Journal of Biotechnology 84 (2000) 197215

Review article

Probiotic bacteria: safety, functional and technologicalproperties

Maria Saarela a, Gunnar Mogensen b, Rangne Fonden c, Jaana Matto a,Tiina Mattila-Sandholm a,*

a VTT Biotechnology, P.O. Box 1500, FIN-02044 VTT, Finlandb GM-Consult, Thorsmose6ej 16, 3200 Helsinge, Denmark

c Arla Foods, 10546 Stockholm, Sweden

Received 19 July 2000; received in revised form 21 September 2000; accepted 4 October 2000

Abstract

During the past two decades probiotic (health promoting) micro-organisms have been increasingly included invarious types of food products, especially in fermented milks. Several aspects, including safety, functional andtechnological characteristics, have to be taken into consideration in the selection process of probiotic micro-organ-isms. Safety aspects include specifications such as origin (healthy human GI-tract), non-pathogenicity and antibioticresistance characteristics. Functional aspects include viability and persistence in the GI-tract, immunomodulation,antagonistic and antimutagenic properties. Before probiotic strains, chosen on the basis of their good safety andfunctional characteristics, can benefit the consumer, they must first be able to be manufactured under industrialconditions. Furthermore, they have to survive and retain their functionality during storage, and also in the foods intowhich they are incorporated without producing off-flavours. Factors related to the technological and sensory aspectsof probiotic food production are of utmost importance since only by satisfying the demands of the consumer can thefood industry succeed in promoting the consumption of functional probiotic products in the future. 2000 ElsevierScience B.V. All rights reserved.

Keywords: Probiotic bacteria; Health; Safety

www.elsevier.com:locate:jbiotec

1. Introduction

Probiotics are live microbial food supplementswhich benefit the health of consumers by main-taining or improving their intestinal microbialbalance (Fuller, 1989). Due to their perceivedhealth benefits probiotic bacteria have been in-

* Corresponding author. Fax: 358-9-4552103.E-mail address: tiina.mattila-sandholm@vtt.fi (T. Mattila-

Sandholm).

0168-1656:00:$ - see front matter 2000 Elsevier Science B.V. All rights reserved.

PII: S0168 -1656 (00 )00375 -8

M. Saarela et al. : Journal of Biotechnology 84 (2000) 197215198

creasingly included in yoghurts and fermentedmilks during the past two decades. Most com-monly they have been lactobacilli such as Lacto-bacillus acidophilus, and bifidobacteria oftenreferred to as bifidus (Daly and Davis, 1998). Amajor development in functional foods pertain tofoods containing probiotics and prebiotics whichenhance health promoting microbial flora in theintestine. There is growing scientific evidence tosupport the concept that the maintenance ofhealthy gut microflora may provide protectionagainst gastrointestinal disorders including gas-trointestinal infections, inflammatory bowel dis-eases, and even cancer (Haenel and Bendig, 1975;Mitsuoka, 1982; Salminen et al., 1998a). The useof probiotic bacterial cultures stimulates thegrowth of preferred micro-organisms, crowds outpotentially harmful bacteria, and reinforces thebodys natural defence mechanisms. Today,plenty of evidence exists on the positive effects ofprobiotics on human health. However, this hasusually been demonstrated in diseased humanpopulations only (Salminen et al., 1998a). Thusthere is an urgent need for evidence for probiotichealth benefits in average (generally healthy)populations.

Before a probiotic can benefit human health itmust fulfil several criteria: It must have goodtechnological properties so that it can be manu-factured and incorporated into food productswithout loosing viability and functionality or cre-ating unpleasant flavours or textures; it must sur-vive passage through the upper gastrointestinal(GI) tract and arrive alive at its site of action; andit must be able to function in the gut environ-ment. To study the probiotic strain in the GI-tract, molecular techniques must be establishedfor distinguishing the ingested probiotic strainfrom the potentially thousands of other bacterialstrains that make up the gastrointestinal ecosys-tem. Additionally, techniques are required to es-tablish the effect of the probiotic strain on othermembers of the intestinal microbiota and impor-tantly on the host. This includes not only positivehealth benefits, but also demonstration that probi-otic strains do not have any deleterious effects.Armed with this knowledge, the probiotics canthen enter human pilot studies that attempt to

assess their health benefits to consumers (Mattila-Sandholm and Salminen, 1998; Mattila-Sandholmet al., 1999).

2. Selecting probiotic strains: important aspects

The theoretical basis for the selection of probi-otic micro-organisms including safety, functionaland technological aspects is illustrated in Fig. 1.

Current safety criteria for successful probioticshave been defined in several reviews (Lee andSalminen, 1995; Donohue and Salminen, 1996;Salminen et al., 1996b, 1998b; Adams, 1999). Thesignificance of human origin has been debatedrecently, but most current successful strains areindicated to be of human origin. It can also beargued that a probiotic strain can function betterin a similar environment (e.g. human GI-tract) towhere it was originally isolated from. Safety as-pects include the following specifications:1. Strains for human use are preferably of human

origin.

Fig. 1. The theoretical basis for selection of probiotic micro-organisms includes safety, functional (survival, adherence,colonisation, antimicrobial production, immune stimulation,antigenotoxic activity and prevention of pathogens) and tech-nological aspects (growth in milk, sensory properties, stability,phage resistance, viability in processes).

M. Saarela et al. : Journal of Biotechnology 84 (2000) 197215 199

2. They are isolated from healthy human GI-tract.

3. They have a history of being non-pathogenic.4. They have no history of association with dis-

eases such as infective endocarditis or GI-disorders.

5. They do not deconjugate bile salts (bile saltdeconjucation or dehydroxylation would be anegative trait in the small bowel; Marteau etal., 1995).

6. They do not carry transmissible antibiotic re-sistance genes.

The functional requirements of probioticsshould be established by using in vitro methodsand the results of these studies should be reflectedin controlled human studies. While selecting apreferable probiotic strain several aspects of func-tionality have to be considered:1. Acid tolerance and tolerance to human gastric

juice.2. Bile tolerance (an important property for sur-

vival in the small bowel).3. Adherence to epithelial surfaces and persis-

tence in the human GI-tract.4. Immunostimulation, but no proinflammatory

effect.5. Antagonistic activity against pathogens such

as Helicobacter pylori, Salmonella sp., Listeriamonocytogenes and Clostridium difficile.

6. Antimutagenic and antigarcinogenicproperties.

Feeding trials with different probiotic strainshave shown that the probiotic strain usually dis-appears from the GI-tract within a couple ofweeks after the ingestion is discontinued(Fukushima et al., 1998; Johansson et al., 1998;Alander et al., 1999; Donnet-Hughes et al., 1999).The role of the probiotic persistence in the humanGI-tract has therefore been questioned. However,even temporary persistence, which has been notedfor several ingested probiotic strains, may en-hance their chances for beneficial functions in theGI-tract, and is therefore considered a desirabletrait.

Even though a probiotic strain fulfils the neces-sary safety and functional criteria the aspectsrelated to probiotic production and processing arealso of utmost importance. Several technological

Fig. 2. Safety aspects of probiotics.

aspects have to be considered in probiotic selec-tion. These include the following:1. Good sensory properties.2. Phage resistance.3. Viability during processing.4. Stability in the product and during storage.

Good viability and activity of probiotics areconsidered prerequisites for optimal functionality.However, several studies have shown that non-vi-able probiotics can have beneficial effects such asimmune modulation and carcinogen binding inthe host (for a review see Ouwehand and Salmi-nen, 1998; Salminen et al., 1999). Thus, for certainprobiotic strains it might be sufficient that theygrow well during initial production steps (to ob-tain high enough numbers in the product) butthey do not necessarily need to retain good viabil-ity during storage.

Safety, functional and technological aspects ofprobiotics are discussed in more detail below. Thereview is limited to Lactobacillus and Bifidobac-terium, since these genera are by far the mostimportant with probiotic strains for human use.

3. Safety aspects of probiotics

The safety of probiotic strains is of prime im-portance and guidelines for the safety assessmentcan be found in several articles (Lee and Salmi-nen, 1995; Donohue and Salminen, 1996; Salmi-nen et al., 1996b, 1998b; Adams, 1999) (Fig. 2).Several approaches are possible in the assessmentof the probiotic safety: (a) studies on the instrinsicproperties of the probiotic strain; (b) studies on

M. Saarela et al. : Journal of Biotechnology 84 (2000) 197215200

the pharmacokinetics of the probiotic strain; and(c) studies on interactions between the probioticstrain and the host.

