Guelph, Guelph, Ontario, Canada

Elsts

II. O
utbreaks Linked to Fresh Produce 157
III. C
haracteristics of Pathogens Recovered
from Salad Vegetables
160
A
. P athogenic E. coli 160
B
. S higella 164
C
. S almonella 164
D
. C ampylobacter 165
E. L
isteria monocytogenes 166
F. A
eromonas hydrophila 166
G
. E ndospore-forming bacteria 166
H
. E nteric viruses 167
I. H
uman pathogenic protozoa 167
IV. T
ransmission of Human Pathogens in Manure, Soil,
and Water to the Vegetable Production Chain
168
A
. M anure and biosolids 169
B
. Ir rigation water 171
C
. S oil 176
D
. T ransport of human pathogens within the
environment
177
V. In
teraction of Pathogens with Fresh Produce 179
A
. S urvival in the phyllosphere 179
evier Inc.reserved.

155

156 Keith Warriner et al.

B
. C olonization of the rhizosphere 180
C
. In ternalization of human pathogens
in growing plants
181
D
. G enetic and physiological factors 186
VI. In
terventions to Enhance the Safety
of Fresh Produce
188
A
. B iocontrol of human pathogens 191
VII. C
onclusions and Future Research 192
Refere
nces 192
Abstract Foodborne illness outbreaks linked to fresh produce are becoming

more frequent and widespread. High impact outbreaks, such as that

associated with spinach contaminated with Escherichia coli O157:H7,

resulted in almost 200 cases of foodborne illness across North

America and >$300 m market losses. Over the last decade there

has been intensive research into gaining an understanding on the

interactions of human pathogens with plants and how microbiologi-

cal safety of fresh produce can be improved. The following review

will provide an update on the food safety issues linked to fresh

produce. An overview of recent foodborne illness outbreaks linked

to fresh produce. The types of human pathogens encounteredwill be

described and how they can be transferred from their normal animal

or human host to fresh produce. The interaction of human pathogens

with growing plants will be discussed, in addition to novel interven-

tionmethods to enhance themicrobiological safety of fresh produce.

I. INTRODUCTION

The fresh-cut market has experienced rapid growth within the last decadeand an estimated 6million packs of bagged produce are sold within NorthAmerica each day (Doyle and Erickson, 2008; Jongen, 2005). The drivingforce behind the rapid growth of the fresh produce is the desire of con-sumers to lead a healthy lifestyle along with the convenience of pre-prepared products. Consumers have become accustom to all-year-aroundavailability of fruit and vegetables with the convenience of prepackedproducts that require minimal preparation. To meet consumer demand,the primary production has shifted away from local farmers to highlycentralized centers, which in the case of North America, are located inMexico, California, and Florida (Doyle and Erickson, 2008). Indeed, Cali-fornia alone supplies over 70% of all the leafy greens (e.g., lettuce, spinach)consumed within North America (FDA, 2002). Centralization of produc-tion has brought many benefits to the consumer such as relatively cheapproduce, consistent quality, and all-year-round availability. However, thedownside of centralized production is that when foodborne illness out-breaks occur, they are typically widespread and involve a high number of

Microbiological Safety of Fresh Produce 157

cases (Gorny, 2006). Although consumers in the course of an outbreakfrequently turn to organic or locally grown produce, there is no evidencethat this poses any less risk compared to ‘conventionally’ produced crops(Arthur et al., 2007; Loncarevic et al., 2005). In this respect, it cannot beconcluded that the rapid rise in foodborne illness outbreaks are linked tocentralized production alone. It is noteworthy that the increase in thefresh-cut (bagged salad) market has coincided with the increase in food-borne illness cases. Here, the produce is cut or shredded which providesentry points for pathogens, in addition to releasing nutrients to supportmicrobial growth.ModifiedAtmospheric Packaging (MAP) reduces spoil-age by aerobes but can enhance the virulence of pathogens such as Escher-ichia coli O157:H7 (Chua et al., 2008). In addition, the higher proportion ofvulnerable, susceptible, people within a population along with increasedsurveillance and high prevalence of virulent pathogens within the envi-ronment are further reasons to consider for the rise in foodborne illnessoutbreaks (Arthur et al., 2008; FDA, 2001; Sewell and Farber, 2001).

II. OUTBREAKS LINKED TO FRESH PRODUCE

There has been a rapid rise in foodborne illness outbreaks linked to freshproduce (Fig. 4.1). The pathogens of main concern are Salmonella andE. coli O157:H7 although, in principle, a diverse range of pathogenicmicrobes can contaminate fresh produce at any point in the chain.

There has been several high profile foodborne illness outbreaks asso-ciated with fresh produce with sprouted seeds, tomatoes, and leafy greensremaining the most prominent (Table 4.1) (Doyle and Erickson, 2008). Theunderlying reasons for why specific produce types have been implicated inthe majority of outbreaks can, in part, be explained by the market volume

8070605040302010

Year

Out

erbr

eaks

Reported cases Outbreaks

90

0

3000

2500

2000

1500

1000

500

Cas

es

3500

0

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

1990

2006

FIGURE 4.1 Foodborne illness outbreaks linked to fresh produce from 1990 to 2006.

Source: Centre for Science in the Public Interest (2008).

TABLE 4.1 Outbreaks linked to fresh produce

Date Pathogen Produce Comments

December 2005 Salmonella Mung bean sprouts Canada, 618 confirmed casesFebruary 2006 Salmonella Alfalfa sprouts Canada, sprout recall due to suspected

contamination

February 2006 Salmonella Alfalfa sprouts Australia, 100 confirmed cases of salmonellosis

June 2006 E. coli O121:H19 Lettuce United States, 4 confirmed cases

July 2006 Salmonella Fruit salad United States and Canada, 41 confirmed cases

August 2006 Salmonella Alfalfa sprouts United States, sprout recall due to suspected

contamination

September 2006 E. coli O157:H7 Spinach United States, 205 confirmed cases; 3 deathsSeptember 2006 Clostridium

botulinum

Pasteurized carrot

juice

United States and Canada; 6 cases

October 2006 E. coli O157:H7 Lettuce Canada; 30 confirmed cases

October 2006 E. coli O157:H7 Lettuce Canada; recall due to suspected contamination

October 2006 Salmonella Tomatoes United States; 183 cases

November 2006 E. coli O157:H7 Lettuce United States; 81 confirmed cases

November 2006 E. coli O157:H7 Lettuce United States; 71 confirmed cases

November 2006 Salmonella Peanut butter United States; 481 confirmed casesApril 2007 Salmonella Lettuce UK, recall for suspected contamination

August 2007 Shigella sonnei Baby carrots Canada, 4 cases

April 2008 Salmonella Cantaloupe Canada, United States and Mexico, 64 confirmed

cases

June 2008 Salmonella Tomatoes/peppers United States and Canada, 1442 confirmed cases

September 2008 E. coli O157:H7 Lettuce United States and Canada; 134 confirmed cases

September 2008 Salmonella Alfalfa sprouts United States, 14 confirmed cases

November 2008 Salmonella Basil UK, 32 confirmed casesDecember 2008 Salmonella Alfalfa sprouts United States, recall for suspected contamination


(Anonymous, 2006; Thunberg et al., 2002; Valentin-Bon et al., 2008).However, there is a growing body of evidence supporting the hypothesisthat certain pathogens are adapted to persist on different produce types(Bassett and McClure, 2008). For example, foodborne illness outbreakslinked to tomatoes have commonly implicated Salmonella (Barak andLiang, 2008; Greene et al., 2008) (Table 4.1). The same pathogen is alsoassociated with cantaloupes, sprouted seeds, and lettuce (Arthur et al.,2007; Mohle-Boetani et al., 2008; Sivapalasingam et al., 2004). E. coli O157:H7 has been associated with sprouted seeds, lettuce, apples ( juice), andspinach ( Jablasone et al., 2005; Sivapalasingam et al., 2004; Valentin-Bonet al., 2008). Parsley is prone to contamination from Shigella and soft fruitwith enteric viruses such as Hepatitis A (Anonymous, 1999; Jacobsonet al., 2004; Mataragas et al., 2008; Naimi et al., 2003; Peterson et al., 1983;Reller et al., 2006; Seymour and Appleton, 2001; Wu et al., 2000). A morecurious associated is the link between basil and Cyclospora (Chacin-Bonilla, 2007; Sivapalasingam et al., 2004). The underlying reasons forsuch associations remain obscure as is many aspects on the interactionsof human pathogens with fresh produce.

Although foodborne illness outbreaks linked to fresh produce have beenrecorded over 30 years, there has been a rapid increase in the number ofcases recorded (Fig. 4.1). In 2005, the largest outbreakof salmonellosis linkedto mung bean sprouts occurred within Ontario (Table 4.1). The implicatedSalmonella serovarwasEnteritidis,which ismore commonly associatedwithpoultry and raw eggs. How the S. Enteritidis strain became associated withthe mung beans sprouts remains open to speculation although one majorsprout producer within the province was targeted as the cause of theoutbreak. The inability to trace human pathogens implicated in fresh pro-duce foodborne illness outbreaks back to the original source is not uncom-mon and the ‘‘smoking gun’’ is rarely found. This can be attributed to therelatively short shelf life of fresh produce, which is often discarded by thetime an outbreak is identified. The lack of traceability of produce is a furthercontributing factor thatmakes identification of specific sources problematic.However, the 2006E. coliO157:H7 outbreak linked to baby spinachAmericawas unique in that the strain of the pathogen was recovered from theinfected people, spinach within unopened bags, and the farm where thecropwas cultivated (Cooley et al., 2007). The route of spinach contaminationwas considered to be through the transfer of E. coli O157:H7 from a cattleranch near the field via infected wild pigs that found access to the cropthrough a broken fence. Yet, it is noteworthy that a survey of the Salinasvalley in the summer of 2006 found a high prevalence of E. coli O157:H7within the area, suggesting the actual route could have been via contami-nated irrigation water (Cooley et al., 2007).

Other notably outbreaks linked to fresh produce occurring in 2006included lettuce, sprouts, cantaloupes, and Clostridium botulinum linked


to pasteurized carrot juice (Doyle and Erickson, 2008) (Table 4.1). Thelatter was of specific interest given that the neurotoxin levels detected incontaminated product were the highest ever recorded (FDA, 2006). Aswith the majority of fresh produce outbreaks the exact sequence of eventsthat led to the six cases of botulism within the United States and Canadaremains unexplained.

One of the largest foodborne illness outbreaks linked to fresh produceoccurred primarily within the southern states of the US in 2008. The initialcause of the Salmonella Saintpaul outbreak was identified as tomatoes(Centers for Disease Control and Prevention (CDC), 2008; Lang, 2008).However, the failure to recover the Salmonella serovar from tomatoesshifted focus to peppers and cilantro (common ingredients of salsa)from Mexico. The outbreak lasted over 80 days and resulted in over1400 cases being recorded (CDC, 2008). The actual number of cases isprobably 10–30 times this figure given that the majority of illness out-breaks go unrecorded. Despite the number of cases involved in the Salmo-nella Saintpaul outbreak, no specific source or ‘‘smoking gun’’ wasidentified. However, the outbreak, like those within recent years, high-lighted key deficiencies within the fresh-cut chain. Specifically:

� Inability to control the dissemination of human pathogens within theenvironment;

� Failure of post-harvest interventions to remove field acquiredcontamination;

� Lack of traceability to track contaminated produce back to the source.

The knowledge gaps associated with the microbiological safety withinthe fresh produce chain are significant. Despite the large body of researchdevoted to fresh produce, there still remains much to be known about thesurvival of pathogens within the environment and nature of interactionswith growing plants. To understand the nature of human pathogen inter-actions with plants and survival in the environment, it is informative toprovide a brief overview of the characteristics of those implicated in themajority of recorded outbreaks (Table 4.2).

III. CHARACTERISTICS OF PATHOGENS RECOVEREDFROM SALAD VEGETABLES

A. Pathogenic E. coli

Nonpathogenic (generic) E. coli is a normal inhabitant of the gastrointes-tinal tract of humans and animals. However, some E. coli strains have nowacquired virulence factors enabling them to cause disease of the gastroin-testinal, urinary, or central nervous system. Pathogenic E. coli can be

TABLE 4.2 List of pathogenic bacteria and symptoms

Pathogen Incubation period Infectious dose and symptoms Significant sources

Bacterial pathogens

Aeromonas

hydrophilia

Unknown,

symptoms can

last for severalweeks

Gastroenteritis, septicemia, cellulitis, colitis,

and meningitis

Water, sewage

Dose: Unknown but thought to be high (>109

cfu)

Bacillus cereus 6–15 h diarrheal

type

Diarrheal type: watery diarrhea, abdominal

cramps

Soil, starchy grains

0.5–6 h emetic Emitic: vomiting, occasional abdominal cramps

and/or diarrhea. Dose >106 cfu

Campylobacter jejuni 2–5 days Diarrhea which may be watery or sticky and

can contain blood. Guillain–Barre syndrome.

Dose: >500 cfu

Manure especially derived

from poultry production.

Raw poultryClostridium

botulinum

18–36 h Neurotoxin affects nervous system leading to

lethargy, weakness and breathing difficulty.

LD50 3 ng/kg

Soil, sediments, water

Escherichia coli

O157:H7

24–48 h Severe abdominal pain and diarrhea which is

initially watery but becomes grossly bloody.

Hemorrhagic colitis. > 100 cells

Manure from ruminants,

sewage, raw beef

Listeria

monocytogenes

1–90 weeks Flu-like symptoms that may develop into

septicemia, meningitis and encephalitis. Stillbirth or abortion in pregnant women. >104

Manure, sewage, soil, silage

Salmonella 24–48 h Nausea, vomiting, abdominal cramps,

diarrhea, fever, and headache. >103Manure, soil, wild and

domestic animals, sewage.

Raw meat especially

poultry

(continued)


Pathogen Incubation period Infectious dose and symptoms Significant sources

Shigella sonnie 12–50 h Abdominal pain; cramps; diarrhea; fever;

vomiting; blood, pus, or mucus in stools;

tenesmus >10 cells

Manure, sewage

Enteric viruses

Norwalk Like Virus(Norovirus)

24–48 h Nausea, vomiting, diarrhea, andabdominal pain

Hepatits A 10–50 days Sudden onset of fever, malaise, nausea,

anorexia, and abdominal discomfort,

followed by jaundice

Sewage, infected food

handlers and water

Dose: >20 virons

Enteric protozoan

Cryptosporidium

parvum

2–4 days Severe watery diarrhea Domestic animals, manure,

sewage, infected foodhandlers and water

Dose: >10 cells

Cyclospora

cayetanensis

7 days Diarrhea which can last up to 6 weeks Sewage and water

Giardia lamblia 7 days Diarrhea which can last 1–2 weeks Sewage, manure, wild and

domestic animals,

infected food handlers

and waterDose: >1 cysts

Adapted from FDA Bad Bug Book (2008).


subdivided into five different categories based on the type of clinicalcondition they cause although all share common linkages.

All pathogenic E. coli stains follow a similar strategy of infection bycolonizing the intestinal mucosal cells. The mode in which illness occursvaries between the different pathogenic E. coli types. ETEC and EaggECproduce enterotoxin, EIEC invades the epithelial cells with EPEC andEHEC adhering to the cell and modifying cellular activity.

Although all pathogenic E. coli represent a significant health risk, thosebelonging to the EHEC group are of most concern, especially E. coli O157:H7 (Weiner and Osek, 2007). The reason for the high virulence of EHEC isthrough the production of Shiga-like toxins (verotoxin or verocytotoxin).The genes for Shiga toxin are believed to have been horizontally trans-ferred to E. coli from Shigella via bacteiophage. There are two toxins(encoded by Stx 1 and Stx 2) that act by cleaving a single adenine residuefrom 28S rRNA belonging to the ribosomal subunit resulting in theshutdown of protein synthesis. The kidney is rich in receptors for attach-ment of E. coli O157:H7 and consequently toxicoinfection by the bacte-rium can be accompanied by renal failure (HUS syndrome) (Table 4.2).

Although E. coli O157:H7 is considered the most significant EHECstrain, it must be noted that other non-O157 Shiga-toxin producingtypes such as O111, O145, O113, O103, O91, O26, and O104 also exist(Bower, 1999). Collectively, all E. coli possessing toxin genes are categor-ized as Shiga-toxin E. coli or STEC. However, the presence of stx genes isonly one of several virulence factors required to cause illness (McNallyet al., 2001). It is interesting to note that E. coli O157 and non-O157serotypes associated with animals contain only half the virulent factorscompared to those of clinical isolates ( Johnson et al., 2004). Therefore, themost virulent STEC have a tendency to be harbored by humans or intro-duced to animals in contact with sewage ( Johnson et al., 2004).

The main source of E. coli O157:H7 is from the manure of ruminants(cattle, sheep) and sewage (Chase-Topping et al., 2008). Other livestockand wildlife have lower frequency of carriage. Although the estimates ofSTEC vary seasonally, and between herds, approximately 2–100% ofcattle harbor E. coli O157:H7 (Hancock et al., 1997). In a 12-month abattoirstudy in Great Britain, Milnes et al. (2008) determined the fecal carriage ofSTEC O157 to be 4.7% in cattle, 0.7% in sheep, and 0.3% in swine.Conversely, in another British study, Hutchison et al. (2005) isolatedE. coli O157:H7 in 13% of fresh cattle manure, 21% of fresh sheep manure,and 12% in fresh swine manure.

ETEC, EIEC, and EAggEC have been previously recovered from con-taminated vegetables (Robins-Browne, 2007; Scavia et al., 2008) and are amajor cause of diarrhea, especially in infants. The three groups of patho-genic E. coli can be water or foodborne although typically transmittedthrough person-to-person contact. EPEC, a further type of pathogenic


E. coli, is almost exclusively transferred via person-to-person contactalthough it has also been implicated in sporadic cases of foodborne illness(Ochoa et al., 2008).

B. Shigella

Shigella sonnei has been implicated in several vegetable related foodborneillness (Table 4.1) outbreaks although it is normally associated withperson-to-person contact (Solodovnikov et al., 2008). Although E. coliO157:H7 and Shigella share pathological traits, the latter is less tolerantto environmental stress (Islam et al., 1996). Therefore, in the majority ofcases, infected food workers are considered the primary source of Shigella.However, outbreaks of foodborne illness associated with lettuce contami-nated at preharvest with Shigella have occurred. The most notable was anoutbreak involving contaminated iceberg lettuce imported from Spaininto the United Kingdom, Norway, and Sweden. Subsequent investiga-tions identified that irrigation water contaminated with human sewagewas the source of the pathogen (Kapperud et al., 1995).

C. Salmonella

The genus Salmonella includes over 2700 serovars, 200 of which are com-monly associated with human illness with S. Typhimurium and S. Enter-itidis being the most prevalent (Franz and van Bruggen, 2008). Salmonellais carried within the gastrointestinal tract of wild animals, poultry, pigs,and humans. However, Salmonella recovered from vegetables typicallybelong to less common serotype groups, for example Newport orMontevideo (Franz and van Bruggen, 2008).

There is concern with regard to the distribution of multidrug-resistantSalmonella within the food chain. It is commonly believed that the use ofantibiotics as animal growth promoters has led to the prevalence ofresistant serovars. Yet, through studying the epidemiology of Salmonella,it has become evident that the overprescription of antibiotics along withmisuse (e.g., failure to complete a course of the drug) has played asignificant role in the emergence of resistant strains (Kelly et al., 2004).The possible use of antibiotics to suppress plant pathogens has beenconsidered as a possible route by which Salmonella and other humanpathogens can acquire resistance. Although this may seem unlikely, it isinteresting to note that streptogramin-resistant Enterobacter faecium hasbeen previously isolated from bean sprouts (Snary et al., 2004).

Similar to E. coli, the main transmission route of Salmonella to vegeta-bles is through fecal contamination, cross-contamination and foodhandling. In a 12-month abattoir study, Milnes et al. (2007) determinedthe fecal carriage of Salmonella to be 23.4% in swine 1.4% in cattle, and


1.1% in sheep. Hutchison et al. (2004) isolated Salmonella from 8% of freshcattle manure, 8% of swine manure, 18% of poultry manure, and 8% ofsheep manure; levels in stored manure were lower.

Salmonella have been isolated from a broad range vegetables especiallysprouted seeds (Brandl, 2006; Johnston et al., 2005). An interesting featureof Salmonella associated with vegetables (and other environmental sources)is the tendency to have low virulence compared to those isolated fromclinical sources (Herikstad et al., 2002; Olsen et al., 2001; Sivapalasingamet al., 2004). Evidence is accumulating to suggest that genes present withinSalmonella enhance the survival of the pathogen outside the host environ-ment. Significantly, mutants of Salmonella lacking such genes have highervirulence than their parent strain (Winfield and Groisman, 2004). There-fore, a number of Salmonella appear to have enhanced their survival in theenvironment at the expense of virulence. However, this does not imply ofcourse that Salmonella associated with fresh produce represents a low risk.

D. Campylobacter

Campylobacter jejuni is a normal commensal of the gastrointestinal tract ofpoultry, pigs, and cattle. In a 12-month abattoir study, Milnes et al. (2007)determined the fecal carriage of thermophilic Campylobacter to be 54.6% incattle, 43.8% in sheep, and 69.3% in swine. Hutchison et al. (2004) isolatedCampylobacter from 13% of fresh cattle manure, 14% of swine manure, 19%of fresh poultry manure, and 21% of fresh sheep manure; levels in storedmanure were significantly lower. Human carriers also represent signifi-cant vehicle by which the pathogen can be transferred to foods.

Campylobacter is notoriously fastidious and has very specific growthconditions. The bacterium can survive for short periods outside the hostenvironment but not to the same extent as Salmonella and E. coli (Alter andScherer, 2006;Garenaux et al., 2008;Mihaljevic et al., 2007).However, despitesuch fragility, C. jejuni, and to a lesser extentCampylobacter coli, has been themain cause of gastroenteritis for several years (Janssen et al., 2008). This islikelydue to the low infectiousdose (<500 cells) required tocause symptomsin susceptible hosts and the high carriage rate in livestock (Ozcakir, 2007).

C. jejuni invade and become established in epithelial cells of the lowerintestine whereupon a cholera-like toxin is secreted. The main symptomassociated with the disease is perfuse diarrhea that can last between 2 and14 days although is rarely life threatening. There has been an increase inthe recovery of antibiotic-resistant C. jejuni from human isolates althoughnot from animals (Pumbwe and Piddock, 2004).

C. jejuni have been recovered from vegetables especially root crops(Brandl et al., 2004). However, evidence to date suggests that the mainsource of Campylobacter recovered from vegetables occurs via cross-contamination events in food service outlets and the domestic environment.


E. Listeria monocytogenes

Unlike typical enteric bacteria, L. monoytogenes has adapted to survive inboth the host and nonhost environment. Because L. monocytogenes iswidely distributed in nature, the pathogen is a common contaminant ofvegetables, especially root crops (Embil et al., 1984; Lianou and Sofos,2007; Swaminathan and Gerner-Smidt, 2007). Hutchison et al. (2004)isolated Listeria from approximately 30% of fresh cattle manure, 20% ofswine manure, 19% of poultry manure, and 29% of sheep manure. Thelatter is of relevance given that the first identified foodborne listeriosisoutbreak was linked to coleslaw (cabbage) fertilized with uncompostedsheep manure (Embil et al., 1984).

The virulence of L. monocytogenes is often under estimated consideringthe pathogen causes serious illness that frequently results in death(Drevets and Bronze, 2008). Similar to other human pathogens, the viru-lence of L. monocytogenes that can tolerate environmental stress is lowerthan those recovered from clinical cases of listeriosis (Chan et al., 2007).

F. Aeromonas hydrophila

Aeromonas are widely distributed in the environment especially water butcan occur in human feces (Campo et al., 2001). Two distinct types of gastro-enteritis have been associatedwithA. hydrophila, a cholera-like illnesswith awatery diarrhea and a dysenteric illness characterized by loose stools con-taining blood and mucus (Pund and Theegarten, 2008). The ability of Aero-monas to cause illness depends on the presence and expression of virulencefactors. Although the bacterium is frequently recovered fromwater in highdensities (105–109/100 ml),very fewstrainshave the capacity tocause illnessin humans. This has been related to the lack of complete virulence factorsand high tolerance of the host (Chifiriuc et al., 2007; Pund and Theegarten,2008). Therefore, althoughwidely distributed on vegetables, the pathogen ismore significant in ready-to-eat foods such as meat and soft-cheeses.

G. Endospore-forming bacteria

Spores of Cl. botulinum and Cl. perfringens can be found both in soil andvegetables. Because clostridia are obligate anaerobes, their growth isrestricted on fresh produce stored aerobically or under hyperoxygenatedmodified atmosphere. However, there is a trend to preserve vegetables inoil which is more conducive to the growth of clostridia (Barker et al., 2005;Carlin and Peck, 1996; Peck, 2006). Indeed, there have been several botu-lism outbreaks linked to vegetable-in-oil products (Lohse et al., 2003).

Bacillus cereus is an aerobic spore-forming bacterium that is widelydistributed in soil and on plant material. Therefore, its occurrence of


vegetables is not uncommon especially with regard to leafy vegetablesand sprouts (Joo et al., 2004; Kim et al., 2004).

H. Enteric viruses

Enteric viruses only reproduce within the human host and all follow thefecal–oral route of transmission (Girard et al., 2006). The most significantcharacteristic of enteric viruses is the low infectious dose to cause illness(<20 virons) and the ease by which they can be transferred from person toperson (Mattison et al., 2007). Enteric viruses are also very stable withresistances to environmental stresses comparable to those associated withbacterial endospores (Mattison et al., 2007).

The majority of foodborne illness associated with enteric viruses areshort-lived and not life threatening (Kurdziel et al., 2001). Tracing sourcesof enteric viruses is problematic due to the lack of routine detection tech-niques combined with under-reporting of outbreaks (Kurdziel et al., 2001).

The majority of outbreaks linked to enteric viruses is typically causedby person-to-person contact although fresh produce can also be a signifi-cant vehicle, especially with regards to soft fruit which are handled andrarely washed prior to consumption (Koopmans, 2008). Both hepatitis Aand Norwalk-like viruses (NLV) have been implicated in cases of food-borne illness associated with contaminated vegetables (Koopmans, 2008).In such outbreaks, the crops had been directly exposed to sewage or hadbeen handled by infected workers (Holtby et al., 2001; Long et al., 2002).

A high profile hepatitis A outbreak associated with green salad onionswas reported in the United States in 2003 (Chancellor et al., 2006). Theinitial outbreaks were centered on a restaurant in Pennsylvania thatresulted in 575 cases of hepatitis A and 1 death (Vale, 2005). Althoughthe restaurant was initially identified as the source of the outbreak,subsequent investigation linked other cases in Tennessee and Georgia.Further inspection of the farm in Mexico was later identified as the mostlikely source of the virus (Vale, 2005).

Interestingly, the persistence of viruses such as polio has been shownto be dependent on the vegetable type. When introduced onto lettuce orcabbage, a 1 log reduction in polio virus was observed over 8 days. Incontrast, viruses introduced onto green onions remained stable for over14 days (Kurdziel et al., 2001). The underlying factors associated with thepersistence of enteric viruses on fresh produce remain to be elucidated.

I. Human pathogenic protozoa

Human pathogenic protozoa such as Giardia, Entamoeba, Toxoplasma,Sarcocystis, Isopora, Cryptosporidium, Eimeria, and Cyclospora can be trans-ferred via fecally contaminated water or vegetables (Armon et al., 2002).


Similar to enteric viruses, protozoa require a suitable host for replicationbut can persist within nonhost environments for significant time periods(Sidhu and Toze, 2009). The main source of human protozoan is fromdirect contact with humans although foodborne sources can also repre-sent a significant vehicle (Sidhu and Toze, 2009; Thompson et al., 2008).All of these human pathogenic protozoa cause diarrhea-like symptomsexcept Toxoplasma, which causes fetal damage and glandular fever-likesyndrome (Dumetre and Darde, 2003).

The prevalence of Cryptosporidium and Giardia from human, agricul-tural, or wildlife sources was examined by Heitman et al. (2002). Theresearchers found that while sewage effluent had the highest prevalenceof the parasites by far, the greatest concentrations were detected in cattlefeces. However, direct linkages, using molecular techniques, betweencryptosporidiosos outbreaks and contamination from livestock sourcesare limited. Cryptosporidium has been isolated from 5% of fresh cattlemanure, 13.5% of swine manure, and 29% of sheep manure. Giardiaprevalence is similar to that of Cryptosporidium being found in 4%, cattle,2% swine, and 21% sheep manure samples (Hutchison et al., 2004;Thompson et al., 2005).

IV. TRANSMISSION OF HUMAN PATHOGENSIN MANURE, SOIL, AND WATER TO THEVEGETABLE PRODUCTION CHAIN

Enteric pathogens have to be introduced into the production chain atsome point in order to contaminate fresh produce. Direct fecal contami-nation of vegetables just prior to consumption represents the greatest risk(Gorny, 2002; Mukherjee et al., 2007). However, other sources of contami-nation such as manure amended soil and irrigation water are morecommonly encountered (Hutchison et al., 2008; Islam et al., 2004).

Enteric pathogens on growing vegetables are generally believed to bein survival mode as opposed to actively growing. Previously, it has beenconsidered that enteric bacteria such as E. coli only survive for 2–3 daysafter being excreted by the animal host. However, such generalizationsare inappropriate considering that E. coli populations within the gastroin-testinal tract of animals can consist of over 1000 distinct genotypes(Gordon et al., 2002). More significantly, the E. coli associated with animalshave a broad range of survival abilities within nonhost environments. Ithas also been found that those genotypes/strains that dominate theenteric environment have relatively poor survival outside the host envi-ronment (Whittam, 1989). Evidence obtained to date would also suggestthat strains of Salmonella and E. coli O157:H7 also exhibit a range ofsurvival abilities during the transition from the gastrointestinal tract to


the environment (Winfield and Groisman, 2003, 2004). This has naturallycomplicated studies attempting to determine the relative survival ofenteric pathogens in the environment. In this respect it is likely thatmany studies previously performed (especially with laboratory strains)underestimated the tolerance of pathogens to environmental stresseshence their persistence in manure, soil and water.

A. Manure and biosolids

Farm and sewage effluents represent the most significant source ofhuman pathogens recovered in water, soil and hence, vegetables(Baloda et al., 2001; Beuchat and Ryu, 1997; Davies et al., 2008; McGrathet al., 1995; Rideout and Teschke, 2004). Manure is predominantly used inorganic cultivation systems but less so by conventional growers (USDA,2001). Although organic produce is thought to represent a significancerisk with regards to carriage of enteric pathogens, no data have beenreported to confirm this view (Loncarevic et al., 2005). The application ofuntreated manure or sewage to growing crops is a direct route by whichvegetables can be contaminated (Culley and Barnett, 1984; Franz andvan Bruggen, 2008). For example, Cieslak et al. (1993) isolated E. coliO157:H7 from lettuce cultivated in a garden in which the soil wasamended with fresh manure. Under normal conditions the direct contactof manure with vegetables should not occur since, either a treatment stepis applied prior to disposal of effluent into soil or water, or there is asignificant time interval between application and crop production. Oftenregulatory agencies will identify an appropriate minimum time delaybetween manure application and fruit or vegetable harvest. For example,Canadian regulations specify 3 months for tree fruits and grapes, 15months for small fruits, and 12 months for vegetables. The USDA‘‘organic production and handling requirements’’ specify that, unlesscomposted, raw animal manure must be incorporated into the soil notless than 120 days prior to harvest of a product whose edible portion hasdirect contact with the soil surface or soil particles, or 90 days if there is nodirect contact (USDA, 2005).

Manure and sewage waste can undergo a variety of treatments such ascomposting, aerobic and anaerobic digestion, alkaline stabilization, con-ditioning, dewatering, and heat drying. The Environmental ProtectionAgency specify that biosolids derived from manure treatment destinedfor general fertilizer must have fecal coliform counts <1000 cfu/g, Salmo-nella < 4 cfu/g, and enteric viruses at <4 plaque-forming units/g bioso-lids (Mechie et al., 1997). Similarly, Canadian regulations specify fecalcoliforms <1000 MPN/g dry weight of total solids or no Salmonella witha detection level <3 MPN/4 g dry weight total solids (CCME, 2005).

In the main, proper manure sewage/manure treatment (e.g., compost-ing) can be sufficient to inactivate human pathogens. However, the


treatment required to assure adequate reduction of enteric pathogens inmanure remains a debatable issue. Current treatment regimes tend to bebased on studies that evaluated the survival of endogenous or artificiallyintroduced human pathogens in manure held under a narrow range ofconditions. Therefore, the reported persistence of human pathogens inmanure tends to vary significantly (Cote and Quessy, 2005). However, it iswidely accepted that the survival of bacterial and viral human pathogensis dependent on temperature, aeration and holding time, dry mattercontent, pH, and bacterial concentration (Ingram and Millner, 2007;Leifert et al., 2008; Mannion et al., 2007; Semenov et al., 2007; You et al.,2006).

E. coli O157:H7 can survive in high moisture content bovine manure forover 70 days at 5 �C that compared to 49 days at 30 �C (Semenov et al., 2007).The persistence of Salmonella in manure is also favored under low tempera-ture and highmoisture conditions (Mannion et al., 2007). However, survivalof both pathogens in manure slurry is restricted to a maximum of 10 days(Cote et al., 2006). Laboratory studies on E. coli O157:H7 and S. Typhimur-ium in cowmanure and cow slurry indicated that death rate is a function oftemperature and depth in the storage container (Himathongkham et al.,1999). Decimal reduction times varied from 6 days to 3 weeks, with thedecline being fastest at 37 �C compared to the colder temperatures.The authors suggested that this data should enable predictions of storageconditions that would lead to predetermined levels of reduction of the twopathogens.

Several studies of inoculated pathogens in stored manure haveattempted to determine survival rates. Hutchison et al. (2005) examinedthe declines of inoculated Salmonella, E. coli, Campylobacter, Listeria, andCryptosporidium in summer and winter storage in 35,000 L tanks of freshwaste (slurries, dirty waters). D-values (days for 1 � log cycle decrease)ranged from 6 to 44 days, and generally in the order Campylobacter (meanof 10.7 days)< Listeria (13.9 days)< Salmonella (14.6 days)< E. coliO157<Cryptosporidium parvum (232 days). Interestingly, mean summer and win-ter decline rates were similar, and slurry dry matter was not found to besignificant. The authors found that less than 6 months passive storage wassufficient to reduce bacterial populations below detectable levels, butinsufficient for significant reduction of C. parvum. In a similar study,Nicholson et al. (2005) examined solid and liquid manures mixed withE. coli O157:H7, Listeria, Campylobacter, and Salmonella (dairy and pigsolid, broiler litter, dairy slurry at 7% and 2% dry matter, and dirtywater). They found that survival in solids varied from 2 to 32 days, andwas lowest in turned and unturned piles when the temperature exceeded50 �C. Survival was found to be in the order Campylobacter < Listeria <Salmonella < E. coli in solid manure, and Salmonella < Listeria ¼ E. coli <Campylobacter in liquid manure and dirty water.


Treatment of manure piles through mixing or composting can increasepathogen kill-off. Nicholsen et al. (2005) demonstrated survival of E. coliO157:H7 in sheep manure in aerated piles up to 4 months, but in staticpiles up to 21 months. In aerated bovine manure piles, the organismpersisted for 47 days. However, Fleming andMacAlpine (2002) did estab-lish rapid, temperature-related E. coli and Salmonella die-off during com-posting of liquid swine manure mixed with various materials.Composting temperatures that exceed 55 �C for 3 days are generallyconsidered to eliminate most pathogens (Duffy, 2003; Grewal et al., 2006;Turner, 2002; Turner et al., 2005).

C. jejuni persistence in manure has traditionally thought to be compar-atively low, being 3 days in cattle manure and 2 days in sewage(Hutchison et al., 2005). However, in field studies, Campylobacter hasbeen detected in undisturbed liquid swine manure held in tanks forseveral weeks, and storage under winter conditions (frozen surface) pro-longs survival (Huber, unpublished results).

Unlike vegetative cells, enteric virus can persist in sewage for up to 4months under low temperatures and high moisture (Divizia et al., 2008;Gino et al., 2007).

B. Irrigation water

Irrigation water is susceptible from direct contamination from sewage ormanure spills. In the course of heavy rainfall, contamination associatedwith manure heaps can leach into water courses (Heaton and Jones, 2008;Lazarova and Savoys, 2004; Ottoson and Stenstrom, 2003). The level ofcontamination from these sources varies significantly. General populationlevels of E. coli in variety of potential contamination sources are presentedin Table 4.3.

Irrigation water used in crop production represents one of the mostsignificant sources of contamination in fresh produce production.Contamination from manure heaps can readily be transferred to thewater course via run off and subsequently disseminated over large dis-tances via streams or rivers. When contaminated, water is used to irrigatecrops, human pathogens can be directly transferred to plant tissue andpersist through to harvest.

Salad crops irrigated with water contaminated with sewage wasresponsible for numerous cases of typhoid fever and hepatitis A inSantiago, Chile. Lettuce irrigated with contaminated water can accumu-late E. coli O157:H7 over repeated exposures (Franz and van Bruggen,2008). Lettuce plants exposed to contamination 7 days or less prior toharvest represent a greater risk than contaminated irrigation water intro-duced early in the cultivation period (Solomon et al., 2002). GeneralCanadian guidelines require that irrigation water for food crops contain

TABLE 4.3 Levels of Escherichia coli encountered in different manure,

effluent and waste water

Contamination source

Level of E. coli

in source (log10cfu/100 ml) Data source

Liquid manure 5–7 Huber (unpublished data)

Runoff from manure piles 4–6 Huber (unpublished data)Raw sewage 6–8 Servais et al. (2007)

Dairy wash water 0–4 Huber (unpublished data)

Fruit and vegetable wash

water

0–3 OMAFRA (unpublished

data)

Streams through

grazing land

4 Servais et al. (2007)

Surface water through

cropland

3 Servais et al. (2007)


less than 100 cfu fecal coliforms/100 ml (CCME, 2005); some provincialrequirements may be lower (e.g., B.C. regulations for E. coli < 77 cfu/100 ml).

Recycling of municipal waste water (gray water) for irrigation pur-poses have been implemented in several countries such as Australia,Germany, Israel, Spain, Holland, and the USA. However, studies haveillustrated that the practice of using recycled water may increase therisk of introducing human pathogens (Gross et al., 2007; O’Toole et al.,2008). For example, onions and garlic cultivated with treated municipalwaste water harbored unacceptable levels of Salmonella and E. coli atharvest (Fasciolo et al., 2002). Regulations for recycled water vary fromaverage nondetectable to 200 cfu/100 ml depending on US state andcrop (Table 4.4). For food crops, recycled water has had to undergotertiary treatment, usually involving disinfection, again depending onthe state. Regulations for other pathogens are not specified, but Floridarequires monitoring for Cryptosporidium and Giardia on a regular basis.

The method of applying irrigation water can also enhance the intro-duction of human pathogens to growing vegetables (FDA, 1998). Thevarious irrigation modes used for vegetables include gravity (flood)irrigation, spray irrigation, drip/trickle irrigation, and subirrigation(FDA, 1998). Many factors, such as water availability and cost, soiltype, slope, depth of water table, economics, and cropping rotations,determine the mode of irrigation rather than food safety issues. Floodand spray irrigation represent the greatest risk as any contaminationwithin the water is directly deposited onto the edible leaves of crops(FDA, 1998).

TABLE 4.4 Recycled water standards in different states with the US

Arizona California Florida Hawaii Nevada Texas Washington

Treatment Secondarytreatment,

filtration and

disinfection

Oxidized,coagulated,

filtered and

disinfected

Secondarytreatment,

filtration and

high-level

disinfection

Oxidized,filtered and

disinfected

Secondarytreatment

and

disinfection

NS Oxidized,coagulated,

filtered and

disinfected

BOD5 NS NS 20 mg/l NS 30 mg/l 5 mg/l 30 mg/l

TSS NS NS 5 mg/l NS NS NS 30 mg/l

Turbidity

(NTU)

2–5 2–5 NS 2 NS 3 5

Fecal Total Fecal Fecal Fecal Fecal TotalColiform

average

ND 2.2/100 ml 75% below level

of detection

2.2/100 ml 200/100 ml 20/100 ml 2.2/100 ml

Coliform

maximum

23/100 ml 23/100 ml 25/100 ml 23/100 ml 400/100 ml 75/100 ml 23/100 ml

Source: Environmental Protection Agency (2004).


There have been numerous studies performed to determine thepersistence of human pathogens within irrigation water (Ajwa et al.,2002; Armon et al., 2002; Bastos et al., 2008; Fasciolo et al., 2002; Islamet al., 2004; Ottoson and Stenstrom, 2003) (Table 4.5). There are diffi-culties in relating the relative persistence of enteric pathogens in watersince no standardized experimental protocol is followed. In addition,the occurrence of viable but noncultureable (VBNC) populations ofpathogens can further complicate survival values (Lee et al., 2007; Liuet al., 2008; Ozcakir, 2007). For example, Salmonella species (includingS. Typhimurium DT 104) introduced into autoclaved river waterdecreased from 8 log to 5 log cfu/ml over a 45-day period at 23 �C(Santo Domingo et al., 2000). However, when the VBNC cells wereexamined, less than 1 log of reduction was observed. Lower tempera-tures also extended the persistence of enteric pathogens in water(Santo Domingo et al., 2000). In general, the persistence of Salmonellais greater than that of E. coli O157:H7 which in turn persists for longertimes than Campylobacter (Krampitz and Hollander, 1999; Ottoson andStenstrom, 2003).

Survival studies of enteric pathogens are typically performed in ster-ilized (autoclaved or filtered) water samples. Although this facilitatesenumeration of the introduced pathogen, it does not provide an assess-ment of survival in natural environments. In this respect it is interesting tonote that survival of enteric pathogens in nonsterile water is significantlyshorter due to the activity of protozoan (Artiz and Killham, 2002). How-ever, protozoan can form protective niches for enteric pathogens, therebyenhancing persistence under certain conditions (Wang et al., 2005).

The persistence of E. coliO157:H7 has also been found to vary depend-ing on the source of water. Artiz and Killhem (2002) evaluated the sur-vival of E. coli O157:H7 in water sourced from four different wells. Theauthors reported that in two water samples, E. coli O157 introduced atlevels of 107 cfu/ml were reduced to below the level of detection within 10days regardless of being suspended in sterile or nonsterile water.Although not confirmed, this low level of persistence was attributed tothe presence of antimicrobial ions such as copper. The study underlinesthe difficulties encountered when attempting to predict the survival ofenteric pathogens in water.

Giardia cysts persist for a shorter period in irrigation water comparedto Cryptosporidium oocysts (Karim et al., 2004; Skraber et al., 2007). Tem-peratures as low as �4 �C inactivate Giardia cysts in water while Crypto-sporidium oocysts remained viable for >12 weeks at 4 �C (Karim et al.,2004). At 25 �C, Giardia cysts were inactivated in water within 2 weeksbut Cryptosporidium oocysts survived for >10 weeks. Factors affectingthe survival of pathogens in water have been reviewed by Bichai et al.(2008).

TABLE 4.5 Survival of human pathogens in different water sources

Pathogen Notes

Temperature

(�C) Survival Reference

E. coli O157 Sterile municipal

water

8 91 days Wang and Doyle (1998)

Sterile municipal

water

25 49 days Wang and Doyle (1998)

Sterile well water 15 1 log reduction in 70

days

Artiz and Killhem (2002)

Well water 15 65 days Artiz and Killhem (2002)

Sterile well water 15 10 days Artiz and Killhem (2002)

Salmonella Sterile municipal

water

23 2 log reduction after 45

days

Santo-Domingo et al.

(2000)River water 23 3 log reduction after 45

days

Santo-Domingo et al.

(2000)

Sterile well water 18 152 days Mitscherlich and Marth

(1984)

Campylobacter Sterile municipal

water

4 8–28 days Terzieva and McFeters

(1991)

Sterile municipal

water

37 22 h Terzieva and McFeters

(1991)Yersinia

enterocolitica

Sterile spring water 4 446 days Karapinar and Gonul

(1991)

River water 16 6 days Chao et al. (1988)

Groundwater 30 10 days Chao et al. (1988)

Rotavirus Groundwater 15 2 log reduction in 5 days Gerba (1999)


C. Soil

Soil is a natural habitat for several human pathogens such as B. cereus,Cl. botulinum, and Cl. perfringens, L. monocytogenes, and Aeromonas (Waageet al., 1999). The aforementioned bacteria have adapted to survival in soilwith spores persisting for indefinite periods.