Knowledge on survival of the probiotics withinthe GI-tract, their translocation and colonizationproperties, and the fate of probiotic-derived activecomponents is important for the evaluation ofpossible positive and negative effects of probioticconsumption. The survival of different probioticstrains in different parts of the GI-tract varies:Some strains are rapidly killed in the stomachwhile others can pass through the whole gut inhigh numbers (Marteau et al., 1993). The pharma-cokinetics of probiotics has been studied in vivousing intubation, perfusion and biopsy techniques(Pochart et al., 1992; Johansson et al., 1993;Nielsen et al., 1994; Alander et al., 1997; Marteauet al., 1997). Some enzymatic properties such asexcessive deconjugation of bile salts or degrada-tion of mucus has been postulated to be poten-tially detrimental (Marteau et al., 1995;Ruseler-van Embden et al., 1995). Such propertiescan be studied in vitro. Platelet aggregation prop-erties (Korpela et al., 1997), enzymes which seemto favour cardiac valve colonization (Pelletier etal., 1996) and formation of unwanted metabolitescan also be studied in vitro.

Assessing the risks of probiotic consumptioncould be a very expensive and time-consumingtask. A low risk may have to be accepted whenrecommended to immunocompromised individu-als, but the risk to benefit ratio needs to be clearlyestablished in such cases. This requires relevantinformation on the efficacy and safety of theproducts. At present, the available information oncurrent probiotic lactic acid bacteria providesgood safety back-up. To our knowledge there areup to date two reports of probiotic bacteriumcausing infection. A L. rhamnosus strain indistin-guishable from L. rhamnosus GG has been iso-lated from a liver abscess from an elderly ladywith a history of hypertension and diabetes melli-tus (Rautio et al., 1999). In another case a probi-otic L. rhamnosus strain (strain or productspecifications were not given) was suggested tohave caused endocarditis in an elderly male(Mackay et al., 1999). However, unlike in the liverabscess case where strain identification was based

on thorough genomic fingerprinting of the L.rhamnosus isolates, the endocarditis case studyrelied on conventional phenotypic strain identifi-cation methods and on pyrolysis mass spectrome-try identification. The discriminatory power ofconventional phenotypic identification methods instrain differentiation is usually poor and there isno data on the ability of pyrolysis mass spec-trometry to differentiate between different lacto-bacillar strains. Therefore the data on theendocarditis case must be considered insufficientfor proving that the ingested probiotic L. rhamno-sus strain and not one of the indogenous L.rhamnosus strains is the causative agent.

These two findings (one proved and one yetpresumptive) indicate that although probioticproducts have been safely consumed in largequantities over the years throughout Europe andJapan, occasional severe infections, especially inimmunocompromised patients, may occur.

3.1. Safety considerations of antibiotic resistancesin lactobacilli and bifidobacteria

Due to the indiscriminate use of antibiotics inhuman and veterinary medicine and in animalgrowth promoters antibiotic resistance has be-come an increasingly common characteristic inmicro-organisms (Austin et al., 1999; Robredo etal., 2000), thus causing serious problems in treat-ment of microbial infections. Antibiotic resistancein bacteria may be intrinsic or acquired. Intrinsicresistance is a naturally occurring trait and maybe considered as a species characteristic, whereasacquired resistance derives either from geneticmutations or acquisition of foreign DNA fromother bacteria.

Lactobacilli display a wide range of antibioticresistances naturally (Charteris et al., 1998b), butin most cases antibiotic resistance is not of thetransmissible type. Lactobacillus strains with non-transmissible antibiotic resistances do not usuallyform a safety concern. Several species of lacto-bacilli including L. rhamnosus and L. casei areintrinsically resistant to vancomycin (Nicas et al.,1989; Swenson et al., 1990; Charteris et al.,1998b). These species have peptidoglycan precur-sors terminating with D-lactate instead of the

M. Saarela et al. : Journal of Biotechnology 84 (2000) 197215 201

target precursor for vancomycin activity terminat-ing with D-alanine (Billot-Klein et al., 1994).Many intrinsically vancomycin resistant strains oflactobacilli have a long history of safe use asprobiotics and there is no indication that van-comycin resistant lactobacilli could transfer theresistance to other bacteria. Tynkkynen et al.(1998) have demonstrated that the vancomycinresistance factor of the probiotic strain L. rham-nosus GG is not closely related to those of entero-cocci, and they could not observe the transfer ofantibiotic resistances between L. rhamnosus GGand enterococci.

Although plasmid-linked antibiotic resistancesare not very common among lactobacilli, they dooccur (Ishiwa and Iwata, 1980; Vescovo et al.,1982; Rinckel and Savage, 1990), and their safetyimplications should be taken into consideration.Since transfer of antibiotic resistant genes mayoccur between phylogenetically distant bacteria(Courvalin, 1994), strains harbouring mobile ele-ments carrying resistance genes should not beused either as human or animal probiotics.Checking the ability of a proposed probioticstrain to act as a donor of conjugative antibioticresistance genes may be a prudent precautionespecially when probiotics are administered dur-ing antibiotic therapy, and at least in the case ofanimal feeding, where the use of antibiotics asgrowth promoters apparently creates selective ad-vantage for spreading of the resistance factors.Antibiotic susceptibility patterns vary greatly be-tween different species of lactobacilli, and strainswith atypical resistance to some clinically impor-tant antibiotics have been detected among lacto-bacilli (Charteris et al., 1998b; Felten et al., 1999),indicating the necessity for susceptibility testing ofeach probiotic strain.

Most bifidobacteria are intrinsically resistant tonalidixic acid, neomycin, polymyxin B,kanamycin, gentamycin, streptomycin andmetronidazole (Miller and Finegold, 1967; Mat-teuzzi et al., 1983; Charteris et al., 1998a). Inearlier studies vancomycin has been found highlyinhibitory against bifidobacteria (Miller and Fine-gold, 1967; Lim et al., 1993), whereas in a recentstudy by Charteris et al. (1998a) vancomycin re-sistance was suggested to be a general characteris-

tic of bifidobacteria. However, differences in thetechniques used in susceptibility testing hindersthe comparison of data. Suppression of fecalcounts of bifidobacteria during vancomycin ther-apy would suggest susceptibility of intestinalbifidobacteria to this agent (Edlund et al., 1997).Studies on the genetics of the antibiotic resistanceof bifidobacteria are warranted for understandingof the mechanisms of resistance in this genus.

4. Functional aspects of probiotics

Clinical effects of some probiotic strains inhumans are shown in Table 1. These include, forexample, immunomodulation, modulation of in-testinal flora, prevention of diarrhoeas, and lower-ing of fecal enzyme activities. Some of thefunctional aspects of probiotics are discussed inmore detail below.

4.1. Adhesion properties

Adhesion of probiotic strains to the intestinalsurface and the subsequent colonization of thehuman GI-tract has been suggested as an impor-tant prerequisite for probiotic action. Adherentstrains of probiotic bacteria are likely to persistlonger in the intestinal tract and thus have betterpossibilities of showing metabolic and im-munomodulatory effects than non-adheringstrains. Adhesion provides an interaction with themucosal surface facilitating the contact with gutassociated lymphoid tissue mediating local andsystemic immune effects. Thus, only adherent pro-biotics have been thought to effectively induceimmune effects and to stabilise the intestinal mu-cosal barrier (Salminen et al., 1996a). Adhesionmay also provide means of competitive exclusionof pathogenic bacteria from the intestinal epithe-lium: Exclusion of pathogens by lactic acid bacte-ria and bifidobacteria has been shown in vitrousing Caco-2 and HT-29-MTX cell lines (Bernetet al., 1993, 1994; Coconnier et al., 1993a,b). Inthe inhibition of pathogen adhesion in vitro bothliving and heat-killed L. acidophilus cells havebeen effective (Coconnier et al., 1993a).

M. Saarela et al. : Journal of Biotechnology 84 (2000) 197215202

Tab

le1

Clin

ical

effe

cts

ofso

me

prob

ioti

cst

rain

san

dyo

ghur

tst

rain

s

Stra

inR

efer

ence

sC

linic

alef

fect

sin

hum

ans

Low

erin

gfa

ecal

enzy

me

acti

viti

es,

redu

ctio

nof

Siit

onen

etal

.(1

990)

,G

oldi

net

al.

(199

2),

Kai

laL

acto

baci

llus

rham

nosu

sG

G(A

TC

C53

103)

anti

biot

ic-a

ssoc

iate

ddi

arrh

oea

inch

ildre

n,et

al.

(199

2),

Hos

oda

etal

.(1

994)

,Is

olau

riet

al.

trea

tmen

tan

dpr

even

tion

ofro

tavi

rus

and

acut

e(1

991,

1994

),M

ajam

aaet

al.