The persistence of enteric pathogens in soil is dependent on severalfactors. For example, the survival of E. coli is prolonged in clay soils whereabsorption of cells to the soil particles provides protection against proto-zoa (Lang and Smith, 2007; Mosaddeghi et al., 2009; Wong et al., 2008).E. coli O157:H7 has been reported to persist for 25 weeks in loam and claysoils, but for 8 weeks in sandy soils (Lang and Smith, 2007). Persistence ofenteric pathogens is also extended in moist soils at cool temperatures(Lang and Smith, 2007). Salmonella has higher persistence in soil com-pared to E. coli O157:H7. When S. Typhimurium was inoculated at 8 log10CFU/g into moist soil, stored at 20 �C, less than 2 log reductions innumbers were observed after 45 days (Guo et al., 2002). However, undernatural environmental conditions, S. Typhimurium introduced via hogmanure only persisted for 14 days (Sengelov et al., 2003). Campylobacter isless persistent compared to both E. coli O157 and Salmonella but neverthe-less can be recovered 20 days after introduction into soil (Nicholson et al.,2005).

The survival of common manure-derived pathogens in soils wasreviewed by Nicholson et al. (2005), who concluded that maximum sur-vival ranged from 45 to 100 days with an average log reduction of 1.94days for E. coli O157, L. monocytogenes, Salmonella, and C. jejuni. The timefor a 1� log reduction in C. parvum levels, however, varied between 8 and31 days.

In addition to stress, enteric bacteria also have to compete with theendogenous microflora to become established within the soil environ-ment. It has hitherto been considered that enteric pathogens competepoorly for nutrients and are susceptible to inhibition by soil-borne bacte-ria. Indeed, E. coli O157 have been shown to decline more rapidly inmanure-amended unautoclaved soils compared to autoclaved soil (Jianget al., 2002). However, Ibekwe et al. (2004) reported that introduction ofE. coli O157:H7 into soil increases the diversity of microbial populations.This would suggest that enteric pathogens, rather than being integratedinto soil microflora, can actually modify the microecology. Whether thiseffect enhances persistence has yet to be elucidated.

Agronomic practices can impact the survival of manure-derivedpathogens in the soil. Generally, bacteria are thought to decline morerapidly when manure is left on the surface rather than incorporated intothe soil immediately after application, presumably due to the eliminationof drying conditions and exposure to UV at the soil surface (Diaz and


Schulze-Makuch, 2006). Huber (unpublished data) determined thedecline of E. coli and Clostridium in soil amended with liquid swinemanure. E. coli declined after 28 days, following a spring application,but was still detected 55 days after a fall application, while Clostridiumremained detectable longer than 55 days after both spring and fall appli-cations. Season and location played a role in the length of time pathogensand fecal indicators were detected, suggesting that soil type or environ-mental factors such as rainfall play a role in survival. Interestingly, E. coliwas detected in tile drainage waters after rainfall up to 77 days afterapplication of manure to the soil, suggesting that these organisms canpersist and may mobilize into waterways after prolonged periods of time.

Enteric viruses can persist for up to 4 months in subsurface soil layers.In contrast, viruses on the surface are typically inactivated within days bythe antimicrobial effects of UV (Gerba et al., 2002; Moll and Vestal, 1992).Under heavy rainfall, viruses can be spread over wide areas (>150 m)especially when introduced into water courses (Santamaria and Toranzos,2003). WhileGiardia is sensitive to freezing of soil, Cryptosporidium is moreresistant. Mahdy et al. (2008) reported Giardia cysts in soil were inacti-vated after 7 days at �4 �C, but Cryptosporidium could survive for >12weeks. However, persistence of both protozoa was reduced to 8 weeks at4 �C and 4 weeks at 25 �C (Mahdy et al., 2008). Cryptosporidium have beenshown to be particularly sensitive to drying. Various studies have shownless than 5% viability following 4 h of air drying at room temperature(Anderson, 1986; Nasser et al., 2007; Robertson et al., 1992).

D. Transport of human pathogens within the environment

Movement of pathogens from manure storages or manure or biosolids-fertilized cropland to surface or groundwater can be a significant sourceof contamination of water that may be used for irrigation purposes (Berryet al., 2007; Hill et al., 2005; Muirhead et al., 2006). This can occur as theresult of over application of liquid manures or biosolids, but more com-monly occurs in response to heavy rainfall shortly following manure orbiosolids application (Muirhead et al., 2006). Potential also exists forrainfall-induced surface transport from treated cropland directly to fruitand vegetable fields.

Astrom et al. (2007) compared the total coliform and E. coli contamina-tion of surface runoff and tile effluent from two no-till field plots: one withsurface-applied liquid dairy manure and a nonmanure-treated plot.These researchers found greater bacterial populations in surface and tilewaters from the manure treated plot, but significant contaminationoccurred from the nonmanured plot attributed to bird and other wildlifefeces. Soupir et al. (2006) examined the release and transport of bacteriafrom manure applied to pastureland. Using rainfall simulations, they


found that from 30% to >100% of the released bacteria were transportedin overland flow to the edge of the field, depending on the manure type.Tyrrel and Quinton (2003) using soil flumes, found that incorporation ofslurries reduced the transport of fecal coliform in overland flow.

A long-term Ontario study by Culley and Phillips (1982) examined thebacterial quality of runoff and subsurface discharge waters (total andfecal coliforms and fecal streptococci). They found that neither the ratenor application time affected the bacterial content of spring surface andsubsurface discharge unless the manure was applied on frozen soil. Inthis case, spring snowmelt dominated runoff and significantly decreaseddischarge water quality, indicating survival over winter and subsequenttransport to surface waters.

Application of manure onto tile drained agricultural fields createssignificant potential for transportation of pathogens into source watersdirectly through tile drains, or by overland flow into surface waters. Anumber of studies have been conducted that address pathogenmovementthrough tile drains (Lapen et al., 2008). Dean and Foran (1992a,b) demon-strated contamination of tile water following liquid manure application in8 out of 12 ‘‘normal’’ farm applications. Greco (2002) demonstrated loss ofa marker E. coli to tile waters via preferential flow through macroporeswithin hours of liquid manure application and a simulated rainfall. Simi-larly, Akhand et al. (2008) demonstrated liquid municipal biosolid andprecipitation induced tile flow on silt (hence potential pathogens) loamsoils. In both studies, flow to tile drains occurred within hours of signifi-cant rainfall and was dominated bymacropore flow. The dissemination ofbacteria into tile drains can occur within hours following manure appli-cation followed by a significant rainfall. However, the transport of E. colican be reduced by pretillage of soil prior to manure application (Hubar,unpublished results).

Preventing contamination of irrigation water is problematic due to theopen nature of animal production and problems associated with manuremanagement. Nevertheless, monitoring the microbiological quality ofwater is a key intervention to reduce the risk of transferring contamina-tion to fresh produce. Furthermore, when contamination is detected inwater there is a need to rapidly identify the source and implementcontainment plans. Microbial source tracking (MST), which is a collectionof chemical, physiological, and genetic methods, can differentiate oridentify sources of fecal contamination (Nayak and Stewart-King, 2008;Parajuli et al., 2009; Reischer et al., 2008). MSTmethods are further dividedinto library-dependent and library-independent methods. Library-basedmethods require a database to be constructed using, for example, E. coli,collected from a diverse range of environmental, human, domestic, andwild animals (Santo Domingo et al., 2007; Stoeckel and Harwood, 2007).The isolate derived from the water sample is then compared with profiles


within the database to identify the possible source of fecal contamination.The underlying basis principle of library-based methods is that stains ofthe target bacterium is specific to animal or human sources, and that thepopulation is stable and has geographical structure (Barnes and Gordon,2004; Gordon et al., 2002). That is, strains, are specific to the host, thepopulation is not turned over (i.e., database is valid over long timeperiods), and strains isolated within one area can be differentiated fromthose located in a different geographical location.

The fingerprint generated byMST can be physiological (e.g., Biolog) ormore commonly using a genetic technique such as rep-polymerase chainreaction (Stoeckel and Harwood, 2007). Library-independent approachesare methods based on quantification of host-associated markers such ascoliphage and Bacteroidales, or chemicals, such as caffeine or fecal sterolsthat signifies human, as opposed to animal sources (Stoeckel andHarwood, 2007).

There have been mixed reports on the success of MST to trace sourcesof fecal contamination (Harwood et al., 2003; Myoda et al., 2003; Stoeckelet al., 2004). In the case of methods based on E. coli, the main limitationswere with respect to the reliability of the database. Specifically, E. colistrains are not entirely host specific, their populations are highly diversewith low temporal stability and poor geographical structure (Barnes andGordon, 2004; Davis and Gordon, 2002; Dixit et al., 2004; Gordon, 2001;Gordon et al., 2002).

It is generally understood that a multitiered approach, combininglibrary-based and library-independent techniques is the most reliableway to identify fecal source pollution. A number of studies have usedthis combination of approaches to identify sources of contamination, inagriculturally contaminated watersheds has been undertaken and shownpromise (Vogel et al., 2007).

V. INTERACTION OF PATHOGENS WITH FRESH PRODUCE

A. Survival in the phyllosphere

The phyllosphere (or aerial) parts of plants represent a challenge for thesurvival of microbes. The exposure to high doses of UV, fluctuations intemperature, and relative humidity all compromise viability (Heaton andJones, 2008; Whipps et al., 2008). Bacteria (epiphytes) that exist within thephyllosphere have evolved specialized mechanisms to improve stresstolerance and nutrient acquisition. Pseudomonas spp. form the predomi-nant bacterial population recovered on the leaves of plants (Brandl andAmundson, 2008; Lindow and Brandl, 2003). Epiphytic pseudomonad’sproduce fluorescent or pigmented compounds that afford protection to UV.


The hydrophobic waxy cuticle of plants can inhibit the movement andaccessibility of nutrients to bacterial cells. However, biosurfactantsproduced by the majority of epiphytic Pseudomonas spp. decreases thewater tension, enabling relatively free movement across the leaf surface tonutrient sources and natural openings such as stomata. Pseudomonas arealso known to release a toxin called syringomycin that can produce holesin the plant cell membrane allowing access to intracellular nutrientswithout necessarily resulting in disease symptoms (Cao et al., 2005).

The association of human pathogens with biofilms formed by residentepithytes is considered to enhance survival on leaf surfaces. It has beenestimated that 10–40% of the total bacteria on the surface of parsley andbroad-leaf endive are associated with biofilms (Lindow and Brandl, 2003).However, studies performed with E. coli O157:H7 or Salmonella wouldsuggest that bacterial cells tend to aggregate between the grooves ofepidermal cells rather than associate with biofilm structures (Warrineret al., 2003a).

There have been relatively few studies with regard to the survival ofhuman pathogens on the surface of leaves over long periods. However,studies using C. jejuni, E. coli O157:H7, and Salmonellawould suggest thatthis is significantly lower compared to the rhizosphere (Brandl andAmundson, 2008; Brandl et al., 2004). Nevertheless, as previously out-lined, contamination of edible leaves immediately prior to harvest wouldrepresent a significant food safety hazard.

B. Colonization of the rhizosphere

The ability to utilize the nutrients released by seeds or roots is considereda prerequisite in becoming established in the rhizosphere of plants priorto internalization (Buyer et al., 2002). In this respect, enteric pathogens,including Campylobacter, can actively grow on exudates released by plants(Brandl et al., 2004; Gagliardi and Karns, 2000). There is also evidence thathuman pathogens have extended persistence within the environment bybecoming integrated into the rhizosphere of plants. Gagliardi and Karns(2002) found that the presence of certain crops increased the persistence ofE. coliO157:H7 in soil. In unplanted, fallow soils, E. coliO157:H7 persistedonly for 25–41 days, but was found up to 92 and 96 days if alfalfa or ryeplants were present, respectively. Ibeweke et al. (2004) also found that thepresence of roots in contaminated soils increased concentrations of E. coliO157:H7. In this study, E. coli introduced through irrigation water wasfound to reach higher densities in rhizosphere soils than in nonrhizo-sphere soils (Ibeweke et al., 2004). Bacterial populations in soil increasedafter the addition of plant material to soil; the bacterial population spikedand then fluctuated in a wave-like fashion that was not found to be theresult of nitrogen shortages or pH (Ibeweke et al., 2004). In contrast to


these results, the presence of maize roots did not affect survival of E. coliin soil (Bernstein et al., 2007). Likewise, the presence of other legume cropsother than alfalfa did not increase persistence of E. coliO157:H7more thanin fallow soils (Gagliardi and Karns, 2002).

C. Internalization of human pathogens in growing plants

The possibility of pathogens such as E. coli O157:H7 and Salmonella tobecome internalized into the vascular system of growing plants hasreceived significant attention (Table 4.6). Once internalized, pathogensare protected from postharvest washing and cannot be readily removedor inactivated. It has been conclusively proved that human pathogens canenter stomata and cut edges of fresh produce (Seo and Frank, 1999;Takeuchi and Frank, 2000; Takeuchi et al., 2000). Damage caused byspoilage bacteria/fungi can also enable human pathogens to enter theinner plant tissue and thereby become protected (Brandl, 2008). However,evidence that human pathogens can infiltrate the roots of intact growingplants and become established as an endophyte has yet to bedemonstrated.

The presence of fungal endophytes within healthy tissue of vegetableswas first described in 1904 (Tan and Zou,2001). Work performed bySamish and Etinger-Tulczynsha (1962) suggested that bacterial endo-phytes also existed within plants although this was disputed for manyyears (Lund, 1992). However, it has only recently been accepted thatbacteria can indeed reside in the internal structures of undamaged plants(Rosenblueth and Martinez-Romero, 2006; Schulz and Boyle, 2005).

The endophytic bacterial population of plants is known to be diverse,comprising both Gram positive and Gram negative cells (Guo et al., 2008;Torres et al., 2008; Tyler and Triplett, 2008). For example, nitrogen-fixingbacteria from the genus Azospirillum,Herbaspirillum, Acetobacter, Azoarcus,and Burkholderia spp. are frequently encountered endophytes in nonle-gume plants (Baldani et al., 1997). Non-nitrogen-fixing endophytesinclude species of Bacillus, Pseudomonas, Corynebacterium, Micrococcus,Erwinia, Streptomyces, Rhodococcus, Microlunatus, and Luteococcus (Jamesand Olivares, 1998). Of relevance to the current review, human pathogenshave also been detected in surveys of endophytic populations of plants.For example, Salmonella, Staphylococcus, Mycobacterium, Klebsiella, andBurkholderia have been recovered from endophytic population of plants,thereby supporting the hypothesis that human pathogens can internalizewithin healthy plant tissue (Rosenblueth and Martinez-Romero, 2006).

However, experimentally demonstrating the internalization of humanpathogens has produced conflicting results primarily because of thedifficulties in proving that infiltration has actually occurred (Table 4.6).

TABLE 4.6 Internalization of pathogens within growing plants

Comments Method used to assess internalization Researchers

Internalization of poliovirus into

growing tomato plants but only at

high >104 tita levels

Sand layer between inoculated soil and

phyllosphere

Oron et al. (1995)

E. coli O157:H7 in radish sprouts Surface sterilization using 0.2% HgCl Itoh et al. (1998)Internalization of Salmonella into

tomatoes by inoculating stems or

blossom of plants

Immersion in 70% ethanol for 2 min Guo et al. (2001)

Colonization of gfp labeled E. coli O157:

H7 at cut edges of leafy greens

No surface sterilization Takeuchi and Frank (2000)

Confocal microscopy to detect gfp label

Internalization of E. coli O157:H7 within

lettuce seedlings cultivated in soil or

hydroponically

No surface sterilization Wachtel et al. (2002)


Internalization of MS2 coliphage into

growing cress plants

Inoculated soil was overlaid with agar Kirkham et al. (2002)

Internalization of E. coli O157:H7 into

lettuce seedlings

No surface sterilization Solomon et al. (2002)


Internalization of nonpathogenic E. coli

into cabbage seedlings

80% ethanol Rafferty et al. (2003)

Internalization of E. coli O157:H7 and

Salmonella into growing Arabidopsis

plants when pathogens were

introduced into soil

No surface sterilization Cooley et al. (2003)

Confocal microscopy to detectinternalized populations

Internalization of E. coli in spinach

plants cultivated in soil or

hydroponically

20,000 ppm sodium hypochlorite Warriner et al. (2003a)

182

Internalization of E. coli and Salmonella

in sprouting mung beans

20,000 ppm sodium hypochlorite Warriner et al. (2003b)

No internalization of E. coli O157:H7 in

spinach plants when subjected to

physical or biological damage

2000 ppm sodium hypochlorite Hora et al. (2005)

Internalization of E. coli O157:H7,

Salmonella and L. monocytogenes in

growing plants

20,000 ppm sodium hypochlorite Jablasone et al. (2005)

No internalization of Salmonella in

tomato fruit when inoculum was

applied to the roots of growing plants

No surface sterilization Jablasone et al. (2004)

Screening for Salmonella in ripened fruit

Aggregation of Salmonella in stomata

and cut cuticle cracks

No surface sterilization Duffy et al. (2005)

Enhancement of internalization by

E. coli O157:H7 in Arabidopsis if

coinoculated with Wausteria paucula

No surface sterilization Cooley et al. (2006)

Internalization of Salmonella in parsley

leaves

No surface sterilization Lapidot and Yaron (2007)

Internalization of Salmonella but not

L. monocytogenes when introduced

onto the roots of 4-week-old barleyplants

1% chloramine T Kutter et al. (2006)

Internalization of E. coli O157:H7 into

mature lettuce plants when

introduced onto the roots of plants

No surface sterilization. Perforated

polypropylene sheet used to

physically separate the inoculation

site (roots) from the aerial leaves

Bernstein et al. (2007)

(continued)

183


Comments Method used to assess internalization Researchers

Internalization of E. coli O157:H7 andSalmonella in hydroponically

cultivated lettuce

1% silver nitrate Franz et al. (2007)

Internalization of different Salmonella

serovars into tomatoes when

introduced onto the plant blossom

2000 ppm calcium hypochlorite Shi et al. (2007)

No internalization of E. coli O157:H7 via

roots when applied to soil

Surface sterilization using 80% ethanol

followed by 0.1% HgCl2

Zheng et al. (2008)

No internalization of E. coli O157:H7 inleaves when inoculated into the

phyllosphere of growing lettuce

plants

Surface sterilization using 80% ethanolfollowed by 0.1% HgCl2

Zheng et al. (2008)

No internalization of E. coli O157:H7 or

Salmonella when introduced into the

water for irrigating 3–33-day

posttransplanted lettuce plants

Surface sterilization using 80% ethanol

followed by 0.1% HgCl2

Erickson et al. (2008)

Internalization of E. coli O157:H7overexpressing curli into lettuce

leaves when applied as a surface

inoculum

70% ethanol and visualization usingbioluminescent marker

Tanner et al. (2008)


Attempting to demonstrate internalization of human pathogenswithin growing plant tissues is problematic. The traditional approach isto surface sterilize plant material with sanitizers (e.g., sodium hypochlo-rite, peracetic acid, ethanol) and recover the subsequent bacteria. How-ever, spores (fungal and bacterial) and biofilms are resistant to sanitizersleading to false positive results (Reissinger et al., 2001). Penetration ofsanitizer into the internal tissue of plants is a further problem and canpotentially lead to an underestimation of endophyte numbers. Therefore,to overcome such limitations, several researchers have taken the approachof carefully inoculating the roots of plants without contacting the leaveswhich are subsequently screened for the presence of the target pathogen(Franz et al., 2007). The addition of a physical layer (e.g., sand) betweenthe roots and phyllosphere is a further approach to prevent contamina-tion being introduced on, as opposed to within, leaves (Franz et al., 2007).However, the potential for contamination being introduced onto theexternal surface of leaves exists thereby reducing the confidence thatpathogens detected are truly internalized.

An alternative to culturing techniques is the application of cell labelingexploiting green fluorescent protein (gfp) in combination with laser con-focal microscopy (Table 4.6). Gfp is a protein originally isolated from thejellyfish Aequorea victoria. The key benefit of gfp is the ability to fluoresceunder UV light in the absence of an energy source or other cellularcofactors, thereby enabling in situ visualization with minimum disruptionto cell physiology. The gene encoding for gfp can be readily inserted andexpressed in bacterial cells using plasmid vectors. However, for theplasmid to be retained and replicated within the host cell, selective pres-sure (typically using an antibiotic) needs to be applied. Therefore, whenstudying plant:microbial interactions over extended periods, selectiveagents cannot be used and hence the gfp phenotype can be readily lost.A further limitation to gfp labeling is the need for a high cell density inorder to visualize bacteria using confocal microscopy. Clearly, if thetagged bacteria are present in low numbers then locating cells withinplant tissue is unlikely.

A further method to visualize the presence of internalized bacteria isthrough the use of glucuronidase (GUS) activity stain. The GUS stain isbased on the cleavage of a chromagenic substrate (e.g., 5-bromo-4-chloro-3-indoyl-b-D-glucuronide; X-GLUC) that can be directly visualized as ablue/green precipitate within plant tissues. Therefore, if the target cell ispresent the chromagen accumulates and hence has greater sensitivitycompared to gfp labels. The GUS technique has been used extensivelyto study plant:bacteria interactions based on gus gene insertion into thetarget bacterium (Wilson et al., 1995). GUS activity is not present in plantsor a wide range of bacteria (Wilson et al., 1995), although it is expressed inthe majority (>96%) of known generic E. coli strains (Liang et al., 2005).


Warriner et al. (2003b,c) have used GUS in situ staining to demonstrate theinternalization of generic E. coli in spinach and bean sprouts.

D. Genetic and physiological factors

The finding of a diverse population of endophytes within plants may beunexpected given that the endogenous plant defenses function to guardagainst microbial invasion. The inducible defenses are commonly activatedin response to phytopathogens to contain the site of infection and prime theplant against further microbial attack. At the site of infection the hypersen-sitive response (HR) is induced, which releases an oxidative burst to causelocalized necrosis. At the same time, the plant hormones (e.g., salicylic acid)are circulated to other parts of the plant to activate the systemic acquiredresistance (SAR) which primes the defenses. It is thought that endophyticbacteria enhance the resistant of plants to phytopathogens by inducingSAR. The ingress of opportunistic saprophytes into the plant activates thelocalized induced resistance (LIR) (Newman et al., 2001) which releasesantimicrobials within the localized area to suppress saprophytic activitywithout leading to necrosis (Esposito et al., 2008). Hence, microbes thatbecome established as endophytes do not induce HA and LIR withinplants. One strategy developed by phytopathogens and endophytes is tolose flagella or shield Lipopolysaccharide (LPS) which prevent detection,hence activation by plant defenses (Gozzo, 2003; Liu et al., 2007).

Enteric bacteria have been found to differ with respect to the ability tobecome integrated into the endophytic microflora of alfalfa roots (Donget al., 2001). A strain isolated from maize, Klebsiella pneumoniae 342, colo-nizes the interior of several host plants in higher numbers compared toS. Typhimurium and E. coli K12 (Dong et al., 2003). However, a Salmonellamutant lacking flagella and Type III secretion system could readily colo-nize the roots of alfalfa and become integrated into the endophytic micro-flora. However, restoration of either phenotype reduced the ability ofSalmonella to colonize roots (Dong et al., 2003). It was thought that thelack of flagella and Type III secretion system prevented the activation ofthe salicylic acid response or independent response which in turn failed toinduce PR1 promoter, and hence release of antimicrobials (Iniguez et al.,2005). The question of if enteric pathogens found naturally in the envi-ronment shed their outer surface structures to enhance internalizationinto plants remains unclear. Nevertheless, it is possible that the ability toprevent activation of plant defensesmay explain the interstrain variabilitythat exists with respect to pathogen interactions with plants.

Although plant and animal pathogens infect different hosts, there areseveral similarities in the strategies employed (Buckhout and Thimm,2003). For example, Type III secretion systems can be found in bothplant and animal pathogens. The Type III secretion system is


essentially a microtube by which the invading bacterium attaches tothe surface of the host cell. Chemicals and proteins are deliveredthrough the Tir III protein to sequester defense mechanisms and repro-gram host cell activity. Of course, this does not imply that a plantpathogen would cause disease in animals. Indeed, to date only Ps.auruginosa PA14 is known to cause disease in both animals and plants(Plotnikova et al., 2000). Nevertheless, it is conceivable that traits inphytobacteria associated with plant interactions could be present inenteric pathogens such as E. coli O157:H7. It is known that broadhost range bacteriophage that infects Pseudomonas and E. coli O157:H7can transfer genetic traits between these two genera (Hendrix, 1999;Muniesa et al., 2003). An example of phytobacteria sharing genes withenteric pathogens has been found in Ps. syringae pv. maculicola. Thephytopathogen posses a b-lactamase which protects the bacteriumagainst preformed defenses in Arabidopsis. The gene encoding for theenzyme, donated as sax (Survival on Arabidopsis eXtract), has beenidentified in a range of Ps. syringae pathovars but absent from non-phytopathogenic strains. From comparative homology studies, the geneshows a high level of similarity to an uncharacterized gene in E. coliO157:H7 (Crooks and Lamb, 2001). Whether the expression of saxwithin E. coli O157:H7 enhances persistence within plants has yet tobe established but does point to an adaptive response of human patho-gens to survive outside of the host environment. However, this viewremains contentious with many researchers in the field, suggesting thatthe interaction of human pathogens with plants is a passive processbeing comparable to any oppertunistic saprophyte (Doyle andErickson, 2008). Yet, as previously indicated, there is accumulatingevidence to support the hypothesis that human pathogens haveevolved specialized mechanisms for becoming established and persist-ing on plants.

It has been observed that pathogens such as Salmonella and E. coliO157:H7 preferentially attach to cut surfaces and natural openings suchas stomata, whereas common epiphytes such as Ps. fluoresecens colonizethe intact plant tissue (Li et al., 2008; Melotto et al., 2006; Seo and Frank,1999; Takeuchi and Frank, 2001). It is known that attachment of humanpathogens is an active process which requires the bacterium to be in aviable state although the actual internalization process can be passive(Solomon and Matthews, 2006).

It is thought that cell surface components such as cellulose, flagella,pilli, and Type III secretion systems all play a role in cell attachment(Barak et al., 2005; Zogaj et al., 2001). Several genes and mechanismshave been identified as being involved in attachment of human pathogensto plants. These mechanisms include curli, fimbriae, adhesins, andcapsule production (Barak et al., 2005, 2007).


From microarray studies it has been demonstrated that virulencegenes are upregulated in S. Typhimurium during colonization inresponse to lettuce root exudates (Klerks et al., 2007) which aid attachmentto the plant tissue. It is also observed that genes involved in sugar–phosphate metabolism are also upregulated which is thought to attractthe enteric pathogen to cut edges of damaged leaves (Klerks et al., 2007).The presence of structures on the plant cell walls has also been proposed.This hypothesis is supported by the fact that attachment of pathogenssuch as E. coli O157:H7 and Salmonella is plant specific. For example,attachment of pathogens is more frequently observed with Brassicaceaecompared to lettuce, carrots, or tomatoes (Barak and Liang, 2008). Collec-tively, studies to date support the view that human pathogens haveadapted to colonize plants as a means of persisting between animal orhuman hosts.

VI. INTERVENTIONS TO ENHANCE THE SAFETYOF FRESH PRODUCE

Postharvest washing of vegetables remains the key intervention toremove field acquired contamination. There have been numerous papersand reviews on the relative performance of different sanitizers includinghypochlorite, chlorine dioxide, and peroxyacetic acid (Allwood et al.,2004; Gonzalez et al., 2004; Ibarra-Sanchez et al., 2004; Koseki and Isobe,2006; Rodgers et al., 2004; Romanova et al., 2002).

From reviewing the literature, typical log reductions achieved for Sal-monella on lettuce cover a wide range: peroxyacetic acid 1.7 (Hellstrom et al.,2006), acidified sodium chlorite 3.1 (Inatsu et al., 2005), chlorine dioxide 1.53(Inatsu et al., 2005), ozone 5.6 (Rodgers et al., 2004), and electrolyzed water1.0 (Koseki and Isobe, 2006; Koseki et al., 2003). However, on naturalcontaminated produce the log count reductions achieved are typically1–2 log cfu/g regardless of the sanitizer applied (Doyle and Erickson,2008). The limitation of postharvest washes can be attributed internalizedpopulations or those within biofilms on the surface of produce (Gonzalezet al., 2004; Koseki et al., 2003). The heavy organic loading of washwater canalso readily neutralize sanitizers such as hypochlorite. This can beaddressed to a degree by operating at a set oxidation–reduction potential(ORP). Here, the amount of chlorine introduced into the water is increasedin the presence of organic matter tomaintain the chlorous acid levels withinthe wash. However, even with ORP-controlled systems the LCR is notsignificantly improved compared to when chlorine washes alone areapplied (Guentzel et al., 2008).

Postharvest washing does not only provide a low level ofconfidence of decontaminating produce but is also thought to result in


cross-contamination between fresh produce batches. It is known thatcontaminated flume water is a potential source of contamination of Sal-monella for fruits such as tomatoes (Zhuang et al., 1995). Luo (2007)reported that lettuce washed in recycled chlorinated water with hightotal solids content spoiled more rapidly compared to samples washedin fresh chlorinated water. More direct evidence for cross-contaminationvia wash water was reported by Ilic et al. (2008). The researchers evalu-ated the effect of commercial wash process on the coliform and E. colicounts associated with spinach. The finding of the study was that theproportion of samples positive for coliforms increased from 53% to 79%following washing (Ilic et al., 2008). It is possible that the increase incoliform prevalence was caused by the uptake of wash water caused bya temperature differential effect. It is known that warm produce placedinto cold water results an influx of wash water (hence contamination) intothe inner vascular system of leafy greens and fruit (Bolton et al., 2002;Fukumoto et al., 2002; Ibarra-Sanchez et al., 2004). To overcome the prob-lem of infiltration, it is recommended to use warm water for washingproduce (FDA, 1998). However, this has the adverse effect of warming theproduce, thereby accelerating plant autolysis and growth of spoilagemicrobes or even human pathogens.

Surface pasteurization of produce using steam, hot water, or chlorinedioxide gas was shown to enhance the reduction of microbial loads onhard surface produce (Stringer et al., 2007). However, delicate producesuch as leafy vegetables can be damaged by the steam process (Allwoodet al., 2004; McWatters et al., 2002; Sy et al., 2005). A study conducted bySapers and Sites (2003) showed that cantaloupe treated with hot (80 �C)5% hydrogen peroxide for 3 min reduced E. coli and Salmonella popula-tions with no signs of tissue damage after 26 days of storage at 4 �C.However, some disadvantages with heat treatments for produce occur.Heat treatment can reduce microbial loads on produce but has little effectif contaminated after heating. A study by Conway et al. (2005) showedthat injured apples had larger microbial loads of pathogens when heattreated. This was most likely due to the damage of enzymes that are theplant’s defense system against invasion of microorganisms. Overall, thequality of commodities such as fresh cut lettuce diminishes duringextended storage after heat treatment of more than 3 min.

Irradiation of produce has been shown to be effective in reducingmicrobial contamination where the maximum dosage level is 1.0 kGyfor fruits and vegetables (Bari et al., 2005; Gomes et al., 2008; Niemira,2007). An irradiation dose of 1.0 kGy treatment decreased mesophilicbacteria in Mexican salads (Erickson, 2008). However, irradiation cancause changes in pectin structure leading to texture loss and hence shelflife. Also, viruses seem to be relatively resistant to irradiation treatmentand relative to vegetative cells, suggesting that doses delivered to inactive


pathogens (e.g., Salmonella) would be insufficient to kill NLV (Gomilaet al., 2008).

High hydrostatic pressure (HHP) processes have been used mainly forsauces or seafood and proven effective at reducing microbial populationswithout adverse effects on product quality (Considine et al., 2008; Brinezet al., 2006). HHP treatment causes bacterial inactivation by damaging thecell membrane, which affects membrane permeability and intracellularenzyme inactivation and possibly ruptures the plant cell wall (Kniel et al.,2007). Although HHP have proven to be one of the most effect decontam-ination intervention strategies, there is the potential of causing disruptionto plant tissue leading to accelerated spoilage (Basak and Ramaswamy,1998; Butz et al., 2002; Prestamo and Arroyo, 1998). Pressure-induceddamage to plant tissue results from the stress and strains imposed onthe cell walls which subsequently results in loss of texture (Fuchigamiet al., 1995; Hartmann and Delgado, 2004; Kidmose and Martens, 1999).However, it has also been reported that HHP treatment can enhance thetexture of vegetables through de-esterifying pectin via the activity ofpectin methyltransferase (PMT) which facilitates calcium binding to stabi-lize plant cell walls (Fennema, 1996; Sila et al., 2004).

UV light has been considered as an alternative to sanitizer-basedsystems for decontaminating fresh produce (Bialka and Demirci, 2008).Unlike chemical sanitizers, UV does not leave residues and the productdoes not need to be dried thereby providing energy savings (Bialka andDemirci, 2008). UV light can be divided into three classes: UV-A(400–320 nm), UV-B (320–280 nm), and UV-C (<280 nm) where the mosteffective range for produce decontamination is within 200–280 nm. UV-Ctreatments are effective at decreasing psychrotrophic bacteria, coliforms,and yeasts and molds. However, bacterial spores and stationary phasecells are much more resistant to UV-C light than vegetative and exponen-tial phase cells. A study by Yaun et al. (2004) reported that a greater than9 mW/cm2 dose of UV-C light reduced microbial populations by 2 log inlettuce and tomatoes and 3 log in apples.

There are some disadvantages associated with UV-C-treated produce,which include possible negative effects on quality at high doses and poorpenetration. Also, UV-C treatment was more effective for produce withsmoother surfaces and did not significantly reduce viruses like norovirusand FCV (Fino and Kniel, 2008).

UV light has been combined with hydrogen peroxide to develop aproduce decontamination that can be used to inactivate pathogens on thesurface and internalized contamination (Hadjok et al., 2008; Xie et al.,2008). The underlying principle of the approach is to generate highlyreactive, antimicrobial, radicals from the UV degradation of hydrogenperoxide in a process termed advanced oxidative process (AOP) (Rodgerset al., 2004; Suty et al., 2004). By using a 50 �C H2O2 (1.5%) simultaneously


applied with UV-C, it was possible to achieve>5 log reductions of humanpathogens and viruses (MS2 phage surrogate) on leafy greens (Hadjoket al., 2008; Xie et al., 2008).

A. Biocontrol of human pathogens

Control of human pathogens at the primary production level is problem-atic due to its open nature. It has been proposed that decontamination ofirrigation water through ozonation or chlorination is an option althoughthe cost and effect on the plant microecology are obvious limitations(Ajwa et al., 2002; Rojas-Valencia et al., 2004). A more practical approachis to use biocontrol strategies whereby antagonistic microbes are intro-duced into the rhizosphere to reduce or inhibit pathogens. To date themajority of biocontrol strategies have been focused on controlling phyto-pathogens (Vassilev et al., 2006). The use of biocontrol strategies to controlhuman pathogens is an emerging area but has yet to be studied in detail. Ithas been previously reported that Enterobacter is antagonistic againstSalmonella introduced onto Arabidopsis and lettuce (Cooley et al., 2003,2006). Although the mechanisms are unknown, it is thought that Enter-obacter can compete more effectively for nutrients utilized by Salmonella tosupport growth and persistence. Fett et al. (2006) introduced a Pseudomo-nas fluorescens strain (isolated from the rhizosphere of wheat) into the soakwater of Salmonella inoculated alfalfa seeds. When the seeds weresprouted that sprouts with Ps. fluorescens had 5 log lower Salmonella levelscompared to controls (Fett et al., 2006). Matos and Garland (2005) intro-duced a cocktail of bacteria (unknown composition) obtained from ger-minated sprouts to alfalfa seeds inoculated with Salmonella. TheSalmonella counts on sprouts derived from seed treated with biocontrolagent were 5.7 logs lower compared to controls. However, because thebacterial cocktail was directly obtained from sprouts, attempting to repro-duce the composition on a commercial scale could be problematic. Theapplication of bacteriophage has been evaluated for controlling humanpathogens although with various degrees of success (Guenther et al., 2008;Hudson et al., 2005). Bacteriophages are viruses that infect and replicatewithin bacterial hosts. Advantages of phages are that they are specific,self-perpetuating, and self-limiting. Although widely used in EasternEurope, bacteriophages have only recently being approved for food appli-cations. In 2006, the Food and Drug Administration (FDA) approved aL. monocytogenes-specific phage preparation (LMP-102) for use as an anti-microbial agent against L. monocytogenes contamination of ready-to-eatfoods (Guenther et al., 2008; Lang, 2006; Leverentz et al., 2001). In relationto fresh produce, Abuladze et al. (2008) evaluated a cocktail (designatedECP-100) of phages to control E. coli O157:H7 on a variety of vegetabletypes. The researchers reported a 94–99% reduction of E. coli O157:H7


introduced onto tomatoes and complete inactivation with spinach inocu-lated with 14,000 cfu of the pathogen. However, it is possible that infec-tion and phage replication occurred during cultivation of survivors giventhat phage numbers on the fresh produce samples remained the same(Abuladze et al., 2008). Leverentz et al. (2004) reported a 2–3 log reductionof Salmonella on melon although complete elimination of the entericpathogen was not observed.

VII. CONCLUSIONS AND FUTURE RESEARCH

Despite the increased awareness of food safety issues surrounding freshproduce, the number and frequency of foodborne illness outbreaks con-tinue to rise. It is evident that the centralization of production coupledwith the growth in the bagged salad market are significant factors toexplain the incidence of foodborne illness linked to fresh cut produce.However, environmental factors also play a significant role especially interms of manure management in disseminating human pathogens towater courses and/or soil.

The adaption of human pathogen strains to grow and persist on plantsremains a relatively unexplored area. In many aspects, the adaption ofhuman pathogens to plants would make sense given that the microbesneed to survive in the environment between infecting hosts.

The finding that human pathogens can contaminate growing crops inthe field and persist through to consumption is of concern and interven-tions above simple postharvest washing have to be considered. Onestrategy of interest is through implementing biocontrol methods thatcan control human pathogens at the primary production level.

With regards to postharvest control, it can be envisaged that moreeffective decontamination treatments will be adopted such as AOP ormore likely irradiation. However, regardless of technological advancesthere will always be a role of GAP (Good Agricultural Practice) and GMP(Good Manufacturing Practice) for enhancing the microbiological safetyof fresh produce.

REFERENCES

Abuladze, T., Li, M., Menetrez, M. Y., Dean, T. M., Senecal, A., and Sulakvelidze, A. (2008).Bacteriophages reduce experimental contamination of hard surfaces, tomato, spinach,broccoli, and ground beef by Escherichia coli O157:H7. Appl. Environ. Microbiol. 74,6230–6238.

Anderson, B. C. (1986). Effect of drying on the infectivity of cryptosporidia-laden calf fecesfor 3- to 7-day old mice. Am. J. Vet. Res. 47, 2272–2273.


Anonymous. (1999). From the Centers for Disease Control and Prevention. Outbreaks ofShigella sonnei infection associated with eating fresh parsley—United States and Canada,July–August 1998. JAMA 281, 1785–1787.

Anonymous. (2006). The FDA: Fresh leafy greens grown in the United States are safe. FDA

Consum. 40, 11.Ajwa, H. A., Trout, T., and Mueller, J. (2002). Application of alternative fumigants through

drip irrigation systems. Phytopathology 92, 1349–1355.Akhand, N., Lapen, D. R., Topp, E., Edwards, M. J., Sabourin, L., Coelho, B. R., Duenk, P. W.,

Payne, M., and Gottschall, N. (2008). Using macro to simulate liquid sewage biosolidtransport to tile drains for several land applicationmethods. Trans. ASABE 51, 1235–1245.

Allwood, P. B., Malik, Y. S., Hedberg, C. W., and Goyal, S. M. (2004). Effect of temperatureand sanitizers on the survival of feline calicivirus, Escherichia coli, and F-specific coliphageMS2 on leafy salad vegetables. J. Food Prot. 67, 1451–1456.

Alter, T. and Scherer, K. (2006). Stress response of Campylobacter spp. and its role in foodprocessing. J. Vet. Med. B Infect. Dis. Vet. Public Health 53, 351–357.

Armon, R., Gold, D., Brodsky, M., and Oron, G. (2002). Surface and subsurface irrigationwith effluents of different qualities and presence of Cryptosporidium oocysts in soil and oncrops. Water Sci. Technol. 46, 115–122.

Arthur, L., Jones, S., Fabri, M., and Odumeru, J. (2007). Microbial survey of selected Ontario-grown fresh fruits and vegetables. J. Food Prot. 70, 2864–2867.

Artiz, R. R. E. and Killham, K. (2002). Survival of Escherichia coli O157:H7 in private drinkingwater wells: Influences of protozoan grazing and elevated copper concentrations’. FEMS

Microbiol. Lett. 216, 117–122.Astrom, J., Pettersson, T. J. R., and Stenstrom, T. A. (2007). Identification and management of

microbial contaminations in a surface drinking water source. J. Water Health 5, 67–79.Baldani, J. I., Caruso, L., Baldani, V. L. D., Goi, S. R., and Dobereiner, J. (1997). Recent

advances in BNF with non-legume plants. Soil Biol. Biochem 29, 911–922.Baloda, S. B., Christensen, L., and Trajcevska, S. (2001). Persistence of a Salmonella enterica

serovar typhimurium DT12 done in a piggery and in an agricultural soil amended withcontaminated slurry. Appl. Environ. Microbiol. 67, 2859–2862.

Barak, J. D. and Liang, A. S. (2008). Role of soil, crop debris, and a plant pathogen inSalmonella enterica contamination of tomato plants. PLoS ONE 3, e1657.

Barak, J. D., Gorski, L., Naraghi-Arani, P., and Charkowski, A. O. (2005). Salmonella enterica

virulence genes are required for bacterial attachment to plant tissue. Appl. Environ.Microbiol. 71, 5685–5691.

Barak, J. D., Jahn, C. E., Gibson, D. L., and Charkowski, A. O. (2007). The role of cellulose andO-antigen capsule in the colonization of plants by Salmonella enterica. Mol. Plant-Microbe

Interact. 20, 1083–1091.Bari, M. L., Nakauma, M., Todoriki, S., Juneja, V. K., Isshiki, K., and Kawamoto, S. (2005).

Effectiveness of irradiation treatments in inactivating Listeria monocytogenes on freshvegetables at refrigeration temperature. J. Food Prot. 68, 318–323.

Barker, G. C., Malakar, P. K., and Peck, M. W. (2005). Germination and growth from spores:Variability and uncertainty in the assessment of food borne hazards. Int. J. Food Microbiol.100, 67–76.

Barnes, B. and Gordon, D. M. (2004). Coliform dynamics and the implications for sourcetracking. Environ. Microbiol. 6, 501–509.

Basak, S. and Ramaswamy, H. (1998). Effect of high pressure processing on the texture ofselected fruits and vegetables. J. Texture Stud. 29, 587–601.

Bassett, J. and McClure, P. (2008). A risk assessment approach for fresh fruits. J. Appl.Microbiol. 104, 925–943.

Bernstein, N., Sela, S., and Neder-Lavon, S. (2007). Assessment of contamination potential oflettuce by Salmonella enterica serovar Newport added to the plant growing medium.J. Food Prot. 70, 1717–1722.


Berry, E. D., Woodbury, B. L., Nienaber, J. A., Eigenberg, R. A., Thurston, J. A., andWells, J. E. (2007). Incidence and persistence of zoonotic bacterial and protozoan patho-gens in a beef cattle feedlot runoff control vegetative treatment system. J. Environ. Qual.

36, 1873–1882.Beuchat, L. R. and Ryu, J. H. (1997). Produce handling and processing practices. Emerg. Infect.

Dis. 3, 459–465.Bialka, K. L. and Demirci, A. (2008). Efficacy of pulsed UV-light for the decontamination of

Escherichia coli O157:H7 and Salmonella spp. on raspberries and strawberries. J. Food Sci.

73, M201–M207.Bichai, F., Paynment, P., and Barbeau, B. (2008). Protection of waterborne pathogens by

higher organisms in drinking water: A review. Can. J. Microbiol. 54, 509–524.Bolton, D. J., Pearce, R. A., Sheridan, J. J., Blair, I. S., McDowell, D. A., and Harrington, D.

(2002). Washing and chilling as critical control points in pork slaughter hazard analysisand critical control point (HACCP) systems. J. Appl. Microbiol. 92, 893–902.