(199

5),

Raz

aet

al.

(199

5),

Sepp

etal

.(1

995)

,B

enne

ttet

al.

(199

6),

diar

rhoe

ain

child

ren,

trea

tmen

tof

rela

psin

gC

lost

ridi

umdi

ffici

ledi

arrh

oea,

imm

une

resp

onse

Mal

inet

al.

(199

6),

Hilt

onet

al.

(199

7),

Maj

amaa

and

Isol

auri

(199

7),

Shor

niko

vaet

al.

mod

ulat

ion,

alle

viat

ion

ofat

opic

derm

atit

issy

mpt

oms

inch

ildre

n(1

997c

),A

land

eret

al.

(199

7,19

99),

Kan

kaan

paa

etal

.(1

998)

,P

elto

etal

.(1

998)

,R

auta

nen

etal

.(1

998)

,A

rvol

aet

al.

(199

9)M

odul

atio

nof

inte

stin

alflo

ra,

imm

une

Lac

toba

cillu

sjo

hnso

nii

(aci

doph

ilus)

LJ-

1(L

a1)

Lin

k-A

mst

eret

al.

(199

4),

Schi

ffri

net

al.

(199

5),

Mar

teau

etal

.(1

997)

,M

iche

tti

etal

.(1

999)

,en

hanc

emen

t,ad

juva

ntin

Hel

icob

acte

rpy

lori

Don

net-

Hug

hes

etal

.(1

999)

trea

tmen

tB

ifido

bact

eriu

mla

ctis

Bb-

12B

lack

etal

.(1

989,

1991

),M

arte

auet

al.

(199

0),

Pre

vent

ion

oftr

avel

ler

sdi

arrh

oea,

trea

tmen

tof

Alm

etal

.(1

993)

,L

ink-

Am

ster

etal

.(1

994)

,vi

ral

diar

rhoe

ain

clud

ing

rota

viru

sdi

arrh

oea,

Saav

edra

etal

.(1

994)

,Sc

hiff

rin

etal

.(1

995)

,m

odul

atio

nof

inte

stin

alflo

ra,

impr

ovem

ent

ofco

nsti

pati

on,

mod

ulat

ion

ofim

mun

ere

spon

se,

Kan

kaan

paa

etal

.(1

998)

,F

ukus

him

aet

al.

(199

8)al

levi

atio

nof

atop

icde

rmat

itis

sym

ptom

sin

child

ren

Lac

toba

cillu

sre

uter

i(B

ioG

aia

Bio

logi

cs)

Wol

fet

al.

(199

5,19

98),

Shor

niko

vaet

al.

Shor

teni

ngof

rota

viru

sdi

arrh

oea

inch

ildre

n,tr

eatm

ent

ofac

ute

diar

rhoe

ain

child

ren,

safe

(199

7a,b

)an

dw

ell-

tole

rate

din

HIV

-pos

itiv

ead

ult

subj

ects

Aso

and

Aka

zan

(199

2),

Oka

wa

etal

.(1

993)

,M

odul

atio

nof

inte

stin

alflo

ra,

low

erin

gfa

ecal

Lac

toba

cillu

sca

sei

Shir

ota

Tan

aka

and

Ohw

aki

(199

4),

Aso

etal

.(1

995)

,en

zym

eac

tivi

ties

,po

siti

veef

fect

son

supe

rfici

alSp

anha

aket

al.

(199

8)bl

adde

rca

ncer

and

cerv

ical

canc

er,

noin

fluen

ceon

the

imm

une

syst

emof

heal

thy

subj

ects

Mod

ulat

ion

ofin

test

inal

flora

,in

crea

sein

faec

alL

acto

baci

llus

plan

taru

mD

SM98

43(2

99v)

Joha

nsso

net

al.

(199

3,19

98)

shor

t-ch

ain

fatt

yac

idco

nten

tSu

raw

icz

etal

.(1

989)

,B

uts

etal

.(1

993)

,P

reve

ntio

nof

anti

biot

ic-a

ssoc

iate

ddi

arrh

oea,

Sac

char

omyc

esbo

ular

dii

McF

arla

ndet

al.

(199

4),

Ble

ichn

eret

al.

(199

7)tr

eatm

ent

ofC

lost

ridi

umdi

ffici

leco

litis

,pr

even

tion

ofdi

arrh

oea

incr

itic

ally

illtu

be-f

edpa

tien

tsN

oef

fect

onro

tavi

rus

diar

rhoe

a,no

imm

une

Gol

din

etal

.(1

992)

,M

ajam

aaet

al.

(199

5),

Yog

hurt

stra

ins

(Str

epto

cocc

usth

erm

ophi

lus

and:

orL

.de

lbru

ecki

isu

bsp.

bulg

aric

us)

Don

net-

Hug

hes

etal

.(1

999)

enha

ncin

gef

fect

duri

ngro

tavi

rus

diar

rhoe

a,no

effe

cton

faec

alen

zym

es,

wea

kef

fect

onre

spir

ator

ybu

rst

acti

vity

ofbl

ood

leuc

kocy

tes

but

not

onov

eral

lph

agoc

ytic

acti

vity

inhe

alth

yad

ults

M. Saarela et al. : Journal of Biotechnology 84 (2000) 197215 203

Controlled comparable studies on in vitromodel systems, such as the human colon car-cinoma cell lines HT-29 and Caco-2, are impor-tant in the assessment of adhesion properties(Chauviere et al., 1992; Lehto and Salminen,1997; Tuomola and Salminen, 1998). HT-29 andCaco-2 cell lines differentiate into enterocytes andcan thus be used as a model for the small intestineepithelium. A mucus (gel covering the intestinalepithelial surface) secreting variant of the HT-29cell line, HT-29-MTX, has also been used inadhesion studies (Tuomola, 1999). In vitro adhe-sion experiments have indicated differences in theadhesion potential of different probiotic strains(Lehto and Salminen, 1997; Tuomola and Salmi-nen, 1998). Lately, glycoproteins obtained fromhuman ileostomy effluent or from faecal materialhave been used as a model for small intestinalmucus in probiotic adhesion assays (Kirjavainenet al., 1998; Tuomola et al., 1999). The mucuslayer is the first contact between ingested bacteriaand intestinal mucous membrane, and thereforeimportant additional data on probiotic adhesionproperties can be obtained by supplementing thedata obtained with cell lines without mucus-se-creting ability (Caco-2 and HT-29) with results ofmucus adhesion assays. Already it has beenshown that probiotic strains vary in their abilityto adhere to mucus, and also that strains adheringto Caco-2 cells do not necessarily adhere to mucusas efficiently (Tuomola et al., 1999). Furthermore,it has been shown that bifidobacteria adhere dif-ferently to mucus isolated from subjects represent-ing different age groups (being poorest to mucusisolated from elderly) (Ouwehand et al., 1999).

In vivo human adhesion of probiotic strainscan be studied by obtaining biopsy material froma subject after a period of probiotic consumption.Since biopsy sampling raises ethical issues, sam-pling is usually limited to subjects going throughroutine diagnostic colonoscopy. From these sub-jects tissue samples can be obtained, not onlyfrom the rectal-sigmoidal region, but also fromother parts of large intestine (ascending, trans-verse, and descending colon). The preferential ad-hesion of a commercial probiotic strain (L.rhamnosus GG) to the descending part of largecolon was detected by using biopsy material

(Alander et al., 1997). Johansson et al. (1993)have also demonstrated the adhesion of differentLactobacillus strains to rectal mucosal biopsysamples obtained from volunteers who had con-sumed fermented oatmeal soup. Biopsy samplingprobably gives the most accurate information onthe adhesion ability of probiotic strains. However,there are severe limitations in the technique: firstand most importantly, ethical considerations limitthe use of the technique. Secondly, the techniqueis very laborious and therefore only few individu-als can be included in trials. Thirdly, the evacua-tion of the colon prior to colonoscopy probablyleads to a loss of large number of adhering bacte-ria, leaving only the bacteria with the strongestadhering ability attached.

Although a lot of research effort has beenexpended on probiotic adhesion studies, the roleof adhesion in successful probiotic function re-mains speculative. It could also be argued thatstrong adhesion ability may increase the risk ofinfection in the host. Some probiotic strains arepoorly adhering in vitro and:or in vivo and stillthey can show positive effects in the hosts. Inaddition, the reproducibility of in vitro adhesionstudies can be poor (especially between differentlaboratories), which also complicates the interpre-tation of the results.

4.2. Immunomodulatory properties

Gut associated lymphoid tissue may have con-tact with adhesive probiotic strains and their com-ponents and therefore adhesion is one way ofprovoking immune effects. Human studies haveshown that probiotic bacteria can have positiveeffects on the immune system of their host. How-ever, differences between probiotic bacteria in re-spect to their immunomodulatory effects havebeen observed (Isolauri et al., 1999). In two sepa-rate trials, L. johnsonii LJ-1 (previously L.acidophilus LA1) and L. sali6arius UCC 118 stim-ulated a mucosal IgA response and increasedphagocytic activity. The immunomodulation me-diated by these strains was not linked to aninflammatory response or general modification ofimmune responsiveness that could potentiallyhave harmful effects, but was rather associated

M. Saarela et al. : Journal of Biotechnology 84 (2000) 197215204

with transient alterations beneficial to the con-sumer (Schiffrin et al., 1997; Mattila-Sandholmand Kauppila, 1998). L. rhamnosus GG andBifidobacterium lactis Bb-12 derived extracts havebeen shown to suppress lymphocyte proliferationin vitro (Mattila-Sandholm and Kauppila, 1998).Further evidence for immunomodulation by thesetwo strains was provided by a trial involvingchildren with severe atopic eczema resulting fromfood allergy. Children fed L. rhamnosus GG andB. lactis Bb-12 showed a significant improvementin clinical symptoms compared to the placebogroup (Mattila-Sandholm and Kauppila, 1998;Alander and Mattila-Sandholm, 2000). L. rham-nosus GG has been shown to down-regulate themilk-induced inflammatory response in milk-hy-persensitive subjects, whereas it had an immunos-timulatory effect in healthy subjects (Pelto et al.,1998). Furthermore, L. rhamnosus GG was shownto promote IgA immune response in Crohns dis-ease patients (Malin et al., 1996) and enhancedthe circulating antibody secreting cell response inchildren with rotavirus diarrhoea (Kaila et al.,1992).

4.3. Antagonistic properties

To have an impact on the colonic flora it isimportant for probiotic strains to show antago-nism against pathogenic bacteria via antimicrobialsubstance production or competitive exclusion.Enormous research efforts have focused on bacte-riocin research. Although probiotic strains mayproduce bacteriocins, their role in the pathogeninhibition in vivo can only be limited, since tradi-tional bacteriocins have an inhibitory effect onlyagainst closely related species such as other Lac-tobacillus or on sporefomers such as Bacillus orClostridium (Holzapfel et al., 1995). However, lowmolecular weight metabolites (such as hydrogenperoxide, lactic and acetic acid, and other aromacompounds) and secondary metabolites may bemore important since they show wide inhibitoryspectrum against many harmful organism likeSalmonella, Escherichia coli, Clostridium, and He-licobacter (Skytta et al., 1992; Helander et al.,1997; Niku-Paavola et al., 1999). L. rhamnosusstrain GG produces in vitro low molecular weight

antimicrobial(s), possibly short chain fatty acid(s)but distinct from lactic and acetic acid, with in-hibitory activity against anaerobes such asClostridium, Bacteroides and Bifidobacterium,against Enterobacteriaceae, Pseudomonas, Staphy-lococcus and Streptococcus, but not against otherlactobacilli (Silva et al., 1987). The antagonisticactivity of L. rhamnosus GG against enteropatho-genic bacteria has also been shown in vivo in S.typhimurium infected mice (Hudault et al., 1997).The spent culture supernatant (SCS) of L.acidophilus strain LB decreased the viability ofStaphylococcus aureus, Listeria monocytogenes, S.typhimurium, Shigella flexneri, Escherichia coli,Klebsiella pneumoniae, Bacillus cereus, Pseu-domonas aeruginosa, and Enterobacter spp. invitro. The unidentified low molecular weight an-timicrobial substance(s) was independent of lacticacid production and did not effect Lactobacillusor Bifidobacterium strains tested. The antibacterialactivity of L. acidophilus LB SCS towards S.typhimurium was also maintained in vivo in theinfected-mouse model (Coconnier et al., 1997). L.acidophilus ( johnsonii ) strain LA1 (LJ-1) pro-duces nonbacteriocin antibacterial substance(s)(unidentified but distinct from lactic acid) thatinhibits in vitro a wide range of gram-negativeand gram-positive pathogens, such as S. aureus,L. monocytogenes, S. typhimurium, S. flexneri, K.pneumoniae, P. aeruginosa and Enterobacter cloa-cae. However, inhibition of lactobacilli andbifidobacteria could not be detected. Inhibitoryactivity of the strain LA1 towards S. typhimuriumwas also shown in vivo in mouse model (Bernet-Camard et al., 1997). Furthermore, L. acidophilusLA1 shows antimicrobial effect against H. pylori,both in vitro and in humans (Michetti et al.,1999).

4.4. Antimutagenic and anticarcinogenic properties

Antimutagenic and anticarcinogenic propertiesof bacteria ingested in foods and:or representingthe microflora of gastrointestinal tract have beenwidely studies for a number of years. Bacterialanticarcinogenic properties are considered to rep-resent one or more of the following types: bindingand degradation of (pro)carcinogens, production

M. Saarela et al. : Journal of Biotechnology 84 (2000) 197215 205

of antimutagenic compounds, modulation of pro-carcinogenic enzymes in the gut, and suppressionof tumours by an immune response mechanism(Fernandes et al., 1987; Lidbeck et al., 1992a;Hirayama and Rafter, 1999). Special emphasis hasbeen put on in vitro antimutagenic (especiallymutagen binding) and anticarcinogenic propertiesof lactic acid bacterial strains (representing Lacto-bacillus, Lactococcus, Leuconostoc, Enterococcus,and Pediococcus species) isolated from fermenteddairy and non-dairy products (Hosono et al.,1986, 1990a,b; Fernandes and Shahani, 1990;Zhang et al., 1990; Thyagaraja and Hosono, 1993,1994; El-Nezami et al., 1998a,b). However, ac-cording to the current criteria, only few of thesestrains can be considered to represent probiotics.In addition, antimutagenic property is by nomeans typical for lactic acid bacteria (or probi-otics) only: This property seems to be distributedamong bacteria representing both gram-positiveand gram-negative species (for a review see Fer-nandes and Shahani, 1990). A wide range ofbacteria (and also yeasts) are capable of mutagenbinding in vitro (Morotomi and Mutai, 1986;Zhang and Ohta, 1993; Orrhage et al., 1994).Furthermore, also non-viable bacterial cells areable to bind mutagens and carcinogens (Zhang etal., 1990; Thyagaraja and Hosono, 1994; El-Nezami et al., 1998a). In addition to in vitrostudies antigenotoxic and antitumor activities ofsome Bifidobacterium and Lactobacillus strainshas also been shown using rat and mice models(Reddy and Rivenson, 1993; Pool-Zobel et al.,1993, 1996; Singh et al., 1997; Takagi et al., 1999).

In human trials probiotic strains have beenassociated with the reduction of fecal mutagenic-ity or fecal enzymatic activities involved in muta-gen or carcinogen activation. Faecal enzymes suchas nitroreductase and b-glucuronidase have theability to convert procarcinogens to carcinogensin the colon. In human studies L. acidophilusNCFB 1748 consumption was shown to decreasefecal and urinary mutagenicity (Lidbeck et al.,1992b). Suppression of urinary mutagenicity hasalso been shown after the consumption of L. caseistrain Shirota (Hayatsu and Hayatsu, 1993). L.rhamnosus GG supplementation decreased fecalb-glucuronidase, nitroreductase and glycocholic

acid hydrolase activities in healthy females (Linget al., 1994). Reduction of faecal enzyme activitieshas also been shown after L. gasseri strain ADHand L. casei strain Shirota consumption in hu-mans (Pedrosa et al., 1995; Spanhaak et al., 1998).L. casei strain Shirota consumption has furtherproved beneficial for some cancer patients byreducing the recurrence of superficial bladder can-cer and by prolonging survival and relapse-freeinterval in cervical cancer (Aso and Akazan, 1992;Aso et al., 1995; Okawa et al., 1993). However,although there is evidence that probiotic bacteriamay show antimutagenic and anticarcinogenicproperties in vitro and in animal models, thepossible role of probiotics in the cancer preven-tion in humans still remains highly controversial.

5. Technological aspects of probiotics

Functional foods with probiotics are now wellestablished on the European market. Startingabout 20 years ago this product range has in-creased (Young, 1998) and is presently known tomost consumers. To succeed in promoting theconsumption of functional probiotic products thefood industry has to satisfy the demands of theconsumer. All probiotic foods should be safe andhave good sensory properties. The probiotic foodsshould also include specific probiotic strains at asuitable level during the storage time. By examin-ing existing products it has been suggested thatthis is not always the case (Hamilton-Miller et al.,1999).