Bower, J. R. (1999). Foodborne diseases: Shiga toxin producing E. coli (STEC). Pediatr. Infect.Dis. J. 18, 909–910.

Brandl, M. T. (2006). Fitness of human enteric pathogens on plants and implications for foodsafety (1). Annu. Rev. Phytopathol. 44, 367–392.

Brandl, M. T. (2008). Plant lesions promote the rapid multiplication of Escherichia coli O157:H7 on postharvest lettuce. Appl. Environ. Microbiol. 74, 5285–5289.

Brandl, M. T. and Amundson, R. (2008). Leaf age as a risk factor in contamination of lettucewith Escherichia coli O157:H7 and Salmonella enterica. Appl. Environ. Microbiol. 74,2298–2306.

Brandl, M. T., Haxo, A. F., Bates, A. H., andMandrell, R. E. (2004). Comparison of survival ofCampylobacter jejuni in the phyllosphere with that in the rhizosphere of spinach andradish plants. Appl. Environ. Microbiol. 70, 1182–1188.

Brinez, W. J., Roig-Sagues, A. X., Herrero, M. M. H., and Lopez, B. G. (2006). Inactivation byultrahigh-pressure homogenization of Escherichia coli strains inoculated into orange juice.J. Food Prot. 69, 984–989.

Buckhout, T. J. and Thimm, O. (2003). Insights into metabolism obtained from microarrayanalysis. Curr. Opin. Plant Biol 6, 288–296.

Butz, P., Edenharder, R., Fernandez Garcia, A., Fister, H., Merkel, C., and Tauscher, B. (2002).Changes in functional properties of vegetables induced by high pressure treatment. FoodRes. Int. 35, 295–300.

Buyer, J. S., Roberts, D. P., and Russek-Cohen, E. (2002). Soil and plant effects on microbialcommunity structure. Can. J. Microbiol. 48, 955–964.

Campo, C., Navarro, V., Perez, C., Gutierrez, I., and Alonso, R. (2001). Aeromonas spp.bacteremia: Study of 12 cases and review of the literature. Enferm. Infecc. Microbiol. Clin.

19, 161–164.Canadian Council of Ministers of the Environment (CCME) (2005). Guideline for compost man-

agement. http://www.ccme.ca/assets/pdf/compostgdlns_1340_e.pdfAccessedDec. 08.Cao, T., Duncan, R. A., McKenry, M. V., Shackel, K. A., Dejong, T. M., and Kirkpatrick, B. C.

(2005). Interaction between nitrogen-fertilized peach trees and expression of syrB, a geneinvolved in syringomycin production in Pseudomonas syringae pv. syringae. Phytopathology95, 581–586.

Carlin, F. and Peck, M. W. (1996). Growth of and toxin production by nonproteolyticClostridium botulinum in cooked pureed vegetables at refrigeration temperatures. Appl.Environ. Microbiol. 62, 3069–3072.

Centers for Disease Control and Prevention (CDC). (2008). Outbreak of salmonella serotypeSaintpaul infections associated with multiple raw produce items—United States, 2008.MMWR Morb. Mortal. Wkly Rep. 57, 929–934.

http://www.ccme.ca/assets/pdf/compostgdlns&lowbar;1340&lowbar;e.pdf

http://www.ccme.ca/assets/pdf/compostgdlns&lowbar;1340&lowbar;e.pdf


Centre for Science in the Public Interest. (2008). Outbreaks linked to fresh produce. http://cspinet.org/new/pdf/cspi_outbreak_alert.pdf. Accessed Dec. 08.

Chacin-Bonilla, L. (2007). Transmission of Cyclospora cayetanensis infection: A review focus-ing on soil-borne cyclosporiasis. Trans. R. Soc. Trop. Med. Hyg. 102, 215–216.

Chan, Y. C., Boor, K. J., and Wiedmann, M. (2007). B-dependent and -independent mechan-isms contribute to transcription of Listeria monocytogenes cold stress genes during coldshock and cold growth. Appl. Environ. Microbiol. 73, 6484–6498.

Chancellor, D. D., Tyagi, S., Bazaco, M. C., Bacvinskas, S., Chancellor, M. B., Dato, V.M, andde Miguel, F. (2006). Green onions: Potential mechanism for hepatitis A contamination.J. Food Prot. 69, 1468–1472.

Chao, W. L., Ding, R. J., and Chen, R. S. (1988). Survival of Yersinia enterocolitica in theenvironment. Can. J. Microbiol 34, 753–756.

Chase-Topping, M., Gally, D., Low, C., Matthews, L., and Woolhouse, M. (2008). Super-shedding and the link between human infection and livestock carriage of Escherichia coliO157. Nat. Rev. Microbiol. 6, 904–912.

Chifiriuc, M. C., Israil, A. M., Larion, C., Alexandru, I., and Dobre, G. (2007). Virulence andresistance markers in clinical and environmental Aeromonas strains isolated in Romania.Roum. Arch. Microbiol. Immunol. 66, 69–79.

Chua, D., Goh, K., Saftner, R. A., and Bhagwat, A. A. (2008). Fresh-cut lettuce in modifiedatmosphere packages stored at improper temperatures supports enterohemorrhagicE. coli isolates to survive gastric acid challenge. J. Food Sci. 73, M148–M153.

Cieslak, P. R., Barrett, T. J., Griffin, P. M., Gensheimer, K. F., Beckett, G., Buffington, J., andSmith, M. G. (1993). Escherichia coliO157:H7 infection from amanured garden. Lancet 342,8867.

Considine, K. M., Kelly, A. L., Fitzgerald, G. F., Hill, C., and Sleator, R. D. (2008). High-pressure processing—Effects on microbial food safety and food quality. FEMS Microbiol.

Lett. 281, 1–9.Conway, W. S., Leverentz, B., Janisiewicz, W. J., Saftner, R. A., and Camp, M. J. (2005).

Improving biocontrol using antagonist mixtures with heat and/or sodium bicarbonate tocontrol postharvest decay of apple fruit. Post-harvest Biol. Technol. 36, 235–244.

Cooley, M. B., Miller, W. G., and Mandrell, R. E. (2003). Colonization of Arabidopsis thalianawith Salmonella enterica and enterohemorrhagic Escherichia coli O157:H7 and competitionby Enterobacter asburiae. Appl. Environ. Microbiol. 69, 4915–4926.

Cooley, M. B., Chao, D., and Mandrell, R. E. (2006). Escherichia coli O157:H7 survival andgrowth on lettuce is altered by the presence of epiphytic bacteria. J. Food Prot. 69,2329–2335.

Cooley, M., Carychao, D., Crawford-Miksza, L., Jay, M. T., Myers, C., Rose, C., Keys, C.,Farrar, J., and Mandrell, R. E. (2007). Incidence and tracking of Escherichia coliO157:H7 ina major produce production region in California. PLoS ONE 2, e1159.

Cote, C. and Quessy, S. (2005). Persistence of Escherichia coli and Salmonella in surface soilfollowing application of liquid hog manure for production of pickling cucumbers. J. FoodProt. 68, 900–905.

Crooks, C. and Lamb, C. (2001). Identification of a host-pathogen specific gene conferringprotection from a preformed antimicrobial activity in Arabidopsis. In ‘‘10th InternationalCongress on Molecular Plant-Microbe Interactions University of Wisconsin’’, Madison,USA.

Culley, J. L. B. and Phillips, P. A. (1982). Bacteriological quality of surface and subsurfacerunoff from manured sandy clay loam soil. J. Environ. Qual. 11, 155–158.

Culley, J. L.B and Barnett, G. M. (1984). Land disposal of manure in the province of Quebec.Can. J. Soil Sci. 64, 75–86.

Davies, R. H., Newell, D. G., Sayers, A. R., Cheasty, T., Cassar, C., Ridley, A., Cook, A. J. C.,Evans, S. J., Teale, C. J., Smith, R. P., McNally, A., Toszeghy, M., et al. (2008). Intestinal

http://cspinet.org/new/pdf/cspi&lowbar;outbreak&lowbar;alert.pdf




carriage of verocytotoxigenic Escherichia coliO157, Salmonella, thermophilic Campylobacter

and Yersinia enterocolitica, in cattle, sheep and pigs at slaughter in Great Britain during2003. Epidemiol. Infect 136, 739–751.

Davis, S. A. and Gordon, D. M. (2002). The influence of host dynamics on the clonalcomposition of Escherichia coli populations. Environ. Microbiol. 4, 306–313.

Dean, D. M. and Foran, M. E. (1992a). The impact of liquid animal waste application on tiledrain water quality. Can. J. Soil Sci. 72, 304–305.

Dean, D. M. and Foran, M. E. (1992b). The effect of farm liquid waste application on tiledrainage. J. Soil Water Con. 47, 368–369.

Diaz, B. and Schulze-Makuch, D. (2006). Microbial survival rates of Escherichia coli andDeinococcus radiodurans under low temperature, low pressure, and UV-Irradiation condi-tions, and their relevance to possible Martian life. Astrobiology 6, 332–347.

Dixit, S. M., Gordon, D. M., Wu, X. Y., Chapman, T., Kailasapathy, K., and Chin, J. J. (2004).Diversity analysis of commensal porcine Escherichia coli—Associations between geno-types and habitat in the porcine gastrointestinal tract. Microbiology 150, 1735–1740.

Dong, S. F., Scagel, C. F., Cheng, L. L., Fuchigami, L. H., and Rygiewicz, P. T. (2001). Soiltemperature and plant growth stage influence nitrogen uptake and amino acid concen-tration of apple during early spring growth. Tree Physiol. 21, 541–547.

Doyle, M. P. and Erickson, M. C. (2008). Summer meeting 2007—The problems with freshproduce: An overview. J. Appl. Microbiol. 105, 317–330.

Drevets, D. A. and Bronze, M. S. (2008). Listeria monocytogenes: Epidemiology, humandisease, and mechanisms of brain invasion. FEMS Immunol. Med. Microbiol. 53, 151–165.

Duffy, G. (2003). Verocytoxigenic Escherichia coli in animal feces, manures and slurries.J Appl. Microbiol. 94, 94S–103S.

Duffy, E. A., Cisneros-Zevallos, L., Castillo, A., Pillai, S. D., Ricke, S. C., and Acuff, G. R.(2005). Survival of Salmonella transformed to express green fluorescent protein on Italianparsley as affected by processing and storage. J. Food Prot. 68, 687–695.

Dumetre, A. and Darde, M. L. (2003). How to detect Toxoplasma gondii oocysts in environ-mental samples?. FEMS Microbiol. Rev. 27, 651–661.

Embil, J. A., Ewan, E. P., and MacDonald, S. W. (1984). Surveillance of Listeria monocytogenes

in human and environmental specimens in Nova Scotia, 1974 to 1981. Clin. Invest. Med. 7,325–327.

Environmental Protection Agency. (2004). Guidelines for water reuse. Available at http://www.epa.gov/ord/NRMRL/pubs/625r04108/625r04108.pdf. Accessed Dec. 08.

Erickson, B. E. (2008). Irradiation of lettuce & spinach. Chem. Eng. News 86, 32–34.Erickson, M. C., Liao, J., Payton, A., Webb, C., Ma, L., Zhang, G., Doyle, M. P., and

Beuchat, L. R. (2008). ‘‘Pre-Harvest Internalization of Zoonotic Pathogens by Lettuce asInfluenced by Environmental Growth Conditions’’. Abstracts of the InternationalAssociation for Food Protection, Columbus, Ohio.

Esposito, N., Ovchinnikova, O. G., Barone, A., Zoina, A., Holst, O., and Evidente, A. (2008).Host and non-host plant response to bacterial wilt in potato: Role of the lipopolysacchar-ide isolated from Ralstonia solanacearum and molecular analysis of plant-pathogen inter-action. Chem. Biodivers. 5, 2662–2675.

Fasciolo, G. E., Meca, M. I., Gabriel, E., and Morabito, J. (2002). Effects on crops of irrigationwith treated municipal wastewaters. Water Sci. Technol. 45, 133–138.

Fennema, O. (1996). ‘‘Food Chemistry.’’ 3rd ed., pp. 14–16. Marcel Dekker Inc., Wisconsin.Fett, W. F., Fu, T. J., and Tortorello, M. L. (2006). Emerging issues in food safety.

In ‘‘Microbiology of Fresh Produce.’’ (K. R. Matthews, ed.), pp. 167–219. Wiley & SonsLtd, London.

Fino, V. R. and Kniel, K. E. (2008). UV light inactivation of hepatitis A virus, Aichi virus, andfeline calicivirus on strawberries, green onions, and lettuce. J. Food Prot. 71, 908–913.

http://www.epa.gov/ord/NRMRL/pubs/625r04108/625r04108.pdf




Fleming, R. and MacAlpine, M. (2002). MARVEL Composting Technology: A system tocompost liquid swine manure. Final Report for Agriculture Adaptation Council, Guelph,Ontario.

Food and Drugs Administration (FDA). (1998). Guide to minimize microbial food safetyhazards for fresh fruits and vegetables. http://www.foodsafety.gov/�dms/prodguid.htmlAccessed Dec. 08.

Food and Drugs Administration (FDA). (2001). Analysis and Evaluation of PreventiveControl Measures for the Control and Reduction/Elimination of Microbial Hazards onFresh and Fresh-Cut Produce. Available at http://www.fda.gov/Food/ScienceResearch/ResearchAreas/SafePracticesforFoodProcesses/ucm090977.htmAccessed June 2009.

Food and Drugs Administration (FDA). (2002).U.S. lettuce production and trade http://www.fas.usda.gov/htp/Hort_Circular/2002/02-09/Lettuce.htmAccessed Dec. 08.

Food andDrugs Administration (FDA). (2006). FDAwarns consumers not to drink bolthousefarms carrot juice due to botulism concerns. http://www.cfsan.fda.gov/�lrd/fpcarro2.htmlAccessed Dec. 08.

Food and Drugs Administration (FDA). (2008). Bad bug book. http://www.cfsan.fda.gov/�mow/intro.htmlAccessed Dec. 08.

Franz, E. and van Bruggen, A. H. C. (2008). Ecology of E. coliO157:H7 and Salmonella enterica

in the primary vegetable production chain. Crit. Rev. Microbiol 34, 143–161.Franz, E., Visser, A. A., Van Diepeningen, A. D., Klerks, M.M., Termorshuizen, A. J., and van

Bruggen, A. H. (2007). Quantification of contamination of lettuce by GFP-expressingEscherichia coli O157:H7 and Salmonella enterica serovar Typhimurium. Food Microbiol.

24, 106–112.Fuchigami, M., Miyazaki, K., and Hyacumoto, N. (1995). Frozen carrots texture and pectic

components as affected by low-temperature-blanching and quick freezing. J. Food Sci. 60,132–136.

Fukumoto, L. R., Toivonen, P. M., and Delaquis, P. J. (2002). Effect of wash water tempera-ture and chlorination on phenolic metabolism and browning of stored iceberg lettucephotosynthetic and vascular tissues. J. Agric. Food Chem. 50, 4503–4511.

Gagliardi, J. V. and Karns, J. S. (2000). Leaching of Escherichia coli O157:H7 in diverse soilsunder various agricultural management practices. Appl. Environ. Microbiol. 66, 877–883.

Gagliardi, J. V. and Karns, J. S. (2002). Persistence of Escherichia coli O157:H7 in soil and onplant roots. Environ. Microbiol. 4, 89–96.

Garenaux, A., Jugiau, F., Rama, F., de Jonge, R., Denis, M., Federighi, M., and Ritz, M. (2008).Survival of Campylobacter jejuni strains from different origins under oxidative stressconditions: Effect of temperature. Curr. Microbiol. 56, 293–297.

Gerba, C. P. (1999). Virus survival and transport in groundwater. J. Ind. Microbiol. Biotechnol.

22, 535–539.Gerba, C. P., Gramos, D. M., and Nwachuku, N. (2002). Comparative inactivation of enter-

oviruses and adenovirus by UV light. Appl. Environ. Microbiol. 68, 5167–5169.Gino, E., Starosvetsky, J., and Armon, R. (2007). Bacteriophage ecology in a small community

sewer system related to their indicative role in sewage pollution of drinking water.Environ. Microbiol. 9, 2407–2416.

Girard, M. P., Steele, D., Chaignat, C. L., and Kieny, M. P. (2006). A review of vaccineresearch and development: Human enteric infections. Vaccine 24, 2732–2750.

Gomes, C., Da Silva, P., Chimbombi, E., Kim, J., Castell-Perez, E., and Moreira, R. G. (2008).Electron-beam irradiation of fresh broccoli heads (Brassica oleracea L. italica). LWT-Food

Sci. Technol. 41, 1828–1833.Gomila, M., Solis, J. J., David, Z., Ramon, C., and Lalucat, J. (2008). Comparative reductions of

bacterial indicators, bacteriophage-infecting enteric bacteria and enteroviruses in waste-water tertiary treatments by lagooning and UV-radiation. Water Sci. Technol. 58,2223–2233.

http://www.foodsafety.gov/~dms/prodguid.html




http://www.fda.gov/Food/ScienceResearch/ResearchAreas/SafePracticesforFoodProcesses/ucm090977.htm



http://www.fas.usda.gov/htp/Hort_Circular/2002/02-09/Lettuce.htm



http://www.cfsan.fda.gov/~lrd/fpcarro2.html




http://www.cfsan.fda.gov/~mow/intro.html




Gonzalez, R. J., Luo, Y., Ruiz-Cruz, S., and McEvoy, J. L. (2004). Efficacy of sanitizers toinactivate Escherichia coli O157:H7 on fresh-cut carrot shreds under simulated processwater conditions. J. Food Prot. 67, 2375–2380.

Gordon, D. M. (2001). Geographical structure and host specificity in bacteria and theimplications for tracing the source of coliform contamination. Microbiology 147,1079–1085.

Gordon, D. M., Bauer, S., and Johnson, J. R. (2002). The genetic structure of Escherichia coli

populations in primary and secondary habitats. Microbiology 148, 1513–1522.Gorny, J. R. (2002). Association of Escherichia coli O157:H7 with preharvest leaf lettuce upon

exposure to contaminated irrigation water. J. Food Prot. 65, 741.Gorny, J. (2006). Microbial contamination of fruit and vegetables. In ‘‘Microbiology of Fruits

and Vegetables.’’ (G. M. Sapers, J. R. Gorny, and A. E. Yousef, eds.), pp. 3–29. CRC Press,Lincoln.

Gozzo, F. (2003). Systemic acquired resistance in crop protection: From nature to a chemicalapproach. J. Agric. Food Chem. 51, 4487–4503.

Greco, R. (2002). Preferential flow in macroporous swelling soil with internal catchment:Model development and applications. J. Hydrol. 269, 150–168.

Greene, S. K., Daly, E.R, Talbot, E. A., Demma, L. J., Holzbauer, S., Patel, N.J, Hill, T. A.,Walderhaug, M. O., Hoekstra, R. M., Lynch, M. F., and Painter, J. A. (2008). Recurrentmultistate outbreak of Salmonella Newport associated with tomatoes from contaminatedfields, 2005. Epidemiol. Infect 136, 157–165.

Grewal, S. K., Rajeev, S., Sreevatsan, S., andMichel, F. C., Jr. (2006). Persistence ofMycobacte-

rium avium subsp. paratuberculosis and other zoonotic pathogens during simulated com-posting, manure packing, and liquid storage of dairy manure.Appl. Environ. Microbiol. 72,565–574.

Guenther, S., Huwyler, D., Richard, S., and Loessner, M. J. (2008). Virulent bacteriophage forefficient biocontrol of Listeria monocytogenes in ready-to-eat foods.Appl. Environ.Microbiol.

75, 93–100.Guentzel, J. L., Liang Lam, K., Callan, M. A., Emmons, S. A., and Dunham, V. L. (2008).

Reduction of bacteria on spinach, lettuce, and surfaces in food service areas using neutralelectrolyzed oxidizing water. Food Microbiol. 25, 36–41.

Guo, X., Chen, J. R., Brackett, R. E., and Beuchat, L. R. (2001). Survival of Salmonellae on andin tomato plants from the time of inoculation at flowering and early stages of fruitdevelopment through fruit ripening. Appl. Environ. Microbiol. 67, 4760–4764.

Guo, X., Chen, J., Brackett, R. E., and Beuchat, L. R. (2002). Survival of Salmonella on tomatoesstored at high relative humidity, in soil, and on tomatoes in contact with soil. J. Food Prot.65, 274–279.

Guo, B., Wang, Y., Sun, X., and Tang, K. (2008). Bioactive natural products from endophytes:A review. Appl. Biochem. Microbiol. 44, 136–142.

Hadjok, C., Mittal, G. S., and Warriner, K. (2008). Inactivation of human pathogens andspoilage bacteria on the surface and internalized within fresh produce by using acombination of ultraviolet light and hydrogen peroxide. J. Appl. Microbiol. 104, 1014–1024.

Hancock, D.D, Rice, D.H, Herriot, D.E, and Besser, T. E. (1997). A longitudinal study ofEscherichia coli O157:H7 in fourteen cattle herds. Epidemiol. Infect. 118, 193–195.

Hartmann, C. and Delgado, A. (2004). Numerical simulation of the mechanics of a yeast cellunder high hydrostatic pressure. J. Biochem 37, 977–987.

Harwood, V. J., Wiggins, B., Hagedorn, C., Ellender, R. D., Gooch, J., Kern, J.,Samadpour, M., Chapman, A. C. H., Robinson, B. J., and Thompson, B. C. (2003).Phenotypic library-based microbial source tracking methods: Efficacy in the Californiacollaborative study. J. Water Health 1, 153–166.


Heaton, J. C. and Jones, K. (2008). Microbial contamination of fruit and vegetables and thebehaviour of enteropathogens in the phyllosphere: A review. J. Appl. Microbiol 104,613–626.

Heitman, T. L., Frederick, L. M., Viste, J. R., Guselle, N. J., Morgan, U. M., Thompson, R. C.,and Olson,M. E. (2002). Prevalence ofGiardia andCryptosporidium and characterization ofCryptosporidium spp. isolated fromwildlife, human, and agricultural sources in the NorthSaskatchewan River Basin in Alberta, Canada. Can. J. Microbiol. 48, 530–541.

Hellstrom, S., Kervinen, R., Lyly, M., Ahvenainen-Rantala, R., and Korkeala, H. (2006).Efficacy of disinfectants to reduce Listeria monocytogenes on precut iceberg lettuce.J. Food Prot. 69, 1565–1570.

Hendrix, R. W., Smith, M. C. M., Burns, R. N., Ford, M. E., and Hatfull, G. F. (1999).Evolutionary relationships among diverse bacteriophages and prophages: All theworld’sa phage. Proc. Natl. Acad. Sci. USA 96, 2192–2197.

Herikstad, H., Motarjemi, Y., and Tauxe, R. V. (2002). Salmonella surveillance: A globalsurvey of public health serotyping. Epidemiol. Infect. 129, 1–8.

Hill, D. D., Owens, W. E., and Tchoounwou, P. B. (2005). Impact of animal waste applicationon runoff water quality in field experimental plots. Int. J. Environ. Res. Public. Health 2,314–321.

Himathongkham, S., Bahari, S., Riemann, H., and Cliver, D. (1999). Survival of Escherichia coliO157:H7 and Salmonella typhimurium in cow manure and cow manure slurry. FEMSMicrobiol. Lett. 178, 251–257.

Holtby, I., Tebbutt, G. M., Green, J., Hedgeley, J., Weeks, G., and Ashton, V. (2001). Outbreakof Norwalk-like virus infection associated with salad provided in a restaurant. Commun.

Dis. Public Health 4, 305–310.Hora, R., Kumar, M., Garcia, L., Schumacher, B., Odumeru, J., and Warriner, K. (2005).

Spatial distribution of Salmonella, Escherichia coliO157:H7, and other bacterial populationsin commercial and laboratory-scale sprouting mung bean beds. J. Food Prot. 68,2510–2518.

Hudson, J. A., Billington, C., Carey-Smith, G., and Greening, G. (2005). Bacteriophages asbiocontrol agents in food. J. Food Prot. 68, 426–437.

Hutchison, M. L., Walters, L. D., Avery, S. M., Synge, B. A., and Moore, A. (2004). Levels ofzoonotic agents in British livestock manures. Lett. Appl. Microbiol. 39, 207–214.

Hutchison, M. L., Walters, L. D., Avery, S. M., and Moore, A. (2005). Analyses of livestockproduction, waste storage, and pathogen levels and prevalences in farm manures. Appl.Environ. Microbiol 71, 1231–1236.

Hutchison, M. L., Avery, S. M., and Monaghan, J. M. (2008). The air-borne distribution ofzoonotic agents from livestock waste spreading andmicrobiological risk to fresh producefrom contaminated irrigation sources. J. Appl. Microbiol. 105, 848–857.

Ibarra-Sanchez, L. S., Alvarado-Casillas, S., Rodriguez-Garcia, M. O., Martinez-Gonzales, N. E., and Castillo, A. (2004). Internalization of bacterial pathogens in tomatoesand their control by selected chemicals. J. Food Prot. 67, 1353–1358.

Ibekwe, A. M., Watt, P. M., Shouse, P. J., and Grieve, C. M. (2004). Fate of Escherichia coli

O157:H7 in irrigation water on soils and plants as validated by culture method and real-time PCR. Can. J. Microbiol. 50, 1007–1014.

Ilic, S., Odomeru, J., and LeJeune, J. T. (2008). Coliforms and prevalence of Escherichia coli andfoodborne pathogens on minimally processed spinach in two packing plants. J. Food Prot71, 2398–2403.

Inatsu, Y., Bari, M. L., Kawasaki, S., Isshiki, K., and Kawamoto, S. (2005). Efficacy of acidifiedsodium chlorite treatments in reducing Escherichia coli O157:H7 on Chinese cabbage.J. Food Prot. 68, 251–255.

Ingram, D. T. and Millner, P. D. (2007). Factors affecting compost tea as a potential source ofEscherichia coli and Salmonella on fresh produce. J. Food Prot. 70, 828–834.


Iniguez, A. L., Dong, Y. M., Carter, H. D., Ahmer, B. M. M., and Triplett, E. W. (2005).Regulation of enteric endophytic bacterial colonization by plant defenses. Mol. PlantMicrobe Interact. 18, 169–178.

Islam, M. S., Rezwan, F. B., and Khan, S. I. (1996). Survival of Shigella flexneri in artificialaquatic environment: Effects of different physicochemical stress factors. J. Diarrhoeal Dis.

Res. 14, 37–40.Islam, M., Doyle, M. P., Phatak, S. C., Millner, P., and Jiang, X. (2004). Persistence of

enterohemorrhagic Escherichia coli O157:H7 in soil and on leaf lettuce and parsleygrown in fields treated with contaminated manure composts or irrigation water. J. FoodProt. 67, 1365–1370.

Islam, M., Morgan, J., Doyle, M. P., Phatak, S. C., Millner, P., and Jiang, X. P. (2004). Fate ofSalmonella enterica serovar Typhimurium on carrots and radishes grown in fields treatedwith contaminated manure composts or irrigation water. Appl. Environ. Microbiol. 70,2497–2502.

Itoh, Y., Sugita-Konishi, Y., Kasuga, F., Iwaki, M.., Hara-Kudo, Y., Saito, N., Noguchi, Y.,Konuma, H, and Kumagai, S (1998). Enterohaemorrhagic Escherichia coliO157:H7 presentin radish sprouts. Appl. Environ. Microbiol. 64, 1532–1535.

Jablasone, J., Brovko, L. Y., and Griffiths, M. W. (2004). A research note: The potential fortransfer of Salmonella from irrigation water to tomatoes. J. Sci. Food Agri. 84, 287–289.

Jablasone, J., Warriner, K., and Griffiths, M. (2005). Interactions of Escherichia coli O157:H7,Salmonella Typhimurium and Listeria monocytogenes plants cultivated in a gnotobioticsystem. Int. J. Food Microbiol. 99, 7–18.

Jacobson, A. P., Thunberg, R. L., Johnson, M. L., Hammack, T. S., and Andrews, W. H. (2004).Alternative anaerobic enrichments to the bacteriological analytical manual culturemethod for isolation of Shigella sonnei from selected types of fresh produce. J. AOAC

Int. 87, 1115–1122.James, E. K. and Olivares, F. L. (1998). Infection and colonization of sugar cane and other

graminaceous plants by endophytic diazotrophs. Crit. Rev. Pant Sci. 17, 77–119.Janssen, R., Krogfelt, K. A., Cawthraw, S. A., van Pelt, W., Wagenaar, J. A., and Owen, R. J.

(2008). Host-pathogen interactions in Campylobacter infections: The host perspective. Clin.Microbiol. Rev. 21, 505–518.

Jiang, X. P., Morgan, J., and Doyle, M. P. (2002). Fate of Escherichia coli O157:H7 in manure-amended soil. Appl. Environ. Microbiol. 68, 2605–2609.

Johnson, L. K., Brown, M. B., Carruthers, E. A., Ferguson, J. A., Dombek, P. E., andSadowsky, M. J. (2004). Sample size, library composition, and genotypic diversityamong natural populations of Escherichia coli from different animals influence accuracyof determining sources of fecal pollution. Appl. Environ. Microbiol. 70, 4478–4485.

Johnston, L. M., Jaykus, L. A., Moll, D., Anciso, J., Mora, B., and Moe, C. L. (2005). A fieldstudy of the microbiological quality of fresh produce. J. Food Prot. 68, 1840–1847.

Jongen, W. (2005). In ‘‘Improving the Safety of Fresh Fruit and Vegetables’’ (W. Jongen, ed.).Woodhead Publishing Limited, Cambridge.

Joo, G. J., Kim, Y. M., Lee, I. J., Song, K. S., and Rhee, I. K. (2004). Growth promotion of redpepper plug seedlings and the production of gibberellins by Bacillus cereus, Bacillusmacroides and Bacillus pumilus. Biotechnol. Lett. 26, 487–491.

Kapperud, G., Rorvik, L. M., Hasseltvedt, V., Hoiby, E. A., Iversen, B. G., Staveland, K.,Johnsen, G., Leitao, J., Herikstad, H., and Andersson, Y. (1995). Outbreak of Shigella sonneiinfection traced to imported iceberg lettuce. J. Clin. Microbiol. 33, 609–614.

Karapinar, M. and Gonul, S. A. (1991). Survival of Yersinia enterocolitica and Escherichia coli inspring water. Int. J. Food Microbiol. 13, 315–320.

Karim,M. R., Manshadi, F. D., Karpiscak, M. M., and Gerba, C. P. (2004). The persistence andremoval of enteric pathogens in constructed wetlands. Water Res. 48, 1191–1197.


Kelly, L., Smith, D. L., Snary, E. L., Johnson, J. A., Harris, A. D., Wooldridge, M., andMorris, J. G. (2004). Animal growth promoters: To ban or not to ban? A risk assessmentapproach. Int. J. Antimicrob. Agents 24, 205–212.

Kidmose, U. and Martens, H. (1999). Changes in texture, microstructure and nutritionalquality of carrot slices during blanching and freezing. J. Food Sci. Agric. 79, 1747–1753.

Kim, H. J., Lee, D. S., and Paik, H. D. (2004). Characterization of Bacillus cereus isolates fromraw soybean sprouts. J. Food Prot. 67, 1031–1035.

Kirkham, N., Adams, M., Knight, A., and Carter, M. (2002). Uptake of enteric viruses inirrigation water by plants. In ‘‘Abstracts from the XIIth International Congress of Virol-ogy,’’ p. 502. International Union of Microbiology Societies, Paris.

Klerks, M. M., Franz, E., van Gent-Pelzer, M., Carolien, Z., and van Bruggen, A. H. C. (2007).Differential interaction of Salmonella enterica serovars with lettuce cultivars and plant-microbe factors influencing the colonization efficiency. ISME J. 1, 620–631.

Kniel, K. E., Shearer, A. E., Cascarino, J. L., Wilkins, G. C., and Jenkins, M. C. (2007). Highhydrostatic pressure and UV light treatment of produce contaminated with Eimeria

acervulina as a Cyclospora cayetanensis surrogate. J. Food Prot. 70, 2837–2842.Koopmans, M. (2008). Progress in understanding norovirus epidemiology. Curr. Opin. Infect.

Dis. 21, 544–552.Koseki, S., Yoshida, K., Kamitani, Y., and Itoh, K. (2003). Influence of inoculation method,

spot inoculation site, and inoculation size on the efficacy of acidic electrolyzed wateragainst pathogens on lettuce. J. Food Prot. 66, 2010–2016.

Koseki, S. and Isobe, S. (2006). Effect of ozonatedwater treatment onmicrobial control and onbrowning of iceberg lettuce (Lactuca sativa L.). J. Food Prot. 69, 154–160.

Krampitz, E. S. and Hollander, R. (1999). Longevity of pathogenic bacteria especially Salmo-nella in cistern water. Zentralbl. Hyg. Umweltmed. 202, 389–397.

Kurdziel, A. S., Wilkinson, N., Langton, S., and Cook, N. (2001). Survival of poliovirus on softfruit and salad vegetables. J. Food Prot. 64, 706–709.

Kutter, S., Hartmann, A., and Schmid, M. (2006). Colonization of barley (Hordeum vulgare)with Salmonella enterica and Listeria spp. FEMS Microbiol. Ecol. 56, 262–271.

Lang, L. H. (2006). FDA approves use of bacteriophages to be added to meat and poultryproducts. Gastroenterology 131, 1370.

Lang, N. L. and Smith, S. R. (2007). Influence of soil type, moisture content and biosolidsapplication on the fate of Escherichia coli in agricultural soil under controlled laboratoryconditions. J. Appl. Microbiol. 103, 2122–2131.

Lang, L. (2008). Investigation of outbreak of infections caused by Salmonella Saintpaul.Gastroenterology 135, 1440–1441.

Lapidot, A. and Yaron, S. (2007). The importance of biofilm formation in subsistance ofSalmonella Typhimurium on parsley. Abstr. Gen. Meet. Am. Soc. Microbiol. 107, 521.

Lazarova, V. and Savoye, P. (2004). Technical and sanitary aspects of wastewater disinfectionby UV irradiation for landscape irrigation. Water Sci. Technol. 50, 203–209.

Lee, S. H., Kim, Y. H., Do, J. H., and Kim, S. H. (2007). Occurrence of Cryptosporidium oocystsandGiardia cysts in the Nakdong river and their removal during water treatment. J. Water

Health 5, 163–169.Leifert, C., Ball, K., and Cooper, J. M. (2008). Control of enteric pathogens in ready-to-eat

vegetable crops in organic and ‘‘low input’’ production systems: A HACCP basedapproach. J. Appl. Microbiol. 105, 931–950.

Leverentz, B., Conway, W. S., Alavidze, Z., Janisiewicz, W. J., Fuchs, Y., Camp, M. J.,Chighladze, E., and Sulakvelidze, A. (2001). Examination of bacteriophage as a biocontrolmethod for Salmonella on fresh-cut fruit: A model study. J. Food Prot. 64, 1116–1121.

Leverentz, B., Conway, W. S., Janisiewicz, W., and Camp, M. J. (2004). Optimizing concen-tration and timing of a phage spray application to reduce Listeria monocytogenes onhoneydew melon tissue. J. Food Prot. 67, 1682–1686.


Liang, W. J., Wilson, K. J., Xie, H., Knol, J., Suzuki, S., Rutherford, N. G., Henderson, P. J. F.,and Jefferson, R. A. (2005). The gusBC genes of Escherichia coli encode a glucuronidetransport system. J. Bacteriol. 187, 2377–2385.

Lianou, A. and Sofos, J. N. (2007). A review of the incidence and transmission of Listeriamonocytogenes in ready-to-eat products in retail and food service environments. J. FoodProt. 70, 2172–2198.

Lindow, S. E. and Brandl, M. T. (2003). Microbiology of the phyllosphere. Appl. Environ.Microbiol. 69, 1875–1883.

Li, H. P., Tajkarimi, M., and Osburn, B. I. (2008). Impact of vacuum cooling on Escherichia coliO157:H7 infiltration into lettuce tissue. Appl. Environ. Microbiol. 74, 3138–3142.

Liu, X. G., Gao, K. X., Kang, Z. S., and He, B. L. (2007). Systemic resistance induced bybiocontrol agents in plants and its biochemical and cytological mechanisms. Ying Yong

Sheng Tai Xue Bao 18, 1861–1868.Lohse, N., Kraghede, P. G., andMolbak, K. (2003). Botulism a 38-year-oldman after ingestion

of garlic in chilli oil. Ugeskr. Laeger 165, 2962–2963.Loncarevic, S., Johannessen, G. S., and Rorvik, L. M. (2005). Bacteriological quality of

organically grown leaf lettuce in Norway. Lett. Appl. Microbiol. 41, 186–189.Long, S. M., Adak, G. K., O’Brien, S. J., and Gillespie, I. A. (2002). General outbreaks of

infectious intestinal disease linked with salad vegetables and fruit, England and Wales,1992–2000. Commun. Dis. Public Health 5, 101–105.

Lund, B. M. (1992). Ecosytems in vegetable foods. Soc. Appl. Bacteriol. Symp. 115S–126S.Luo, Y. G. (2007). Fresh-cut produce wash water reuse affects water quality and packaged

product quality and microbial growth in Romaine lettuce. Hortscience 42, 1413–1419.Mahdy, A. K. M., Lim, Y. A. L., Surin, J., Wan, K. L., and Al-Mekhlafi, M. S. H. (2008). Risk

factors for endemic giardiasis: Highlighting the possible association of contaminatedwater and food. Trans. R. Soc. Trp. Med. Hyg. 102, 465–470.

Mannion, C., Lynch, P. B., Egan, J., and Leonard, F. C. (2007). Seasonal effects on the survivalcharacteristics of Salmonella Typhimurium and Salmonella Derby in pig slurry duringstorage. J. Appl. Microbiol. 103, 1386–1392.

Mataragas, M., Skandamis, P. N., and Drosinos, E. H. (2008). Risk profiles of pork andpoultry meat and risk ratings of various pathogen/product combinations. Int. J. FoodMicrobiol. 126, 1–12.

Matos, A. and Garland, J. L. (2005). Effects of community versus single strain inoculants onthe biocontrol of Salmonella and microbial community dynamics in alfalfa sprouts. J. FoodProt. 68, 40–48.

Mattison, K., Karthikeyan, K., Abebe, M., Malik, N., Sattar, S. A., Farber, J. M., andBidawid, S. (2007). Survival of calicivirus in foods and on surfaces: Experiments withfeline calicivirus as a surrogate for norovirus. J. Food Prot. 70, 500–503.

McGrath, S. P., Chaudri, A. M., and Giller, K. E. (1995). Long-term effects of metals in sewagesludge on soils, microorganisms and plants. J. Ind. Microbiol. 14, 94–104.

McNally, A., Roe, A. J., Simpson, S., Thomson-Carter, F. M., Hoey, D. E., Currie, C,Chakraborty, T., Smith, D. G., and Gally, D.L (2001). Differences in levels of secretedlocus of enterocyte effacement proteins between human disease-associated and bovineEscherichia coli O157. Infect. Immun. 69, 5107–5114.

McWatters, L. H., Chinnan, M. S., Walker, S. L., Doyle, M. P., and Lin, C. M. (2002).Consumer acceptance of fresh-cut iceberg lettuce treated with 2% hydrogen peroxideand mild heat. J. Food Prot. 65, 1221–1226.

Mechie, S. C., Chapman, P. A., and Siddons, C. A. (1997). A fifteen month study of Escherichiacoli O157:H7 in a dairy herd. Epidemiol. Infect. 118, 17–25.

Melotto, M., Underwood, W., and He, S. (2006). A novel function of stomata in innateimmunity against plant and human pathogenic bacteria. Phytopathobiol. 96, S77.


Mihaljevic, R. R., Sikic, M., Klancnik, A., Brumini, G., Mozina, S. S., and Abram, M. (2007).Environmental stress factors affecting survival and virulence of Campylobacter jejuni.Microb. Pathog. 43, 120–125.

Milnes, A. S., Stewart, I., Clifton-Hadley, F. A., Davies, R. H., Newell, D. G., Sayers, A. R.,Cheasty, T., Cassar, C., Ridley, A., Cook, A. J., Evans, S. J., Teale, C. J., et al. (2008).Intestinal carriage of verocytotoxigenic Escherichia coli O157, Salmonella, thermophilicCampylobacter and Yersinia enterocolitica, in cattle, sheep and pigs at slaughter in GreatBritain during 2003. Epidemiol. Infect. 136, 739–746.

Mitscherlich, E. and Marth, E. H. (1984). Microbial survival in the environment. In ‘‘Bacteriaand Rickettsiae Important in Human and Animal Health’’ (E. Mitscherlich andE. H. Marth, eds.), pp. 802–854. Springer-Verlag, New York.

Mohle-Boetani, J. C., Farrar, J., Bradley, P., Barak, J.D, Miller, M., Mandrell, R., Mead, P.,Keene, W. E., Cummings, K., Abbott, S., Werner, S. B., Barak, J. D., et al. (2008). Salmonellainfections associated with mung bean sprouts: Epidemiological and environmentalinvestigations. Epidemiol. Infect. 137(3), 357–366.

Moll, D. M. and Vestal, J. R. (1992). Survival of microorganisms in smectite clays: Implica-tions for Martian exobiology. Icarus 98, 233–239.

Mosaddeghi, M. R., Mahboubi, A. A., Zandsalimi, S., and Unc, A. (2009). Influence of organicwaste type and soil structure on the bacterial filtration rates in unsaturated intact soilcolumn. J. Environ. Manage. 90, 730–739.

Muirhead, R. W., Collins, R. P., and Bremer, P. J. (2006). Numbers and transported state ofEscherichia coli in runoff direct from fresh cowpats under simulated rainfall. Lett. Appl.Microbiol. 42, 83–87.

Mukherjee, A., Speh, D., and Diez-Gonzalez, F. (2007). Association of farm managementpractices with risk of Escherichia coli contamination in pre-harvest produce grown inMinnesota and Wisconsin. Int. J. Food Microbiol. 120, 296–302.

Muniesa, M., Moce-Llivina, L., Katayama, H., and Jofre, J. (2003). Bacterial host strains thatsupport replication of somatic coliphages. Ant. Van Leeuwenhoek Int. J. Gen. Mol. Microbiol.83, 305–315.

Myoda, S. P., Carson, C. A., Fuhrmann, J. J., Hahm, B.-K., Hartel, P. G., Yampara-Iquise, H.,Johnson, L.-A., Kuntz, R. L., Nakatsu, C. H., Sadowsky, M. J., and Samadpour, M. (2003).Comparison of genotypic-based microbial source tracking methods requiring a hostorigin database. J. Water Health 1, 167–180.

Naimi, T. S., Wicklund, J. H., Olsen, S. J., Krause, G., Wells, J. G., Bartkus, J. M., Boxrud, D. J.,Sullivan, M., Kassenborg, H., Besser, J. M., Mintz, E. D., Osterholm, M. T., et al. (2003).Concurrent outbreaks of Shigella sonnei and enterotoxigenic Escherichia coli infectionsassociated with parsley: Implications for surveillance and control of foodborne illness.J. Food Prot. 66, 535–541.

Nasser, A. M., Tweto, E., and Nitzan, Y. (2007). Die-off of Cryptosporidium parvum in soil andwastewater effluents. J. Appl. Microbiol. 102, 169–176.

Nayak, R. and Stewart-King, T. (2008). Molecular epidemiological analysis and microbialsource tracking of Salmonella enterica serovars in a preharvest turkey production environ-ment. Foodborne Pathog. Dis. 5, 115–126.

Newman, M.A, Dow, J. M., and Daniels, M. J. (2001). Bacterial lipopolysaccharides andplant-pathogen interactions. Eur. J. Plant Pathol. 107, 95–102.

Nicholson, F. A., Groves, S. J., and Chambers, B. J. (2005). Pathogen survival during livestockmanure storage and following land application. Bioresource Technol. 96, 135–143.

Niemira, B. A. (2007). Relative efficacy of sodium hypochlorite wash versus irradiation toinactivate Escherichia coli O157:H7 internalized in leaves of romaine lettuce and babyspinach. J. Food Prot. 70, 2526–2532.


Ochoa, T. J., Barletta, F., Contreras, C., and Mercado, E. (2008). New insights into theepidemiology of enteropathogenic Escherichia coli infection. Trans. R. Soc. Trop. Med.Hyg. 102, 852–856.

Olsen, S. J., Bishop, R., Brenner, F. W., Roels, T. H., Bean, N., Tauxe, R. V., and Slutsker, L.(2001). The changing epidemiology of Salmonella: Trends in serotypes isolated fromhumans in the United States, 1987–1997. J. Infect. Dis. 183, 753–761.

Oron, G., Goemans, M., Manor, Y., and Feyen, J. (1995). Poliovirus distribution in the soil-plant system under reuse of secondary wastewater. Water Res. 29, 1069–1078.

O’Toole, J., Sinclair, M., Diaper, C., and Leder, K. (2008). Comparative survival of entericindicators, E. coli and somatic and F-RNAbacteriophages on turf-grass.Water Sci. Technol.

58, 513–518.Ottoson, J. and Stenstrom, T. A. (2003). Faecal contamination of grey water and associated

microbial risks. Water Res. 37, 645–655.Ozcakir, O. (2007). Viable but non-culturable form of bacteria. Mikrobiyol. Bul. 41, 477–484.Parajuli, P. B., Mankin, K. R., and Barnes, P. L. (2009). Source specific fecal bacteria modeling

using soil and water assessment tool model. Bioresour. Technol. 100, 953–963.Peck, M. W. (2006). Clostridium botulinum and the safety of minimally heated, chilled foods:

An emerging issue? J. Appl. Microbiol. 101, 556–570.Peterson, D. A., Hurley, T. R., Hoff, J. C., andWolfe, L. G. (1983). Effect of chlorine treatment

on infectivity of hepatitis A virus. Appl. Environ. Microbiol. 45, 223–227.Plotnikova, J.M., Rahme, L. G., andAusubel, F.M. (2000). Pathogenesis of the human opportu-

nistic pathogen Pseudomonas aeruginosa PA14 in Arabidopsis. Plant Physiol. 124, 1766–1774.Pumbwe, L. and Piddock, L. J. (2004). Antibiotics in media for isolation of Campylobacter spp.

do not enhance resistance. J. Antimicrob. Chemother. 54, 961–962.Pund, R. P. and Theegarten, D. (2008). The importance of aeromonads as a human pathogen.

Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz 51, 569–576.Prestamo, G. and Arroyo, G. (1998). High hydrostatic pressure effect on vegetable structure.

J. Food Sci 63, 878–881.Rafferty, S. M., Cassells, A. C., and Falkiner, F. R. (2003). Investigations into the persistence of

Escherichia coli endophytically, in cabbage (Brassica oleracea var. capitata L.) and associatedalteration in host proteins and chitinase activity. J. Food Agric. Environ. 1, 184–189.

Reischer, G. H., Haider, J. M., Sommer, R., Stadler, H., Keiblinger, K. M., Hornek, R. L.,Mach, R. L., and Farnleitner, A. H. (2008). Quantitative microbial faecal source trackingwith sampling guided by hydrological catchment dynamics. Environ. Microbiol. 10,2598–2608.

Reissinger, A., Vilich, V., and Sikora, R. A. (2001). Detection of fungi in planta: Effectivenessof surface sterilization methods. Mycol. Res. 105, 563–566.

Reller, M. E., Nelson, J. M., Molbak, K., Ackman, D. M., Schoonmaker-Bopp, D. J., Root, T. P.,and Mintz, E. D. (2006). A large, multiple-restaurant outbreak of infection with Shigellaflexneri serotype 2a traced to tomatoes. Clin. Infect. Dis. 42, 163–169.

Rideout, K. and Teschke, K. (2004). Potential for increased human foodborne exposure toPCDD/F when recycling sewage sludge on agricultural land. Environ. Health Perspect.

112, 959–969.Robertson, L. J., Campbell, A. T., and Smith, H. V. (1992). Survival of Cryptosporidium parvum

oocysts under various environmental pressures. Appl. Environ. Microbiol. 58, 3494–3500.Robins-Browne, R. M. (2007). The rise and rise of enteropathogenic Escherichia coli. S. Afr.

Med. J. 97, 1182–1186.Rodgers, S. L., Cash, J. N., Siddiq, M., and Ryser, E. T. (2004). A comparison of different

chemical sanitizers for inactivating Escherichia coli O157:H7 and Listeria monocytogenes insolution and on apples, lettuce, strawberries, and cantaloupe. J. Food Prot. 67, 721–731.

Rojas-Valencia, M. N., Orta-de-Velasquez, M. T., Vaca-Mier, M., and Franco, V. (2004).Ozonation by-products issued from the destruction of microorganisms present in waste-waters treated for reuse. Water Sci. Technol. 50, 187–193.


Rosenblueth, M. andMartinez-Romero, E. (2006). Bacterial endophytes and their interactionswith hosts. Mol. Plant-Microbe. Interact. 19, 827–837.

Samish, Z. and Etinger-Tulczynsha, R. (1962). ‘Distribution of bacteria within the tissue ofhealthy tomatoes’. Appl. Microbiol. 11, 7–10.

Santamaria, J. and Toranzos, G. A. (2003). Enteric pathogens and soil: A short review. Int.Microbiol. 6, 5–9.

Santo Domingo, J. W., Harmon, S., and Bennett, J. (2000). Survival of Salmonella species inriver water. Curr. Microbiol. 40, 409–417.

Santo Domingo, J. W., Bambic, D. G., Edge, T. A., and Wuertz, S. (2007). Quo vadis sourcetracking? Towards a strategic framework for environmental monitoring of fecal pollu-tion. Water Res. 41, 3539–3552.

Sapers, G. M. and Sites, J. E. (2003). Efficacy of 1% hydrogen peroxide wash in decontaminat-ing apples and cantaloupe melons. J. Food Sci. 68, 1793–1797.

Scavia, G., Staffolani, M., Fisichella, S., Striano, G., Colletta, S., Ferri, G., Escher, M.,Minelli, F., and Caprioli, A. (2008). Enteroaggregative Escherichia coli associated with afoodborne outbreak of gastroenteritis. J. Med. Microbiol. 57, 1141–1146.

Schulz, B. and Boyle, C. (2005). The endophytic continuum. Mycol. Res. 109, 661–686.Semenov, A. V., van Bruggen, A. H. C., van Overbeek, L., Termorshuizen, A. J., and

Semenov, A. M. (2007). Influence of temperature fluctuations on Escherichia coli O157:H7 and Salmonella enterica Serovar Typhimurium in cowmanure. FEMSMicrobiol. Ecol. 60,419–428.

Seo, K. H. and Frank, J. F. (1999). Attachment of Escherichia coliO157:H7 to lettuce leaf surfaceand bacterial viability in response to chlorine treatment. J. Food Prot. 62, 3–9.

Servais, P., Garcia-Armisen, T., George, I., and Billen, G. (2007). Fecal bacteria in the rivers ofthe Seine drainage network (France): Sources, fate and modeling. Sci. Total Environ. 375,152–167.

Sewell, A. M. and Farber, J. M. (2001). Foodborne outbreaks in Canada linked to produce.J. Food Prot. 64, 1863–1877.

Seymour, I. J. and Appleton, H. (2001). Foodborne viruses and fresh produce. J. Appl.Microbiol. 91, 759–773.

Shi, X., Namvar, A., Kostrzynska, M., Hora, R., and Warriner, K. (2007). Persistence andgrowth of different Salmonella serovars on pre- and postharvest tomatoes. J. Food Prot. 70,2725–2731.

Sidhu, J. P. and Toze, S. G. (2009). Human pathogens and their indicators in biosolids:A literature review. Environ. Int. 35, 187–201.

Sila, D., Smout, C., Vu, S., and Hendrickx, M. (2004). Effects of high-pressure pretreatmentand calcium soaking on the texture degradation kinetics of carrots during thermalprocessing. J Food Sci. 69, E205–E211.

Sivapalasingam, S., Friedman, C. R., Cohen, L., and Tauxe, R. V. (2004). Fresh produce:A growing cause of outbreaks of foodborne illness in the United States, 1973 through1997. J. Food Prot. 67, 2342–2353.

Skraber, S., Helmi, K., Willame, R., Ferreol, M., Gantzer, C., Hoffmann, L., andCauchie, H. M. (2007). Occurrence and persistence of bacterial and viral faecal indicatorsin wastewater biofilms. Water Sci. Technol. 55, 377–385.

Snary, E. L., Kelly, L. A., Davison, H. C., Teale, C. J., and Wooldridge, M. (2004). Antimicro-bial resistance: A microbial risk assessment perspective. J. Antimicrob. Chemother. 53,906–917.

Solodovnikov, I., Ivanenko, A. V., Ustiuzhanin, I., Filatov, N. N., Efremova, N. V.,Glinenko, V. M., Samchuk, G. F., and Ushakova, N. S. (2008). Sonnei dysentery—Thenext in stage of the shigelloses’ evolution. Zh. Mikrobiol. Epidemiol. Immunobiol. 3, 121–124.


Solomon, E. B., Yaron, S., andMathews, K. R. (2002). Transmission of Escherichia coliO157:H7from contaminatedmanure and irrigation water to lettuce plant tissue and its subsequentinternalization. Appl. Environ. Microbiol. 68, 397–400.

Solomon, E. B. and Matthews, K. R. (2006). Interaction of live and dead Escherichia coli O157:H7 and fluorescent microspheres with lettuce tissue suggests bacterial processes do notmediate adherence. Lett. Appl. Microbiol. 42, 88–93.

Soupir, M. L., Mostaghimi, S., Yagow, E. R., Hagedorn, C., and Vaughan, D. H. (2006).Transport of faecal bacteria from poultry litter and cattle manures applied to pastureland.Water Air Soil Pollut. 169, 125–136.

Stoeckel, D. M. and Harwood, V. J. (2007). Performance, design, and analysis in microbialsource tracking studies. Appl. Environ. Microbiol. 73, 2405–2415.

Stringer, S. C., Plowman, J., and Peck, M.W. (2007). The microbiological quality of hot water-washed broccoli florets and cut green beans. J. Appl. Microbiol. 102, 41–50.

Suty, H., De Traversay, C., and Cost, M. (2004). Applications of advanced oxidation pro-cesses: Present and future. Water Sci. Technol. 49, 227–233.

Swaminathan, B. and Gerner-Smidt, P. (2007). The epidemiology of human listeriosis.Microbes Infect. 9, 1236–1243.

Sy, K. V., Murray, M. B., Harrison, M. D., and Beuchat, L. R. (2005). Evaluation of gaseouschlorine dioxide as a sanitizer for killing Salmonella, Escherichia coli O157:H7, Listeria mono-

cytogenes, and yeasts and molds on fresh and fresh-cut produce. J. Food Prot. 68, 1176–1187.Takeuchi, K. and Frank, J. F. (2000). Penetration of Escherichia coliO157:H7 into lettuce tissues

as affected by inoculum size and temperature and the effect of chlorine treatment on cellviability. J. Food Prot. 63, 434–440.

Takeuchi, K. and Frank, J. F. (2001). Quantitative determination of the role of lettuce leafstructures in protecting Escherichia coli O157:H7 from chlorine disinfection. J. Food Prot.

64, 147–151.Takeuchi, K., Matute, C. M., Hassan, A. N., and Frank, J. F. (2000). Comparison of the

attachment of Escherichia coli O157: H7, Listeria monocytogenes, Salmonella typhimurium,and Pseudomonas fluorescens to lettuce leaves. J. Food Prot. 63, 1433–1437.

Tan, R. X. and Zou, W. X. (2001). Endophytes: A rich source of functional metabolites. Nat.

Product Rep. 18, 448–459.Tanner, P. E., Turco, R., Reuhs, B., Applegate, B., and Cousin, M. (2008). ‘‘Visualization of

Attachment and Internalization of Bioluminescent Derivative of Escherichia coli O157:H7ATCC 43895 on Lettuce Leaves.’’ Abstracts of the International Association for FoodProtection, Columbus, Ohio.

Terzieva, S. I. andMcFeters, G. A. (1991). Survival and injury of Escherichia coli, Campylobacter

jejuni, and Yersinia enterocolitica in stream water. Can. J. Microbiol. 37, 785–790.Thompson, R. C., Olson, M. E., Zhu, G., Enomoto, S., Abrahamsen, M. S., and Hijjawi, N. S.

(2005). Cryptosporidium and cryptosporidiosis. Adv. Parasitol. 59, 77–158.Thompson, R. C., Palmer, C. S., and O’Handley, R. (2008). The public health and clinical

significance of Giardia and Cryptosporidium in domestic animals. Vet. J. 177, 18–25.Thunberg, R. L., Tran, T. T., Bennett, R. W., Matthews, R. N., and Belay, N. (2002). Microbial

evaluation of selected fresh produce obtained at retail markets. J. Food Prot. 65, 677–682.Torres, A. R., Araujo, W. L., Cursino, L., Hungria, M., Plotegher, F., Mostasso, F. L., and

Azevedo, J. L. (2008). Diversity of endophytic enterobacteria associated with differenthost plants. J. Microbiol. 46, 373–379.

Turner, C. (2002). The thermal inactivation of E. coli in straw and pig manure. BioresourceTechnol. 96, 521–529.

Turner, C., Williams, A., White, R., and Tillett, R. (2005). Inferring pathogen inactivationfrom the surface temperatures of compost heaps. Bioresource Technol. 95, 521–529.

Tyler, H. L. and Triplett, E. W. (2008). Plants as a habitat for beneficial and/or humanpathogenic bacteria. Ann. Rev. Phytopathol. 46, 53–73.


Tyrrel, S. F. and Quinton, J. N. (2003). Overland flow transport of pathogens from agricul-tural land receiving faecal wastes. Soc. Appl. Microbiol. Symp. Ser. 32, 87S–93S.

United States Department of Agriculture (USDA). (2001). Agricultural environmental man-agement. http://www.nal.usda.gov/wqic/manure.shtmlAccessed Dec. 08.

United States Department ofAgriculture (USDA). (2005). 7 CFR 205.203—Soil fertility and cropnutrient management practice standard. http://vlex.com/vid/19902699Accessed Dec. 08.

Vale, S. (2005). Hepatitis A associated with green onions. N. Engl. J. Med. 353, 2300–2301.Valentin-Bon, I., Jacobson, A., Monday, S. R., and Feng, P. C. (2008). Microbiological quality

of bagged cut spinach and lettuce mixes. Appl. Environ. Microbiol. 74, 1240–1242.Vassilev, N., Vassileva, M., and Nikolaeva, I. (2006). Simultaneous P-solubilizing and bio-

control activity of microorganisms: Potentials and future trends. Appl. Microbiol. Biotech-

nol. 71, 137–144.Vogel, J. R., Stoeckel, D. M., Lamendella, R., Zelt, R. B., Domingo, J. W. S., Walker, S. R., and

Oerther, D. B. (2007). Identifying fecal sources in a selected catchment reach usingmultiple source-tracking tools. J. Environ. Qual. 36, 718–729.

Waage, A. S., Vardund, T., Lund, V., and Kapperud, G. (1999). Detection of small numbers ofCampylobacter jejuni and Campylobacter coli cells in environmental water, sewage, and foodsamples by a seminested PCR assay. Appl. Environ. Microbiol. 65, 1636–1643.

Wachtel, M. R., Whiteland, L. C., and Mandrell, R..E. (2002). Prevalence of Escherichia coli

associated with a cabbage crop inadvertently irrigated with partially treated sewagewastewater. J. Food Prot. 65, 471–475.

Wang, G. D. and Doyle, M. P. (1998). Survival of enterohemorrhagic Escherichia coli\p O157:H7 in water. J. Food Prot. 61, 662–667.

Wang, Y., He, T., Wen, X., Li, T., Waili, T. T., Zhang, W., Zhou, H., Zheng, H., Wen, H.,Davaadorj, N., Gambolt, L., Mukhar, T., et al. (2005). Human cystic echinococcosis in twoMongolian communities in Hobukesar (China) and Bulgan (Mongolia). Trans. R. Soc.Trop. Med. Hyg. 99, 692–698.

Warriner, K., Ibrahim, F., Dickinson,M., Wright, C., andWaites, W.M. (2003a). Interaction ofEscherichia coli with growing salad spinach plants. J. Food Prot. 66, 1790–1797.

Warriner, K., Spaniolas, S., Dickinson, M., Wright, C., and Waites, W. M. (2003b). Internali-zation of bioluminescent Escherichia coli and Salmonella Montevideo in growing beansprouts. J. Appl. Microbiol. 95, 719–727.

Warriner, K., Ibrahim, F., Dickinson, M., Wright, C., and Waites, W. M. (2003c). Internaliza-tion of human pathogens within growing salad vegetables. Biotechnol. Genet. Eng. Rev. 20,117–134.

Weiner, M. and Osek, J. (2007). Shiga toxin producing E. coli-the actual state of knowledge.Medycyna Weterynaryjna 63, 758–762.

Whipps, J. M., Hand, P., Pink, D. A., and Bending, G. D. (2008). Human pathogens and thephyllosphere. Adv. Appl. Microbiol. 64, 183–221.

Whittam, T. S. (1989). Clonal dynamics of Escherichia coli in its natural habitat. Antonie VanLeeuwenhoek 55, 23–32.

Wilson, K. J., Sessitsh, A., Corbo, J. C., Giller, K. E., Akkermans, A. D. L., and Jefferson, R. A.(1995). Beta-glucuronidase (GUS) transposons for ecological and genetic-studies ofrhizobia and other Gram-negative bacteria. Microbiol. 141, 1691–1705.

Winfield, M. D. and Groisman, E. A. (2003). Role of nonhost environments in the lifestyles ofSalmonella and Escherichia coli. Appl. Environ. Microbiol. 69, 3687–3694.

Winfield, M. D. and Groisman, E. A. (2004). Phenotypic differences between Salmonella andEscherichia coli resulting from the disparate regulation of homologous genes. Proc. Natl.

Acad. Sci. USA 101, 17162–17167.Wong, T. P., Byappanahalli, M., Yoneyama, B., and Ray, C. (2008). An evaluation of the

mobility of pathogen indicators, Escherichia coli and bacteriophage MS-2, in a highlyweathered tropical soil under unsaturated conditions. J. Water Health 6, 131–140.

http://www.nal.usda.gov/wqic/manure.shtml

http://www.nal.usda.gov/wqic/manure.shtml

http://vlex.com/vid/19902699

http://vlex.com/vid/19902699


Wu, F. M., Doyle, M. P., Beuchat, L. R., Wells, J. G., Mintz, E. D., and Swaminathan, B. (2000).Fate of Shigella sonnei on parsley and methods of disinfection. J. Food Prot. 63, 568–572.

Xie, Y., Hajdok, C., Mittal, G. S., and Warriner, K. (2008). Inactivation of MS2 F(þ) coliphageon lettuce by a combination of UV light and hydrogen peroxide. J. Food Prot. 71, 903–907.

Yaun, B. R., Sumner, S. S., Eifert, J. D., and Marcy, J. E. (2004). Inhibition of pathogens onfresh produce by ultraviolet energy. Int. J. Food Microbiol. 90, 1–8.

You, Y. W., Rankin, S. C., Aceto, H. W., Benson, C. E., Toth, J. D., and Dou, Z. X. (2006).Survival of Salmonella enterica serovar Newport in manure and manure-amended soils.Appl. Environ. Microbiol. 72, 5777–5783.

Zheng, G., Ma, L., Beuchat, L. R., Erickson, M. C., Phelan, V. H., and Doyle, M. P. (2008).‘‘Potential Internalization of Escherichia coli O157:H7 in Pre-harvest Iceberg Lettuce(Lactuca sativa L.).’’ Abstracts of the International Association for Food Protection, Colum-bus, Ohio.

Zhuang, R. Y., Beuchat, L. R., and Angulo, F. J. (1995). Fate of SalmonellaMontevideo on andin raw tomatoes as affected by temperature and treatment with chlorine. Appl. Environ.Microbiol. 61, 2127–2131.

Zogaj, X., Nimtz, M., Rohde, M., Bokranz, W., and Romling, U. (2001). The multicellularmorphotypes of Salmonella typhimurium and Escherichia coli produce cellulose as thesecond component of the extracellular matrix. Mol. Microbiol. 39, 1452–1463.

CHAPTER 5

Understanding Oil AbsorptionDuring Deep-Fat Frying

Pedro Bouchon

Contents I. Food Deep-Fat Frying: A General Overview 210

A. The deep-fat frying process 210

B. Heat and mass transfer during deep-fat frying 214

C. Structure development during frying 217

II. Nutritional Aspects of Food Deep-Fat Frying 218

A. Frying oils and oil degradation 219

B. Consumption of fried food and human health 220

III. Oil Absorption 222

A. Kinetics of oil absorption 222

B. Factors affecting oil absorption 226

C. Oil absorption reduction 229

References 231

Abstract One of the most important quality parameters of fried food is the

amount of fat absorbed during the process, which undermines

recent consumer trends toward healthier food and low-fat

products. In order to obtain a product with a low fat content, it

is essential to understand the mechanisms involved during the

frying process, so that oil migration into the structure can be

minimized. To get such an understanding, this chapter briefly

describes the frying process from technological and scientific

perspectives. First, it gives a general overview of the frying process

and describes the most important quality attributes of fried food.

Thereafter, it centers on key nutritional aspects, particularly on the

effect of excessive oil consumption on human health, oil degradation,


Department of Chemical and Bioprocess Engineering, Pontificia Universidad Catolica de Chile,Santiago, Chile

209

210 Pedro Bouchon

and toxic compounds generation in fried food. Finally, this chapter

discusses the most important factors affecting oil absorption, oil

absorption kinetics, and different strategies that may be adopted to

decrease oil content.

I. FOOD DEEP-FAT FRYING: A GENERAL OVERVIEW

Deep-fat frying, also known as immersion frying, is one of the oldest andmost common unit operations used in the preparation of food. The pro-cess was first developed around the Mediterranean area due to the influ-ence of olive oil there, but today numerous processed foods are deep-fatfried because of the unique flavor–texture combination imparted to thefood (Varela, 1988). Certainly, fried products are of great importance tothe food industry because of their popularity among consumers and thehuge quantities of fried food and oils that are used at industrial andcommercial levels. A critical aspect of deep-fat fried food is the highamount of oil that is absorbed during the process, reaching in somecases 40% of the total food product weight. Numerous studies haverevealed that excess consumption of fat is a key dietary contributor tocoronary heart disease and perhaps cancer of the breast, colon, andprostate (Browner et al., 1991), imposing an alert to human consumption.Despite this, consumption of oils and fats is still high. For instance, in theUnited States, consumers eat four or more snacks a day and consumemore than 6.5 billion pounds of snack food annually. As such, salty snacksaccount for slightly over half of the total snack sales and are consequentlya large part of the American diet (Mintel International Group Ltd, 2006).A wide variety of food materials can be used to produce fried products,including vegetables, meat, dairy, and grains. Key growth categories arethose that offer the most product variations adhering to convenience,flavor, and health trends. In terms of health, interests in salty snackproducts that are organic or all natural, low-calorie, low-fat, low-carbohydrate, low-sodium, or offer health-promoting benefits such aselimination of trans fat are in greater demand by consumers. Althoughconsumers are interested in healthier snack products, they are not willingto sacrifice flavor. Intense and full-flavor snacks remain an importanttrend in the salty snack market (Mariscal and Bouchon, 2008).

A. The deep-fat frying process

Deep-fat frying can be defined as a process of cooking food by immersingthem in edible oil at a temperature above the boiling point of water, andtherefore, may be classified as a dehydration process (Farkas, 1994).Frying temperatures usually range between 130 and 190 �C, but most

Understanding Oil Absorption During Frying 211

common frying temperatures are in the 170–190 �C range. Deep-fat fryingis a complex unit operation involving high temperatures, significantmicrostructural changes to both the surface and the body of the food,and simultaneous heat and mass transfer resulting in flows in oppositedirections of water vapor (bubbles) and oil at the surface of the piece asdepicted in Fig. 5.1 (Bouchon et al., 2003). The high temperatures of thefrying oil lead to the evaporation of water at the surface of the food. Dueto evaporation, water in the external layers of the product moves tothe surrounding oil and surface drying occurs, inducing crust formation.Additionally, oil is absorbed by the food, replacing part of the water(Mellema, 2003). One of major aim of deep-fat frying is to seal the foodsurface while immersing the food into the oil bath so that its flavor andjuices can be successfully retained within the food. As a matter of fact,most of the desirable characteristics of fried food are derived from theformation of a composite structure: a dry, porous, crisp, and oily outerlayer or crust and a moist cooked interior (Bouchon and Aguilera, 2001).

Frying technology is important to many sectors of the food industry,including the suppliers of oils and ingredients, fast-food shop and restau-rant operators, industrial producers of fully fried, par-fried, and snackfoods, and manufacturers of frying equipment (Blumenthal, 1991). Deep-fat fryers basically consist of a chamber where heated oil and food areplaced and the size depends on their use. Accordingly, the frying equip-ment is divided into two broad categories: (1) batch frying equipment,normally used in catering restaurants and small plants and (2) continuousfryers, which are used on an industrial scale to process large amounts of

Mass transferHeat transfer

Core

Crust

Conduction:

Convection: Water vapor:

Oil:

Water:

FIGURE 5.1 Schematic diagram of simultaneous heat transfer (left-hand side of the

figure) and mass transfer (right-hand side of the figure) during deep-fat frying (with the

courtesy of M. C. Moreno).

212 Pedro Bouchon

food. Fryers normally operate under atmospheric conditions; howeverlow- or high-pressure conditions may be used.

1. Fried foodAwide spectrum of fried food is available in the market. They are usuallyclassified into three categories: (1) thin products, such as potato, tortilla,and banana chips, (2) thick products, such as French fries, and (3) bat-tered/breaded food, such as fish fingers. Thin products are nearly fullydehydrated (moisture content lower than 5%), a requirement for shelf-lifestability, which is around 2months. Their fat content is high, achieving upto 40% w.b. (Dobraszczyk et al., 2006). Oil stability is the key factor duringstorage, rather than fungal spoilage, because of the development of off-flavors. Thick as well as buttered/breaded products have a higher watercontent (ranging between 30% and 50% w.b.) and lower oil content.Frozen par-fried French fries may have oil content as low as 5% w.b.;however, these products must be either oven-cooked or go through asecond frying step before consumption, which necessarily increases thefat intake per portion (there is further dehydration and, if fried, additionaloil intake). In fact, when using a second frying stage, the final oil content isgenerally higher than in fresh fried products. Doughnuts are also anextremely popular fried food category. They have a high oil content thatranges from 15% to 20% w.b., but about 10% of the fat is used in thepreparation of the dough. Battered and breaded foods (fish/chicken)contain similar oil contents of around 15–20% w.b. A critical aspect ofthese products is the contrast between the crispy and oily outer layer andthe soft cooked interior (Dobraszczyk et al., 2006).

2. Frying equipmentModern batch fryers are constructed with high-grade stainless steel toavoid oxidation catalysis. Usually, the operators immerse and remove thebaskets manually from the oil, but new equipment may include an auto-matic basket-lift system. The device may consist of one or more chamberswith an oil capacity ranging from 5 to 25 l, and oil may be directly heatedthrough electricity, gas, or fuel (Dobraszczyk et al., 2006). Importantfactors to consider when selecting a batch fryer are power source, speedof temperature recovery, and safety. The simplest heating system consistsof gas flames directly placed underneath the bottom of the vessel. Oil canalso be directly heated through an electrical resistance heater that may beinstalled few inches above the bottom of the fryer, allowing the arrange-ment for a cool zone at the bottom of the vessel where debris can fall,minimizing oil damage. This is a clear advantage compared to the previ-ous heating system, where the provision of a cold zone under the heatersis not possible (Rossell, 1998). New developed high-efficiency fryersinclude turbo-jet infrared burners that use up to 40% less energy than


the standard gas-fired fryers with the same capacity (Moreira et al., 1999).In order to increase shelf life, avoid smoking, charring, and off-flavordevelopment, oil filtration and removal of food scraps are essential prac-tices that need to be carried out every day. New equipment can alsohave a built-in pump filtration unit for the removal of sediments(Kochhar, 1998).

Continuous fryers are used to process large amounts of food, having athroughput that varies from 250 to 25,000 kg product per hour (Moreira,2006). These are automated machines that consist of a frying vessel whereoil is maintained at the desired temperature, a conveyor belt that trans-ports the food through the oil (the product is often pushed through thebath by means of a screen and/or paddles), and an extraction system thateliminates the fumes, primarily made up of moisture and a fine mist offatty acids (Dobraszczyk et al., 2006; Moreira et al., 1999). The oil may beheated directly using a battery of gas burners or an electric heater in thefrying vessel, or by means of an external heat exchanger where oil iscontinuously pumped through. Some continuous fryers are designedwithmultiple heating zones along the fryer that can be adjusted separately,providing optimal temperature control to improve product quality.Continuous fryers may be also provided with an indirect oil heatingsystem unit. In those systems, oil is heated by pumping a heated thermalfluid into a tube arrangement immersed in the oil bath (Dobraszczyk et al.,2006; Moreira et al., 1999). It is important to mention that in continuousfryers, the oil that is constantly absorbed by the fried product needs to bereplaced with fresh oil continuously. The amount of fresh oil added to thevessel is the oil turnover, defined as (weight of oil in the fryer)/(weight of oiladded per hour) (Banks, 1996), and therefore represents the time neededto replace all the oil contained in the equipment. Fast oil turnoversare desired since they preserve the oil quality better. Normally, the oilturnover is kept between 3 and 8 h (Kochhar, 1998).

Deep-fat fryers may also operate at a higher pressure (9�32 psi). Thesedevices have been developed to meet particular needs primarily in certaincatering outlets, especially those devoted to chicken frying because of theuniform color and improved texture (higher moisture content) conferredto products. Pressure fryers may reduce the frying time considerably, butthey can also increase frying oil deterioration rate since steam retentionwithin the fryer increases free fatty acids content (Moreira et al., 1999).

Another technology that is being increasingly adopted is vacuumfrying, which consists of a deep-fat frying process carried out in a closedsystem under pressure well below the atmospheric levels (preferablylower than 1 psi), making it possible to reduce substantially the fryingtemperature due to water boiling-point depression. The low temperaturesemployed and minimal exposure to oxygen account for most of itsbenefits, which include natural color, flavor, and nutrient preservation

214 Pedro Bouchon

(Mariscal and Bouchon, 2008; Shyu and Hwang, 2001), as well as oilquality protection (Shyu et al., 1998) and reduction of toxic compoundgeneration (Granda et al., 2004). The equipment was first developed byFlorigo B.V. during the sixties to produce high-quality chips; however,due to the improvement in blanching technology and in raw materialquality, the use of this technology almost disappeared (Moreira et al.,1999). Nowadays, vacuum frying technology is being used to maintainnatural colors, flavors, and nutrients in high added-value products, suchas vegetables and fruits (Dueik and Bouchon, 2009).

B. Heat and mass transfer during deep-fat frying

From an engineering perspective, deep-fat frying can be defined as a unitoperation where heat and mass transport phenomena occur simulta-neously. Convective heat is transferred from the frying media to thesurface of the product, which is thereafter conducted within the food.Mass transfer is characterized by the loss of water from the food as watervapor and the movement of oil into the food (Singh, 1995).

Water evaporation initiates at the surface of the product after initialheating occurs, and the boiling point of the interstitial liquid is reached,which is slightly higher than the boiling point of water. After the initialsurface water is lost, water starts escaping vigorously and heat trans-ferred through natural convection gives path to a forced convectionregime due to the high turbulence associated with nucleate boiling.As frying progresses, the evaporation front moves toward the interior ofthe product and a dehydrated crust layer is formed, whose temperaturerises above the boiling point of water. It is important to note that themaximum surface temperature only approaches that of the frying oil,remaining 10–15 �C below it, because of the heat transfer resistance atthe boundary oil layer in contact with the surface of the food (Fig. 5.2).On the other hand, the temperature inside the food material (known ascore region), where liquid water is still there, is restricted to valuesaround the boiling point of the liquid. As frying proceeds, the waterloss rate decreases (falling-rate period) leading to bubble-end point, thatis, when water escape stops (Singh, 1995). This is actually the processingcondition that must be fulfilled in chip like products (crisps in the U.K.),where a maximum of 5% moisture content is permitted.

Mass transfer during frying is not only characterized by the movementof water in the form of vapor from the food into the oil, but also by themovement of oil into the food. Frying is a dehydration process wherewater escape leaves empty spaces within the crust structure, whichin turn determines the volume available for oil absorption. In fact,the amount of oil uptake has been shown to be directly proportional to

100 2000 300

60

100

140

20

180

Time (s)

100 2000 300Time (s)

100 2000 300Time (s)

Tem

pera

ture

(�C

)

60

100

140

20

180

Tem

pera

ture

(�C

)

60

100

140

20

180

Tem

pera

ture

(�C

)

R2= 0.886

R2= 0.992

R2= 0.989

R2= 0.983

R2= 0.934

R2= 0.918

Surface T�: ExperimentalModel

Model ExperimentalCentre T�:

FIGURE 5.2 Predicted and experimentally determined temperatures (mean values)

when frying a raw potato cylinder at 155 �C (top), 170 �C (middle), and 185 �C (bottom);

from Bouchon and Pyle (2005a).


the amount of moisture lost, as will be discussed in the following sections(Gamble et al., 1987).

One of the key parameters that distinguishes frying from other unitoperations is the high heat transfer rates that are achieved, which are farhigher than those found during baking and drying. Heat transfer rates tothe surface of the food will depend on the thermal properties and chemi-cal composition of the frying medium, and on the turbulence generatedby the vigorous vapor escape. Several authors have attempted to measurenatural convective heat transfer coefficients mainly using a metal trans-ducer. Evidently, the absence of water vapor surrounding the metal piece

216 Pedro Bouchon

yields a natural convective heat transfer coefficient that is different fromthat when a food is undergoing frying and that is only meaningful duringthe early stages of frying when few bubbles are present. Using the lumpedcapacity method with a spherical aluminum transducer, Miller et al.(1994) estimated natural convective heat transfer coefficients for canolaoil, palm oil, corn oil, and soybean oil at 170, 180, and 190 �C. Theyobtained values ranging from 250 to 280 W/(m2 K). Using a similarapproach, Moreira et al. (1995a) estimated the natural convective heattransfer coefficient for soybean oil, which was 280 W/(m2 K) when heatedat 190 �C. Comparable values were obtained by Bouchon and Pyle(2005a), who determined natural convective heat transfer coefficients of262, 267, and 282 W/(m2 K) when heating palm olein at 155, 170, and185 �C, respectively.

Several studies have attempted to estimate boiling convective heattransfer coefficients for immersion frying. Hubbard and Farkas (1999)obtained maximum average values of 610, 650, and 890 W/(m2 K), whenfrying potato cylinders at 120, 150, and 180 �C, respectively, which arefar higher than natural convective ones. In addition, they found that thetime to reach these maxima decreased as the frying temperatureincreased and that the convective heat transfer coefficient graduallydecreased to 300 W/(m2 K) over the duration of the process. Costaet al. (1999) reported maximum average values of 443 and 650 W/(m2

K) and average values of 353 and 389 W/(m2 K), after approximately5 min when frying French fries at 140 and 180 �C, respectively. Interest-ingly, they explained that the heat transfer coefficient might be expectedto be position dependent due to the difference in turbulence occurring atdifferent locations in the product. In fact, Sahin et al. (1999) founddifferences when determining the boiling convective heat transfer coef-ficient at the top and bottom surfaces of potato slices during frying(150–190 �C). Contrary to what might be expected, they determinedhigher coefficients at the bottom surface (450–480 W/(m2 K)) as com-pared to the top surface (300–335 W/(m K)) until crust was formed. Theyattributed this to the fact that a strong insulating effect was produced bythe vigorous escape of bubbles at the top surface, while at the bottomsurface bubbles remained in a single layer, providing a lower resistance.Bouchon and Pyle (2005a) estimated boiling convective heat transfercoefficients for increasing frying times at different temperatures, whichranged approximately from 260 to 600 W/(m2 K), similar to those foundby Costa et al. (1999) and Sahin et al. (1999). They adjusted a first-orderkinetic model to experimental data to describe the change with fryingtime, which they used when testing a mathematical model of the pro-cess. Overall, all studies have found that convective heat transfer coeffi-cients are up to two or three times greater that those measured in theabsence of bubbling. This research has made it possible to get a better


understanding of the different heat transfer rates and vapor escapingregimes encountered during the different periods of frying.

C. Structure development during frying

The structure of the crust and core regions of fried food products is theresult of several alterations, most of which occur at the cellular andsubcellular levels. Aguilera and Gloria (1997), using fast freezing afterfrying and cryosectioning, demonstrated that three distinct microstruc-tures exist in finished fried commercial French fries: (1) a thin outer layer(approx. 250 mm) formed by the remnants of cell walls of cells damagedby cutting; (2) an intermediate layer of shrunken intact cells, whichextends to the evaporation front; (3) the core with fully hydrated intactcells containing gelatinized starch. Microstructural changes in the coreregion are similar to those occurring during simmering since this innerstructure is restricted to temperatures below the boiling point of water.Main changes include starch gelatinization (in starchy products), soften-ing of the middle lamellae (which is greatly responsible for the so-calledmealy texture), and protein denaturation.

Microstructural changes at the crust are certainly more aggressivethan those occurring in the inner structure due to the exposure to tem-peratures well above 100 �C (in atmospheric deep-fat frying). Besides thephysical damage produced when the product is cut, chemical and physi-cal changes include starch gelatinization and subsequent dehydration,protein denaturation, breakdown of cellular adhesion, water evaporationand rapid dehydration of cells located in the forming crust, and oil uptakeitself (Bouchon and Aguilera, 2001).

Desired organoleptic properties, particularly textural ones, are a directconsequence of these microstructural changes. A chip must be firm andsnap easily when deformed, emitting a crunchy sound (Krokida et al.,2001a), whereas in thick products the better the contrast between a richand soft inner structure and a crispy outside, the better the product(Moreira, 2006). Firmness is often related to starch swelling and gelatini-zation, as well as to the stability of pectic substances of the cell wall andmiddle lamellae.

The importance of microstructural changes occurring during deep-fatfrying has been greatly recognized when studying oil absorption mechan-isms. In fact, as commented by Baumann and Escher (1995), the explana-tion of factors affecting oil uptake needs to be validated by a structureanalysis in relation to the location of oil deposition and to the mechanismof oil adhesion to the surface. Numerous studies have shown that oiluptake during deep-fat frying is confined to the surface region of thefried product and in cavities and open pores within the two outer layersthat constitute the crust (Bouchon et al., 2001; Farkas et al., 1992; Keller

150mmA

C

C

A

B

B90mm

FIGURE 5.3 Orthogonal sections (A, B, and C) of a reconstituted image obtained by CLSM,

showing the oil location in the crust of a fried potato; from Bouchon and Aguilera (2001).

218 Pedro Bouchon

et al., 1986; Saguy et al., 1997). Bouchon and Aguilera (2001) and Pedreschiet al. (1999) used noninvasive confocal laser scanning microscopy to studyoil location directly in fried potatoes, where they observed that oil seemedto flow through the passages that imposed the lowest resistance and wasconcentrated in concave shells around the cells, with no presence of oil intheir interior (Fig. 5.3). Microstructural evidence plus the fact that oiluptake is related to the amount of moisture lost, are key aspects toconsider the microstructure of the crust region (mean pore size, connect-edness, and permeability) as the single most important product-relateddeterminant for the final oil uptake into the food (Bouchon et al., 2001).Specific aspects related to oil absorption kinetics will be discussed inSection III.

II. NUTRITIONAL ASPECTS OF FOOD DEEP-FAT FRYING

One of the most important quality parameters of fried food is the amountof fat absorbed during the process, which is incompatible with recentconsumer trends toward healthier food and low-fat products (Bouchon


and Pyle, 2004). In fact, current nutrition recommendations point to areduction of total dietary fats, including trans and saturated fatty acids.In addition, important nutritional compounds degrade during the pro-cess, and toxic molecules may generate either in the foodstuff or in thefrying oil itself, whose intake should be at least limited.

A. Frying oils and oil degradation

Food can be fried in a wide range of fats and oils, which include vegetableoils, shortenings, animal fats, or a mixture thereof. Most important criteriaused to select frying oils are long frying stability, fluidity, bland flavor,low tendency to foam or form smoke, low tendency to gum (polymerize),oxidative stability of the oil in the fried food during storage, and certainlyprice (Kochhar, 1999). Saturated fatty acids provide a greater stability infrying applications, but they are undesirable from a nutritional stand-point (Sanibal and Mancini-Filho, 2004). Conversely, oils high in polyun-saturated fatty acids show lower thermo-oxidative stability than richmonoenoic unsaturated fatty acids or saturated fatty acids oils (Kitaet al., 2005). Antioxidants such as tertiary butyl hydroquinone (TBHQ),butylated hydroxyanisole (BHA), and butylated hydroxytoluene (BHT)may be added to improve oil stability; however, some of them are limitedor prohibited under certain regulations. TBHQ is regarded as the bestantioxidant for protecting frying oils against oxidation and, like others, itprovides carry-through protection to the finished fried product(Dobraszczyk et al., 2006).

Most popular oils used for frying are palm oil and its fractions, sun-flower oil (especially high-oleic sunflower oil), rapeseed (canola), andsoybean oils. The last two have a high level of linolenic acid (8–10%),making them vulnerable to oxidation and off-flavor development, andtherefore, can be slightly hydrogenated for industrial frying. This proce-dure can also be applied to sunflower oil and may be attractive whenrequiring a high polyunsaturated-to-saturated ratio for dietary purposes(Rossell, 1998). Olive oil has excellent attributes, which make it suitablefor frying, that is, a low level of polyunsaturated fatty acids and a mixtureof phenolic antioxidants that make it resistant to oxidation. However,extravirgin and virgin olive oils are far too expensive for industrial use,converting refined solvent-extracted olive oil as a plausible candidate forcertain industrial frying operations (Dobraszczyk et al., 2006). Animal fats,despite their high level of saturated fatty acids, may also be used in deep-fat frying due to the characteristic flavor imparted to the food and/or lowcost. In turn, fish oils are rarely used as frying medium, since their highlevel of long-chain polyunsaturated fatty acids makes it prone to oxida-tion (Rossell, 1998).

220 Pedro Bouchon

Care must be taken when selecting frying oils, since they undergothermal, oxidative and hydrolytic degradation due to their exposure toelevated temperatures in the presence of air and moisture (Kita et al.,2005). Water, which is released from the foodstuff, attacks ester linkagesof triacylglycerols giving rise to di- and monoglycerides, glycerols, andfree fatty acids, free molecules that are more susceptible to oxidative andthermal degradation than when esterified to the glycerol molecule (Choeand Min, 2007). Oil thermal oxidation to form peroxides takes place byloss of hydrogen in the presence of trace metals, heat, and light, givingrise to hydroperoxides. Hydroperoxides are compounds that are notstable under deep-fat frying conditions and may undergo fission toproduce awide variety of secondary lipid peroxidation products, includingaldehydes, ketones, and other carbonyl-containing compounds (Mahunguet al., 1999). These compounds contribute to the volatile fraction of degradedfrying oils, determining the development of off-flavors in the fried product(Melton et al., 1994; Subramanian et al., 2000). In addition, dimers, oligomers,and polymersmay be formed, giving rise to excess darkening, increasing oilviscosity, and decreasing smoke point of the frying oil (Choe andMin, 2007;Mahungu et al., 1999).

No-calories fat substitutes, such as sucrose polyesters (Olestra), whichare synthesized from sucrose and fatty acid methyl esters, have beenwidely studied and several snacks fried in this medium are available inthe market place. This product has no calories since digestive enzymes arenot able to break it down due to structural impairment. A major disad-vantage that prevents a wide acceptance of this product is related to thegastrointestinal discomfort that may be caused to some individuals(Dobraszczyk et al., 2006, p. 104).

B. Consumption of fried food and human health

Fat has a strong influence on the palatability of fried foods. The inclusionof cooking fat into the crusty surface, which is developed during thefrying process, helps in building up the crunchiness that is highlyappreciated by consumers. On the other hand, the linkage betweenoverconsumption of fat and several diseases has been well-documented.Oil consumption, especially saturated fat, is considered to be one of thekey dietary contributors to diseases like obesity, coronary heart disease,cancer, diabetes, and hypertension (Saguy and Dana, 2003). In addition,several studies have provided strong evidence that trans fatty acidsincrease plasma concentration of low-density lipoproteins and reducethe concentration of high-density ones (Ascherio and Willett, 1997).Trans fatty acids are produced during hydrogenation, a process that iscommonly used to increase thermal stability of frying oils, but they can


be also generated during high thermal processing, such as deep-fatfrying (Choe and Min, 2007). It is estimated that diet-related diseasescost the society over US$250 billion annually in medical expenses andloss of productivity (Anand and Basiotis, 1998). As a consequence,consumer trends are moving toward healthier food and low-fat pro-ducts, creating the need to reduce the amount of oil in end products.Despite such market forces, the consumption of snack food is increasingin developed and developing countries, and fried products still containlarge amounts of fat varying from 5% in frozen French fries to up to 40%in potato chips (Dobraszczyk et al., 2006). Due to the large contributionof fried foods to the total saturated and trans fatty acids intake, the use ofhealthier oil sources offers immense potential to favorably alter popula-tion fat intake (Minihane and Harland, 2007).