Before probiotic strains can be delivered toconsumers, they must first be able to be manufac-tured under industrial conditions, and then sur-vive and retain their functionality during storageas frozen or freeze dried cultures, and also in thefood products into which they are finally formu-lated. Additionally, they must be able to be incor-porated into foods without producing off-flavoursor textures. The packaging materials used and thestorage conditions under which the products arestored, are important for the quality of productscontaining probiotic bacteria.

Foods used for dissemination of probiotics areusually fermented foods even if probiotics also

M. Saarela et al. : Journal of Biotechnology 84 (2000) 197215206

could be present in infant formula, fruit drinks,whey drinks and sweet milk. Fermented milk andcheese are the most common foods with probi-otics (Svensson, 1999). Fermented foods are pro-duced by a microbial fermentation in whichfermentable carbohydrates are transformed intoethanol and:or organic acids mainly acetic, lacticand propionic acid. Yeast and lactic acid bacteriaare the microbes commonly used in food fermen-tation. In fermented dairy products mainly Lacto-bacillus and Lactococcus species, but also yeastand propionic acid bacteria, are used (Driessenand Loones, 1992).

5.1. Manufacturing processes for fermentedprobiotic products

Most leading starter culture manufacturers to-day produce common GI lactic acid bacteria andbifidobacteria commercially (Spork, 1994;Mogensen and Friis, 1997). Commercially avail-able probiotic cultures may consist of a singlestrain or a mixture of several strains. In mostcases the probiotic properties are affected by theway of which the strain or culture has been pro-duced (for a review see German et al., 1999).Therefore specific information on strain-specificproperties should be available for the processoptimisation. Probiotic cultures may be used inspecial formulations like capsules or tablets, orthey may be used in the production of a largevariety of fermented food products. In some casesthe cultures may be added to a food to contributespecific probiotic or functional properties.

Most commercial probiotic culture prepara-tions are supplied in highly concentrated form,and most of them are constructed for DVS (directvat set) application (Honer, 1995). Use of thesehighly concentrated DVS cultures is common dueto the difficulties involved in propagating probi-otic micro-organisms at the production site. TheDVS cultures are supplied either as highly concen-trated frozen cultures or as freeze-dried cultures.The cultures should be filled in gas and light proofcontainers to protect the cultures against light andhumidity. Most often alu-foil coated cartons orpouches are used. As the cultures are sensitive, itis important to handle and store the cultures

according to the manufacturers instructions.Usually deep-frozen cultures contain more than1010 cfu g1, whereas freeze-dried cultures typi-cally contain more than 1011 cfu g1 (Obermanand Libudzisz, 1998). The cell concentration pergram of product varies with the culture and thetype of organisms used.

In fermented probiotic products it is importantthat the probiotic culture used contributes togood sensory properties. Therefore it is quitecommon to use probiotic bacteria mixed togetherwith other types of bacteria suited for the fermen-tation of the specific product. For milk-basedproducts the probiotic strains are often mixedwith S. thermophilus and L. delbrueckii to achievethe desired flavour and texture. The main flavourcomponents of species often used in probioticformulations are as shown in the Table 2. Inmany cases the consumers find products fer-mented with L. delbrueckii too acidic and with tooheavy acetaldehyde flavour (yoghurt flavour).Therefore probiotic cultures have been developedto bring out the preferred flavours in the productsin which they are used. Examples of such culturesare the so called ABT cultures (ABT standing forL. acidophilus, Bifidobacterium and S. ther-mophilus). The manufacturing technology for pro-ducing fermented milks containingBifidobacterium strains has been extensively re-viewed by Tamime et al. (1995).

In selecting starter micro-organisms reliableacid-forming ability is the most important charac-teristics. However, when selecting probiotics the

Table 2Main flavour characteristics of some strains commonly used inprobiotic mixtures

Culture Main flavourcomponents

Lactobacillus acidophilusa Lactate (DL)Lactate (L), acetateBifidobacterium spp.a

Lactate (L),Streptococcus thermophilusb

acetaldehyde, diacetylLactobacillus delbrueckii subsp. Lactate (D),

acetaldehyde, diacetylbulgaricusb

a Probiotic strains.b Non-probiotic strains.

M. Saarela et al. : Journal of Biotechnology 84 (2000) 197215 207

criteria should be connected to the impact onhuman health and well-being. As the environmentwithin the GI-tract and within the food might bequite different the probiotic is often not suitableas a starter organism (Oberman and Libudzisz,1998; German et al., 1999). The growth rate mightbe too slow and they might give off-flavours(Svensson, 1999). This could partly be overcomeby using specific aseptic processes as used in pro-ducing fermented acidophilus milk with levels ofprobiotic reaching 109 cfu g1 (Fonden, 1989).Another possibility is to improve the suitability ofthe food as a substrate for the probiotic by addingenergy sources (e.g. glucose), growth factors (e.g.yeast extract and protein hydrolysates) or suitableantioxidants, minerals or vitamins (Kurmann,1988; Ishibashi and Shimamura, 1993; Dave andShah, 1998; Gomes et al., 1998). However, even ifsuch an adjustment may improve the performanceof the probiotic as a starter, it is often notenough. By use of a help starter in addition to aprobiotic preparation this problem can usually besolved (Fonden et al., 2000).

When producing probiotic-containing fer-mented products a fermentation temperature of3740C is usually recommended since these tem-peratures span the range in which most probioticstrains multiply well. Probiotic strains can be anintegral part of the starter and grow in a symbi-otic relation with the other strains composing theculture (Saxelin et al., 2000). There are exampleswhere the probiotic strain or strains are added tothe fermented milk after fermentation (e.g. theDanish product Cultura). In special milk prod-ucts (e.g. sweet products) probiotics are added tothe product in a such a way that they will retaintheir viability but are prevented from multiplyingby cooling the product.

5.1.1. Probiotic interaction with starter bacteriaThe interactions between probiotic and starter

might have an impact on the quality of theproduct. It has been shown that it is possible toproduce fermented dairy products with excellentsensory properties and good survival of the bacte-ria by using starter and probiotic organisms to-gether (Fonden et al., 2000). Even if severalnegative interactions have been proposed it seems

to be possible to find a suitable starter for mostprobiotics by testing the presently availablestarters. Suitable starters might be S. ther-mophilus, yoghurt cultures and mesophilic starterswith different combinations of Lactococcusstrains. The most suitable combination of starterand a specific probiotic bacteria has to be deter-mined by a screening process evaluating the im-pact of different starters on the sensory propertiesand on the survival of the probiotic strain(Ishibashi and Shimamura, 1993; Samona et al.,1996).

In this screening process some general princi-ples should be followed. If possible, the probioticshould be able to grow during the fermentation. Itwill increase the total number of the probioticresulting in a lower process cost and increasedadaptation of the probiotic to the fermented food.As most probiotics grow well at 37C a ther-mophilic starter might be preferable to amesophilic one (Svensson, 1999). The growth rateof the starter should be moderate allowing somegrowth of the probiotic bacteria. It is also impor-tant to add the probiotic before or at the sametime as the starter (Reddy, 1989; Kailasapathyand Rybka, 1997). Addition of the probiotic afterfermentation does not allow any growth but in-stead might result to a reduced viability as shownwhen L. acidophilus was mixed with yoghurt (Hullet al., 1984). The starter might improve thegrowth conditions of probiotic by producing sub-stances favourable to the growth of the probioticor by reducing the oxygen pressure. Bifidobacteriaare quite sensitive to oxygen but by selection ofspecific strains of S. thermophilus it was possibleto increase survival of B. longum (Ishibashi andShimamura, 1993). L. delbrueckii subsp. bulgari-cus strains might also increase growth ofbifidobacteria by their proteolytic activity result-ing in increased availability of valine, glycine andhistidine (Misra and Kulia, 1994).

In selecting a suitable starter the negative im-pact on probiotic survival in vitro and in vivoshould also be taken in consideration. The sur-vival of the probiotic bacteria might be influencedby the metabolites formed by the starter such aslactic acid, hydrogen peroxide and bacteriocins.L. delbrueckii subsp. bulgaricus, a part of an

M. Saarela et al. : Journal of Biotechnology 84 (2000) 197215208

Fig. 3. Changes in CFU (10log) of seven probiotic strains(AG) in fermented milk products after 14 days storage at4C. Milks were fermented either with the probiotic strainonly (None), or together with a supporter culture St 20 (S.thermophilus), YC 380 (S. thermophilus and L. delbrueckiisubsp. bulgaricus) or YC 280 (S. thermophilus and L. del-brueckii subsp. bulgaricus). The figure shows the strain tostrain variation between probiotics in regard to their survivalin the presence of different supporter cultures.

was ten times higher with the sweet milk (Pet-tersson et al., 1983).

5.1.2. Interactions between probiotics andprebiotics

Functional foods with both probiotics and pre-biotics are called synbiotics (Roberfroid, 1998).Most prebiotics as known today are fermentable,non-digestible carbohydrates with different num-ber of sugar moieties from two up to severalhundreds. Some examples are lactulose, galacto-and fructooligosaccharides and resistant starch.As the concept of synbiotics is quite new there arenot many specific studies of interactions betweenpro- and prebiotics. Due to their general proper-ties prebiotics might influence the growth andsurvival of the probiotic by influencing the growthand metabolites of both the probiotic and thestarter. This has to be kept in mind while consid-ering interactions between probiotics and starters.

Interaction in vivo might be favoured by anadaptation of the probiotic to the prebiotic. Byadapting its metabolism to a substrate givensimultaneously with the probiotic might result ina competitive advantage for the probiotic. How-ever, as far as we know most studies have beendone with probiotic grown without such anadaptation.

5.2. Manufacturing of non-dairy probiotic foods

Application of probiotic cultures in non-dairyproducts and environments represents a challenge(Andersen, 1998). Probiotic viability in the foodmatrix depends on factors such as pH, storagetemperature, presence of competing micro-organ-isms and inhibitors (e.g. NaCl). In products likeprobiotic-containing baby foods or confectionery(e.g. chocolate) it is important that the formula-tion maintains the activity and viability of theprobiotic for extended periods of time. Since theprobiotic cultures are added as additives to thesekind of products, they do not usually multiply,which sets great demands for the probiotic stabil-ity. Factors like water activity, oxygen tensionand temperature becomes increasingly importantwhen dealing with these kinds of products. Stor-age at room temperature, which is common for

ordinary yoghurt culture, produces high levels ofD-lactic acid also when the product is stored afterfermentation at low temperature. The impact ofD-lactate on the probiotic varies from strain tostrain (Dave and Shah, 1997; Kailasapathy andRybka, 1997; Rybka and Fleet, 1997). Samona etal. (1996) showed that bifidobacterial strainscould not grow in the presence of youghurt cul-tures but by choosing the right combination ofprobiotic and starter strains the decline inbifidobacterial counts could be prevented. Starterswith only S. thermophilus might function bettertogether with probiotics sensitive to acids and lowpH. They are forming less acid both during fer-mentation and storage (Okongi et al., 1984;Ishibashi and Shimamura, 1993) (Fig. 3).

The impact of the metabolites formed by thestarter might be enhanced in vivo. When thefunctional food reaches the stomach the pH de-creases and the level of undissociated lactic acidincreases. As the concentration of the unionisedacid has the greatest impact on survival of theprobiotic strain, the viability might be influencedby the amount of hydrochloric acid from thegastric juice and the amount of lactic acid fromthe product. The importance of this is indicatedby the difference in survival of L. acidophilusNCFB 1748 through the GI-tract when compar-ing sweet milk with fermented milk. Survival rate

M. Saarela et al. : Journal of Biotechnology 84 (2000) 197215 209

several types of non-dairy products such as cerealproducts, drinks, chocolate etc. can create anoverwhelming challenge for probiotic stability.This problem can sometimes be solved by usingprobiotic encapsulation technology to ensure theviability and stability of probiotic cultures (Myl-larinen et al., 1998). Stable probiotic-containingbaby food formulations and confectioneries havebeen developed and are currently on the market(Langhendries et al., 1995; Fukushima et al.,1997).

6. Future trends

Diet is a major focus of public health strategyaimed at maintaining optimum health throughoutlife, preventing early onset of chronic diseasessuch as gastrointestinal disorders, cardiovasculardisease, cancer, osteoporosis, as well as promotinghealthier ageing. Although the highly complexrelationship of food and health is still poorlyunderstood, recent research advances in differentdisciplines provide promising new approaches toimprove our understanding. The growing demandfor healthy foods is stimulating innovation andnew product development in the food industryinternationally. The food industry has a centralrole in facilitating healthier eating practicesthrough the provision and promotion of healthyfoods.

Continuously increasing consumer health con-sciousness and expenditure are socio-economicfactors responsible for the expanding Europeanand world-wide interest in functional foods. Con-siderable confusion and scepticism, however, ex-ists between consumers, consumer organisations,scientific communities and media about the claimsassociated with probiotic products. Recent EUprojects have demonstrated that, with co-ordi-nated efforts towards communication and a scien-tific approach to selecting and applyingprobiotics, functional food products can be devel-oped with measurable health benefits for con-sumers. Probiotic strains can be successfullymanufactured and incorporated into highly ac-ceptable food products where they can retain theirviability and functionality. There are many strain

to strain variations, not only in their technologicalproperties, but also in their effects on humanhealth (Alander and Mattila-Sandholm, 2000).

The probiotic concept is today widely spread inthe scientific and industrial fields. However, fur-ther scientific input is required. Important targetresearch areas, including GI-tract diagnostics andimmunology, methodology, biomarkers, andfunctionality, will lead to tools and scientificallysound methods for well-designed informative hu-man studies. Controlled human studies are essen-tial for the socio-economic success of probioticfunctional foods, and they should be tailored forspecific population groups such as the elderly andbabies. Future research on probiotic bacteria willcentre on selecting new and more specific strainsfor the well-being of the host (age groups, healthypopulations, disease specific).

The future scientific and technological researchtrends will be: to inter-link the expertise and scientific knowl-

edge on food, GI-tract functionality and hu-man health.

to study the mechanisms of action of probioticsin the GI-tract, and develop diagnostic toolsand biomarkers for their assessment.

to evaluate the role of probiotics in healthyconsumer groups.

to examine the effects of probiotics on GI-dis-eases, GI-infections, and allergies.

to address the consumer aspects and trade-offs. to ensure the stability and viability of probiotic

products by developing technologies (e.g.bioencapsulation)

to develop technology for non-dairy probioticapplications (i.e. cereal based materials).

Acknowledgements

This work was supported by European Con-certed Action PL98 4230.

References

Adams, M.R., 1999. Safety of industrial lactic acid bacteria. J.Biotechnol. 68, 171178.

M. Saarela et al. : Journal of Biotechnology 84 (2000) 197215210

Alander, M., Mattila-Sandholm, T. (Eds.), 2000. Functionalfoods for EU-Health in 2000, 4th Workshop, FAIR CT96-1028, PROBDEMO, VTT Symposium 198, Rovaniemi,Finland.

Alander, M., Korpela, R., Saxelin, M., Vilpponen-Salmela, T.,Mattila-Sandholm, T., von Wright, A., 1997. Recovery ofLactobacillus rhamnosus GG from human colonic biopsies.Lett. Appl. Microbiol. 24, 361364.

Alander, M., Satokari, R., Korpela, R., Saxelin, M., Vilp-ponen-Salmela, T., Mattila-Sandholm, T., von Wright, A.,1999. Persistence of colonization of human colonic mucosaby a probiotic strain, Lactobacillus rhamnosus GG, afteroral consumption. Appl. Environ. Microbiol. 65, 351354.

Alm, L., Ryd-Kjellen, E., Setterberg, G., Blomquist, L., 1993.Effect of a new fermented milk product Cultura onconstipation in geriatric patients. 1st Lactic Acid BacteriaComputer Conference.

Andersen, L., 1998. Pre- and probiotics sausage science.Functional Foods June, 2629.

Arvola, T., Laiho, K., Torkkeli, S., Mykkanen, H., Salminen,S., Maunula, L., Isolauri, E., 1999. Prophylactic Lacto-bacillus GG reduces antibiotic-associated diarrhea in chil-dren with respiratory infections: a randomized study.Pediatrics 104, e64.

Aso, Y., Akazan, H., 1992. Prophylactic effect of a Lactobacil-lus casei preparation on the recurrence of superficial blad-der cancer. Urol. Int. 49, 125129.

Aso, Y., Akazan, H., Kotake, T., Tsukamoto, T., Imai, K.,Naito, S., 1995. Preventive effect of a Lactobacillus caseipreparation on the recurrence of superficial bladder cancerin a double-blind trial. Eur. Urol. 27, 104109.

Austin, D.J., Kristinsson, K.G., Anderson, R.M., 1999. Therelationship between the volume of antimicrobial con-sumption in human communities and the frequency ofresistance. PNAS 96, 11521156.