In relation to cancer, there is some evidence that highly oxidized andheated fats may have carcinogenic characteristics. HNE (4-hydroxy-2-trans-nonenal), a secondary lipid peroxidation product derived fromlinoleic acid oxidation, has assumed particular interest because it hasshown cytotoxic and mutagenic properties. Its toxicity, as well othersecondary lipid peroxidation products (HHE: 4-hydroxy-2-trans-hexenaland HOE: 4-hydroxy-2-trans-octenal), is explained through the high reac-tivity with proteins, nucleic acids, DNA, and RNA. Research links them todifferent diseases such as atherosclerosis, Alzheimer’s, and liver diseases(Seppanen and Csallany, 2006). Research is rapidly progressing, butresults are still not conclusive.

In addition to generation of toxic compounds in the frying oil,toxic molecules may be generated in foodstuff. In April 2002, Swedishscientists sounded an alarm when they discovered that certain cookedfood, particularly potato chips and French fries, contained high levelsof acrylamide, a chemical compound that is listed by the WorldHealth Organization (WHO) as a probable human carcinogen (Mitka,2002). This substance has been shown to be produced when food isheated above 120 �C due to a reaction between amino acids andreducing sugars (Mottram et al., 2002). WHO has not yet called forany reduction in food containing high levels of acrylamide; however,several studies that aim at reducing its content in fried food are nowavailable (Dueik and Bouchon, 2009). Acrylamide may be convertedinto glycilamide by living organisms, a compound that is thought tobe considerably more toxic than acrylamide; however, little research isyet available in the scientific literature (Besaratinia and Pfeifer, 2004;Koyama et al., 2006). Other heat-induced harmful compounds may befound in certain food. Among them, we can find droxymethylfurfuralin carbohydrate-rich foods and heterocyclic amines in protein-richfoods. An in-depth discussion about toxic compound generation maybe found in Dueik and Bouchon (2009).

222 Pedro Bouchon

III. OIL ABSORPTION

As explained in previous sections, frying is a complex unit operationinvolving simultaneous heat and mass transfer, leading to the removalof water from the food and oil absorption at the surface of the piece. It ischaracterized by the existence of two different regions, the crust and thecore, separated by a moving evaporation front, which propagates inwardas frying progresses. A key quality factor of fried food is the amount of oiluptake, which should be minimized. In order to obtain food productswith a low fat content, it is essential to understand the mechanismsinvolved during the frying process, especially the kinetics aspects, sothat oil migration into the structure can be minimized.

A. Kinetics of oil absorption

Even though it is not fully understood when and how the oil penetratesinto the food structure, it has been shown that most of the oil is confinedto the surface region of the fried product (Bouchon et al., 2001; Farkas et al.,1992; Keller et al., 1986; Saguy et al., 1997) and there is strong evidence thatit is mostly absorbed during the cooling period (Aguilera and Gloria,1997; Bouchon et al., 2003; Moreira et al., 1997; Ufheil and Escher, 1996).For that reason, it is believed that during frying, after initial heatingoccurs, the vigorous escape of water vapor would generate a barrier toprevent oil migration into the porous crust and as a consequence oilabsorption would be limited during most of the immersion period. As aresult, oil uptake would mainly result from the competition betweendrainage and suction into the porous crust once the food is removedfrom the oil and cools down, being essentially a surface-related phenom-enon. The mechanism of oil absorption was first explained by Gambleet al. (1987). They suggested that the largest amount of oil was pulled intothe product when it was removed from the fryer because of the vacuumeffect due to steam condensation. Accordingly, they suggested that oilabsorption depended on the amount of water removed and on the waymoisture was lost. In 1996, Ufheil and Escher (Ufheil and Escher, 1996)studied the dynamics of oil uptake during deep-fat frying of potato slicesusing a fat-soluble and heat-stable dye (Sudan Red B). They determinedthat most of the oil was absorbed when the product was removed fromthe oil bath and proposed that oil uptake was primarily a surface phe-nomenon, involving equilibrium between adhesion and drainage of oilupon removal of the product from the oil. Matz (1993), when focusing onpostfrying cooling kinetics, determined that potato chips only absorbed15% of the total oil when they were rapidly removed from the fryer, whiletheir temperature was still rising. Moreira et al. (1997) determined that


only 20% of the total oil content of tortilla chips was absorbed while theywere immersed in the oil bath and that almost 64% of the total oil contentwas absorbed during postfrying cooling, the rest remaining on the surface(outer layers of the product). Later, Bouchon et al. (2003) combined andadapted the methods developed by Ufheil and Escher (1996) and Moreiraet al. (1997) and were able to distinguish three different oil fractions whenfrying potato cylinders (155, 170, and 185 �C), that is, (1) structural oil(STO), which represents the amount of oil absorbed during frying, (2)penetrated surface oil (PSO), which represents the amount of oil suctionedinto the food during cooling following its removal from the fryer, and (3)surface oil (SO), that is, the oil that remains on the surface. A schematicdiagram showing these oil fractions is presented in Fig. 5.4. Resultsshowed that only a small amount of oil was able to penetrate duringfrying since most of the oil was picked up at the end of the process,suggesting that oil uptake and water removal were not synchronousphenomena. After cooling, oil was located either on the surface of theproduct or was suctioned into the porous crust microstructure, with aninverse relationship between them for increasing frying times. Accordingto experimental facts, several approaches have been used to describe andmodel oil absorption. Moreira and Barrufet (1998) explained the mecha-nism of oil absorption during cooling in tortilla chips in terms of capillaryforces. This hypothesis was supported by experimental results, wherethey determined that oil uptake occurred during the first 20 s of cooling,that is, when the temperature was still above the condensation tempera-ture (�100 �C). Ni and Datta (1999) developed amultiphase porous mediamodel to predict energy transfer, water loss, and oil absorption duringfrying. They assumed that vapor and air transport took place throughconvection and diffusion, whereas liquid phase transport (water and oil)was mediated by convective and capillary flows. In their model, they

Surface oil

Structural oilPenetrated surface oil

FIGURE 5.4 Diagram showing the three locations of oil in the product microstructure

after deep-fat frying; from Bouchon et al. (2003).

224 Pedro Bouchon

considered that oil absorption could take place during the immersionperiod as water moved out from the food, and therefore, they did nottake into account oil absorption during postfrying cooling, as determinedexperimentally by several authors. In recent years, Bouchon et al. (2005b)developed a modified form of the Washburn equation, where theyexpressed the pressure difference needed to initiate oil infiltration duringpostfrying cooling (Patm – Ppore > 0), as a function of capillary pressureand vapor pressure, enriching the total driving pressure. A schematicdiagram showing the capillary penetration phenomena when havingdifferent arrangements is shown in Fig. 5.5.

If a reference datum plane (h ¼ 0) is set at the bottom of each of thecapillaries shown in the previous Fig. 5.5, the expression for the totaldriving force, that is the piezometric pressure difference along the pene-tration length h for a capillary with an upward (þrgh cos a) or a down-ward (�rgh cos a) orientation, can be represented by following equation(Bouchon and Pyle, 2005b),

DP� ¼ P�2 � P�

1 ¼ Patm � Ppore

¼ Patm � PV � 2s cos yr

� rgh cos a

0@

1A (1)

where P�2 is piezometric pressure at the bottom of the capillary (Pa);P�

1,piezometric pressure at the liquid side of the meniscus (Pa); Pv, vaporpressure (Pa); Patm, atmospheric pressure (Pa); r, radius of the capillary(m); s, oil surface tension (N/m); y, contact angle (rad); r, oil density(kg/m3); g, acceleration due to gravity (m/s2); h, oil penetration distance(m); and a, angle between normal and vertical axes (rad).

1

2

2

1

Pv

Pv

Patm

Patm

hh

θ

θ

α

α

FIGURE 5.5 Schematic diagram showing the capillary penetration phenomena when

having different arrangements. Left: upward configuration, the action of gravity restricts

capillary penetration. Right: downward configuration, the action of gravity enhances

capillary penetration; from Bouchon and Pyle (2005b).


Since Ppore depends heavily on vapor pressure, this is expected tooccur after some cooling takes place. As Ppore decreases and vapor con-denses, the pressure difference is expected to force the oil within thestructure. The condensation mechanism should predominate in thicksamples and short frying times, since in thinner samples or longer fryingtimes, moisture loss rate may diminish considerably (and therefore Ppore),allowing oil absorption to begin early.

Oil drainage upon removal of the food from the oil bath certainly playsan important role, since this defines the surface oil layer to be suckedupon cooling. It has been suggested that surface roughness may impor-tantly increase the surface area, enhancing oil absorption (Saguy et al.,1998). In an effort to quantify the irregular conformation of the surface,Pedreschi et al. (2000) and Rubnov and Saguy (1997) have used fractalgeometry, confirming the significant role of crust roughness in oilabsorption.

Wetting properties are certainly important since they affect the oilcapability to drain. Pinthus and Saguy (1994) used a fundamentalapproach based on surface chemistry to describe the relationship betweenthe initial interfacial tension between a restructured potato product andvarious frying media, and the medium uptake during deep-fat frying.They found that total oil uptake was higher for lower initial interfacialtensions, showing a power relationship. This result suggests that a lowerinterfacial tension between the fluid and the solid would increase wettingadhesion and, therefore, increase the total oil content, reflecting theimportance of the wetting phenomena. In addition, the authors found alinear relationship between medium uptake and s.cosy, suggesting theimportance of capillary displacement in the mechanism of mediumuptake.

As can be seen, the surface properties of the product and the physicaland chemical properties of the frying media are extremely relevant to theoil uptake mechanisms. Blumenthal (1991) noticed the importance of oilsurface tension during deep-fat frying, and developed what he called the‘‘surfactant theory of frying.’’ He explained that several classes of surfac-tants are formed during frying of food as a result of the degradation of thefrying oil itself or as a result of the reactions occurring between the foodcomponents and the oil. These compounds act as wetting agents, reduc-ing the interfacial tension between the food and the frying oil, causingincreased contact between the food and the oil and finally producingexcessive oil absorption by the fried product. In fact, Tseng et al. (1996)evaluated the effect of oil degradation on the thermal and physical prop-erties of soybean oil, and they determined how the quality attributes oftortilla chips were affected by oil degradation. They found that surfacetension decreased and viscosity increased significantly with oil degrada-tion, a fact that may well affect oil tendency to drain. Fracturability,

226 Pedro Bouchon

moisture content, and oil uptake of tortilla chips were not significantlyaffected by the oil degradation time. However, after allowing the chip tocool down, only 19% of the total oil was on the surface of the chips fried infresh oil, while 49% remained on the surface of those fried in degraded oil.Also as frying oil degrades, polymer formation increases oil viscosity,affecting its tendency to drain. Extended discussions about oil absorptionmechanisms are reviewed by Mellema (2003) and Zaiifar et al. (2008).

B. Factors affecting oil absorption

There has been much research to examine the different factors affectingoil absorption during frying and many empirical studies have correlatedoil absorption measurements with process and/or product characteris-tics. According to the oil absorption mechanisms explained in the previ-ous section, some factors that may be relevant to the amount of oilabsorbed will be now presented.

1. Moisture contentThe amount of oil uptake has been shown to be directly proportional tothe amount of moisture lost. Several studies claim that higher initialmoisture content results in an increased oil uptake; however, oil absorp-tion seems to be better related to the amount of water loss than to theinitial moisture content (Gamble et al., 1987). As explained in the previoussection, it is well-established that oil absorption will occupy the emptyspace left by water, which in turn determines the maximum availablevolume for oil absorption. The effective water vapor transport throughthe forming crust is, therefore, an important parameter that affects waterescape and probably oil uptake, and as explained by Saguy et al. (1998),diffusion rate is markedly affected by the mechanical properties of theproduct and the crust. Because of the aforementioned relationshipbetween moisture loss and oil absorption, many studies aim at reducinginitial water content in order to decrease the uptake. The effectiveness ofthese pretreatments though, which is usually achieved through drying, isnot due to a reduction of the moisture content on its own (as commonlybelieved), but due to the structural changes occurring at the surface of thefood, which reduce surface permeability (Moreno and Bouchon, 2008).

2. Crust microstructureIt has been found that a decrease in the initial porosity in the food mayreduce oil absorption (Pinthus et al., 1995). However, as explained by theauthors, crust formation plays an additional and fundamental role assoon as frying commences. As the moisture turns to steam and exits theproduct, it leaves behind a sponge-like tunnel network, which constitutesthe oil reservoir. In accordance, the microstructure of the crust region,


which is formed while the food is cooking in the frying oil, has beenpointed out as the single most important product-related determinant forthe final oil uptake into the product (Bouchon et al., 2001). In fact, poredevelopment (Thanatuksorn et al., 2007) and pore size distribution (Saguyet al., 1998) have been found to directly influence oil absorption duringfrying. Some natural ingredients are added to reduce oil uptake becauseof their film-forming capability and/or because they reduce the porosityof the external layers. In formulated products, the permeability of theouter layer of the product depends on the thickness of the sheeted doughsince it determines the structural resistance to vapor escape. A strongerand more elastic network can result in a less permeable outer layerthat may act as an effective barrier against oil absorption (Bouchon andPyle, 2004).

3. Product geometryThe surface area of the food plays an important part in oil uptake.As explained previously, oil absorption is a surface phenomenon involvingequilibrium between adhesion and oil drainage as the product is removedfrom the fryer. Therefore, products with a greater surface-to-volume ratiowill absorb more oil as revealed by the linear relationship found betweenexposed surface area and amount of oil uptake (Gillat, 2001). Also, severalstudies have shown that oil absorption decreases with increasing productthickness (Baumann and Escher, 1995; Gamble and Rice, 1988; Selmanand Hopkins, 1989). Surface roughness is another factor that can result inan increased oil uptake, since it not only impairs oil drainage but alsoincreases overall surface area (Saguy et al., 1998).

4. Frying oil temperature and frying timeThese two process parameters are closely related since products must befried until they reach certain final moisture content, so a lower oil tem-perature implies a longer frying time. A clear influence of oil temperatureon oil absorption has not been found. Gamble et al. (1987) found nocorrelation between oil temperature and oil content when frying potatoslices, but concluded that a lower oil temperature resulted in lower oilcontent in the early stages of frying with a greater difference between 145and 165 �C than between 165 and 185 �C. Similarly, Moreira et al. (1997)determined higher differences in oil absorption between 130 and 160 �Cthan between 160 and 190 �C. In addition, Moreira et al. (1995b) deter-mined that the oil absorption rate was unaffected by the oil temperaturewhen frying tortilla chips and that a frying temperature of 190 �C gave ahigher oil content (3�5%) than a frying temperature of 155 �C.Nonaka et al.(1977) also found that oil content in French fries increased with increasingfrying temperature. Bouchon et al. (2003), in deep-fat frying potato cylin-ders, showed that total oil absorption is a temperature-independent

228 Pedro Bouchon

process for short frying times (1 min at 155, 170, and 185 �C). For longerfrying times, they found that oil content of potato cylinders fried at 155 �Cwas significantly lower than those fried at 170 and 185 �C, but no differ-ence was found between the two higher temperatures. Krokida et al.(2000), when frying potato strips for increasing time (from 0.3 to20 min), also concluded that a lower oil temperature resulted in a loweroil content for long frying times (over 3 min), the difference being higheras frying proceeded. They also found that equilibrium moisture contentvaried as the oil temperature increased from 150 to 190 �C. In accordance,they concluded that the lower oil content could be explained by the lowermoisture loss and not necessarily as an effect of the oil temperature itself.They also determined that oil content increased for increasing fryingtimes, especially for thinner products. The effect of frying time in theamount of oil absorbed may be related to the microstructure developedduring frying, as previously explained. Pinthus et al. (1995) concludedthat crust porosity increased linearly with frying time and, as the cruststructure has been demonstrated to play a significant role in the oiluptake, a thicker porous crust would lead to higher oil content.

It is important to point out that some conclusions about the effect of oiltemperature on oil uptake may be biased by the way results areexpressed. Some researchers have reported that oil absorption results asa percentage of the total weight of the product, that is, a wet basis.Conclusions must be analyzed with care in these situations since whenfrying at a higher temperature for the same frying time, a higher dehy-dration results. When results are expressed on wet basis (w.b.), there is asystematic reduction in the basis as the water content diminishes. Whenoil uptake results are measured as a percentage on a dry-weight basis (d.b.)and the solids remain constant throughout the whole process, it mayprovide a consistent basis for comparison (Moreno and Bouchon, 2008).

5. Oil type and deteriorationThe influence of the oil type and quality on oil absorption and residuesabsorbed by fried foods is widely documented (e.g., Blumenthal, 1991;Blumenthal and Stier, 1991; Krokida et al., 2000; Nonaka et al., 1977;Pokorny, 1980). No relationship has been found between oil type and oilabsorption; however, it has been shown that an increase in the initialinterfacial tension between oil and restructured potato productsdecreases oil absorption (Pinthus and Saguy, 1994). Further, as mentionedearlier, oil degradation produces surfactants that act as wetting agents,which also increase the absorption (Blumenthal, 1991).

Food materials leaching into the oil, breakdown of the oil itself, andoxygen absorption at the oil�air interface contribute to change the puretriglyceride oil into a mixture of hundreds of compounds. These materialsincrease heat transfer and also reduce the surface tension between the


food and the oil (Blumenthal, 1991). These surface-active agents may havea pronounced effect on fat absorption. By improving the wetting capabil-ities of oil and reducing the surface tension, these surfactants may lead toa higher oil uptake (Saguy et al., 1998). Also, oil viscosity increases as aresult of dimer and polymer formation in aging oils (Blumenthal, 1991).A higher viscosity would make oil drainage from the product surfacedifficult, increasing the oil taken up.

C. Oil absorption reduction

Oil reduction in deep-fat-fried products may be obtained through prefry-ing and/or postfrying treatments. Prefrying treatments are mainly basedon the marked effect that the crust microstructure has in oil absorption,and mainly intend to reduce surface permeability. Postfrying treatmentsaim to remove surface oil before postcooling suction begins.

Drying (microwave, hot-air treatment, baking) prior to frying is shownto be effective in oil uptake reduction in several food products (Gambleand Rice, 1987; Krokida et al., 2001a; Moreno and Bouchon, 2008). It isimportant to note that the effectiveness of these pretreatments is not dueto a reduction of the moisture content on its own (as commonly believed),but due to the structural changes occurring at the surface of the food,which reduce surface permeability. Interestingly, osmotic dehydrationhad also been extensively reported as an effective pretreatment in oilabsorption reduction, whose effectiveness greatly depends on the solu-tion employed (Bunger et al., 2003; Krokida et al., 2001b; Moyano andBerna, 2002). However, as revealed in a recent study, the decrease in oilabsorption has been shown to be really due to the increase in solid contentoccurring during the osmotic dehydration process, rather than a reduc-tion in the amount of oil taken up (Moreno and Bouchon, 2008). Actually,the study demonstrated that osmotic predehydrated samples may absorbas much oil as freeze-dried samples.

In addition, during the last decade, much attention has been given tothe use of hydrocolloids with thermal gelling or thickening properties,such as methylcellulose (MC), hydroxypropyl methylcellulose (HPMC),long-fiber cellulose, corn zein, and alginates, among others, to inhibit oiluptake (Albert and Mittal, 2002; Garcıa et al., 2002; Mellema, 2003). Thehydrocolloid mixture can be added to the product in several ways:(1) added directly in the formula, such as in doughnuts and restructuredpotato products, (2) included in the batter or breading, (3) sprayed on theproduct as a solution (Pinthus et al., 1993).

A pioneering work in relation to direct incorporation was carried outby Pinthus et al. (1993), where they determined that the addition of HPMCand powdered cellulose to doughnuts and falafel balls reduced oilabsorption, being more effective with HPMC as an oil barrier. In a more

230 Pedro Bouchon

recent study, Bouchon and Pyle (2004) examined the oil-absorption capac-ity of three different restructured potato chips during deep-fat fryingusing low-leach potato flake and native or pregelatinized potato starch.Interestingly, they found that the product containing native potato starchas an ingredient picked up the lowest amount of oil when sheeted into athick chip, whereas it absorbed the largest amount of oil when sheetedinto a thin chip. They explained that thin chips are vulnerable to highlevels of stretching, as they result in highly permeable and brittle outersurfaces susceptible to oil infiltration. In contrast, thick restructuredpotato chips could withstand higher steam pressures due to their strongersolid structure, which was not ruptured. In addition, a flat and smoothouter surface was obtained, which allowed oil to drain easily from thesurface andwas less vulnerable to oil absorption, as revealed qualitativelyby SEM and quantitatively by reflective confocal microscopy. Recently,Gazmuri and Bouchon (2009) studied the oil absorption capacity of arestructured matrix made with different proportions of native wheatstarch and vital wheat gluten, which were either directly fried or friedafter a predrying step. Results showed that gluten had a predominant rolein the structure, making the dough more elastic and less permeable to oilabsorption. Interestingly, they found that even though predried productswith high gluten content had higher moisture content before frying, theyabsorbed the lowest amount of oil, suggesting that oil uptake is not clearlyrelated to the amount of moisture lost but rather to the productmicrostructure.

Even though some results can be found in relation to restructuredsheeted products, most research has been focused on coatings and batters.The effectiveness of a coating depends on its mechanical and barrierproperties, which in turn depend on its microstructure and composition,along with food substrate characteristics (Garcıa et al., 2002).Mallikarjunan et al. (1997) explained that when edible films made ofcellulose derivatives are used, a protective surface layer is formed dueto thermally induced gelation above 60 �C, which inhibits fat absorption.They determined that an edible coating made of MC was more effectivethan corn zein and HPMC when applied on mashed potato balls.Williams and Mittal (1999) also determined a higher reduction in oiluptake when applying an edible film of MC, compared to HPMC andgellan gum films, on a pastry mix. Similarly, Garcıa et al. (2002) deter-mined that MC was more effective than HPMC when applied on thesurface of potato strips and dough discs. Albert and Mittal (2002) eval-uated the effect of several edible coatings applied to the surface of a pastrymix. They determined that soy protein isolate (SPI), whey protein isolate(WPI), and MC were the most effective coatings.

Postfrying treatments, such as hot air (Nonaka et al., 1977) and super-heated steam drying (Kochhar, 1999) have been shown to be effective in


oil uptake reduction. These processes are aimed to remove surface oilbefore postcooling suction takes place. The equipment, know as ‘‘low-fatbox,’’ is mounted at the discharge end of the fryer and it removes the fatexcess from the recently fried food normally using superheated steam at150–160 �C. The oil�vapor mixture that is obtained is then filtered, andthe oil is pumped back to the fryer. Units range from batch strippers forpilot plant to continuous production units and they can reduce oil contentby up to 25% (Kochhar, 1999). Following the same principle, in commoncontinuous production lines the excess fat is removed by passing theproduct immediately after emerging from the fryer over a vibratingscreen, which allows the fat to drain off (Dobraszczyk et al., 2006). Theeffect of vacuum frying in terms of oil uptake reduction is still not clear,but it appears to be instrumental in reducing oil absorption (Mariscal andBouchon, 2008).

REFERENCES

Aguilera, J. M. and Gloria, H. (1997). Determination of oil in fried potato products bydifferential scanning calorimetry. J. Agric. Food Chem. 45, 781–785.

Albert, S. and Mittal, G. S. (2002). Comparative evaluation of edible coatings to reduce fatuptake in a deep-fat fried cereal product. Food Res. Int. 35, 445–458.

Anand, R. S. and Basiotis, P. P. (1998). Is total fat consumption really decreasing? Fam. Econ.

Nutr. Rev. Retrieved July 2008 from. http://findarticles.com/p/articles/mi_m0EUB/is_3_11/ai_53885191.

Ascherio, A. and Willett, W. (1997). Health effects of trans fatty acids. Am. J. Clin. Nutr. 66,1006S–1010S.

Banks, D. (1996). Food service frying. In ‘‘Deep Frying’’ (E. G. Perkins and M. D. Erickson,eds.), pp. 245–257. AOAC Press, Champaign.

Baumann, B. and Escher, E. (1995). Mass and heat transfer during deep fat frying of potatoslices: I. Rate of drying and oil uptake. Lebensmittel-Wissenschaft Technologie. 28, 395–403.

Besaratinia, A. and Pfeifer, G. (2004). Genotoxicity of acrylamide and glycidamide. J. Natl.Cancer Inst. 96, 1023–1029.

Blumenthal, M. M. (1991). A new look at the chemistry and physics of deep-fat frying. FoodTechnol. 45, 68–71.

Blumenthal, M. M. and Stier, R. F. (1991). Optimization of deep-fat frying operations. Trend.Food Sci. Technol. 2, 144–148.

Bouchon, P. and Aguilera, J. M. (2001). Microstructural analysis of frying of potatoes. Int.J. Food Sci. Technol. 36, 669–676.

Bouchon, P. and Pyle, D. L. (2004). Studying oil absorption in restructured potato chips.J. Food Sci. 69, E115–E122.

Bouchon, P. and Pyle, D. L. (2005a). Modelling oil absorption during post-frying cooling.II Solution of the mathematical model, model testing and simulations. Trans. Inst. Chem.

Eng. C Food Bioproducts Process. 83, 253–260.Bouchon, P. and Pyle, D. L. (2005b). Modelling oil absorption during post-frying cooling.

I Model development. Trans. Inst. Chem. Eng. C Food Bioproducts Process. 83, 261–272.Bouchon, P., Hollins, P., Pearson, M., Pyle, D. L., and Tobin, M. J. (2001). Oil distribution in

fried potatoes monitored by infrared microspectroscopy. J. Food Sci. 66, 918–923.

http://findarticles.com/p/articles/mi_m0EUB/is_3_11/ai_53885191



232 Pedro Bouchon

Bouchon, P., Aguilera, J. M., and Pyle, D. L. (2003). Structure oilabsorption relationshipsduring deep-fat frying. J. Food Sci. 68(9), 2711–2716.

Browner, W. S., Westenhouse, J., and Tice, J. A. (1991). What if American ate less fat?A quantitative estimate of the effect on mortality. J. Am. Med. Assoc. 265, 3285–3291.

Bunger, A., Moyano, P., and Rioseco, V. (2003). NaCl soaking treatment for improving thequality of french-fried potatoes. Food Res. Int. 36, 161–166.

Choe, E. and Min, D. (2007). Chemistry of deep-fat frying oils. J. Food Sci. 72, R77–R86.Costa, R. M., Oliveira, F. A. R., Delaney, O., and Gekas, V. (1999). Analysis of the heat transfer

coefficient during potato frying. J. Food Eng. 39, 293–299.Dobraszczyk, B. J., Ainsworth, P., Ibanoglu, S., and Bouchon, P. (2006). Baking, extrusion and

frying. In ‘‘Food Processing Handbook’’ ( J. G. Brennan, ed.), pp. 237–290. Wiley-VCH,Weinheim.

Dueik, V. and Bouchon, P. (2009). Vacuum frying: An advanced technology able to overcomemost of detrimental changes occurring during atmospheric frying. Criti. Rev. Food Sci.

Nutr. (submitted).Farkas, B. E (1994). Modeling immersion frying as a moving boundary problem. Ph.D.

dissertation, University of California, Davis.Farkas, B. E., Singh, R. P., and McCarthy, M. J. (1992). Measurement of oil/water interface in

foods during frying. In ‘‘Advances in Food Engineering’’ (R. P. Singh andA. Wirakartakusumah, eds.), pp. 237–245. CRC Press Inc., Boca Raton, FL.

Gamble, M. H. and Rice, P. (1987). Effect of pre-fry drying on oil uptake and distribution inpotarto crisp manufacture. Int. J. Food Sci. Technol. 22, 535–548.

Gamble, M. H. and Rice, P. (1988). The effect of slice thickness on potato crisp yield andcomposition. J. Food Eng. 8, 31–46.

Gamble, M. H., Rice, P., and Selman, J. D. (1987). Relationship between oil uptake andmoisture loss during frying of potato slices from c.v. Record U.K. tubers. Int. J. Food Sci.

Technol. 22, 233–241.Garcıa, M. A., Ferrero, C., Bertola, N., Martino, M., and Zaritzky, N. (2002). Edible coatings

from cellulose derivatives to reduce oil uptake in fried products. Innovative Food Sci.

Emerg. Technol. 3, 391–397.Gazmuri, A. M. and Bouchon, P. (2009). Analysis of wheat gluten and starch matrices during

deep-fat frying. Food Chem. doi:10.1016/j.foodchem.2009.01.020.Gillat, P. (2001). Flavour and aroma development in frying and fried food. In ‘‘Frying:

Improving Quality’’ ( J. B. Rossell, ed.), pp. 266–366. CRC Press Inc., Boca Raton, FL.Granda, C., Moreira, R. G., and Tichy, S. E. (2004). Reduction of acrylamide formation in

potato chips by low-temperature vacuum frying. J. Food Sci. 69, 405–411.Hubbard, L. J. and Farkas, B. E. (1999). A method for determining the convective heat

transfer coefficient during immersion frying. J. Food Process Eng. 22, 201–214.Keller, C. H., Escher, F., and Solms, J. A. (1986). Method of localizing fat distribution in deep-

fat fried potato products. Lebensmittel-Wissenschaft Technologie. 19, 346–348.Kita, A., Lisinska, G., and Powolny, M. (2005). The influence of frying medium degradation

on fat uptake and texture of French fries. J. Sci. Food Agric. 85, 1113–1118.Kochhar, S. P. (1998). Security in industrial frying processes. Grasas Aceites. 49, 282–302.Kochhar, S. P. (1999). Safety and reliability during frying operations: Effects of detrimental

components and fryer design features. In ‘‘Frying of Food’’ (D. Boskou and I. Elmadfa,eds.), pp. 253–269. Technomic Publishing Inc., Lancaster.

Koyama, N., Sakamoto, H., Sakuraba, M., Koizumi, T., Takashima, Y., Hayashi, M.,Matsufuji, H., Yamagata, K., Masuda, S., Kinae, N., and Honma, M. (2006). Genotoxicityof acrylamide and glycidamide in human lymphoblastoid TK6 cells. Mutat. Res. 603,151–158.

Krokida, M. K., Oreopoulou, V., and Maroulis, Z. B. (2000). Water loss and oil uptake as afunction of frying time. J. Food Eng. 44, 39–46.


Krokida, M. K., Oreopoulou, V., Maroulis, Z. B., and Marinos-Kouris, D. (2001a). Effect ofpre-treatment on viscoelastic behavior of potato strips. J. Food Eng. 50, 11–17.

Krokida, M. K., Oreopoulou, V., Maroulis, Z. B., and Marinos-Kouris, D. (2001b). Effect ofosmotic dehydration pretreatment on quality of French fries. J. Food Eng. 49, 339–345.

Mahungu, S. M., Artz, W., and Perkins, E. G. (1999). Oxidation products and metabolicprocesses. In ‘‘Frying of Food’’ (D. Boskou and I. Elmadfa, eds.), pp. 25–45. TechnomicPublishing Inc., Lancaster.

Mallikarjunan, P., Chinnan, M. S., Balasubramaniam, V. M., and Phillips, R. D. (1997). Ediblecoatings for deep-fat frying of starchy products. Lebensmittel-Wissenschaft Technologie. 30,709–714.

Mariscal, M. and Bouchon, P. (2008). Comparison between atmospheric and vacuum fryingof apple slices. Food Chem. 107(4), 1561–1569.

Matz, S. A. (1993). ‘‘Snack Food Technology.’’ Van Nostrand Reinhold/AVI, New York.Mellema, M. (2003). Mechanism and reduction of fat uptake in deep-fat fried foods. Trend.

Food Sci. Technol. 14, 364–373.Melton, S. L., Jafar, S., Sykes, D., and Trigiano, M. K. (1994). Review of stability measure-

ments for frying oils and fried food flavor. J. Am. Oils Chem. Soc. 71, 1301–1308.Miller, K. S., Singh, R. P., and Farkas, B. E. (1994). Viscosity and heat transfer coefficients for

canola, corn, palm, and soybean oil. J. Food Process. Preservation. 18, 461–472.Minihane, A. M. and Harland, J. I. (2007). Impact of oil used by the frying industry on

population fat intake. Criti. Rev. Food Sci. Nutr. 47, 287–297.Mintel International Group Ltd. (2006). ‘‘Salty Snacks—US’’. Accessed August 2006. http://

www.marketresearch.com.Mitka, M. (2002). Fear of frying: Is acrylamide in foods a cancer risk? J. Am. Med. Assoc. 288,

2105–2106.Moreira, R. G. (2006). The engineering aspects of deep-fat frying. In ‘‘Handbook of Food

Science, Technology, and Engineering’’ (Y. H. Hui, ed.). Vol. 3, Chap. 111. CRC PressInc., Boca Raton, FL.

Moreira, R. G. and Barrufet, M. A. (1998). A new approach to describe oil absorption in friedfoods: A simulation study. J. Food Eng. 35, 1–22.

Moreira, R. G., Palau, J., Sweat, V., and Sun, X. (1995a). Thermal and physical properties oftortilla chips as a function of frying time. J. Food Process. Preservation 19, 175–189.

Moreira, R. G., Palau, J., and Sun, X. (1995b). Simultaneous heat andmass transfer during thedeep fat frying of tortilla chips. J. Food Process Eng. 18, 307–320.

Moreira, R. G., Sun, X., and Chen, Y. (1997). Factors affecting oil uptake in tortilla chips indeep-fat frying. J. Food Eng. 31, 485–498.

Moreira, R. G., Castell-Perez, M. E., and Barrufet, M. A. (1999). ‘‘Deep-Fat Frying: Funda-mental and Applications.’’ Aspen Publication, Gaithersburg.

Moreno, M. C. and Bouchon, P. (2008). A different perspective to study the effect of freeze,air, and osmotic drying on oil absorption during potato frying. J. Food Sci. 73, E122–E128.

Mottram, D. S., Wedzicha, B. L., and Dodson, A. T. (2002). Food chemistry: Acrylamide isformed in the Maillard reaction. Nature. 419, 448–449.

Moyano, P. and Berna, A. (2002). Modeling water loss during frying of potato strips: Effect ofsolute impregnation. Drying Technol. 20(7), 1303–1318.

Ni, H. and Datta, A. K. (1999). Moisture, oil and energy transport during deep-fat frying offood materials. Trans. Inst. Chem. Eng. C Food Bioproducts Process. 77, 194–204.

Nonaka, M., Sayre, R. N., and Weaver, M. L. (1977). Oil content of French fries as affected byblanch temperatures, fry temperatures and melting point of frying oils. Am. Potato J.

54, 151–159.Pedreschi, F., Aguilera, J. M., and Arbildua, J. J. (1999). CLSM study of oil location in fried

potato slices. Microsc. Anal. 37, 21–22.

http://www.marketresearch.com



234 Pedro Bouchon

Pedreschi, F., Aguilera, J. M., and Brown, C. (2000). Characterization of food surfaces usingscale-sensitive fractal analysis. J. Food Process Eng. 23, 127–143.

Pinthus, E. J. and Saguy, I. S. (1994). Initial interfacial tension and oil uptake by deep-fat friedfoods. J. Food Sci. 59, 804–807, 823.

Pinthus, E. J., Weinberg, P., and Saguy, I. S. (1993). Criterion for oil uptake during deep-fatfrying. J. Food Sci. 58, 204–205, 222.

Pinthus, E. J., Weinberg, P., and Saguy, I. S. (1995). Oil uptake in deep-fat frying as affectedby porosity. J. Food Sci. 60, 767–769.

Pokorny, J. (1980). Effect of substrates on changes of ftas and oils during frying. La RivistaItaliana delle Sostanze Grasse. 57, 222–225.

Rossell, J. B. (1998). Industrial frying process. Grasas y Aceites. 49, 282–295.Rubnov, M. and Saguy, I. S. (1997). Fractal analysis and crust water diffusivity of a restruc-

tured potato product during deep-fat frying. J. Food Sci. 62, 135–137, 154.Saguy, S. and Dana, D. (2003). Integrated approach to deep fat frying: Engineering, nutrition,

health and consumer aspects. J. Food Eng. 56, 143–152.Saguy, I. S., Gremaud, E., Gloria, H., and Turesky, R. J. (1997). Distribution and quantifica-

tion of oil uptake in french fries utilizing a radiolabeled 14C palmitic acid. J. Agric. FoodChem. 45, 4286–4289.

Saguy, I. S., Ufheil, G., and Livings, S. (1998). Oil uptake in deep-fat frying: Review.Oleagineux, Corps Gras, Lipids 5, 30–35.

Sahin, S., Sastry, S. K., and Bayindirli, S. K. (1999). Heat transfer during frying of potato slices.Lebensmittel-Wissenschaft Technologie. 32, 19–24.

Sanibal, E. A. A. and Mancini-Filho, J. (2004). Frying oil and fat quality measured bychemical, physical, and test kit analyses. J. Am. Oils Chem. Soc. 81, 847–852.

Selman, J. D. and Hopkins, M. (1989). Factors affecting oil uptake during the production offried potato products. Technical Memorandum 475. Campdem Food and Drink ResearchAssociation, Gloucestershire, UK.

Seppanen, C. M. and Csallany, A. (2006). The effect of Intermittent and continuous heating ofsoybean oil at frying temperature on the formation of 4-hydroxy-2-trans-nonenal andother a-, b-unsaturated hydroxyaldehydes. J. Am. Oils Chem. Soc. 83, 121–127.

Shyu, S. L. andHwang, L. S. (2001). Effects of processing conditions on the quality of vacuumfried apple chips. Food Res. Int. 34, 133–142.

Shyu, S., Hau, L., and Hwang, L. (1998). Effect of vacuum frying on the oxidative stability ofoils. J. Am. Oils Chem. Soc. 75, 1393–1398.

Singh, R. P. (1995). Heat and mass transfer in foods during deep-fat frying. Food Technol. 49,134–137.

Subramanian, R., Nandini, K. E., Sheila, P. M., Gopalakrishna, A. G., Raghavarao, K. S. M. S.,Nakajima, M., Kimura, T., and Maekawa, T. (2000). Membrane processing of used fryingoils. J. Am. Oils Chem. Soc. 77, 323–328.

Thanatuksorn, P., Kajiwara, K., and Suzuki, T. (2007). Characterization of deep-fat frying in awheat flour-water mixture model using a state diagram. J. Sci. Food Agric. 87, 2648–2656.

Tseng, Y., Moreira, R., and Sun, X. (1996). Total frying-use time effects on soybean-oildeterioration and on tortilla chip quality. Int. J. Food Sci. Technol. 31, 287–294.

Ufheil, G. and Escher, F. (1996). Dynamics of oil uptake during deep-fat frying of potatoslices. Lebensmittel-Wissenschaft Technologie. 29, 640–644.

Varela, G. (1988). Current facts about the frying of food. In ‘‘Frying of Food: Principles,Changes, New Approaches’’ (G. Varela, A. E. Bender, and I. D. Morton, eds.), pp. 9–25.Ellis Horwood, Chichester.

Williams, R. andMittal, G. S. (1999). Low-fat fried foods with edible coatings: Modelling andsimulation. J. Food Sci. 64, 317–322.

Zaiifar, A. M., Achir, N., Courtois, F., Trezzani, I., and Trystram, G. (2008). Review ofmechanisms, conditions, and factors involved in the oil uptake phenomenon during thedeep-fat frying process. Int. J. Food Sci. Technol. 43, 1410–1423.

CHAPTER 6


* Bureau of Chemical SafeOntario, Canada

{ Faculty of Medicine, Univ{ Professional Advisory Bo} Faculty of Medicine, Dalh} Faculty of Medicine, Univ

Introduction of Oats in theDiet of Individuals with CeliacDisease: A Systematic Review

Olga M. Pulido,*,† Zoe Gillespie,* Marion Zarkadas,‡

Sheila Dubois,* Elizabeth Vavasour,*

Mohsin Rashid,‡,} Connie Switzer,‡,} and

Samuel Benrejeb Godefroy*


rition016/S

ty, Fo

ersityard, Cousieersity

Research, Volume 57 # 20091043-4526(09)57006-4 All righ

od Directorate, Health Products and Food Branch, Health Cana

of Ottawa, Ottawa, Ontario, Canadaanadian Celiac Association, Ottawa, Ontario, CanadaUniversity, Halifax, Nova Scotia, Canadaof Alberta, Edmonton, Alberta, Canada

Elsts

da,

II. M
ethods 239
A
. P ivotal in vivo clinical studies on the effect of
oats in patients with celiac disease and dermatitis

herpetiformis (Table 6.1)
250
B
. N onpivotal studies testing the effect of oats in
patients with celiac disease by in vitro methods

or serology (Table 6.2)
250
III. R
esults 250
A
. P ivotal in vivo clinical studies on the effect of
oats in patients with celiac disease and dermatitis

herpetiformis
250
B
. N onpivotal studies testing the effect of oats in
patients with celiac disease using in vitro

methods or serology
254
C
. O ther studies relevant to the effect of oats in
patients with celiac disease
255
evier Inc.reserved.

Ottawa,

235

236 Olga M. Pulido et al.

IV. D
iscussion 255
A
. E vidence based on pivotal and nonpivotal studies 255
B
. E vidence based on other reviews on the safety
of oats
257
C
. B iochemistry and taxonomy of oats relevant to
its potential toxicity
259
D
. B enefits of the consumption of oats 261
V. C
onclusions 261
VI. A
ppendix I 262
A
. S ummary of pivotal in vivo clinical studies testing
the safety of oats in patients with celiac disease

or dermatitis herpetiformis (Table 6.1)
262
B
. S ummary of nonpivotal studies testing the effect
of oats in patients with celiac disease by in vitro

methods or serology (Table 6.2)
274
C
. O ther studies relevant to the effect of oats in
patients with celiac disease
279
Ackno
wledgments 279
Refer
ences 279
Abstract Celiac disease is an immune-mediated disease, triggered in geneti-

cally susceptible individuals by ingested gluten from wheat, rye,

barley, and other closely related cereal grains. The only treatment

for celiac disease is a strict gluten-free diet for life. This paper

presents a systematic review of the scientific literature on the

safety of pure oats for individuals with celiac disease, which histor-

ically has been subject to debate.

Limitations identified within the scientific database include:

limited data on long-term consumption, limited numbers of parti-

cipants in challenge studies, and limited reporting about the rea-

sons for withdrawals from study protocols. Furthermore, some

evidence suggests that a small number of individuals with celiac

disease may be intolerant to pure oats and some evidence from

in vitro studies suggests that an immunological response to oat

avenins can occur in the absence of clinical manifestations of celiac

disease as well as suggesting that oat cultivars vary in toxicity.

Based on the majority of the evidence provided in the scientific

database, and despite the limitations, Health Canada and the Cana-

dian Celiac Association (CCA) concluded that the majority of peo-

ple with celiac disease can tolerate moderate amounts of pure

oats. The incorporation of oats into a gluten-free diet provides

high fiber and vitamin B content, increased palatability, and benefi-

cial effects on cardiovascular health. However, it is recommended

that individuals with celiac disease should have both initial and

long-term assessments by a health professional when introducing

pure oats into a gluten-free diet.

Introduction of Oats in the Diet of Individuals with Celiac Disease 237

I. INTRODUCTION

Celiac disease is an autoimmune disorder in which ingestion of glutencauses damage to the small-intestinal mucosa in genetically susceptibleindividuals (Fasano et al., 2008; Green and Cellier, 2007; Losowsky, 2008;Presutti et al., 2007). Celiac disease is also known as celiac sprue or gluten-sensitive enteropathy. The clinical presentation of celiac disease is highlyvariable. In addition to the intestinal symptoms, celiac disease is asso-ciated with various extraintestinal complications. It is, therefore, consid-ered as a multisystem disorder (Briani et al., 2008). Patients with celiacdisease also have an increased risk of developing other autoimmunedisorders, such as type I diabetes mellitus (Barton and Murray, 2008;Gianfrani et al., 2008). In children, celiac disease can be associated withgrowth failure and delayed puberty (Tully, 2008). Furthermore, the symp-toms and associated conditions of celiac disease can vary greatly innumber and severity resulting in frequent delays in diagnosis, and/ormisdiagnosis. Common examples of misdiagnoses include: irritablebowel syndrome, chronic fatigue syndrome, and fibromyalgia (Catassiand Fasano, 2008; Cranney et al., 2007; Rashid et al., 2005).