Bennett, R.G., Gorbach, S.L., Goldin, B.R., Chang, T.-W.,Laughon, B.E., Greenough, W.B., Bartlett, J.G., 1996.Treatment of relapsing Clostridium difficile diarrhea withLactobacillus GG. Nutr. Today 31 (Suppl.), 35S38S.

Bernet, M.-F., Brassart, D., Neeser, J.-R., Servin, A.L., 1993.Adhesion of human bifidobacterial strains to cultured hu-man intestinal epithelial cells and inhibition of en-teropathogencell interactions. Appl. Environ. Microbiol.59, 41214128.

Bernet, M.-F., Brassart, D., Neeser, J.-R., Servin, A.L., 1994.Lactobacillus acidophilus LA 1 binds to cultured humanintestinal cell lines and inhibits cell attachment and cellinvasion by enterovirulent bacteria. Gut 35, 483489.

Bernet-Camard, M.-F., Lievin, V., Brassart, D., Neeser, J.-R.,Servin, A.L., Hudault, S., 1997. The human Lactobacillusacidophilus strain LA1 secretes a nonbacteriocin antibacte-rial substance(s) active in vitro and in vivo. Appl. Environ.Microbiol. 63, 27472753.

Billot-Klein, D., Gutmann, L., Sable, S., Guittet, E., vanHeijenoort, J., 1994. Modification of peptidoglucan precur-sors is a common feature of the low-level vancomycin-re-sistant VANB-type Enterococcus D366 and of the naturally

glycopeptide-resistant species Lactobacillus casei, Pediococ-cus pentosaceus, Leuconostoc mesenteroides, and Enterococ-cus gallinarum. J. Bacteriol. 176, 23982405.

Black, F., Einarsson, K., Lidbeck, A., Orrhage, K., Nord,C.E., 1991. Effect of lactic acid producing bacteria on thehuman intestinal microflora during ampicillin treatment.Scand. J. Infect. Dis. 23, 247254.

Black, F.T., Andersen, P.L., Orskov, J., Orskov, F., Gaarslev,K., Laulund, S., 1989. Prophylactic efficacy of lactobacillion travellers diarrhoea. Travel Med. 333335.

Bleichner, G., Blehaut, H., Mentec, H., Moyse, D., 1997.Saccharomyces boulardii prevents diarrhea in critically illtube-fed patients. A multicenter, randomized, double-blindplacebo-controlled trial. Intensive Care Med. 23, 517523.

Buts, J.P., Corthier, G., Delmee, M., 1993. Saccharomycesboulardii for Clostridium difficile-associated enteropathiesin infants. J. Pediatr. Gastroenterol. Nutr. 16, 419425.

Charteris, W.P., Kelly, P.M., Morelli, L., Collins, J.K., 1998a.Antibiotic susceptibility of potentially probiotic Bifidobac-terium isolates from the human gastrointestinal tract. Lett.Appl. Microbiol. 26, 333337.

Charteris, W.P., Kelly, P.M., Morelli, L., Collins, J.K., 1998b.Antibiotic susceptibility of potentially probiotic Lacto-bacillus species. J. Food Prot. 61, 16361643.

Chauviere, G., Cocoinnier, M.-H., Kerneis, S., Fourniat, J.,Servin, A.L., 1992. Adhesion of human Lactobacillusacidophilus strain LB to human enterocyte-like Caco-2cells. J. Gen. Microbiol. 138, 16891696.

Coconnier, M.-H., Bernet, M.-F., Chauviere, G., Servin, A.L.,1993a. Adhering heat-killed human Lactobacillusacidophilus, strain LB, inhibits the process of pathogenicityof diarrhoeagenic bacteria in cultured human intestinalcells. J. Diarrh. Dis. 11, 235242.

Coconnier, M.-H., Bernet, M.-F., Kerneis, S., Chauviere, G.,Fourniat, J., Servin, A.L., 1993b. Inhibition of adhesion ofenteroinvasive pathogens to human intestinal Caco-2 cellsby Lactobacillus acidophilus strain LB decreases bacterialinvasion. FEMS Microbiol. Lett. 110, 299306.

Coconnier, M.-H., Lievin, V., Bernet-Camard, M.-F., Hu-dault, S., Servin, A.L., 1997. Antibacterial effect of theadhering human Lactobacillus acidophilus strain LB. An-timicrob. Agents Chemother. 41, 10461052.

Courvalin, P., 1994. Transfer of antibiotic resistance genesbetween gram-positive and gram-negative bacteria. Antimi-crob. Agents Chemother. 38, 14471451.

Daly, C., Davis, R., 1998. The biotechnology of lactic acidbacteria with emphasis on applications in food safety andhuman health. Agric. Food Sci. Finland 7, 219250.

Dave, R., Shah, N., 1997. Viability of yoghurt and bacteria inyoghurts made from commercial starter cultures. Int.Dairy J. 7, 3141.

Dave, R.I., Shah, N.P., 1998. Ingredient supplementation ef-fects on viability of probiotic bacteria in yougurt. J. DairySci. 81, 28042816.

Donnet-Hughes, A., Rochat, F., Serrant, P., Aeschlimann,J.M., Schiffrin, J.E., 1999. Modulation of nonspecificmechanisms of defense by lactic acid bacteria: effectivedose. J. Dairy Sci. 82, 863869.

M. Saarela et al. : Journal of Biotechnology 84 (2000) 197215 211

Donohue, D.C., Salminen, S.J., 1996. Safety of probioticbacteria. Asia Pacific J. Clin. Nutr. 5, 2528.

Driessen, F.M., Loones, A., 1992. Developments in the fer-mentation process. In: New Technologies for FermentedMilks, IDF Bullentin 277. IDF, Brussels, pp. 2840.

Edlund, C., Barkholt, L., Olsson-Liljequist, B., Nord, C.E.,1997. Effect of vancomycin on intestinal flora of patientswho previously received antimicrobial therapy. Clin. In-fect. Dis. 25, 729732.

El-Nezami, H., Kankaanpaa, P., Salminen, S., Ahokas, J.,1998a. Ability of dairy strains of lactic acid bacteria tobind a common food carcinogen, aflatoxin B1. FoodChem. Toxicol. 36, 321326.

El-Nezami, H., Kankaanpaa, P., Salminen, S., Ahokas, J.,1998b. Physicochemical alterations enhance the ability ofdairy strains of lactic acid bacteria to remove aflatoxinfrom contaminated media. J. Food Prot. 61, 466468.

Felten, A., Barreau, C., Bixet, C., Lagrange, H., Philippon, A.,1999. Lactobacillus species identification, H2O2 production,and antibiotic resistance and correlation with human clini-cal status. J. Clin. Microbiol. 37, 729733.

Fernandes, C.F., Shahani, K.M., 1990. Anticarcinogenic andimmunological properties of dietary lactobacilli. J. FoodProt. 53, 704710.

Fernandes, C.F., Shahani, K.M., Amer, M.A., 1987. Thera-peutic role of dietary lactobacilli and lactobacillic fer-mented dairy products. FEMS Microbiol. Rev. 46,343356.

Fonden, R., 1989. Development of new fermented milk prod-ucts. Food Technol. Int. Eur. 158160.

Fonden, R., Grenov, B., Reniero, R., Saxelin, M., Birkeland,S.-E., 2000. Industrial panel statements: technological as-pect. In: Alander, M., Mattila-Sandholm, T. (Eds.), Func-tional Foods for EU-Health in 2000, 4th Workshop, FAIRCT96-1028, PROBDEMO, VTT Symposium 198.Rovaniemi, Finland, pp. 4350.

Fukushima, Y., Kawata, Y., Hara, H., Terada, A., Mitsuoka,T., 1998. Effect of a probiotic formula on intestinal im-munoglobulin A production in healthy children. Int. J.Food Microbiol. 42, 3944.

Fukushima, Y., Shou-Tou, L., Hara, H., Terada, A., Mit-suoka, T., 1997. Effect of follow-up formula containingbifidobacteria (NAN BF) on fecal flora and fecal metabo-lites in healthy children. Biosci. Microflora 6, 6572.

Fuller, R., 1989. Probiotics in man and animals. J. Appl. Bact.66, 365378.

German, B., Schiffrin, E.J., Reniero, R., Mollet, B., Pfeifer,A., Neeser, J.-R., 1999. The development of functionalfoods: lessons from the gut. TIBTECH 17, 492499.

Goldin, B., Gorbach, S.L., Saxelin, M., Barakat, S., Gualtieri,L., Salminen, S., 1992. Survival of Lactobacillus species(strain GG) in human gastrointestinal tract. Dig. Dis. Sci.37, 121128.

Gomes, A.M., Malcata, F.X., Klaver, F.A., 1998. Growthenhancement of Bifidobacterium lactis Bo and Lactobacillusacidophilus Ki by milk hydrolyzates. J. Dairy Sci. 81,28172825.