Dermatitis herpetiformis1 is a condition of the skin that is also trig-gered by the ingestion of gluten in genetically susceptible individuals andis considered a dermatological form of celiac disease (Abenavoli et al.,2006; Alaedini and Green, 2005; Briani et al., 2008; Losowsky, 2008).

Gluten is a generic name given to storage proteins in wheat, barley,rye, and other closely related cereal grains. It is the gluten in wheat flourthat binds and gives structure to bread, baked goods, and other foods,making it widely used in the production of many processed and packagedfoods. For individuals with celiac disease, these proteins trigger aninflammatory injury in the absorptive surface of the small intestine result-ing in malabsorption of protein, fat, carbohydrate, fat-soluble vitamins,folate, and minerals, especially iron and calcium (Koning, 2008; Tye-Dinand Anderson, 2008; Wieser and Koehler, 2008).

Celiac disease is a lifelong condition. If celiac disease is not diagnosedearly and treated with a strict gluten-free diet, it can be associated withserious complications, including: osteoporosis, lymphoma, and infertilityin both men and women (Bianchi and Bardella, 2008; Fasano et al., 2008;Freeman, 2008; Green and Cellier, 2007; Malandrino et al., 2008;Mangione, 2008; Pastore et al., 2008; Pellicano et al., 2007; Pope andSheiner, 2009).

A small-intestinal biopsy is necessary to confirm the diagnosis of celiacdisease (Dickson et al., 2006; Haines et al., 2008). However, the advent of

1 Throughout this publication, when not otherwise specified, dermatitis herpetiformis is included under thegeneral term celiac disease.


newdiagnostic serological tests, particularly for anti-endomysial and anti-tissue transglutaminase antibodies (Alaedini and Green, 2008; Fasanoet al., 2008; Gianfrani et al., 2008; Hill and Holmes, 2008; Hopper et al.,2008), has now estimated the worldwide prevalence of celiac disease to bebetween 1 in 100 and 200 individuals (Catassi, 2005; Catassi et al., 2007b;Fasano et al., 2003; Harrison et al., 2007; National Institute of Health, 2004).Certain groups of people have markedly elevated risks of developingceliac disease. First-degree relatives of individuals diagnosed with celiacdisease have a 10–20% increased risk of developing celiac disease ( Jolobe,2008). A high prevalence of celiac disease is also found in individuals withDown syndrome and IgA deficiency (Presutti et al., 2007).

Presently, the only treatment of celiac disease is a strict lifelong exclu-sion of wheat, rye, barley, and other related cereal grains2 from the diet(Akobeng and Thomas, 2008; Buchanan et al., 2008; Catassi et al., 2007a;Ciclitira et al., 2005; Collin et al., 2007; Guandalini, 2007; Hopman et al.,2008; Kupper, 2005; Zarkadas and Case, 2005; Zarkadas et al., 2006). Theamount of gluten that can be tolerated varies amongst people with celiacdisease. Some patients tolerate an average of 34–36 mg gluten/day with-out any clinical manifestations of celiac disease, while others who con-sume approximately 10 mg gluten/day developedmucosal abnormalities(Akobeng and Thomas, 2008; Catassi et al., 2007a). Although there is noevidence to suggest a single definitive threshold for a tolerable glutenintake, there is evidence that a daily gluten intake of<10 mg is unlikely tocause significant histological abnormalities (Akobeng and Thomas, 2008;Buchanan et al., 2008; Catassi et al., 2007a; Collin et al., 2007; FDA, 2006;Hopman et al., 2008).

Whether or not individuals with celiac disease can safely consume oatshas been an issue of interest and investigation in recent scientific litera-ture. Using a systematic approach, we conducted a comprehensivereview of the available scientific literature from 1995 to 2008 that wasrelevant to the safety of oats in the diet of individuals with celiac disease,including those with dermatitis herpetiformis. It is agreed in the literaturethat the first large controlled trial on the safety of oats in celiac disease wasconducted in 1995 ( Janatuinen et al., 1995), marking the starting point ofinvestigations in the field. Earlier publications were uncontrolled andtheir limitations were previously reviewed (Garsed and Scott, 2007).Hence, 1995 was selected as starting point.

2 Cereal grains that are known to trigger celiac disease/dermatitis herpetiformis reactions include the fol-lowing: wheat (including durum wheat or ‘‘durum,’’ spelt wheat or ‘‘spelt’’, kamut), barley, rye, triticale, atta,bulgur, einkorn, emmer, and farro. Also of concern are: wheat bran, wheat farina, wheat flour, wheat germ,wheat-based semolina, wheat starch (in some countries ‘‘gluten-free’’ can be made with wheat starch), andgraham flour. Commercial oats that may be contaminated with the foregoing grains are also of concern(Case, 2008; Zarkadas and Case, 2005).


II. METHODS

A computer-assisted search of available English literature databases wasconducted covering January 1995 to November 2008, using Ovid, Med-line, and Pub Med. The search was performed in Food Science andTechnology Abstracts, Medline, Embase, CINAHL, Global Health, andCurrent Contents. The following search terms were used: ‘‘Celiac dis-ease’’ and ‘‘oats,’’ oat challenge or rechallenge, oats and the immunesystem, oats clinical trials, and oats dietary consumption. The followingkeywords were used with adaptations for each database platform: glutentolerance or gluten intolerance or coeliac or celiac, oat or oats or avena oravenin. The reference list of publications generated by the database searchterms were reviewed and compared for further identification of poten-tially relevant papers. Search results were managed using ‘‘ReferenceManagerÒ Soft Ware.’’ Other publications, which were not directly rele-vant to celiac disease and the consumption of oats such as oats atopicsensitization and allergy, were excluded.

Two independent investigators evaluated the publications and theresults were tabulated (Tables 6.1 and 6.2). The safety assessmentincluded all publications identified in the database search as: (A) pivotalin vivo clinical studies conducted in patients, adults, or children, withceliac disease or dermatitis herpetiformis, who were challenged with oatsand underwent a small bowel or skin biopsy, with or without additionaldiagnostic serological tests (anti-endomysial; anti-tissue transglutami-nase) in order to assess the biological response to the introduction ofoats to an otherwise gluten-free diet (Table 6.1); (B) nonpivotal studiesconducted in patients, adults, or children, with celiac disease or dermati-tis herpetiformis using in vitro methods or serology without a biopsy totest the biological response to oats (Table 6.2); (C) other relevant studiesnot fulfilling either of the above selection criteria (not tabulated).

The following documents were also considered in the safety/benefitevaluation: Threshold Working Group on ‘‘Approaches to EstablishThresholds for Major Food Allergens and for Gluten’’ (Food And DrugAdministration, 2006); Gluten-free Labelling of Foods (Food and DrugAdministration, 2007); Health Claims from Certain Foods and Risk ofCoronary Heart Disease (Food And Drug Administration, 2008); publica-tions assessing the benefits of oats for celiac individuals (Peraaho et al.,2004a,b; Sadiq Butt et al., 2008); and for the general population(Andon and Anderson, 2008; Makelainen et al., 2007; Maki et al., 2007a,b;Reyna-Villasmil et al., 2007).

The available information is summarized in Appendix 1 and inTables 6.1 and 6.2, and is organized accordingly.

TABLE 6.1 Pivotal in vivo clinical studies of the effects of oats in patients with celiac disease and dermatitis herpetiformis

Reference

authors/

year

Study

design

Number of

subjects

tested

Study

duration

Amount

of oats

added to

a GFD

Purity

verified Withdrawal

Clinical and

lab tests

Intestinal

biopsy

findings after

challenge Summary comments

Adults

Janatuinen

et al. (1995)

Randomized

controlled

92 CD subjects Not

reported

11 subjects

dropped out

No serology (refer to

Janatuinen et al.,

2000)

Normal villous

architecture at

6 and 12 months

Most subjects tolerated oats in

the amounts of 50–70 g/day

in an otherwise GFD

52 in remission 6 months Mean: 49.9 �14.7 g/day

6 from remission group

(3 control

(1 þ, 2 �),

3 oats (2þ, 1 �))

One out of 21 new Dx

controls did not

enter remission

after

12 months26 oats/26

control

40 new Dx

of CD

12 months Mean: 46.6 �13.3 g/day

5 from new Dx group

(2 control (1þ, 1�),

3 oats (1þ, 2�))

(þ) Reported symptoms of

itching and/or

abdominal bloating

19 oats/21

control

(�) Refused to continue, no

reason provided

Srinivasan

et al. (1996)

Baseline

controlled

10 CD subjects

in remission

12 weeks 50 g/day Yes None No change in gliadin

and EM Abs

No morphological

damage, no change

in IEL counts or

enterocyte height

Oats are not toxic to celiac

patients

Hardman

et al. (1997)

Baseline

controlled

10 DH subjects in

remission

12 weeks Mean: 62.5 �10.8 g/day

Yes None Gliadin, Ret, and EM

Abs

Normal histology after

Oats challenge, no

change in IEL and

enterocyte height

Adults with DH tolerate

moderate amounts of oats

Not detected after oats

challenge

No change in dermal

IgA

Reuanala

et al. (1998)

Controlled 22 DH subjects in

remission

6 months Median: 53.2 g/

day (range

30–66 g/

day)

Yes 2 subjects dropped out:

both from oats group (1

due to persistent mild

rash, 1 no

reason provided)

8 no symptoms with

oats

Unaltered villous

architecture,

IELs, and IH

markers

Absence of oats toxicity in small

bowel and suggest that the

rash observed in DH

subjects is not activated by

eating oats11 oats/11 control 2 oats and 3 control

developed

transient rash EM

Abs negative in all

subjects

Skin biopsy not

different between

groups

Janatuinen

et al. (2000)

(*continuation

of Janatuinen

et al., 1995)

Randomized

controlled

*Same study

population:

Serum from 92

CD subjects

Yes *N/A Gliadin and Ret Abs

not significantly

different between

oats and controls

No increase in IELs

among subjects in

remission with and

without oats

Results strengthen the view

that most subjects tolerated

oats in the amounts of

50–70 g/day in an otherwise

GFD52 in remission 6 months Mean: 49.9 �14.7 g/day

IELs decreased among

new Dx subjects

with and without

oats

26 oats/26 control

40 new Dx of CD 12 months Mean: 46.6 �13.3 g/day19 oats/21 control

Janatuinen

et al. (2002)

(*continuation

of Janatuinen

et al., 1995)

Follow-up;

controlled

*Follow-up with CD

subjects from

previous study

5 years after

original

trial

subjects

reexamined

Median: 34 g/

day (range

10–70 g/

day)

Yes at 6–12

months,

but not

thereafter

*12 subjects no longer

eating oats after 5 years

citing fears about safety

Gliadin, EM, and Ret

Abs not

significantly

different between

oats and controls

Villous architecture

and inflammatory

cell infiltration not

significantly

different between

oats and control

group

Supports the view that long-

term (up to 5 years)

consumption of oats in

moderate amounts are

tolerated by CD subjects

Limitations: higher number

of drop outs; purity of oats

not tested after long term

35 oats/28 control

Lundin

et al. (2003)

Baseline

controlled

19 CD subjects 12 weeks 50 g/day Yes 1 subject dropped out after

2 weeks due to GI

symptoms

Normal gliadin, EM,

and tTG Abs

18 patients with

normal histology

Some concerns remain with

respect to the safety of oats

1 villous atrophy

5 with IFN-g mRNA

Follow-up;

baseline

controlled

Follow-up with

12 subjects who

continued with

oats regularly

1.5 years Less than 50 g/

day

No Normal serology Normal histology

2 with IFN-g

mRNA

Most CD patients tolerate

uncontaminated oats in the

diet

(continued)


Reference

authors/

year

Study

design

Number of

subjects

tested

Study

duration

Amount

of oats

added to

a GFD

Purity

verified Withdrawal

Clinical and

lab tests

Intestinal

biopsy

findings after


St�rsrud et al.

(2003a,b)

Baseline

controlled

20 CD subjects in

remission

2 years Median: 93 g/

day (range

70–100 g/

day)

Yes 5 subjects dropped -out:

2 because of

abdominal distension/

flatulence

(histology and serology

normal)

Normal gliadin and

EM Abs

13 subjects no change

from baseline after

oats

Adult patients in remission can

tolerate 70–100 g/day of

oats with nutritional and

compliance benefits3 partial atrophy at

baseline had

histological

improvement and

inflammation

remained

unchanged or

improved after oats

3 nonmedical 1 slight inflammation

increase

Peraaho et al.

(2004a)

Randomized

controlled

39 CD subjects with

mucosal

recovery

1 year Median: 31.5 g/

day (range

0–70 g/day)

Not

reported

3 subjects dropped out

from oats group due to

GI symptoms (serology

normal, biopsy

incomplete recovery)

EM and tTG Abs not

significantly

different between

oats and controls

Biopsy available from

18 treated and 13

control

Inclusion of oats did not disturb

intestinal integrity but did

cause more GI symptoms

Limitations: it is not clear if

oats used were

uncontaminated

(Others refused)23 oats/16 control

No morphology

differences

between groups

IEL higher in oats

versus control

Kemppainen

et al. (2007)

(*continuation

of Janatuinen

et al., 2002)

Follow-up;

controlled

*Follow-up with

22 CD subjects

from oat group in

previous study

5 years Median: 30 g/

day (range

10–70 g/

day)

Yes at 6–12

months,

but not

thereafter

*10 subjects

no longer eating oats

after 5 years citing fears

about safety

No serology Intestinal biopsy

indicated no

differences in IELs

biomarkers

Long-term use of oats does not

stimulate local

immunologic response in

the small bowel

20 control subjects

with strict GFD

and no history

with oats

Kemppainen

et al. (2008)

Randomized

controlled

trial

32 CD subjects in

remission who

followed GFD þoats

1 year Increased from

20 (range

0–89 g/day)

to 93 g/day

(range 35–

158 g/day)

at 6 months

Yes 2 drop outs; 1

due to reported GI

symptoms after 1 week

and 1 due to unrelated

circumstances

Normal levels of EM

Abs throughout

the study

No marked differences

in the

histopathology of

small-intestinal

biopsies

The lack of reactivity of

patients’ sera against

avenin, suggests that oats is

not harmfully antigenic in

CD, even in an unkilned

form16 kilned oats/16

unkilned oats

Duodenal villous

architecture and

mucosal

inflammation

improved

throughout the

study

Self controlled

groups switched

treatments after 6

months

Children

Hoffenberg et al.

(2000)

Baseline

controlled

10 CD subjects with

New Dx

6 months 24 g/day Yes None Decrease in tTG Abs Villous architecture

and IEL count

improved during

study

Commercially available oats,

tested for purity was

tolerated by newly Dx celiac

children

Hogberg et al.

(2004)

Double-blind

randomized

controlled

116 CD subjects with

new Dx

1 year Median: 15 g/

day (range

5–40 g/day)

Yes 22 subjects dropped out Gliadin, EM, and tTG

Abs not

significantly

different between

oats and controls

All with normal

mucosal

architecture except

2 control subjects

after 1 year

Addition of oats to GFD is

tolerated by newly Dx CD

children15 from oats (5 because of

GI symptoms, 1 due to

poor growth, 7 no

symptoms)

57 oats/59 control

No significant

difference in IELs

count between

groups

7 from control (2 with GI

symptoms, 5 no

symptoms)

Hollen et al.

(2006a)

(*continuation

of Hogberg

et al., 2004)

Controlled *Same study

population 86

subjects from

original study

sampled

1 year *Subjects

consuming

at least

10 g/day

were

selected

from

original

study

population

Yes N/A IgA antiavenin Abs

decline after 3

months, IgG Abs

significantly

decreased but

remained high in

majority of

patients both

groups; nitric

oxide levels high in

4 urine samples

No biopsy Oats was not producing a

humoral immune reaction;

however, findings do not

exclude the possibility that

some CD patients are

susceptible to oatsSerum from 38 CD

subjects on

GFD þ oats

Control: serum from

48 CD subjects on

GFD

(continued)


Reference

authors/

year

Study

design

Number of

subjects

tested

Study

duration

Amount

of oats

added to

a GFD

Purity

verified Withdrawal

Clinical and

lab tests

Intestinal

biopsy

findings after


Hollen et al.

(2006b)

(*continuation

of Hogberg

et al., 2004)

Controlled *Same study-

population 87

subjects who

completed

original study

sampled

1 year Median: 15 g/

day (range

5–40 g/day)

Yes N/A No significant

differences

observed between

groups

No biopsy Children with CD consuming

oats have similar reduction

in urinary nitrite/nitrate

excretion as GFD children;

yet for some, nitrite/nitrate

levels remain high, long-

term follow-up needed

Nitrite/nitrate values

of 9 GFD þ oats

and 8 GFD

remained high

after 1 year

Urine from 39 CD

subjects on GFD

þ oats

Control: Urine from

48 CD subjects on

GFD

Holm et al. (2006) Baseline

controlled

32 CD subjects 2 years Median for new

Dx subjects

Yes 3 remission subjects

dropped out: 2 from

oats challenge due to GI

symptoms (serology

negative, villous

morphology normal,

and IELs decreased)

and 1 from gluten

challenge due to

laborious study

protocol

(asymptomatic)

EM, gliadin, and tTG

Abs were negative

except during

gluten challenge

Relapse after gluten

but not oats

challenge

Pure oats can be safely added to

the GFD of most children

with CD7 years9 new Dx on GFD þ

oats

Recovery of all groups

on GFD þ oats as

per histology and

IELs biomarkers

43 g/day

(range 19–

64 g/day)

Consumption of oats does not

result in mucosal

deterioration or immune

activation

23 in remission

Median for

subjects in

remission

13/23 challenged

with oats and

10/23 challenged

with gluten, after

relapse with

gluten placed on

GFD þ oats

45 g/day

(range 13–

81 g/day)

22 long-term clinical

follow-up

Abs, Antibodies; CD, celiac disease; DH, dermatitis herpetiformis; Dx, Diagnosis; EM, endomysial antibodies; GI, gastrointestinal; GFD, gluten-free diet; IELs, intraepithelial lymphocytes; IFN, interferon; Lab, laboratory;tTG, tissue transglutaminase antibodies; Ret, reticulin; IH, immunohistochemistry.

TABLE 6.2 Nonpivotal studies testing the effects of oats in patients with celiac disease by in vitro methods or serology

Reference authors/year

Number of subjects

tested Method of assessment Results Summary comments

In vitro—duodenal mucosal cultures

Srinivasan et al. (1999)

(*continuation of

Srinivasan et al.,

1996)

9 CD and 11 non-CD

controls

Lactase expression of

in vitro duodenal

mucosal cultures

Lactase was expressed in

11 control samples

from subjects with

normal histology

Results corroborate

the lack of oats

toxicity in adult CD

subjects

*Additionally samestudy population: 10

CD subjects in

remission who

supplemented GFD

with 50 g oats/day

for 3 months

Lactase was notexpressed in samples

from 9 subjects with

active CD

Lactase was expressed in

samples from 10 CD

subjects in remission

who were challenged

with oats

Picarelli et al. (2001) 13 treated CD subjects Immune responses of

in vitro duodenal

mucosal culture to

gliadin and avenin

EM Abs produced in

response to gliadin but

not avenin by celiac

mucosa. No EM Absfrom controls

Oats appear to have no

harmful effect on

CD

(continued)



Number of subjects


Kilmartin et al. (2003) 17 CD and 16 non-CD

controls

Immune responses of

in vitro duodenal

mucosal culture to

gliadin and avenin

Gliadin caused increase

in IFN-g mRNA in all

CD subject samples

Immunogenic

sequences in gliadin

are not present in

avenin. These

results support that

oats are safe forconsumption by CD

subjects

Increased IFN-g protein

in 4 CD subject samples

Smaller increases in IL-2cytokine mRNA

detected in 6 CD

subject samples and

increased IL-2 cytokine

protein in 2 CD subject

samples

No significant response

with aveninNo response to gliadin or

avenin among controls

Arentz-Hansen et al.

(2004)

(*continuation of

Lundin et al., 2003)

9 CD subjects who had

history of oats

exposure (*5/9 from

same study

population)

Derivation of

polyclonal T cell

lines

Avenin-reactive T cell

lines recognized avenin

peptides in the context

of HLA-DQ2

Some CD patients

have avenin-reactive

mucosal T cells that

can cause mucosal

inflammationIn vitro duodenal

mucosal cultures

were challengedwith either pepsin or

A substantial proportion

of the avenin-reactive

T cell appears to bespecific to avenin

trypsin digest or a

chymotrypsin digest

of avenin

T cell lines challenged

with avenin responded

more strongly to

avenin than to gluten in

4 of 5 samples

Srinivasan et al. (2006)

(*continuation of

Srinivasan et al.,

1996)

*Same study

population: 10 CD

subjects who

supplemented GFDwith 50 g oats/day

for 3 months

Immune responses of

in vitro duodenal

mucosal culture

treated with Absagainst: HLA

D-related, Ki-67,

CD25, CD54,

ICAM-1, and mast

cell tryptase

None of the patients

developed clinical or

lab evidence of adverse

effects

No evidence of

immune activation

from oats

supplementation

Distribution of intestinal

HLA-DR expression

was not affected

Number of CD25 and

tryptase positive cells

was not altered

Distribution and intensityof ICAM-1 staining

unchanged

Kilmartin et al. (2006) 7 CD subjects Immune response ofgliadin-reactive

mucosal T cell lines

to wheat, barley, rye,

and oat-related

cereals

All 5 T cell linesdemonstrated

immunoreactivity to

protein fractions from 4

related cereals

Confirms similar T cellantigenic reactivity

of 4 related cereals

Some cell lines reactivity

to wheat, barley, and

rye was evident only

when cereal fractionswere pretreated with

tTG

(continued)



Number of subjects


Measuring

proliferation or

cytokine production

Despite oats stimulation

of T cell lines, it did not

activate a mucosal

lesion in most subjects

In Vitro—other

Silano et al. (2007) 10 children with

verified CD

Peripheral lymphocyte

samples exposed to

avenins from 4

different oatvarieties: Lampton,

Astra, Ava, and

Nave

All varieties of oats were

immunogenic, with

Lampton and Ava

avenins inducinglymphocyte activation

similar to gliadin

More evidence is

needed to show the

safety of oats and

varieties of lowtoxicity

Lymphocyte

proliferation and

IFN-g release in the

culture medium

were measured asindexes on immune

activation

Astra and Nave avenins

showed less

immunogenicity

Silano et al. (2007) N/A 3 varieties of oats were

tested by 2 assaysbased on the known

ability of PT digests

of celiac-active

Avenins from Italian

variety Astra and theAustralia variety

Mortlook were more

active than the

Results indicate that

some varieties ofoats may be

potentially harmful

proteins to

agglutinate K562

cells and to disrupt

lysosomes

Australian variety

Lampton. Gliadin,

digested in the same

way, had more activitythan all 3 avenins

to individuals with

CD

Serology

Guttormsen et al.

(2008)136 CD subjects Serum samples were

tested for levels of

IgA to wheat

gliadin, oats avenin,

and tTG

Gliadin, avenin, and tTGdid not differ among

CD subjects

Findings support thenotion that most

adult CD patients

can tolerate oats

82 CD subjects oats

(24 g/day (mean);

6 months minimum)

and 54 CD subjects

GFD without oats

Ingestion of oats does

not cause increased

levels of IgA against

oats in CD patients

on GFDControl: 139 non-CD

subjects

Abs, Antibodies; CD, celiac disease; EM, endomysial antibodies; GFD, gluten-free diet; HLA-DQ2, human leukocyte antigen; ICAM, intercellular adhesion molecule; IFN, interferon; IL, interleukin; Lab,laboratory; tTG, tissue transglutaminase antibodies.


A. Pivotal in vivo clinical studies on the effect of oats inpatients with celiac disease and dermatitis herpetiformis(Table 6.1)

Studies were grouped by subject (adults vs. children) and organizedchronologically by publication date. The following parameters were tabu-lated from each publication: reference (authors and year); study design(randomized controlled trials, observational changes from baseline in asingle group, cohort study); number of subjects tested (newly diagnosedpatients vs. those in remission on a gluten-free diet prior the oats chal-lenge); study duration; amount of oats added to the gluten-free diet;purity of test material verified (if the oats used in the challenge weretested for gluten contamination); withdrawals; clinical and laboratorytests, biopsy, and histopathology findings after the oats challenge; andsummary comments.

B. Nonpivotal studies testing the effect of oats in patients withceliac disease by in vitro methods or serology (Table 6.2)

Studies were grouped according to the in vitro method used (duodenalmucosal cultures, other, serology) and organized chronologically bypublication date. The following parameters were tabulated from eachpublication: reference (authors and year); number of subject tested (sub-jects with celiac disease vs. controls); method of assessment; results; andsummary comments.

III. RESULTS

A. Pivotal in vivo clinical studies on the effect of oatsin patients with celiac disease and dermatitis herpetiformis

Table 6.1 summarizes the clinical dietary challenge studies in both adultsand children that assess the effect of oats on patients with celiac diseaseand dermatitis herpetiformis. These studies are considered pivotal basedon the use of biopsy results (intestinal/skin) as the key endpoint follow-ing the oats challenge. Within this parameter, 11 oats-challenge studies inadults and 3 in children were identified. Some of the pivotal challengestudies had clinical follow-up studies reported in separate publications.These follow-up publications are identified (*) in Table 6.1 given that thesame pool of patients from the original challenge studies were evaluatedusing other laboratory methods to assess their biological response to oats.Further details of each study are provided in Appendix 1.


Janatuinen et al. (1995) were the first to evaluate the possible toxicity ofoats in a large controlled study. Since then, a number of studies (Table 6.1)have assessed the safety of oats consumption by individuals diagnosedwith celiac disease and dermatitis herpetiformis. Most studies were con-ducted on adults, with a smaller number of studies performed on children.

Of the 14 studies identified as pivotal in adults and children, five wererandomized controlled trials, five were observational changes from base-line in a single group, and three were cohort studies. Furthermore, onestudy conducted with children utilized three study designs (randomizedcontrolled, observational changes from baseline in a single group, and acohort study follow-up) with different groups in the study population(refer to Table 6.1 or Appendix 1 for further details). The duration of allthe studies within the database, ranged from 12 weeks to 5 years in adultsand 6 months to 7 years in children (Table 6.1). In adults, there was onestudy with a 5-year follow-up period ( Janatuinen et al., 2002) and anadditional follow-up analysis of the local immunological response of thesmall-intestinal mucosa (Kemppainen et al., 2007). In children, one studyhad a 7-year follow-up period (Holm et al., 2006).

The amount of oats included in the gluten-free diets ranged from 30 to93 g for adults and from 15 to 45 g for children (Table 6.1). In a smallnumber of adult subjects, the median amount of oats tested over a 2-yearperiod was 93 g (up to 100 g/day) (St�rsrud et al., 2003a,b) and in onestudy (Holm et al., 2006) children ingested amedian of 45 g (up to 81 g/day)oats daily for 2 years.

In this review, emphasis has been placed on the purity of the oatstested because the contamination of commercial oats by wheat, barley,and rye cereals has been reported as an intermittent problem that canpotentially skew trial results (Gelinas et al., 2008; Hernando et al., 2008;Thompson, 2004). Cross contact of oats with these cereals is considered akey reason for apparent oat intolerance reported in the past and continuesto be an issue of concern for commercially available oats (Dickey, 2008;Ellis and Ciclitira, 2008; Garsed and Scott, 2007). Hence, informationabout the purity of oats included in the gluten-free diets of study partici-pants was identified in all the publications included in the database(Table 6.1). All of the pivotal studies reported the purity of oats testedwith only two exceptions (Janatuinen et al., 1995; Peraaho et al., 2004b).However, many of the studies failed to indicate the basis upon which theoat samples were deemed to be gluten-free. It was assumed that unlessotherwise specified, the oat samples that were reported as free of contam-ination with wheat, barley, and rye had gluten levels that were below thedetection limit of the applied testingmethod(s). However, some studies didnot specify the cut-off values used to determine whether the oat sampleswere considered gluten-free. St�rsrud et al. (2003a,b) reported a 20 parts per


million (ppm) limit of detection and Holm et al. (2006) reported that 29 outof 30 oat samples tested contained gliadin levels below 28 ppm.

1. Safety of oats in adultsAmong the studies conductedwith adults (Table 6.1), therewas a combinedtotal of 170 patients with either celiac disease or dermatitis herpetiformis,either in remission or newly diagnosed, who were challenged with oatswhile consuming an otherwise gluten-free diet. Of the 170 adults who hadan intestinal biopsy after they were challenged with oats, only one (1/170)case in one study (Lundin et al., 2003) showed histological evidence ofintestinal mucosal injury associated with exposure to oats. Althoughsome individuals in these studies reported some adverse gastrointestinalsymptoms, the symptoms were considered mild and transient and therewas no evidence to associate oats with the development of histopathologi-cal lesions of the small-intestinal mucosa, the definitive diagnostic criterionof celiac disease.

The majority of adults included in these studies were following agluten-free diet and reported to be in remission as evidenced by thenormal small-intestinal mucosa histology at the initiation of the trial.One study ( Janatuinen et al., 1995) included 40 newly diagnosed patientsof which 21 were used as control and 19 were challenged with oats.Of these newly diagnosed patients, one control subject of the 21 did notenter remission after 1 year of consuming a gluten-free diet. The inclusionof oats in the gluten-free diet of 19 newly diagnosed patients did notprevent symptomatic and mucosal healing ( Janatuinen et al., 1995).

A total of 22 patients with celiac disease (22/170) had evidence ofabnormal bowel histopathology (Kemppainen et al., 2008; St�rsrud et al.,2003a,b) prior to a dietary challenge with oats (baseline). These patientswere on a gluten-free diet for at least 6 months prior to entering the trial.Patients with a normal bowel biopsy at baseline remained normal after theinclusion of oats into their gluten-free diet and patients with an abnormalbowel biopsy at (n ¼ 22) baseline showed either no evidence of furtherhistopathological deterioration after the oats challenge or histopathologicalimprovement. Furthermore, the results of a 5-year follow-up study withadult patients with celiac disease who included oats in their diet indicatedthat exposure to oats did not alter the small-intestinalmucosamorphologyor stimulate an immunological response locally in the mucosa of the smallintestine ( Janatuinen et al., 2002; Kemppainen et al., 2007).

2. Safety of oats in childrenAmong the studies conducted with children (Table 6.1), there were89 children with celiac disease, either in remission (n ¼ 13) or newlydiagnosed (n¼ 76), who were challenged with oats added to an otherwisegluten-free diet. All of the children who were in remission at the time of


entering the trial responded well to the test diet. Follow-up results forthese children were comparable to those of the corresponding controlsand there was no histological evidence of deterioration in the intestinalmucosa after the introduction of oats into their gluten-free diet.

Similarly, newly diagnosed children recovered well on a gluten-freediet that included oats (Hoffenberg et al., 2000; Hogberg et al., 2004; Hollenet al., 2006b; Holm et al., 2006). A combined total of 76 newly diagnosedpatients were challenged with oats in these studies. Children were placedon the test diet (gluten-free diet including oats) after diagnosis and wereassessed for disease remission at the end of the study. A randomizedcontrolled study (Hogberg et al., 2004) conducted with children who werenewly diagnosed with celiac disease showed that the consumption of anoat-containing gluten-free diet for 1 year did not interfere with clinical,serological, or small-intestinal mucosal recovery. In another study,22 children with celiac disease, 9 newly diagnosed and 13 in remissionconsumed a gluten-free diet plus oats for 7 years without clinical orserological evidence of relapse (Holm et al., 2006).

3. Withdrawal of adults and children from pivotal clinical studiesWithin the adult database (Table 6.1), 41 out of 170 individuals who werechallenged with oats withdrew from the trials. However, only 11 of these41 individuals reported adverse side effects associated with the consump-tion of oats. Therefore, of the 170 adults who were challenged with oats,only 6% reported adverse effects (11/170). Of the 89 children who werechallengedwith oats, 17 individuals withdrew from the trials and only 9%reported adverse side effects from the consumption of oats (8/89). In bothadults and children, the side effects reported were mostly gastrointestinalin nature. Information regarding the reasons for the withdrawal of theother cases was limited.

4. Gastrointestinal symptoms in adults and childrenAs reported in the clinical trials (Appendix 1 and Table 6.1), some patientswith celiac disease, both adults and children, experienced gastrointestinalsymptomsmore often when consuming an oat-containing diet than with astandard gluten-free diet. However, in general, such symptoms werereported as transient, mild, and were explained as the effect of anincreased intake of fiber from oat products rather than the reoccurrenceof clinical manifestations of celiac disease. Except for the withdrawalsnoted above, most patients continued to participate in the trials. Thereports of mild gastrointestinal symptoms, without clinical or serologicalevidence of relapse, were not considered sufficiently adverse to warrantthe exclusion of oats from the gluten-free diet of patients with celiacdisease.


B. Nonpivotal studies testing the effect of oats in patients withceliac disease using in vitro methods or serology

In addition to the pivotal studies, several publications (Table 6.2) usedother methods to test the response of individuals with celiac disease whowere introduced to oats. These studies did not fulfill the selection criteriaof pivotal studies namely an in vivo oats challenge with an intestinal/skinbiopsy to assess the biological response to the introduction of oats into anotherwise gluten-free diet. Instead, they used various in vitro techniquesto assess the immune response to avenin, or serology without an intestinalmucosal biopsy. Most of the methods used duodenal mucosal culturesprepared from biopsies or intestinal T cell lines obtained from individualswith celiac disease. Other studies measured the immunogenic reaction inperipheral lymphocytes or measured the presence of various antibodiesin individuals with verified celiac disease who included oats in their diet,in comparison with a reference group (Table 6.2). Some of these studiesused patients that were previously included in pivotal studies. Thesestudies are identified with an asterisk (*) in Table 6.2.

1. In vitro—duodenal mucosal culturesDespite the diversity of the methodology used, the overall results fromthese in vitro nonpivotal studies are consistent with the opinion that oatsare well tolerated by the majority of patients with celiac disease. How-ever, there is some evidence that avenin from oats can elicit an in vitroimmunogenic response. One study that reported an in vitro immunogenicresponse to avenin (Arentz-Hansen et al., 2004) established oat avenin-specific and reactive intestinal T cell lines from three patients who hadclinical symptoms while consuming an oat-containing diet, as well asfrom two other patients who appeared to tolerate oats. Some of thesepatients were recruited from the same pool of patients participating in adesignated pivotal study. Of these participants, one patient was known tobe intolerant to oats with associated intestinal pathology (Lundin et al.,2003). In contrast, other in vitro studies (Kilmartin et al., 2003, 2006) reportthat purified avenin from oats is not immunogenic to the intestinalmucosa of patients with celiac disease.

2. In vitro—otherRecent studies using a peripheral lymphocyte method to test the immu-nogenic properties of oats (avenin) reported that avenins from differentoat cultivars have different levels of toxicity when tested in vitro (Silanoet al., 2007a,b). All the varieties of oats tested (Lampton, Astra, Ava, andNave) by these investigators were immunogenic with differences in theircapacity to induce a response. All avenin from oat cultivars had lessactivity than gliadin (wheat), but more activity than rice prolamins.


3. SerologyGuttormsen et al. (2008) assessed several serological parameters(IgA against oat avenin and wheat gliadin, and tissue transglutaminase) insamples collected from 136 adult patients, who had a biopsy-confirmeddiagnosis of celiac disease and eating a strict gluten-free diet for at least2 years. Resultswere compared to 139 nonceliac controls. Although patientswere not challenged with oats, 82 patients with celiac disease had beentaking oats as part of their gluten-free diet for 6 months or more andrecorded their oats intake during a 3-week period. Patients were instructedto consume only ecologically grown oats specifically produced for celiacdisease patients. Other than this, therewas no report of testing results of thepurity of oats. No significant differences were found in IgA against oats inoats-eating and non-oats-eating patients with celiac disease.

C. Other studies relevant to the effect of oats in patientswith celiac disease

One study (Peraaho et al., 2004b) conducted a retrospective evaluation,beginning in 1997, on the inclusion of oats within a gluten-free diet bypatients with celiac disease and dermatitis herpetiformis. The inclusion ofoats, the effect of oats on symptoms of illness, and the quality of life wereinvestigated in 1000 randomly selected members of the Finland CeliacSociety. Altogether, 710 patients responded to the questionnaire: 423(73%) with celiac disease and 70 (55%) with dermatitis herpetiformiswere currently consuming oats. Subjects reported an appreciation of thetaste, the ease of use, and the low cost of oats. Of the respondents, 94%believed that oats diversified the gluten-free diet. This study providessupport to the beneficial effects of oats for individuals with celiac disease.However, 15% of patients with celiac disease and 28% of patients withdermatitis herpetiformis reported that they had stopped eating oats. Themost common reasons reported for avoiding oats were the fear of anadverse effect or of contamination of the oats with cereals excludedfrom the gluten-free diet, for example, wheat, barley, and rye.

IV. DISCUSSION

A. Evidence based on pivotal and nonpivotal studies

An overall assessment of the studies published since 1995 (Tables 6.1 and6.2) supports the opinion that the majority of adults and children who haveceliac disease or dermatitis herpetiformis, regardless of whether the condi-tion is newly diagnosed or in remission, can tolerate moderate amounts(20–25 g/day dry rolled oats for children; 50–70 g/day for adults) of pure


oats, uncontaminated with gluten from wheat, rye, and barley. The follow-ing issues were taken into consideration when determining the safety ofintroducing oats into the gluten-free diet of individuals with celiac disease:

Comparisons of the various studies are complicated by the differentstudy designs, the different conditions used in the testing, and in thereporting of the purity of the oat products used in the clinical trials. Allthe studies reviewed had some limitations; however, regardless of theselimitations most studies indicated that subjects could tolerate the inclu-sion of oats in a gluten-free diet, providing it was free of contaminationwith gluten-containing cereals.

The key limitations of the study designs are the duration of the studyand the number of subjects. Since most studies conducted are short term,the potential effect of a lifetime exposure to oats requires further investi-gation. Long-term compliance to a gluten-free diet is one of the majordifficulties in the management of celiac disease, and the collection of long-term data related to the consumption of oats is also very challenging.

The number of subjects included in each challenge study was limited,most likely due to the difficulty of recruiting enough subjects willing tocomply with the long and laborious study protocols including endoscopicbiopsies. More specifically, in studies involving children, particularlythose including very young children, an additional challenge was thepossible reluctance of the children and/or their parents to participate,which would interfere with compliance and the desire to continue untilthe end of the study. Training the families of the children to maintaincompliance to the study protocol is an additional essential requirement toensure accurate results. Furthermore, patients with celiac disease willingto participate in challenge studies are aware that they may face reactiva-tion of the clinical symptoms or a delay in their recovery. For ethicalreasons, most studies that recruited newly diagnosed patients hadexclusion criteria to eliminate patients with the most severe pathologies.It is important to recognize the limitations that are encountered in celiacdisease challenge clinical trials, as the information gleaned from thesetrials must be interpreted and evaluated within such limitations.

The differences in the oat products used, the testing, and the reportingof the purity of oats further limited a comprehensive safety assessment.Evaluation of the purity of the oats used in each study is paramount tobeing able to distinguish between possible toxicity of oats versus adverseeffects induced by cross contamination. A standardization of the reportingof the purity of oats utilized within the studies would increase the accuracyof safety assessments and assist in the establishment of a threshold for thetolerance to gluten from wheat, barley, and rye. Cross validation, standar-dization, and international agreement on the test methodologies to assessthe purity of oats will also need reevaluation as further informationbecomes available.


In the current database, there is no evidence that the inclusion of oats inthe gluten-free diet induces adverse effects in newly diagnosed patients.Using the baseline biopsy and clinical assessment for comparison, clinicalstudies in both adults and children that examined newly diagnosed indivi-duals, showed that patients either improved or did not deteriorate after theinclusion of oats to their gluten-free diet. It was noted within the databasethat 22 out of 170 adult patients had abnormal bowel mucosal histologyconsistent with active disease at baseline. This observation is of particularconcernbecause thesepatientswere identifiedasbeing in remissionandonagluten-freedietprior to their inclusion into theoats-challenge trial.Althoughno explanation was given for these findings, the following considerationswere taken into account: (a) compliance to gluten-free diet (Martin, 2008),(b) cross contamination of gluten-free diet and threshold of gluten tolerance(Akobeng and Thomas, 2008; Gelinas et al., 2008; Troncone et al., 2008b),(c) lack of response to gluten-free diet or refractory celiac disease (Abdallahet al., 2007), and (d) histopathology criteria for diagnosis (Upton, 2008).

From currently available data, one adult patient (1/170) challengedwith oats was reported to have a severe adverse reaction to oats. Approxi-mately 6%of adults and 9%of childrenwithdrew from clinical trials due toreported adverse effects from the inclusion of oats in their diet. Thisevidence, along with the indication from some in vitro studies of an immu-nological response to avenin in the absence of clinical manifestations ofceliac disease and the limitations of the database (previously discussed),supports a cautionary approach for the introduction of oats into a gluten-free diet until the prevalence of oats intolerance among people with celiacdisease is well established (Dickey, 2008; Garsed and Scott, 2007).

Despite the noted challenges of conducting a comprehensive evalua-tion of the most recent clinical trials, the majority of the evidence indi-cated that most people with celiac disease could tolerate the inclusion of amoderate amount (20–25 g/day; 65 ml or 1/4 cup) dry rolled oats forchildren and (50–70 g/day; 125–175 ml or 1/2–3/4 cup dry rolled oats)for adults of pure oats (uncontaminated with other gluten-containingcereal grains) in a gluten-free diet (Table 6.1, Appendix 1). Based onsome evidence that a possible sensitivity to pure oats exists, most inves-tigators in the field recommend a clinical follow-up when introducingpure oats to the gluten-free diet (Haboubi et al., 2006; Rashid et al., 2007).This includes both initial and long-term assessments.

B. Evidence based on other reviews on the safety of oats

Earlier published reviews on the safety of oats for patients with celiacdisease were included in this evaluation (Dor and Shanahan, 2002; Kumarand Farthing, 1995; Schmitz, 1997; Thompson, 1997, 2003). Two


systematic reviews were identified and compared to our review (Garsedand Scott, 2007; Haboubi et al., 2006).

Haboubi et al. (2006) conducted a systematic review of the literaturerelated to the inclusion of oats in the gluten-free diet for patients withceliac disease in order to assess whether oats can be recommended tothese patients. In this report a computerized search of the scientificliterature up to 2005 was carried out using specific search and selectioncriteria. Haboubi et al. (2006) identified 17 primary studies, six of whichmet their inclusion criteria. We compared our database to that of Haboubiet al. (2006) and identified the following differences: (a) the time periodevaluated, (b) selection criteria, and (c) data analysis. A key differencebetween databases was that although Haboubi et al. used rigorous selec-tion criteria, whether or not the purity of oats used in the studies weretested was not taken into consideration in their evaluation. By contrast,our analysis of the literature included all the studies identified by Haboubiet al. (2006) as well as relevant studies published up to November 2008.Attention was specifically given to the purity of the oats used in the studiesbecause this parameter is considered paramount to the ability to distinguishbetween possible toxicity of oats versus adverse effects induced by crosscontamination. Another noted difference between databases was thatHaboubi et al. (2006) excluded all studies where the same patient servedas control. It is, however, well accepted that in patients with celiac diseasethere is high variability in the susceptibility to gluten injury. Studies usingthe same subjects as controls provide valuable data. Hence, we includedthem in our assessment, but identified them as ‘‘observational changesfrom baseline in a single group’’ in the Table 6.1 under the study designcolumn.

None of these six studies selected by Haboubi et al. (2006) found anysignificant difference in the serology between the groups consuming oatsand the control groups. However, they observed that two of the six studiesreported significant differences (p < 0.001; p ¼ 0.039) in intraepitheliallymphocyte counts (IELs) between the oat-consuming and control groups(Peraaho et al., 2004a; Reunala et al., 1998). Although we concur with thefindings, it is important to note that the purity of oatswasnot tested inone ofthe selected publications (Peraaho et al., 2004a). Hence, it is possible that thedifferencebetweenoats-challengedandcontrol individualswasdue to crosscontamination. In the other study, Reunala et al. (1998) reported differencesin the expression of gd T cell receptors (p < 0.001) when comparing oat-consuming tocontrol groups, but thesebiomarkerswereelevatedatbaselineand improved after the oats challenge. Furthermore, other biomarkersassessed byReunala et al. did not show a difference between oat-consumingand control groups. Moreover, an additional study by Kemppainen et al.(2007) did not find evidence of local immune activation after an oats chal-lenge using similar immunohistochemical biomarkers.


The systematic review conducted by Garsed and Scott (2007) includeduncontrolled studies published prior to 1995, their cut-off point wasFebruary 2006, and the data are presented all in one table. Our reviewestablished a starting point of 1995 based on the publication byJanatuinen et al. (1995) which was the first clinical trial to evaluate thepossible toxicity of oats in a large controlled study. We also segregatedthe studies conducted in adults versus children and grouped them aspivotal and nonpivotal. The rationale for grouping the studies as pivotaland nonpivotal was that the evidence provided by the pivotal studies (i.e.,studies that used biopsy results (intestinal/skin) as the key end pointfollowing the oats challenge) contributed more strongly to the overallweight of evidence. We compared our database with all studies publishedafter 1995 that were identified byGarsed and Scott (2007) and found thatwehad selected the same studies.

Garsed and Scott (2007) identified a total of 165 patients in the data-base as being challenged with oats and having a biopsy as an end point, ofthese patients only one individual (1/165) had histological evidences ofintestinal mucosal injury associated to oats exposure. In our review, wereported a total of 170 adults with celiac disease challenged with oats andhaving a biopsy as an end point. Our tabulation includes challengestudies published after the Garsed and Scott (2007) cut-off point of Febru-ary 2006 and separated children from the tabulation. Despite these differ-ences, we identified the same individual as the only subject (1/170) tohave a histological response to the inclusion of oats in a gluten-free diet.The results and conclusions of the Garsed and Scott (2007) systematicreview are comparable to our review.

C. Biochemistry and taxonomy of oats relevant to itspotential toxicity

The biochemistry and taxonomy of cereals is relevant to their potentialdifferential toxicity. The proteins in the cereal grains known as ‘‘gluten’’activate celiac disease. Gluten is a complex mixture of hundreds of relatedbut distinct proteins. The grains considered capable of producing adverseeffects in individuals with celiac disease include different species ofwheat (e.g., durum, spelt, kamut), barley, rye, and their cross-bredhybrids (e.g., triticale, which is a cross between wheat and rye)(Ciclitira et al., 2005; Cornell et al., 2002; Dewar et al., 2006; Howdle,2006; Koning, 2008; Lester, 2008; Moron et al., 2008; Thompson, 2000,2001; Troncone et al., 2008a; Vader et al., 2003; Wieser and Koehler, 2008).

Gluten includes two major protein types, the gliadins and glutenins,both of which contain activating toxic peptides. Gliadins can be subdi-vided into a/b-gliadins, g-gliadins, and o-gliadins, whereas the gluteninsconsist of low molecular weight and high molecular weight glutenins.


Several gluten proteins are involved in the pathogenesis of celiac disease(Ciclitira et al., 2005) and both prolamins (i.e., gliadins and glutenins) havebeen implicated (Londei et al., 2005; Piper et al., 2002; Shan et al., 2005;Troncone et al., 2008a; Vader et al., 2003; Wieser and Koehler, 2008).Prolamins are the alcohol-soluble protein fractions in wheat and otherrelated cereals and are of the most concern to individuals with celiacdisease. Prolamins are typically rich in glutamine and proline making itresistant to the normal digestive breakdown (Shan et al., 2002; Wieser andKoehler, 2008). After the digestion of prolamins by gastric, duodenal, andpancreatic enzymes, a 33-amino acid peptide molecule (33mer) and otherimmunogenic peptides remain. It is this nondigested fragment that isconsidered the key in eliciting the immune response in susceptible indi-viduals with celiac disease (Shan et al., 2002).

Wheat, rye, and barley have a common ancestral origin in the grassfamily. Oats are more distantly related to the analogous proteins in wheat,rye, and barley and the oat prolamins (avenin) have substantially lowerproline content. Avenin accounts for 5–15% of the total protein in oats,whereas in wheat, barley, and rye, prolamins constitute 40–50% of thetotal protein (Kilmartin et al., 2006). Some investigators believe that thereare similarities between the protein structure of oats and some wheat-likesequences, which may indicate that large amounts of oats could poten-tially be toxic to patients with celiac disease. However, the putative toxicamino acid sequences are less frequent in avenin than in other prolamins,which explains the less toxic nature of oats (Arentz-Hansen et al., 2004;Ellis and Ciclitira, 2001, 2008; Shan et al., 2005; Vader et al., 2002, 2003).

Recent studies have shown that avenins from different varieties of oatsshow different levels of toxicity when tested in vitro (Silano et al., 2007a,b).All the varieties of oats tested (Lampton, Astra, Ava, and Nave) by theseinvestigators were immunogenic, but some varieties were less immuno-genic than others. All avenins from various oat cultivars had less activitythan gliadin, but more activity than rice prolamins (Silano et al., 2007a,b).The authors considered the possibility that contamination could have beena factor and indicated that the less toxic variety (Lampton oats) wasobtained from an isolated farm where only this crop and buckwheat(nontoxic to individuals with celiac disease) were grown (Silano et al.,2007a,b). Overall, the significance of these results needs further clarifica-tion, as the immunogenic ability of avenins does not necessarily mean thatoats are toxic for patients with celiac disease when added to the gluten-freediet in moderate amounts. In the in vitro experimental models, the encoun-ter between prolamins and lymphocytes is abnormal and the testing of onlythe alcohol-soluble protein fraction of oats does not reflect the in vivosituation (Troncone et al., 2008a,b). Factors within oats that may exert aprotective effect against avenin toxicity or interfere with avenin absorptionhave not been accounted for in these experimental models. More studies


are needed to assess the impact of possible differential toxicities of oatcultivars in the overall safety of oats for individuals with celiac disease.

D. Benefits of the consumption of oats

Avena santiva L. is the traditional oat of choice around the world, and theagricultural properties of hulless or naked oat seeds have recently beenreviewed in detail (Burrows, 2005). The interest in oats for human con-sumption has increased in recent years because of their more widelyrecognized nutritional and health benefits. The nutrient composition ofoats and its potential health benefits have been the subject of extensiveinvestigation and recent reviews (Andon and Anderson, 2008; Food AndDrug Administration, 2008; Maki et al., 2007a,b; Mantovani et al., 2008;Ryan et al., 2007; Sadiq Butt et al., 2008). Oat and its by-products are a goodsource of dietary fiber, especially b-glucan, and nutrients, including pro-teins, minerals, and B complex vitamins. A gluten-free diet can be low infiber, and oats will help to provide the much needed fiber and nutrients(Malandrino et al., 2008; St�rsrud et al., 2003a). Other beneficial effects ofoats include attenuation of postprandial plasma glucose and insulinresponses, increased transport of bile acids toward lower parts of theintestinal tract, increased excretion of bile acids, and decreased serumcholesterol levels. The incorporation of oats into a gluten-free diet wouldnot only improve the nutritional value of the diet but for many alsoimprove compliance as it increases palatability, and provides a greatervariety of food choices in a restrictive diet ( Janatuinen et al., 2002; Peraahoet al., 2004a,b) resulting in better health and overall quality of life.

V. CONCLUSIONS

Available scientific data evaluating the introduction of pure oats in thegluten-free diet of patients with celiac disease and dermatitis herpetifor-mis indicates that moderate amounts of pure oats are well tolerated by themajority of these individuals who are either in remission or newlydiagnosed. The term ‘‘pure oats’’ is used to indicate oats uncontaminatedwith gluten from other cereal grains, like wheat, barley, and rye, asdetected by current test methods. Based on pivotal clinical trials in thepublished literature, the amount of pure oats considered within safelimits is 50–70 g/day for adults and 20–25 g/day for children.

The benefit of the introduction of oats into a gluten-free diet for adultsand children with celiac disease outweighs the possible risk to the fewindividuals with celiac disease who may exhibit intolerance for oats.However, the previously discussed limitations of the database requirethe need for some caution when introducing oats into a gluten-free diet.


Although the majority of evidence supports the inclusion of oats in agluten-free diet, the evidence that some people with celiac disease cannottolerate even pure oats requires further investigation and cautious prog-ress. Individuals with celiac disease or dermatitis herpetiformis interestedin introducing oats to their diet are advised to consult their physician anddietician. It is also advisable to add oats into a gluten-free diet only whensuch a diet is well established, so that any possible adverse reactions can bereadily identified (Rashid et al., 2007). Enhanced education and informationfor the consumer with celiac disease, industry, health care providers, reg-ulators, and public at large are essential for an increased awareness of celiacdisease and dietary management of this increasingly recognized sector ofthe population (Lohi et al., 2007). Availability of pure uncontaminated oatsminimizes the risks and maximizes the food choices for individuals withceliac disease and dermatitis herpetiformis.

VI. APPENDIX I

A. Summary of pivotal in vivo clinical studies testing the safetyof oats in patients with celiac disease or dermatitisherpetiformis (Table 6.1)

1. Adults

1. Janatuinen et al. (1995) were the first to evaluate the possible toxicity ofoats in a large controlled study in adult patients with celiac disease.In this randomized trial, which lasted up to 12 months, they comparedthe effects of gluten-free diets with and without oats. They studied twogroups of patients: those with previously diagnosed celiac disease whowere in remission and those newly diagnosed. The previously diag-nosed group was selected based on records of the recovery of thesmall-intestinal mucosa while on a gluten-free diet for at least12 months. For the newly diagnosed, endoscopy with a duodenalbiopsy was performed and the diagnosis of celiac disease was basedon the presence of the subtotal or total villous atrophy of the duodenalmucosa prior to the introduction of a gluten-free diet. Fifty-two adultswith celiac disease in remission were followed for 6 months and40 adults with newly diagnosed celiac disease for 12 months. Patientswere randomly assigned according to sex to either the oats or thecontrol group. The oats group included 26 patients in remission and19 newly diagnosed. The control group included 26 patients in remis-sion and 21 newly diagnosed. The controls received a gluten-free dietcontaining 0.74 mg of gluten/g of foodstuff. The oat groups supple-mented the same basic gluten-free diet with 50–70 g/day of oats in theform of wheat starch flour mixed with an equal amount of oats, muesli-


containing 60% oats, and rolled-oat breakfast cereal. The authorsneglected to report whether the oats added to the gluten-free dietswere tested for purity. The mean (� SD) intake of oats in the oatgroup was 49.9� 14.7 g/day for 6 months among patients in remissionand 46.6� 13.3 g/day for 12 months among newly diagnosed patients.The oat and control groups did not differ significantly in nutritionalstatus, symptoms, or laboratory measures. The authors reported thatpatients in remission, regardless of diet, did not have a worsening ofthe duodenal villous architecture or increased mononuclear cellularinfiltration. All the patients with a new diagnosis were in remissionwithin 1 year, except for one subject in the control group. The rate ofwithdrawal was comparable: six patients in the oat group and five inthe control group withdrew from the study. The investigators con-cluded that moderate amounts of oats could be included in a gluten-free diet for most adult patients with celiac disease without adverseeffects. In this study, the basic gluten-free diet consumed by partici-pants contained 0.74 mg of gluten/g of foodstuff, and although theoats were not tested for purity, there were no significant differencesbetween the control and the test group in remission as per the duode-nal biopsies and clinical findings.

2. Srinivasan et al. have several publications relevant to the safety of oatsin patients with celiac disease (Srinivasan et al., 1996, 1999, 2006).In 1996, they studied the safety of oats in 10 adult patients with celiacdisease in clinical and histological remission. Each patient consumed50 g of oats (as porridge) daily for 12 weeks while maintaining a strictgluten-free diet. The oat cereal used in the study was tested for evi-dence of gluten contamination using reverse-phase high-performanceliquid chromatography (HPLC), enzyme linked immunosorbent assay(ELISA), and polymerase chain reaction (PCR) techniques. The oatswere determined to be entirely gluten-free. Details were not providedas to the levels of detection for these techniques or the cut-off valuesused to determine whether the oat samples were considered gluten-free. The patients were assessed clinically at 0, 1, 4, and 12 weeks.At each assessment, the following laboratory investigations were per-formed: full hematological and biochemical profiles and serologicaltests for antibodies to gliadin and endomysium. Duodenal biopsieswere obtained before the start of the oats challenge and after the12-week trial period. All patients complied fully with the study proto-col. Throughout the oat challenge, all patients remained asymptomaticwith normal hematological and biochemical indices. Endomysial andgliadin antibody values were unaltered by the oats supplementationand no morphological damage was evident using a standard histolog-ical evaluation. Quantitative histological examinations showed no


significant changes. Subsequently, two patients were given a gluten‘‘microchallenge’’ consisting of 500 mg of gluten daily for 6 weeks:both developed histological evidence of relapse and one patient testedpositive by antibody production tests. These investigators furtherassessed the toxicity of oats in patients with celiac disease by immuno-histochemical techniques detecting the presence of lactase enzyme inthe intestinal biopsy. This enzyme is lost in active celiac disease, butwas unaffected by the oats challenge (Srinivasan et al., 1999).

3. Hardman et al. (1997) studied seven men and three women (mean age:58 years) with biopsy-confirmed dermatitis herpetiformis. The subjectshad followed a strict gluten-free diet for a mean of 15.8 years and hadcontrolled the rash and enteropathy. The subjects added oats to theirdiets for 12 weeks (mean (� SD) daily intake, 62.5 � 10.8 g). The purityof the oats was tested by ELISA and PCR reaction. Details were notprovided as to the levels of detection of these techniques, or the cut-offvalues used to determine whether the oat samples were consideredgluten-free. All patients underwent duodenal and skin biopsies at thebeginning and end of the study. None of the patients reported anyadverse effects. Serologic tests for anti-gliadin, anti-reticulin, and anti-endomysial antibodies were negative both before and after the trialperiod. Villous architecture remained normal after the 12-week period:the mean (� SE) ratio of the height of villi to the depth of crypts was3.59 � 0.11 before the diet and 3.71 � 0.09 afterward (normal, 3–5), andthe mean enterocyte heights were 31.36 � 0.58 and 31.75 � 44 mm,respectively (normal range, 29–34). Duodenal IEL counts all remainedwithin normal limits (mean, 13.8 � 1.03 per 100 enterocytes before thediet and 14.2� 1.2 per 100 enterocytes afterward; normal range, 10–30).Dermal IgA in skin biopsies showed no significant changes.

4. Reunala et al. (1998) conducted an oat challenge study with 11 patientswho had dermatitis herpetiformis. Another 11 patients with dermatitisherpetiformis were used as a control group. At diagnosis all patientshad skin and duodenal biopsies. All patients were in remission ongluten-free diets for at least 5.5 years and free of a rash for 14 months.Test subjects consumed 50 g oats/day for 6 months within an other-wise gluten-free diet. The oats used in this challenge were free ofgluten contamination as tested by ELISA and PCR. In this study, thespecifications for the amount of gluten allowed to be consumed fromwheat starch flours, was up to 0.3 g protein, equivalent to 50 mg ofgluten per 100 g flour which was in accordance with Codex Alimentar-ius 118-1981. Clinical symptoms, serum, skin, and small bowel biopsieswere assessed before and after the oat challenge. Eight patients


challengedwith oats remained asymptomatic, two developed transientrashes, and one withdrew because of the appearance of a more persis-tent but mild rash. Three of the 11 controls also developed transientrashes. IgA endomysial antibodies remained negative in all patientsand the small-intestinal villous architecture remained unaltered afterthe oat challenge. The densities of intraepithelial CD3 and ab and gdT cell receptor positive lymphocytes and crypt epithelial cell DRexpression were assessed by immunohistochemistry. These biomar-kers are considered to be sensitive indicators of immune response togluten. Except for g T cell receptor, these biomarkers were not alteredby oats challenge. The expression of gd T cell receptor showed asignificant difference (p < 0.001) when comparing oat challenged tocontrols. However, it should be noted that this biomarker was elevatedat baseline and improved after the oats challenge.

5. Janatuinen et al. (2000) conducted a randomized controlled interven-tion study over a 6–12-month period with 40 adults who were newlydiagnosed with celiac disease and 52 adults who were in remission.Patients were randomized by sex into either the oat-consuming or thecontrol groups. The control groups received gluten-free cereal. The oat-consuming groups received products supplemented with oats: twotypes of gluten-free wheat starch flour including 50% oats, muesliincluding 60% oats, and rolled-oat breakfast cereal. Some of the oatproducts used in this study were commercially available. The dailyintake of oats was 50–70 g. The purity of the oats was regularly moni-tored. Gluten was analyzed by a quantitative enzyme immunoassayusing a specific monoclonal antibody to a-gliadin. This antibodydetects all prolamins in wheat and rye, only some of the prolamins inbarley, and none of the prolamins in oats. All oat samples were consid-ered gluten-free but no specification was provided with regard to thelimit of detection. All patients were evaluated using serum levels ofgliadin and reticulin antibodies. In the intestinal biopsies, the numberof IELs in the intestinal mucosa was examined before and after theintervention. The rate of disappearance of gliadin and reticulin anti-bodies did not differ between the diet groups of the patients withnewly diagnosed celiac disease. Oats also had no effect on gliadin orreticulin antibody levels in the patients who were in remission. Thenumber of IELs decreased similarly regardless of the diet of newlydiagnosed patients and no increase in the number of IELs was found inthe patients who were in remission and consuming diets with orwithout oats. In summary, there were no significant differencesbetween the clinical symptoms, laboratory measures, and histologyof duodenal biopsies among the test groups who received oats or


those who did not receive oats. The authors concluded that this studyfurther strengthens the view that adults with celiac disease can toleratemoderate amounts of oats.

6. Janatuinen et al. (2002) assessed the safety of the long-term ingestion ofoats in the diet of patients with celiac disease. In a previous study, theeffects of a gluten-free diet and a gluten-free diet including oats werecompared in a randomized trial involving 92 adult patients with celiacdisease (45 in the oats group, 47 in the control group). Oat productswere obtained from commercial sources. The mean amount of oatsadded to the gluten-free diet was 34 g/day. The purity of the oatsand the gluten-free products were monitored during the 6–12-monthintervention (Janatuinen et al., 2000). After the initial phase of6–12 months, patients in the oats group were encouraged to eat oatsfreely in conjunction with an otherwise gluten-free diet. However,there was no systematic monitoring of the purity of these oat supple-ments. A follow-up was conducted after a 5-year period; there were 23patients still on an oats diet. Of the original oat-consuming group,12 subjects dropped out for reasons including uncertainty about safety,flatulence, and rash. In the control group, 28 patients on a conventionalgluten-free diet were examined. In addition to the clinical and nutritionalassessment, the following parameters were evaluated: duodenal biop-sies histopathology and histomorphometry, and measurement of anti-endomysial, antireticulin, and antigliadin antibodies. There were nosignificant differences between controls and those patients consumingoats with respect to duodenal villous architecture, inflammatory cellinfiltration of the duodenal mucosa, or antibody titers after the 5-yearfollow-up. The authors reported that in both groups, histological andhistomorphometric indexes improved equally over time. Despite thehigh withdrawal rate and the fact that neither the oats nor the gluten-free products were monitored for gluten, making it impossible to assessthe level of gluten contamination in the diets, 23 out of 35 patientsremained in the study for 5 years and were without signs of diseaserelapse as evaluated by clinical symptoms, serology, or histopathology.

7. Lundin et al. (2003) conducted an oat challenge study in 19 adultpatients with celiac disease who were in remission on gluten-freediets. The gluten-free diets were supplemented with 50 g/day of oatsfor 12 weeks. Oat samples were obtained from a single manufacturerwho upheld strict practices to avoid contamination with wheat andother gluten-containing grains. Twenty-five oat samples from thismanufacturer were analyzed in-house and a reference laboratory ana-lyzed an additional 120 samples. No contamination was detected inany of these samples using ELISA and mass spectrometry (MS) with alimit of detection for gluten contamination reported as 20 ppm. Six oat


samples were later analyzed by western blot, an ELISA test using acocktail of antibodies, and MALDI-TOF MS with a limit of detectionreported as 5 ppm. The level of gluten detected in five of these sampleswas estimated to be between <1.5 and 23 ppm and considered nega-tive for gluten contamination. The sixth sample was considered con-taminated (>400 ppm by western blot, ELISA, and MS). MS wasunable to determine the source of the contamination due to the largeamount of oat avenins in the sample and the fact that barley and oatsare not distinguishable by MS. An additional sample was analyzedfrom the bottom of the same bag of contaminated oats and<1.5 ppm ofgluten was detected. The authors assumed that a single or a few seedsof barley might have been present in the bag but determined that theoats used in the study were sufficiently pure and adhered to thelimitations by the revised suggested Codex standard limit of 20 ppmfor natural gluten-free products. All patients were evaluated byserological testing including IgA antigliadin, antiendomysium, andanti-tissue transglutaminase antibodies. Endoscopic small-intestinalbiopsies were obtained before and after the oat challenge. Biopsieswere scored using Marsh score and levels of mRNA specific for inter-feron (IFN) were determined by reverse transcription-PCR analysis.IFN was tested as a marker for T cell activation. Oats were welltolerated by most patients but several reported transient abdominaldiscomfort and bloating at the beginning of the challenge. This couldhave resulted from increased fiber consumption in patients not used tolarge intakes of fiber in their diets. One patient withdrew due togastrointestinal symptoms and another patient developed partialvillous atrophy and a rash during the first oats challenge. This patientsubsequently improved on an oat-free diet but developed subtotalvillous atrophy and dramatic dermatitis during a second challengewith oats. There was no evidence of contamination in the oats con-sumed by the patient exhibiting intolerance. Five of the subjects hadpositive levels of IFN-mRNA but no corresponding histologicalabnormalities after the challenge. The significance of this finding isnot fully understood but the authors’ suspect that T cell activationmight not be directly responsible for the villous atrophy seen in celiacdisease. The major difference observed in this study is that one of thepatients was intolerant to oats. The authors reported that after thecompletion of this challenge study they became aware of other patientsidentified as clinically intolerant to oats (dermatitis, abdominal pain,and general anaphylactoid-like reactions) but not further confirmed bysmall-intestinal biopsy. However, none of the subjects were willing toingest oats again. Therefore, the authors concluded that despite theability of most patients to tolerate the incorporation of oats into theirdiets, some individuals might be intolerant to oats.


8. St�rsrud et al. (2003a,b) studied the effects of adding uncontaminatedrolled oats to the daily diets of 20 adult patients with celiac disease whowere in remission. Although there are two reports, they represent onemain study. The core results are based on the 15 patients who com-pleted the study. These subjects added oats to their gluten-free diet for2 years. The median intake of oats was 93 g/day. The oats were freefrom wheat, rye, and barley as tested by ELISA that detects gliadin.This method detects high molecular weight proteins in wheat, rye, andbarley, but not oats (avenin). The method is quantitative with a detec-tion limit of 20 ppm. The examinations of the subjects were performedfour times during the study period and included endoscopic small-intestinal biopsies, blood samples (nutritional status, serological anal-ysis), height and body weight, gastrointestinal symptoms, and dietaryrecords. Histopathology of duodenal biopsies was assessed at baseline(n¼ 20), 6 months (n¼ 17) and 2 years (n¼ 14). One patient refused thebiopsy. Villous architecture and inflammatory infiltrate were assessed.Evidence of partial atrophy and inflammation of the small intestinewas present in some patients at baseline and remained after oatsexposure (n ¼ 3). None of the patients with normal histology deterio-rated after exposure to oats (n ¼ 11). None of the parameters evaluatedindicated any evidence of reactivation of the disease in response to theintroduction of pure oats into the gluten-free diet after the 2-year trialperiod. Five patients withdrew from the study, two due to gastrointes-tinal symptoms, and the other three for nonmedical reasons. Examina-tions of the patients after their withdrawal did not reveal anydeterioration in small-intestinal histology, nutritional status, or raisedlevels of antibodies. The other report that was based on the subjectswho completed this 2-year trial study focused on the benefits and thenutritional status of the participants. The mean intakes of iron anddietary fiber increased (p< 0.001) with the consumption of oats, as wellas the intakes of thiamine and zinc (p < 0.02). Temporary increasedflatulence was also experienced during the first few weeks of consum-ing oats, as well as improved bowel function. All participants whocompleted the study period reported a desire to continue to eat oatsafter the study because they found that the addition of oats to thegluten-free diet gave more variety, better taste, and satiety. The con-sumption of oats improved the nutritional value of the gluten-free diet,did not have negative effects on nutritional status and was appreciatedby the subjects. The authors suggested that including oats could helppeople improve their compliance to a strict gluten-free diet.

9. Peraaho et al. (2004a) studied 39 patients with celiac disease whoconsumed gluten-free diets. Patients were randomized to a gluten-free diet with 50 g oats/day (23 patients) or without oats (16 patients)


for 1 year. The purity of the oats used in this study was not specified inthe report. The following parameters were evaluated: quality of life,gastrointestinal symptoms, small-intestinal histopathology, and serumendomysial and tissue transglutaminase antibodies. The quality of lifedid not differ between the groups, but there were more gastrointestinalsymptoms in the oats-consuming group. Patients consuming oats suf-fered significantly more often from diarrhea, but there was also asimultaneous trend toward more severe constipation symptoms.Three patients on the oats diet dropped out of the study due to gastro-intestinal symptoms. The villous structure did not differ between thegroups, but the density of IELs was slightly significantly higher in theoat group. The severity of the gastrointestinal symptoms did notappear to be dependent on the degree of inflammation. Antibodylevels did not increase during the study period. The authors concludedthat the oat-containing gluten-free diet caused more gastrointestinalsymptoms than the traditional diet. Although the mucosal integritywas not disturbed, there was more inflammation evident in the oatgroup. However, the sources and purity of the oats used in this studyare unknown so it is possible that the gastrointestinal symptoms andthe increased inflammation among the oat-consuming group were dueto gluten contamination from other cereal grains such as wheat, barley,and rye. Despite the limitations of this study and of the effectsobserved in the oat group, the authors suggested that oats can providean alternative within a gluten-free diet but that patients should beaware of the possible effects on the gastrointestinal system. Further-more, these investigators (Peraaho et al., 2004b) also conducted ananalysis on the effect of oats on symptoms and quality of life in 1000randomly selected members of their Celiac Society. Altogether, 710patients responded: 423 (73%) with celiac disease and 70 (55%) withdermatitis herpetiformis were currently consuming oats. Patientsappreciated the taste, the ease of use, and the low costs; 94% believedthat oats diversified the gluten-free diet; 15% of patients with celiacdisease; and 28% of patients with dermatitis herpetiformis hadstopped eating oats. The most common reasons for avoiding oatswere fear of adverse effects or contamination. The authors suggestthat there is a market demand for oats, and celiac societies and dieti-cians should make efforts to promote the development of products freeof gluten contamination.

10. Kemppainen et al. (2007) presented additional data from their 5-yearfollow-up study on the safety of oats in patients with celiac disease(Janatuinen et al., 2002). In the present study, they assessed the localcellular immunological responses using immunohistochemical bio-markers. Forty-two patients with celiac disease took part in an earlier


oats intervention study for 6–12 months. Twenty-two of these patientsoriginally consumed oats as part of their gluten-free diet. During the5-year follow-up, 10 patients had felt uncertain about the safety oflong-term consumption of oats and gave up this part of their diet.Finally, 12 of the 22 patients consumed oats for the whole 5-yearperiod. The control group consisted of the remaining 20 patientswith celiac disease using a strict, conventional, gluten-free diet with-out oats. Intraepithelial CD3, gd T cell receptors (gd IEL), and ab T cellreceptors (ab IEL) T cells were counted after specific staining of small-intestinal biopsy specimens. There were no differences in the densitiesof CD3, ab IELs, and gd IELs between the oat and the control groups.This study provides additional evidence that long-term use of oatsincluded in the gluten-free diets of patients with celiac disease doesnot stimulate an immunological response locally in the mucosa of thesmall intestine. The high number of dropouts demonstrates the diffi-culties of conducting long-term studies in patients with celiac diseaseas the main reason for giving up was feeling of uncertainty about thelong-term safety of oats. Further studies in this regard will help toincrease patients’ compliance and confidence about the safety of oats.

11. Kemppainen et al. (2008) conducted a randomized clinical trial toinvestigate the suitability of large amounts of unkilned oats in com-parison with kilned oats in adult patients with celiac disease. Kilninggenerally processes oats, which in principle may change the antigenicproperties and may be the reason that kilned oats, are tolerated bypatients with celiac disease. The study included 32 patients (19 F,13 M) who previously consumed oats as part of their gluten-freediet. Patients were randomized to two groups: group A was startedon kilned oats and group B on unkilned oats. After 6 months, thepatients changed the treatment groups. Patients had followed agluten-free diet for 8.3 years and had used oats for approximately5 years. The purity of the oats was controlled during the 12-monthfollow-up. Electrophoresis, immunoblotting of the prolamins, and thetotal protein extracts of the oat product samples by polyclonal anti-gliadin antibody did not show any contaminants. At the end of thestudy no marked differences were reported in the histopathology ofsmall-intestinal biopsies between the two groups using eitherunkilned or kilned oats. At baseline, 10 patients had partial villousatrophy and 9 had mild mucosal inflammation. Biopsies from theremaining patients were interpreted as normal. In both groups, duo-denal villous architecture and mucosal inflammation did not worsen,but rather improved during the first 6 months exposure to oats.At 12 months, five patients had partial villous atrophy and four hadmild inflammation. All patients had negative endomysial antibodies,


at the beginning and during the follow-up. Only 1 woman out of 32subjects withdrew from the study because of abdominal symptomsafter 1 week. After 6 months, another female patient withdrewbecause of pregnancy. The authors concluded that the inclusion oflarge amounts of either kilned or unkilned oats in the diet did notcause a worsening of the healed duodenal mucosa, an increase innormalized antibody levels or symptoms, or affect the general well-being of the celiac patients in remission indicating that the kilning ofoats by food processing is not a prerequisite for oats to be tolerated.

2. Children

1. Hoffenberg et al. (2000) conducted a self-controlled, open-label,6-month trial of the consumption of a commercial oat breakfast cerealamong children newly diagnosed with celiac disease. The childrenwere placed on a gluten-free diet plus commercially available oats.Over 6.6 � 0.7 months, they consumed 24 g oat cereal/day, or 1.2 �0.9 g/kg/day. The gliadin contamination of the oat cereal used in thisstudy was tested by an ELISA, which detects gliadin. This method waschosen because it detects high molecular weight proteins in wheat, rye,and barley, but not oats avenin. The samples were considered negativeat a level of <0.01% (100 ppm). The 10 children who completed thestudy were 6.8 � 4.0 (mean � SD) years of age. The children who areincluded in the study had at least one sign or symptom suggestive ofceliac disease at the beginning of the study. Patients were evaluatedclinically, including small bowel histology and anti-tissue transgluta-minase IgA antibody titer. Compared with the start of the study, therewas a significant decrease in biopsy score (p < 0.01), IELs (p < 0.005),anti-tissue transglutaminase IgA antibody titer (p < 0.01), and numberof symptoms (p < 0.01) at the completion of the study. This studywas limited because the children were newly diagnosed and thefamilies were still learning how to manage a gluten-free diet so hiddensources of gluten may still have been present in the diet. However, theoverall results showed improvement with the test diet.

2. Hogberg et al. (2004) reported the results from a double blindmulticen-ter study, which included 116 children with newly diagnosed celiacdisease. Children were randomized into two groups, 59 received agluten-free diet and 57 received a gluten-free diet plus oats. The studyperiod was 1 year. The oats used in the study were specially grown,milled, andpackaged to avoid contaminationwithwheat, rye, or barley.The oat products were tested by an ELISA assay to ensure absence ofgluten contamination. The daily oat intakewas 25–50 g. Small-intestinalbiopsies were performed at the beginning and end of the study. Serum


IgA antigliadin, antiendomysium, and anti-tissue transglutaminaseantibodies were monitored at 0, 3, 6, and 12 months. Ninety-threepatients completed the study, 42 on gluten-free diet plus oats, and50 gluten-free diet controls. The remaining subjects withdrew from thestudy, 15 in the oats group and 7 in the standard gluten-free diet and amajority of these children were from the youngest age groups in thestudy. However, all patients were in clinical remission after the studyperiod and there were no significant differences between control andtest groups for all of the above-noted parameters. It should be noted thatavenin antibodies were also tested in the same group of patients andwere reported separately by Hollen et al. (2006a,b).

3. Hollen et al. (2006a) reported the results from the serology tests in thedouble blind study described above (Hogberg et al., 2004). The focus ofthe study was to evaluate the antibodies to oat prolamins (avenins). Serawere obtained from the study participants. IgA and IgG antiaveninantibodies were monitored at 0, 3, 6, and 12 months. Nitric oxide meta-bolites were measured in seven patients with deviating antibody results.There was a significant decrease in antiavenin antibodies in both groups(standard gluten-free diet and gluten-free diet plus oats) at the end of thestudy as compared to the beginning of the study (p < 0.001) but nodifference was found between the two groups, suggesting that oats wasnot producing a humoral immune reaction. IgA titers declined after 3months. IgG titers, although significantly decreased, remained high inthe majority of patients in both groups. Nitric oxide levels were high infour of the analyzed samples. The authors indicated that this study doesnot exclude the possibility that some patients with celiac disease aresusceptible to oats based on the sero-positive results and evidence ofhigh levels of nitric oxide metabolites in some subjects.

4. Hollen et al. (2006b) reported the results from the urinalysis test in thedouble blind study described above (Hogberg et al., 2004). Urine sam-ples were collected from 87 children and urinary nitrite/nitrate con-centrations were monitored at 0, 3, 6, 9, and 12 months. There was arapid decline in urinary nitrite/nitrate concentrations in both groups asearly as 3 months. No differences were seen between the study groupsat any of the checkpoints. However, at the end of the study, the nitrite/nitrate values of nine children in the gluten-free diet including oats groupand eight children in the standard gluten-free diet group hadnot normal-ized. Childrenwith celiac disease on a gluten-free dietwith oats display asimilar reduction in urinary nitrite/nitrate as those on a traditionalgluten-free diet. Some children, however, still demonstrate high nitrite/nitrate excretion after 1 year on either diet, indicating that long-termfollow-up studies of children on an oats-containing diet are needed.


5. Holm et al. (2006) conducted a 2-year controlled clinical trial. A total of36 children who were over 7 years of age and were either previouslydiagnosed or had newly detected celiac disease were recruited for thisstudy. Of these children, 32 consented and 4 refused because theyfound the protocol too laborious. In all patients, the diagnosis of celiacdisease was based on the presence of small-intestinal mucosal severe,partial, or subtotal villous atrophy with crypt hyperplasia, and initiallyall had been serum endomysial antibody (EMA) positive. Twenty-three out of the 32 children were previously diagnosed with celiacdisease and had been treated with a conventional strict gluten-freediet (avoiding wheat, rye, barley, and oats) for at least 2 years beforethey were included in the study. All children exhibited disease remis-sion. These 23 patients were randomized either to undergo an open oatchallenge or a gluten challenge which allowed the consumption ofwheat, rye, and barley in addition to oats. The intake of oats was50 g/day and patients in the gluten-challenge group ingested 20 ggluten/day. The purity of the oats (gluten-free) was confirmed byELISA and PCR. In addition to the patients in remission, nine newlydiagnosed children with celiac disease were included in the trial con-suming a comparable gluten-free diet including oats. In all patients thefollowing parameters were assessed: small-intestinal mucosal mor-phology, IELs, human leukocyte antigen D-related (HLA-DR) expres-sion, and celiac serology. During the first 2 years on an oat-containingdiet, clinical, nutritional, and serological assessments were carried outat 0, 1, 3, 6, 12, 18, and 24 months. Small bowel mucosal biopsies wereevaluated at baseline and after 6 and 24 months. If the small bowelmucosal biopsy confirmed a relapse among the gluten-challengegroup, the patients reverted to a gluten-free diet (avoiding wheat,rye, and barley) including the consumption of oats. Follow-up exam-inations of the gluten-challenge group were carried out similarly to theoats group until small bowel mucosal histological relapse was evident.After the relapse and commencement of an oat-containing gluten-freediet, examinations continued to be carried out in the sameway as in theoats-challenge group. After the 2-year trial, patients continued to sup-plement their diets with commercially available oat products. Thepurity of these products was previously analyzed: 29 out of 30 testedsamples had gliadin levels below 28 mg/kg (¼ 28 ppm) and only onewas clearly wheat contaminated in excess of 200 ppm gluten. Follow-up visits after the 2-year trial included nutritional and serologicalassessments once a year or every other year for 7 years. In the long-term follow-up, small-intestinal biopsies were considered only if thepatient’s clinical condition or serology implied a relapse of the disease.The authors reported that oats had no detrimental effect on intestinalhistology or serology of the children with celiac disease who were


in remission, during the 2-year trial. In contrast, the gluten-challengegroup relapsed after 3–12 months. Complete recovery from the diseasewas accomplished in all patients on gluten-free diet plus oats. Afterthe 2-year trial, 86% of the children preferred to continue to consumeoats and they all remained in remission for the duration of the 7-yearfollow-up. This study permitted higher levels of oat ingestion (median43 g/day and up to 81 g/day) than other studies conducted in children(median 15–24 g/day). The trial was conducted for 2 years andpatients on the gluten-free diet plus oats were followed-up clinicallyfor 7 years thereafter. The authors concluded that uncontaminated oatscould be safely included in a gluten-free diet in the majority of childrensuffering from celiac disease. In their view, oats diversifies the gluten-free diet and children preferred it in their diet.

B. Summary of nonpivotal studies testing the effect of oats inpatients with celiac disease by in vitro methods or serology(Table 6.2)

1. In vitro—duodenal mucosal cultures

1. This study was a continuation of the previous study (Srinivasan et al.,1996). Duodenal biopsies from 26 patients were stained for lactaseexpression using an indirect immunoperoxidase method. Eleven dis-ease control patients had normal architecture and nine had features ofactive celiac disease. Ten patients, who had celiac disease in clinicaland histological remission, underwent oats challenge for 12 weeks.Confluent expression of lactase was observed in the 11 control patientswith normal histology, whereas staining was absent in the ninepatients with active celiac disease. All 10 patients with treated celiacdisease had normal lactase expression after exposure to oats. Thepreservation of lactase enzyme after oats challenge further supportstheir previous findings of lack of oats toxicity.

2. Picarelli et al. (2001) used an in vitro model to test whether oats induceendomysial antibody production in the supernatant fluid of culturedduodenal mucosal specimens which were collected from 13 treatedpatients with celiac disease. The biopsy specimens were cultured withand without a peptic-tryptic digest (PT) of gliadin and avenin (fromoats) and in medium alone. Samples from 5 of the 13 patients werecultured with the C fraction of PT-avenin. Indirect immunofluores-cence was used to detect EMAs. These antibodies were detected inspecimens from all 13 patients after the challenge with gliadin but not


after culture in medium alone or with PT-avenin or its C fraction. Theauthors concluded that PT-avenin and its C fraction did not induceEMAs in patients with celiac disease.

3. Kilmartin et al. (2003) investigated the immunogenicity of avenin usingcytokines IFN-gamma (IFN-g) and interleukin (IL)-2 as markers ofimmunological activity. In this study, duodenal biopsies from patientswith celiac disease were cultured with 5 mg/ml of PT gliadin (n¼ 9) or5 mg/ml of PT-avenin (n ¼ 8) for 4 h. These biopsies were comparedagainst control biopsies cultured with the medium alone and biopsiesfrom nonceliac patients cultured with PT gliadin (n ¼ 8) or avenin(n ¼ 8). Cytokine mRNA was quantified by TaqMan PCR. Secretedcytokine protein was measured in the culture supernatant by ELISA.The authors found that after culture with PT gliadin, an increase inIFN-g mRNA was observed in all nine patients with celiac disease.Increased IFN-g protein was also found in four of these patients andsmaller increases in IL-2 mRNA were detected in six of the subjectswith celiac disease with a corresponding increase of IL-2 protein foundin two of these patients. In contrast, the biopsies of patients with celiacdisease cultured with PT-avenin did not show a significant response ofIFN-g or IL-2. Similarly, the biopsies from normal controls did notrespond to either gliadin or avenin stimulation. The authors suggestthat the immunogenic sequences in gliadin are not present in avenin.This suggestion supports the results of in vivo studies reporting thatoats are safe for consumption by patients with celiac disease.

4. Arentz-Hansen et al. (2004) conducted an in vitro study using intestinalT cell lines. The study included nine adults with celiac disease who hada history of oats exposure. The oats were derived from a quality controlproduction line and were shown to be free of contamination from othercereals. The selection of the study participants was not random. Five ofthe patients had also participated in a clinical challenge study consist-ing of 19 adults with celiac disease who ate 50 g of oats for 12 weeks(Lundin et al., 2003). Four of the patients had clinical symptoms on anoat-containing diet and three of these four patients had intestinalinflammation typical of celiac disease at the time of oats exposure.The investigators established oats avenin-specific and -reactive intesti-nal T cell lines from these three patients, as well as from two otherpatients who appeared to tolerate oats. The avenin-reactive T cell linesrecognized avenin peptides in the context of HLA-DQ2. These pep-tides have sequences rich in proline and glutamine residues closelyresembling wheat gluten epitopes. The authors concluded that somepatients with celiac disease have avenin-reactive mucosal T cells that


can cause mucosal inflammation. The authors point out that the T cellresponse to the avenin epitopes were found in T cell lines derived fromintestinal biopsies of patients with celiac disease that were stimulatedwith gliadin (Vader et al., 2003). It is unknown whether any of thepatients from whom these T cells were isolated had clinical symptomsor mucosal inflammation related to ingestion of oats. The authorssuggest that it will only be possible to establish the frequency ofintolerance and possible complications with extended clinical follow-up of patients with celiac disease consuming oats. The authors indicatethat their observations demonstrate that even if oats seem to be welltolerated by many patients with celiac disease, there are patients whohave intestinal T cell responses to oats. They suggest that until theprevalence of oats intolerance in patients with celiac disease is wellestablished, clinical follow-ups of patients eating oats is advisable.

5. Srinivasan et al. (2006) evaluated the response of the small intestine tooats by assessing the activation of the gastrointestinal immune sys-tem. This study involved 10 adults who ingested 50 g of oats daily inconjunction with an otherwise gluten-free diet for a 12-week period(Srinivasan et al., 1996). The oat cereal used in this study was reportedto be entirely free of contamination as tested by various methodolo-gies, including reverse-phase HPLC, ELISA, and PCR. Patient com-pliance, clinical symptoms, and serology (IgG antigliadin and IgAanti-endomysial antibodies) were monitored throughout the studyperiod. Duodenal biopsies were obtained by endoscope at the begin-ning and end of the study. After the 12-week trial, four of the patientswere challenged with 500 mg gluten/day and another two patientswith 10 g gluten/day for a period of 6 weeks. Duodenal biopsiesobtained before and after the inclusion of oats in the diet were stainedwith a series of antibodies directed against the following molecules:HLA-DR, Ki67, CD25, and CD54 (intercellular adhesion molecule 1(ICAM-1)) and mast cell tryptase. These detailed immunohistologicalstudies of the biopsies did not reveal evidence of immune activationor morphological damage following the consumption of oats. On theother hand, all patients challenged with gluten (500 mg or 10 g/day)showed evidence of reactivation of the disease with various degreesof gastrointestinal symptoms, serological responses, and evidence ofmorphological changes in the duodenal biopsy. The authors con-cluded that under these trial conditions, the ingestion of oats didnot show evidence of immunogenic or toxic effects on the duodenalmucosa of individuals with celiac disease and supported the view-point that oats is well tolerated by the majority of patients with celiacdisease.


6. Kilmartin et al. (2006) conducted investigations on the etiological roleof the wheat-related cereals, barley, rye, and oats, by examining theimmune response of gliadin-reactive mucosal T cell lines from patientswith celiac disease to fractions from all four cereals. The oats used inthis study were free from wheat contamination. Cell stimulation wasdetermined by measuring proliferation (employing 3H-thymidineincorporation) or cytokine (IL-2, IFN-g) production. All five T celllines demonstrated immunoreactivity to protein fractions from thefour related cereals. In some cell lines, reactivity to wheat, barley,and rye was evident only when these cereal fractions had been pre-treated with tissue transglutaminase. The authors concluded that thisstudy confirms the similar T cell antigenic reactivity of these fourrelated cereals, which has implications for their exclusion fromgluten-free diets. However, they also indicated that despite oat stimu-lation of T cell lines, this cereal does not activate a mucosal lesion inmost patients with celiac disease. The authors indicate that in thisstudy, even when 5 mg/ml of avenin was added to the biopsy culture,there was no evidence of cytokine production. An equivalent amountof gliadin activated the celiac mucosa. Because avenin accounts foronly 5–15% of the total protein in oats (whereas wheat, barley, and ryeprolamins constitute 40–50%), it has been suggested by other investi-gators that large amounts of oats still may be toxic to patients withceliac disease. The authors argued against this and demonstrate thatpurified avenin is not immunogenic to the intestinal mucosa of patientswith celiac disease. The authors noted from their results that theimmunogenic sequences of gliadin were not present in avenin, whichfurther supports the conclusion that oats are safe for consumption bypatients with celiac disease.

2. In vitro—other

1. Silano et al. (2007a,b) investigated the immunogenic effect of aveninsfrom four different cultivars of oats (Lampton from Australia, andAstra, Ava, and Nave from Italy), using peripheral lymphocytes from10 children with celiac disease. Results were compared with the immu-nogenic response induced by wheat gliadin and rice prolamins. Lym-phocyte proliferation and IFN-g release in the culture medium weremeasured as indices of immune activation. The authors report that allthe varieties of oats tested were immunogenic, with Lampton and Avaavenins inducing lymphocyte activation similar to that activated bywheat gliadin, while Astra and Nave avenins showed less immunoge-nicity, but still with a measurable effect. Sliano et al. (2007a) hadpreviously tested some of these oat cultivars using an in vitro model


based on the ability of PT digests of celiac-active proteins to agglutinateK562 cells and to disrupt lysosomes, respectively. As discussed inthese publications, the significance of these results to patients withceliac disease needs further clarification. The authors further indicatethat the immunogenic ability of avenins does not necessarily mean thatthe oat-containing foods are toxic for patients with celiac diseaseand that in fact, in these experimental models, the gliadin encounterwith lymphocytes is abnormal and only the alcohol-soluble proteinfraction of oats has been tested. Therefore, the study cannot rule outthat factors exerting a protective effect against avenin toxicity or inter-fering with avenin absorption have not been tested. The authors con-clude that until there is more evidence to show the safety of oats andvarieties of low-toxicity oats, patients with celiac disease consumingoats-containing foods should be carefully monitored.

3. Serology

1. In a study byGuttormsen et al. (2008) serumwas collected from 136 adultpatients with treated celiac disease and 139 controls. Patients wererecruited from the Norwegian Celiac Disease Association and had adefinite diagnosis of celiac disease, which was confirmed by small-intestinal biopsy. The study participants had been eating a strictgluten-free diet for at least 2 years. Eighty-two of the patients with celiacdisease had been consuming oats as part of their gluten-free diet for6 months or more. The participants with celiac disease completed aquestionnaire that focused on: diagnosis, dietary habits, and clinicalsymptoms. Patients with celiac disease, who were eating oats, registeredtheir intake of oats during a 3-week period. Patients were instructed toconsume only ecologically grown oats specifically produced for patientswith celiac disease. Otherwise there is no report of testing of the purity ofoats. In addition to the questionnaire, IgA against oats avenin, wheatgliadin, and tissue transglutaminase were tested with ELISA. Theauthors report that no significant differences were found in IgA againstoats in oats-eating and nonoats-eating patients with celiac disease. Bothgroups had increased levels of IgA against wheat, oats, and tissuetransglutaminase compared to healthy controls. A significant positivecorrelation was found between antiavenin and antigliadin IgA(p < 0.0001), and between antiavenin and anti-tissue transglutaminaseIgA (p ¼ 0.0012). The authors conclude that ingestion of oats does notcause increased levels of IgA against oats in adult patients with celiacdisease on a gluten-free diet. The findings support the notion that mostadult patients with celiac disease can tolerate oats.


C. Other studies relevant to the effect of oats in patients withceliac disease

1. Peraaho et al. (2004b) conducted a retrospective evaluation beginningin 1997, on the consumption of oats within a gluten-free diet bypatients with celiac disease and dermatitis herpetiformis in Finland.The use of oats, the effect of oats on symptoms of the illness, andquality of life were investigated in 1000 randomly selected membersof the Finnish Celiac Society. Altogether, 710 patients responded to thequestionnaire: 423 (73%) with celiac disease and 70 (55%) with derma-titis herpetiformis were currently consuming oats in their diets.Patients reported appreciating the taste, the ease of use, and the lowcost of oats. Of the respondents, 94% believed that oats diversified thegluten-free diet. However, 15% of the patients with celiac disease and28% of the patients with dermatitis herpetiformis reported that theyhad stopped eating oats. The most common reasons for avoiding oatswere fear of adverse effects or contamination.

ACKNOWLEDGMENTS

The authors wish to thank the Canadian Celiac Association (CCA) and the FondationQuebecoise De La Maladie Coeliaque for their contribution and dedication to the well-being of Canadianswith Celiac Disease and those following a gluten-free diet. Special thanksto the Professional Advisory Board of the CCA for their expert advice and exchange ofinformation during the preparation of this document. Special thanks to Shelley Case forproviding invaluable information. The authors also wish to acknowledge the contributionsof Dr. Vern Burrows of Agriculture Canada for his dedicated pioneering work in thedevelopment of pure, unadulterated oats in Canada, and Canadian farmers making possiblethe availability of this product in Canada.

REFERENCES

Abdallah, H., Leffler, D., Dennis, M., and Kelly, C. P. (2007). Refractory celiac disease. Curr.Gastroenterol. Rep. 9, 401–405.

Abenavoli, L., Proietti, I., Leggio, L., Ferrulli, A., Vonghia, L., Capizzi, R., Rotoli, M.,Amerio, P. L., Gasbarrini, G., and Addolorato, G. (2006). Cutaneous manifestations inceliac disease. World J. Gastroenterol. 12, 843–852.

Akobeng, A. K. and Thomas, A. G. (2008). Systematic review: Tolerable amount of gluten forpeople with coeliac disease. Aliment. Pharmacol. Ther. 27, 1044–1052.

Alaedini, A. and Green, P. H. (2005). Narrative review: Celiac disease: Understanding acomplex autoimmune disorder. Ann. Intern. Med. 142, 289–298.

Alaedini, A. and Green, P. H. (2008). Autoantibodies in celiac disease. Autoimmunity 41,19–26.

Andon, M. B. and Anderson, J. W. (2008). The oatmeal-cholesterol connection: 10 years later.Am. J. Lifestyle Med. 2(1), 51–57.


Arentz-Hansen, H., Fleckenstein, B., Molberg, O., Scott, H., Koning, F., Jung, G.,Roepstorff, P., Lundin, K. E., and Sollid, L. M. (2004). The molecular basis for oatintolerance in patients with celiac disease. PLoS Med. 1, e1.

Barton, S. H. and Murray, J. A. (2008). Celiac disease and autoimmunity in the gut andelsewhere. Gastroenterol. Clin. N. Am. 37, 411–428.

Bianchi, M. L. and Bardella, M. T. (2008). Bone in celiac disease.Osteoporos. Int. 19, 1705–1716.Briani, C., Samaroo, D., and Alaedini, A. (2008). Celiac disease: From gluten to autoimmu-

nity. Autoimmun. Rev. 7, 644–650.Buchanan, R., Dennis, S., Gendel, S., Acheson, D., Assimon, S. A., Beru, N., Bolger, P.,

Carlson, D., Carvajal, R., Copp, C., Falci, K., Garber, E., et al. (2008). Approaches toestablish thresholds for major food allergens and for gluten in food. J. Food Prot. 71,1043–1088.

Burrows, V. D. (2005). Hulless Oats. In ‘‘Specialty Grains for Food and Feed’’ (E-S. M. Abdel-Aal and P. Wood, eds.), pp. 223–251. American Association of Cereal Chemists, St. Paul,MN.

Case, S. (2008). Gluten-free diet. In ‘‘A Comprehensive Resource Guide’’ pp. 19–72. CaseNutrition Consulting, Regina, Canada.

Catassi, C. (2005). The world map of celiac disease. Acta Gastroenterol. Latinoam. 35, 37–55.Catassi, C. and Fasano, A. (2008). Is this really celiac disease? Pitfalls in diagnosis. Curr.

Gastroenterol. Rep. 10, 466–472.Catassi, C., Fabiani, E., Iacono, G., D’Agate, C., Francavilla, R., Biagi, F., Volta, U.,

Accomando, S., Picarelli, A., De, V. I., Pianelli, G., Gesuita, R., et al. (2007a). A prospective,double-blind, placebo-controlled trial to establish a safe gluten threshold for patientswith celiac disease. Am. J. Clin. Nutr. 85, 160–166.

Catassi, C., Kryszak, D., Louis-Jacques, O., Duerksen, D. R., Hill, I., Crowe, S. E.,Brown, A. R., Procaccini, N. J., Wonderly, B. A., Hartley, P., Moreci, J., Bennett, N., et al.(2007b). Detection of Celiac disease in primary care: A multicenter case-finding study inNorth America. Am. J. Gastroenterol. 102, 1454–1460.

Ciclitira, P. J., Ellis, H. J., and Lundin, K. E. (2005). Gluten-free diet—what is toxic? Best Pract.Res. Clin. Gastroenterol. 19, 359–371.

Collin, P., Maki, M., and Kaukinen, K. (2007). Safe gluten threshold for patients with celiacdisease: Some patients are more tolerant than others. Am. J. Clin. Nutr. 86, 260–261.

Cornell, H. J., McLachlan, A., and Cullis, P. G. (2002). Extraction of cereal prolamins and theirtoxicity in coeliac disease. J. Biochem. Mol. Biol. Biophys. 6, 151–158.

Cranney, A., Zarkadas, M., Graham, I. D., Butzner, J. D., Rashid, M., Warren, R., Molloy, M.,Case, S., Burrows, V., and Switzer, C. (2007). The Canadian Celiac Health Survey. Dig.

Dis. Sci. 52, 1087–1095.Dewar, D. H., Amato, M., Ellis, H. J., Pollock, E. L., Gonzalez-Cinca, N., Wieser, H., and

Ciclitira, P. J. (2006). The toxicity of high molecular weight glutenin subunits of wheat topatients with coeliac disease. Eur. J. Gastroenterol. Hepatol. 18, 483–491.

Dickey, W. (2008). Making oats safer for patients with coeliac disease. Eur. J. Gastroenterol.Hepatol. 20, 494–495.

Dickson, B. C., Streutker, C. J., and Chetty, R. (2006). Coeliac disease: An update forpathologists. J. Clin. Pathol. 59, 1008–1016.

Dor, R. and Shanahan, D. J. (2002). Oats and coeliac disease. Gut 51, 757–758.Ellis, H. J. and Ciclitira, P. J. (2001). In vivo gluten challenge in celiac disease. Can. J.

Gastroenterol. 15, 243–247.Ellis, H. J. and Ciclitira, P. J. (2008). Should coeliac sufferers be allowed their oats? Eur. J.

Gastroenterol. Hepatol. 20, 492–493.Fasano, A., Berti, I., Gerarduzzi, T., Not, T., Colletti, R. B., Drago, S., Elitsur, Y., Green, P. H.,

Guandalini, S., Hill, I. D., Pietzak, M., Ventura, A., et al. (2003). Prevalence of celiac


disease in at-risk and not-at-risk groups in the United States: A large multicenter study.Arch. Intern. Med. 163, 286–292.

Fasano, A., Araya, M., Bhatnagar, S., Cameron, D., Catassi, C., Dirks, M., Mearin, M. L.,Ortigosa, L., and Phillips, A. (2008). Federation of International Societies of PediatricGastroenterology, Hepatology, and Nutrition consensus report on celiac disease.J. Pediatr. Gastroenterol. Nutr. 47, 214–219.

Food And Drug Administration (2006). USA. Threshold Working Group On ‘‘ApproachesTo Establish Thresholds For Major Food Allergens And For Gluten. March 2006. Http://Www.Cfsan.Fda.Gov/�Dms/Alrgn.Html.

Food And Drug Administration (2008). USA. Food Labelling:Health Claims;Soluble Fiberfrom Certain Foods and Risk of Coronary Disease. Federal Register 73, No. 85.

Food and Drug Administration., U.D.o.H.a.H.S. (2007). USA. Food Labeling, Gluten-freelabelling of foods. Federal Register, Vol. 72, No 14.

Freeman, H. J. (2008). Neurological disorders in adult celiac disease. Can. J. Gastroenterol. 22,909–911.

Garsed, K. and Scott, B. B. (2007). Can oats be taken in a gluten-free diet? A systematicreview. Scand. J. Gastroenterol. 42, 171–178.

Gelinas, P., McKinnon, C. M., Mena, M. C., and Mendez, E. (2008). Gluten contamination ofcereal foods in Canada. Int. J. Food Sci. Technol. 43(7), 1245–1252.

Gianfrani, C., Troncone, R., and La, C. A. (2008). Autoimmunity and celiac disease.Mini Rev.Med. Chem. 8, 129–134.

Green, P. H. and Cellier, C. (2007). Celiac disease. N. Engl. J. Med. 357, 1731–1743.Guandalini, S. (2007). The influence of gluten: Weaning recommendations for healthy chil-

dren and children at risk for celiac disease.Nestle Nutr. Workshop Ser. Pediatr. Program. 60,139–151.

Guttormsen, V., Lovik, A., Bye, A., Bratlie, J., Morkrid, L., and Lundin, K. E. (2008). Noinduction of anti-avenin IgA by oats in adult, diet-treated coeliac disease. Scand.

J. Gastroenterol. 43, 161–165.Haboubi, N. Y., Taylor, S., and Jones, S. (2006). Coeliac disease and oats: A systematic review.

Postgrad. Med. J. 82, 672–678.Haines, M. L., Anderson, R. P., and Gibson, P. R. (2008). Systematic review: The evidence

base for long-term management of coeliac disease. Aliment. Pharmacol. Ther. 28,1042–1066.

Hardman, C. M., Garioch, J. J., Leonard, J. N., Thomas, H. J., Walker, M. M., Lortan, J. E.,Lister, A., and Fry, L. (1997). Absence of toxicity of oats in patients with dermatitisherpetiformis. N. Engl. J. Med. 337, 1884–1887.

Harrison, M. S., Wehbi, M., and Obideen, K. (2007). Celiac disease: More common than youthink. Cleve. Clin. J. Med. 74, 209–215.

Hernando, A., Mujico, J. R., Mena, M. C., Lombardia, M., and Mendez, E. (2008). Measure-ment of wheat gluten and barley hordeins in contaminated oats from Europe, the UnitedStates and Canada by Sandwich R5 Elisa. Eur. J. Gastroenterol. Hepatol. 20(6), 545–554.

Hill, P. G. and Holmes, G. K. (2008). Coeliac disease: A biopsy is not always necessary fordiagnosis. Aliment. Pharmacol. Ther. 27, 572–577.

Hoffenberg, E. J., Haas, J., Drescher, A., Barnhurst, R., Osberg, I., Bao, F., and Eisenbarth, G.(2000). A trial of oats in children with newly diagnosed celiac disease. J. Pediatr. 137,361–366.

Hogberg, L., Laurin, P., Falth-Magnusson, K., Grant, C., Grodzinsky, E., Jansson, G.,Ascher, H., Browaldh, L., Hammersjo, J. A., Lindberg, E., Myrdal, U., andStenhammar, L. (2004). Oats to children with newly diagnosed coeliac disease: A rando-mised double blind study. Gut 53, 649–654.

Hollen, E., Forslund, T., Hogberg, L., Laurin, P., Stenhammar, L., Falth-Magnusson, K.,Magnusson, K. E., and Sundqvist, T. (2006a). Urinary nitric oxide during one year of

http://www.Cfsan.Fda.Gov/~Dms/Alrgn.Html

http://www.Cfsan.Fda.Gov/~Dms/Alrgn.Html


gluten-free diet with or without oats in children with coeliac disease. Scand. J.

Gastroenterol. 41, 1272–1278.Hollen, E., Holmgren, P. K., Sundqvist, T., Grodzinsky, E., Hogberg, L., Laurin, P.,

Stenhammar, L., Falth-Magnusson, K., and Magnusson, K. E. (2006b). Coeliac childrenon a gluten-free diet with or without oats display equal anti-avenin antibody titers. Scand.J. Gastroenterol. 41, 42–47.

Holm, K., Maki, M., Vuolteenaho, N., Mustalahti, K., Ashorn, M., Ruuska, T., andKaukinen, K. (2006). Oats in the treatment of childhood coeliac disease: A 2-yearcontrolled trial and a long-term clinical follow-up study. Aliment. Pharmacol. Ther. 23,1463–1472.

Hopman, E. G., von Blomberg, M. E., Batstra, M. R., Morreau, H., Dekker, F. W., Koning, F.,Lamers, C. B., and Mearin, M. L. (2008). Gluten tolerance in adult patients with celiacdisease 20 years after diagnosis? Eur. J. Gastroenterol. Hepatol. 20, 423–429.

Hopper, A. D., Hadjivassiliou, M., Hurlstone, D. P., Lobo, A. J., McAlindon, M. E., Egner, W.,Wild, G., and Sanders, D. S. (2008). What is the role of serologic testing in celiac disease?A prospective, biopsy-confirmed study with economic analysis. Clin. Gastroenterol.

Hepatol. 6, 314–320.Howdle, P. D. (2006). Gliadin, glutenin or both? The search for the Holy Grail in coeliac

disease. Eur. J. Gastroenterol. Hepatol. 18, 703–706.Janatuinen, E. K., Pikkarainen, P. H., Kemppainen, T. A., Kosma, V. M., Jarvinen, R. M.,

Uusitupa, M. I., and Julkunen, R. J. (1995). A comparison of diets with andwithout oats inadults with celiac disease. N. Engl. J. Med. 333, 1033–1037.

Janatuinen, E. K., Kemppainen, T. A., Pikkarainen, P. H., Holm, K. H., Kosma, V. M.,Uusitupa, M. I., Maki, M., and Julkunen, R. J. (2000). Lack of cellular and humoralimmunological responses to oats in adults with coeliac disease. Gut 46, 327–331.

Janatuinen, E. K., Kemppainen, T. A., Julkunen, R. J., Kosma, V. M., Maki, M., Heikkinen, M.,and Uusitupa, M. I. (2002). No harm from five year ingestion of oats in coeliac disease.Gut 50, 332–335.

Jolobe, O. (2008). The changing face of coeliac disease and shared genetic traits. Br. J. Hosp.

Med. (Lond.) 69, 299.Kemppainen, T., Janatuinen, E., Holm, K., Kosma, V. M., Heikkinen, M., Maki, M.,

Laurila, K., Uusitupa, M., and Julkunen, R. (2007). No observed local immunologicalresponse at cell level after five years of oats in adult coeliac disease. Scand. J. Gastroenterol.42, 54–59.

Kemppainen, T. A., Heikkinen, M. T., Ristikankare, M. K., Kosma, V. M., Sontag-Strohm, T. S., Brinck, O., Salovaara, H. O., and Julkunen, R. J. (2008). Unkilned andlarge amounts of oats in the coeliac disease diet: A randomized, controlled study.Scand. J. Gastroenterol. 43, 1094–1101.

Kilmartin, C., Lynch, S., Abuzakouk, M., Wieser, H., and Feighery, C. (2003). Avenin fails toinduce a Th1 response in coeliac tissue following in vitro culture. Gut 52, 47–52.

Kilmartin, C., Wieser, H., Abuzakouk, M., Kelly, J., Jackson, J., and Feighery, C. (2006).Intestinal T cell responses to cereal proteins in celiac disease. Dig. Dis. Sci. 51, 202–209.

Koning, F. (2008). Celiac disease: Sandwiched between innate and adaptive immuneresponses induced by gluten. J. Pediatr. Gastroenterol. Nutr. 46(Suppl. 1), E8–E9.

Kumar, P. J. and Farthing, M. G. (1995). Oats and celiac disease. N. Engl. J. Med. 333,1075–1076.

Kupper, C. (2005). Dietary guidelines and implementation for celiac disease.Gastroenterology128, S121–S127.

Lester, D. R. (2008). Gluten measurement and its relationship to food toxicity for celiacdisease patients. Plant Methods 4, 26.


Lohi, S., Mustalahti, K., Kaukinen, K., Laurila, K., Collin, P., Rissanen, H., Lohi, O., Bravi, E.,Gasparin, M., Reunanen, A., andMaki, M. (2007). Increasing prevalence of coeliac diseaseover time. Aliment. Pharmacol. Ther. 26, 1217–1225.

Londei, M., Ciacci, C., Ricciardelli, I., Vacca, L., Quaratino, S., and Maiuri, L. (2005). Gliadinas a stimulator of innate responses in celiac disease. Mol. Immunol. 42, 913–918.

Losowsky, M. S. (2008). A history of coeliac disease. Dig. Dis. 26, 112–120.Lundin, K. E., Nilsen, E. M., Scott, H. G., Loberg, E. M., Gjoen, A., Bratlie, J., Skar, V.,

Mendez, E., Lovik, A., and Kett, K. (2003). Oats induced villous atrophy in coeliac disease.Gut 52, 1649–1652.

Makelainen, H., Anttila, H., Sihvonen, J., Hietanen, R. M., Tahvonen, R., Salminen, E.,Mikola, M., and Sontag Strohm, T. (2007). The effect of beta-glucan on the glycemic andinsulin index. Eur. J. Clin. Nutr. 61(6), 779–785.

Maki, K. C., Davidson, M. H., Witchger, M. S., Dicklin, M. R., and Subbaiah, P. V. (2007a).Effects of high-fiber oat and wheat cereals on postprandial glucose and lipid responses inhealthy men. Int. J. Vitam. Nutr. Res. 77, 347–356.

Maki, K. C., Galant, R., Samuel, P., Tesser, J., Witchger, M. S., Ribaya-Mercado, J. D.,Blumberg, J. B., and Geohas, J. (2007b). Effects of consuming foods containing oat beta-glucan on blood pressure, carbohydrate metabolism and biomarkers of oxidative stress inmen and women with elevated blood pressure. Eur. J. Clin. Nutr. 61, 786–795.

Malandrino, N., Capristo, E., Farnetti, S., Leggio, L., Abenavoli, L., Addolorato, G., andGasbarrini, G. (2008). Metabolic and nutritional features in adult celiac patients. Dig.

Dis. 26, 128–133.Mangione, R. A. (2008). Celiac disease and osteoporosis. Am. J. Health Syst. Pharm. 65,

1601–1602.Mantovani, M. S., Bellini, M. F., Angeli, J. P. F., Oliveira, R. J., Silva, A. F., and Ribeiro, L. R.

(2008). b-Glucans in promoting health: Prevention against mutation and cancer. Mutat.

Res. Rev. Mutat. Res. 658, 154–161.Martin, S. (2008). Against the grain: An overview of celiac disease. J. Am. Acad. Nurse Pract.

20, 243–250.Moron, B., Cebolla, A., Manyani, H., Alvarez-Maqueda, M., Megias, M., Thomas Mdel, C.,

Lopez, M. C., and Sousa, C. (2008). Sensitive detection of cereal fractions that are toxic toceliac disease patients by using monoclonal antibodies to a main immunogenic wheatpeptide. Am. J. Clin. Nutr. 87, 405–414.

National Institute of Health (2004). NIH consensus development conference on celiacdisease. NIH Consens. State Sci. Statements 21, 1–23.

Pastore, L., Carroccio, A., Compilato, D., Panzarella, V., Serpico, R., and Lo,M. L. (2008). Oralmanifestations of celiac disease. J. Clin. Gastroenterol. 42, 224–232.

Pellicano, R., Astegiano, M., Bruno, M., Fagoonee, S., and Rizzetto, M. (2007). Women andceliac disease: Association with unexplained infertility. Minerva Med. 98, 217–219.

Peraaho, M., Collin, P., Kaukinen, K., Kekkonen, L., Miettinen, S., andMaki, M. (2004a). Oatscan diversify a gluten-free diet in celiac disease and dermatitis herpetiformis. J. Am. Diet.

Assoc. 104, 1148–1150.Peraaho, M., Kaukinen, K., Mustalahti, K., Vuolteenaho, N., Maki, M., Laippala, P., and

Collin, P. (2004b). Effect of an oats-containing gluten-free diet on symptoms and qualityof life in coeliac disease. A randomized study. Scand. J. Gastroenterol. 39, 27–31.

Picarelli, A., Di, T. M., Sabbatella, L., Gabrielli, F., Di, C. T., Anania, M. C., Mastracchio, A.,Silano, M., and De, V. M. (2001). Immunologic evidence of no harmful effect of oats inceliac disease. Am. J. Clin. Nutr. 74, 137–140.

Piper, J. L., Gray, G. M., and Khosla, C. (2002). High selectivity of human tissue transgluta-minase for immunoactive gliadin peptides: Implications for celiac sprue. Biochemistry 41,386–393.


Pope, R. and Sheiner, E. (2009). Celiac disease during pregnancy: To screen or not to screen?Arch. Gynecol. Obstet. 279, 1–3.

Presutti, R. J., Cangemi, J. R., Cassidy, H. D., and Hill, D. A. (2007). Celiac disease. Am. Fam.

Physician 76, 1795–1802.Rashid, M., Cranney, A., Zarkadas, M., Graham, I. D., Switzer, C., Case, S., Molloy, M.,

Warren, R. E., Burrows, V., and Butzner, J. D. (2005). Celiac disease: Evaluation of thediagnosis and dietary compliance in Canadian children. Pediatrics 116, e754–e759.

Rashid, M., Butzner, D., Burrows, V., Zarkadas, M., Case, S., Molloy, M., Warren, R.,Pulido, O., and Switzer, C. (2007). Consumption of pure oats by individuals with celiacdisease: A position statement by the Canadian Celiac Association. Can. J. Gastroenterol. 21,649–651.

Reunala, T., Collin, P., Holm, K., Pikkarainen, P., Miettinen, A., Vuolteenaho, N., andMaki, M. (1998). Tolerance to oats in dermatitis herpetiformis. Gut 43, 490–493.

Reyna-Villasmil, N., Bermudez-Pirela, V., Mengual-Moreno, E., Arias, N., Cano-Ponce, C.,Leal-Gonzalez, E., Souki, A., Inglett, G. E., Israili, Z. H., Hernandez-Hernandez, R.,Valasco, M., and Arraiz, N. (2007). Oat-derived beta-glucan significantly improvesHDLC and diminishes LDLC and non-HDL cholesterol in overweight individuals withmild hypercholesterolemia. Am. J. Ther. 14, 203–212.

Ryan, D., Kendall, M., and Robards, K. (2007). Bioactivity of oats as it relates to cardiovascu-lar disease. Nutr. Res. Rev. 20(2), 147–162.

Sadiq Butt, M., Tahir-Nadeem, M., Khan, M. K., Shabir, R., and Butt, M. S. (2008). Oat:Unique among the cereals. Eur. J. Nutr. 47, 68–79.

Schmitz, J. (1997). Lack of oats toxicity in coeliac disease. BMJ 314, 159–160.Shan, L., Molberg, O., Parrot, I., Hausch, F., Filiz, F., Gray, G. M., Sollid, L. M., and Khosla, C.

(2002). Structural basis for gluten intolerance in celiac sprue. Science 297, 2275–2279.Shan, L., Qiao, S. W., Arentz-Hansen, H., Molberg, O., Gray, G. M., Sollid, L. M., and

Khosla, C. (2005). Identification and analysis of multivalent proteolytically resistantpeptides from gluten: Implications for celiac sprue. J. Proteome Res. 4, 1732–1741.

Silano, M., Dessi, M., De Vincenzi, M., and Cornell, H. (2007a). In vitro tests indicate thatcertain varieties of oats may be harmful to patients with coeliac disease. Eur. J.

Gastroenterol. Hepatol. 22, 528–531.Silano, M., Di Benedetto, R., Maialetti, F., De Vincenzi, A., Calcaterra, R., Cornell, H. J., and

De Vincenzi, M. (2007b). Avenins from different cultivars of oats elicit response by coeliacperipheral lymphocytes. Scand. J. Gastroenterol. 42, 1302–1305.

Srinivasan, U., Leonard, N., Jones, E., Kasarda, D. D., Weir, D. G., O’Farrelly, C., andFeighery, C. (1996). Absence of oats toxicity in adult coeliac disease. BMJ 313, 1300–1301.

Srinivasan, U., Jones, E., Weir, D. G., and Feighery, C. (1999). Lactase enzyme, detectedimmunohistochemically, is lost in active celiac disease, but unaffected by oats challenge.Am. J. Gastroenterol. 94, 2936–2941.

Srinivasan, U., Jones, E., Carolan, J., and Feighery, C. (2006). Immunohistochemical analysisof coeliac mucosa following ingestion of oats. Clin. Exp. Immunol. 144, 197–203.

St�rsrud, S., Hulthen, L. R., and Lenner, R. A. (2003a). Beneficial effects of oats in the gluten-free diet of adults with special reference to nutrient status, symptoms and subjectiveexperiences. Br. J. Nutr. 90, 101–107.

St�rsrud, S., Olsson, M., Arvidsson, L. R., Nilsson, L. A., Nilsson, O., and Kilander, A.(2003b). Adult coeliac patients do tolerate large amounts of oats. Eur. J. Clin. Nutr. 57,163–169.

Thompson, T. (1997). Do oats belong in a gluten-free diet? J. Am. Diet. Assoc. 97, 1413–1416.Thompson, T. (2000). Questionable foods and the gluten-free diet: Survey of current recom-

mendations. J. Am. Diet. Assoc. 100, 463–465.


Thompson, T. (2001). Case problem: Questions regarding the acceptability of buckwheat,amaranth, quinoa, and oats from a patient with celiac disease. J. Am. Diet. Assoc. 101,586–587.

Thompson, T. (2003). Oats and the gluten-free diet. J. Am. Diet. Assoc. 103, 376–379.Thompson, T. (2004). Gluten contamination of commercial oat products in the United States.

N. Engl. J. Med. 351, 2021–2022.Troncone, R., Auricchio, R., and Granata, V. (2008a). Issues related to gluten-free diet in

coeliac disease. Curr. Opin. Clin. Nutr. Metab. Care 11, 329–333.Troncone, R., Ivarsson, A., Szajewska, H., and Mearin, M. L. (2008b). Review article: Future

research on coeliac disease—A position report from the European multistakeholderplatform on coeliac disease (CDEUSSA). Aliment. Pharmacol. Ther. 27, 1030–1043.

Tully, M. A. (2008). Pediatric celiac disease. Gastroenterol. Nurs. 31, 132–140.Tye-Din, J. and Anderson, R. (2008). Immunopathogenesis of celiac disease. Curr.

Gastroenterol. Rep. 10, 458–465.Upton, M. P. (2008). ‘‘Give us this day our daily bread’’—Evolving concepts in celiac sprue.

Arch. Pathol. Lab. Med. 132, 1594–1599.Vader, L. W., de, R. A., van der, W. Y., Kooy, Y. M., Benckhuijsen, W., Mearin, M. L.,

Drijfhout, J. W., van, V. P., and Koning, F. (2002). Specificity of tissue transglutaminaseexplains cereal toxicity in celiac disease. J. Exp. Med. 195, 643–649.

Vader, L. W., Stepniak, D. T., Bunnik, E. M., Kooy, Y. M., de, H. W., Drijfhout, J. W., VanVeelen, P. A., and Koning, F. (2003). Characterization of cereal toxicity for celiac diseasepatients based on protein homology in grains. Gastroenterology 125, 1105–1113.

Wieser, H. and Koehler, P. (2008). The biochemical basis of celiac disease. Cereal Chem. 85(1),1–13.

Zarkadas, M. and Case, S. (2005). Celiac disease and the gluten-free diet. Clin. Nutr. 20,127–138.

Zarkadas, M., Cranney, A., Case, S., Molloy, M., Switzer, C., Graham, I. D., Butzner, J. D.,Rashid, M., Warren, R. E., and Burrows, V. (2006). The impact of a gluten-free diet onadults with coeliac disease: Results of a national survey. J. Hum. Nutr. Diet. 19, 41–49.

INDEX

A

Advanced glycation endproducts(AGEs), 97

Alzheimer’s disease (AD)APP and, 114cell culture studies, 115glycoxidation events, 114–115

Amyloid precursor protein (APP), 114

B

Buffer model system, 68

C

Caloric restriction (CR), 103Carnosine

age-related pathology, 110, 111aging

causes, 94mechanisms and antiaging activities,

95–96protein synthesis regulation, 108–109proteotoxicity, 95

aging, dietary restriction-mediated delayADP-ribose metabolism, 104AGEs, 105cAMP levels, 107fasting periods, 103mitochondria synthesis, 108NAD consumption and availability,

106NADþ/NADH levels, 104–105

alcoholic beverages consumption,124–125

altered proteins, 99–100Alzheimer’s disease

causing events, 114cell culture studies, 115CSF homocarnosine levels, 114–115dementia, 116

anserine concentrations, 90anticonvulsants and aging

isoaspartate residues, 102

PIMT overexpression and levels,102–103

selective methylation, 103autistic spectrum disorders and, 123–124blood pressure and, 124cancer and, 121–122cataractogenesis, 120–121corticosteroids, 109–110deafness and, 121deleterious effects, 131diabetes

AGE formation, 111cardiac and circulatory disorders, 110complications, 112

dialysis fluids, 125–126dietary changes, 129–130fructose foods and drinks, 125gene expression, 100–101heart failure, 123homeostatic properties, 92, 93immune function and, 122–123ischemia, 119–120metabolic regulations, 91neurodegenerative diseases

features, 113neurotoxic agents, 112–113neurotoxic effects, 113–114

neurological functions, 91nonenzymic protein glycosylation

antiaging/rejuvenating effects,97–98

browning reaction, 97carbonyl compounds, 98glucose degradation products, 99

osteoporosis, 120oxygen free radicals and oxidative stress,

96–97Parkinson’s disease

ATP synthesis, 116dopaminergic neurons, 116dopamine supply, 118L-dopa therapy, 118–119

pH control, 92structures, 89

287

288 Index

Carnosine (cont.)tissue levels

dietary supplementation, 128–129physiological regulation, 126–128

tissues and functions, 92vegetarian diets, 130–131wound healing and, 122zinc and copper ions, 93

Castor-oil fish, 19Cellulose acetate membrane electrophoresis

(CAME), 25–26Cerebral spinal fluid (CSF)

AD patients, 130homocarnosine levels, 114–115, 130protein AGEs, 114

Chill haze, 57CR. See Caloric restrictionCSF. See Cerebral spinal fluid

D

Dermatitis herpetiformisceliac disease

gluten-free diet, 255, 279oats safety, 238, 258–270in vivo clinical studies, 239–244,

250–252description, 237

Diacylglyceryl ether (DAGE)-rich fishS. maculatus lipid analysis, 39species, 40

Dimethylformamide (DMF), 66

E

Endomysial antibody (EMA), 273, 274

F

Fish-induced keriorrheaanimal study, 40–41biochemistry and toxicity

animal tests, 21–22castor-oil fish, 19escolar consumption, 21human studies, 22–23muscle oil, 20–21wax esters, 18–20

DAGE, 39–40description, 3detection and inspection

authentication methods, 43government actions, 44

DNA analysis

neighbor-joining (NJ) trees, 28species identification, 26

escolar and oilfishbiology, 6–8chemistry and cooking effect, 8global concerns, 14–15harmful effects, 9litigation, 15, 18mislabeling and mishandling, 13–14policies on, 16–17regulation, 15supply, 11–13uses, 9–11

food-poisoning, 40labeling, 41lipid analysis

escolar and oilfish, fatty acids, 27TLC, 27–30

morphological and anatomical analyses,oilfish

characteristic integument, 24–25escolar sashimi and cutlets, 26fillet and cutlets, 24musculature, 23–24

poisoning, 3protein analysis, 25–26reports and symptoms

in continents, 3–4diarrhea, 6escolar, oil discharge, 4gastrointestinal illness, 4, 6oilfish and escolar, consumption, 5outbreaks of, 3–4

risk assessment and educationpublic and health professionals, 41seafood and catering industry, 41–42

sources, 2–3warnings and handling, 42–43wax-ester-rich fish

epipelagic species, 30, 32functional differences, 39Gempylidae family, 30lipid components, 38–39orange roughy, 32, 38species, 33–37

Food deep-fat fryingheat transfer

characterization, 214–215coefficients, boiling convective, 216–217rates, 215–216water evaporation, 214

mass transfer characteristics, 214

Index 289

nutritional aspectsconsumption and human health,

220–221fat absorption, 218–219frying oils and oil degradation, 219–220

oil absorptionaffecting factors, 226–229kinetics of, 222–226reduction, 229–231

oil location after, 223process

definition, 210fried food, 212frying equipment, 212–214heat and mass transfer, 211technology, food industry, 211–212temperatures, 210–211

structure developmentcrust and core regions, 217fried potato, oil location, 218oil absorption mechanisms, 217–218

surface tension, oil, 225–226Fraunhofer theory, 56Fresh fruits and vegetables, microbial safety

foodborne illness outbreaks
deficiencies, 160fresh produce, 157–158Salmonella and E.coli, 159
human pathogens, vegetable productionchain

irrigation water, 171–175manure and biosolids, 169–171soil, 176–177transport, 177–179

interventionsHPP process, 190human pathogens biocontrol, 191–192irradiation, 189–190postharvest washing, 188surface pasteurization, 189UV light, 190–191

MAP packaging, 157pathogens characteristics

bacteria and symptoms, 161–162endospore-forming bacteria, 166–167enteric viruses, 167human pathogenic protozoa, 167–168pathogenic E. coli, 160–166

pathogens interaction, fresh producegenetic and physiological factors,

186–188internalization, 181–186

phyllosphere, 179–180rhizosphere, 180–181

production growth, 156Fresh produce safety

HPP, 190human pathogens control, 191–192irradiation, 189–190postharvest washing, 188surface pasteurization, 189UV light, 190–191

Frying equipmentautomatic basket-lift system, 212continuous fryers, 213deep-fat fryers, 213–214

G

Gas chromatography. See Thin-layerchromatography (TLC)

Gempylidae familycharacteristics, 30species, 31–32

H

Haze-active (HA) polyphenolsbeverage in, 74binding energy, 63flavan-3-ols

(þ)-catechin and (–)-epicatechinstructure, 63–64

gallocatechin and epigallocatechinstructure, 65

proanthocyanidins, 64–65structure, 63

prominent dimer structure, 66–67response surface model, 68

Haze-active (HA) proteinbeer, silica binding, 79beverage in, 74foam active effects and, 78haze-forming activity, 60–61hydroxyproline, 61–62peptide bond, 62–63PRPs, 61response surface model, 68

Haze, beveragescauses

grape juice, wine and grains, 58microorganisms and sucrose syrups,

58–59polysaccharides and proteins, 59

diagnosis

290 Index

Haze, beverages (cont.)
chemical analysis and enzyme
treatment, 60microscopy, 59–60

formation, protein–polyphenol, 73–74HA polyphenol, 75–76haze tests and HA protein, 75

HA polyphenolsdimer structure, 66–67flavan-3-ols, 63–64molecular features, 63proanthocyanidins, 64–65

HA proteinamino acids and DNA code, 61–62barley hordein amino acid sequence, 61codes, proline vs. amino acids, 62–63polyphenol and, 74proline, 60–61

particle size effectshaze intensity and, 70–71sedimentation and filtration

operations, 72pH and alcohol effects

ethanol concentration, 72haze intensity, 72–73

physics incalibration, turbidimeter, 56–57Mie theory, 56photometer, light scattering, 54–55temperature, 57turbidimeter light path, 55–56

polyphenol polymerizationmechanisms, 74

preventionadsorbents, 77–80cold maturation, 76–77enzymes, 80–81ultrafiltration, 77

protein–polyphenol interactionconceptual model, 68–70gliadin and TA, 70haze formation, 66–67hydrophobic and hydrogen

bonding, 67–68influence concept, 69mechanism, 65–66surface models, 68

visual perceptionsuprathreshold particle, 57–58thresholds determination, 57

High hydrostatic pressure (HPP), 190Hypersensitive response (HR), 186

I

Intraepithelial lymphocyte counts (IELs)duodenal counts, 264intestinal biopsies, 265oat-consuming and control

groups, 258T cell receptors, 270

Invisible hazes, 56

L

Localized induced resistance (LIR), 186

M

Magnitude estimation (ME), 57Microbial source tracking (MST), 178Mie theory, 56

N

Nephelometric turbidity units (NTU), 57Norwalk-like viruses (NLV), 167

O

Oats, celiac diseasebenefits, 261dermatitis herpetiformis, 238diagnosis, 237–238gluten

description, 237free diet, 254, 261intake, 238

nonpivotal in vitro studiesavenins, immunogenic effect, 277–278dietary challenges, 245–249duodenal mucosal cultures, 254,

274–277peripheral lymphocyte, 255serology, 255, 278

pivotal in vivo studiesadults and children, 250–252challenge group, gluten, 273–274contamination, 266–267controlled intervention, 265–266dermatitis herpetiformis, 240–244,

250–252, 264–265double blind study, 271–272duodenal biopsies, 264gastrointestinal symptoms,

252–254, 269gliadin contamination, 271

Index 291

gluten contamination, 263immunological responses, 269–270intolerance, 267kilned and unkilned, 270–274long-term ingestion, 266mass spectrometry (MS), 266patients group, 262–263uncontaminated rolled oats, 266urinalysis test, 272villous architecture, 264

potential toxicityavenins, 260–261gliadins and glutenins, 260gluten, 259–260

pure oats, 261–262retrospective evaluation, 279safety

in adults, 252assessment, 239biopsy, 259in children, 252–253database differences, 258IELs and, 259–260

Oil absorptionaffecting factors

crust microstructure, 226–227food surface area, 227frying temperature and time,

227–228moisture content, 226oil type and deterioration,

228–229kinetics of

cooling, oil suction, 223–224fractions and mechanism, 222–223penetration, 222physical and chemical properties,

225–226postfrying cooling, 224–225wetting properties, 225

reductioncoatings and batters, 230drying, 229hydrocolloids and thermal gelling,

229–230postfrying treatments, 230–231

Orange roughy deep-sea fishwax esters, 19animal tests, 22potential hazards, 32

Oxidation-reduction potential(ORP), 188

Oxidized protein hydrolase (OPH), 100

P

Parkinson’s disease (PD)ATP synthesis, 116carnosine treatment, 118mitochondrial dysfunction, 116

Pathogensbacteria and symptoms, 161–162endospore-forming bacteria, 166–167enteric viruses, 167human pathogenic protozoa, 167–168internalization, growing plants

culturing techniques, 185methods, 181–183postharvest washing, 181

pathogenic E. coliAeromonas hydrophila, 166Campylobacter, 165Listeria monocytogenes, 166Salmonella, 164–165Shigella, 164types, 163

Pectin methyltransferase (PMT), 190Peptic-tryptic (PT) digest, 270Phyllosphere, 179–180Polyacrylamide gel electrophoresis (PAGE),

25–26Polyphenol adsorbents, 78Polyvinylpyrrolidone (PVPP)

action, beer, 81binding, 78–79segment structure and usage, 80

Proline-rich proteins (PRPs), 61Protein-isoaspartate-methyltransferase

(PIMT), 102

R

Reactive nitrogen species (RNS), 88, 99Reactive oxygen species (ROS), 88, 96Rhizosphere, 180–181

S

12S rRNA gene sequences, 28Surfactant theory of frying, 17Systemic acquired resistance (SAR), 186

T

Thin-layer chromatography (TLC), escolarand oilfish

differentiation, 28–29characteristic spot, 29–30wax esters levels, 27–28

292 Index

V

Vegetable production chainenteric pathogens, 168human pathogens

library-independent approaches, 179movement, 177subsurface discharge waters, 178water contamination, 178–179

irrigation watercrop production, 171E.coli levels, 172enteric pathogens, 174human pathogens, 175recycling, 172–173

manure and biosolidsE. coli, 170organic cultivation systems, 169sewage waste, 169–170

soil, 176–177

W

Washburn equation, 224Wax esters

animal tests, 22DAGE, 39–40

description, 18detection of, 14epipelagic fish species, 30, 32fatty acids, 27fish species, 33–37gempylotoxin, 9gravities and viscosities, 32lipase hydrolysis, 19–20long chain, classification, 27in muscles and roes, 39orange roughy

fillet, 38levels of, 32, 38storage in, 19

roe oil, 38–39seborrhea, 21–22source of, 18–19structure, 18thin-layer chromatography (TLC)

application, 27–28fatty acids and alcohols, 29characteristic spot, 29–30use, 28–29

toxicity, 21variability, 20–21vertical migration, 19