Haenel, H., Bendig, J., 1975. Intestinal flora in health anddisease. Progr. Food Nutr. Sci. 1, 2164.

Hamilton-Miller, J., Shah, S., Winkler, J., 1999. Public healthissues arising from microbiological and labelling quality offoods and supplements containing probiotic microrgan-isms. Publ. Hlth. Nutr. 2, 223229.

Hayatsu, H., Hayatsu, T., 1993. Suppressing effect of Lacto-bacillus casei administration on the urinary mutagenicityarising from ingestion of fried ground beef in human.Cancer Lett. 73, 173179.

Helander, I., von Wright, A., Mattila-Sandholm, T., 1997.Potential of lactic acid bacteria and novel antimicrobialsagainst gram-negative bacteria. Trends Food Sci. Technol.8, 146150.

Hilton, E., Kolakowski, P., Singer, C., Smith, M., 1997.Efficacy of Lactobacillus GG as a diarrhea preventative. J.Travel Med. 4, 4143.

Hirayama, K., Rafter, J., 1999. The role of lactic acid bacteriain colon cancer prevention: mechanistic considerations.Antonie van Leeuwenhoek 76, 391394.

Holzapfel, W.H., Geisen, R., Schillinger, G.U., 1995. Biologi-cal preservation of foods with reference to protective cul-tures, bacteriocins and food-grade enzymes. Int. J. FoodMicrobiol. 24, 343362.

Honer, C., 1995. Culture shift. Dairy Field 178, 5458.Hosoda, M., He, F., Hiramatu, M., Hashimoto, H., Benno,

Y., 1994. Effects of Lactobacillus GG intake on fecalmicroflora and defecation in healthy volunteers. Bifidus 8,2128.

Hosono, A, Kashina, T., Kada, T., 1986. Antimutagenic prop-erties of lactic acid cultured milk on chemical and fecalmutagens. J. Dairy Sci. 69, 22372242.

Hosono, A., Tanabe, T., Otani, H., 1990a. Binding propertiesof lactic acid bacteria isolated from kefir milk with muta-genic amino acid pyrolyzates. Milchwissenschaft 45, 647651.

Hosono, A., Wardojo, R., Otani, H., 1990b. Binding of aminoacid pyrolyzates by lactic acid bacterial isolated fromDahih. Lebensm. Wiss. Technol. 23, 149153.

Hudault, S., Lievin, V., Bernet-Camard, M.-F., Servin, A.L.,1997. Antagonistic activity exerted in vitro and in vivo byLactobacillus casei (strain GG) against Salmonella ty-phimurium infection. Appl. Environ. Microbiol. 63, 513518.

Hull, R., Roberts, V., Mayes, J., 1984. Survival of L.acidophilus in yoghurt. Aust. J. Dairy Technol. 39, 164166.

Ishibashi, N., Shimamura, S., 1993. Bifidobacteria: researchand development in Japan. Food Technol. June, 126135.

Ishiwa, H., Iwata, M., 1980. Drug resistance plasmids inLactobacillus fermentum. J. Gen. Appl. Microbiol. 26, 7174.

Isolauri, E., Salminen, S., Mattila-Sandholm, T., 1999. Newfunctional foods in the treatment of food allergy. Ann.Med. 31, 299302.

Isolauri, E., Juntunen, M., Rautanen, T., Sillanaukee, P.,Koivula, T., 1991. A human Lactobacillus strain (Lacto-

M. Saarela et al. : Journal of Biotechnology 84 (2000) 197215212

bacillus casei sp. strain GG) promotes recovery from acutediarrhea in children. Pediatrics 88, 9097.

Isolauri, E., Kaila, M., Mykkanen, H., Ling, W.H., Salminen,S., 1994. Oral bacteriotherapy for viral gastroenteritis. Dig.Dis. Sci. 39, 25952600.

Johansson, M.-L., Molin, G., Jeppson, B., Nobaek, S., Ahrne,S., Bengmark, S., 1993. Administration of different Lacto-bacillus strains in fermented oatmeal soup. In vivo colo-nization of human intestinal mucosa and effect on theindigenous flora. Appl. Environ. Microbiol. 59, 1520.

Johansson, M.-L., Nobaek, S., Berggren, A., Nyman, M.,Bjorck, I., Ahrne, S., Jeppsson, B., Molin, G., 1998. Sur-vival of Lactobacillus plantarum DSM 9843 (299v), andeffect on the short-chain fatty acid content of faeces afteringestion of a rose-hip drink with fermented oats. Int. J.Food Microbiol. 42, 2938.

Kaila, M., Isolauri, E., Soppi, E., Virtanen, E., Laine, S.,Arvilommi, H., 1992. Enhancement of the circulating anti-body secreting cell response in human diarrhea by a humanLactobacillus strain. Pediatr. Res. 32, 141144.

Kailasapathy, K., Rybka, S., 1997. L. acidophilus andBifidobacterium ssp their theurapeutic potential andsurvival in yoghurt. Aust. J. Dairy Technol. 52, 2835.

Kankaanpaa, P., Sutas, Y., Salminen, S., Isolauri, E., 1998.Results on clinical demonstration of probiotics in children.In: Mattila-Sandholm, T., Kauppila, T. (Eds.), FunctionalFood Research in Europe, 3rd Workshop, FAIR CT96-1028, PROBDEMO, VTT Symposium 187. Haikko, Fin-land, pp. 3839.

Kirjavainen, P.V., Ouwehand, A.C., Isolauri, E., Salminen,S.J., 1998. The ability of probiotic bacteria to bind humanintestinal mucus. FEMS Microbiol. Lett. 15, 185189.

Korpela, R., Moilanen, E., Saxelin, M., Vapaasalo, H., 1997.Lactobacillus rhamnosus GG (ATCC 53013) and plateletaggregation in vitro. Int. J. Food Microbiol. 37, 8386.

Kurmann, J.A., 1988. Starters for fermented milks. In: Fer-mented Milks Science and Technology, IDF Bulletin227. IDF, Brussels, pp. 4155.

Langhendries, J.P., Detry, J., Van Hees, J., Lamboray, J.M.,Darimont, J., Mozin, M.J., Secretin, M.C., Senterre, J.,1995. Effect of a fermented infant formula containingviable bifidobacteria on the fecal flora composition and pHof healthy full-term infants. J. Pediat. Gastroenterol. Nutr.21, 177181.

Lee, Y.-K., Salminen, S., 1995. The coming of age of probi-otics. Trends Food Sci. Technol. 6, 241245.

Lehto, E.M., Salminen, S., 1997. Adhesion of two Lactobacil-lus strains, one Lactococcus and one Propionibacteriumstrain to cultured intestinal Caco-2 cell line. Biosci. Mi-croflora 16, 1317.

Lidbeck, A., Nord, C.E., Gustafsson, J.-A., Rafter, J., 1992a.Lactobacilli, anticarcinogenic activities and human intesti-nal microflora. Eur. J. Cancer Prev. 1, 341353.

Lidbeck, A., O8 vervik, E., Rafter, J., Nord, C.E., Gustafsson,J.-A., 1992b. Effect of Lactobacillus acidophilus supple-ments on mutagen excretion in faeces and urine in humans.Microb. Ecol. Health Dis. 5, 5967.

Lim, K.S., Huh, C.S., Baek, Y.J., 1993. Antimicrobial suscep-tibility of bifidobacteria. J. Dairy Sci. 76, 21682174.

Ling, W.H., Korpela, R., Mykkanen, H., Salminen, S., Hanni-nen, O., 1994. Lactobacillus strain GG supplementationdecreases colonic hydrolytic and reductive enzyme activi-ties in healthy female adults. J. Nutr. 124, 1823.

Link-Amster, H., Rochat, F., Saudan, K.Y., Mignot, O.,Aeschlimann, J.M., 1994. Modulation of a specific hu-moral immune response and changes in intestinal floramediated through fermented milk intake. FEMS Immunol.Med. Microbiol. 10, 5563.

Mackay, A.D., Taylor, M.B., Kibbler, C.C., Hamilton-Miller,J.M.T., 1999. Lactobacillus endocarditis caused by a probi-otic organism. Clin. Microbiol. Infect. 5, 290292.

Majamaa, H., Isolauri, E., 1997. Probiotics: a novel approachin the management of food allergy. J. Allergy Clin. Im-munol. 99, 179185.

Majamaa, H., Isolauri, E., Saxelin, M., Vesikari, T., 1995.Lactic acid bacteria in the treatment of acute rotavirusgastroenteritis. J. Pediatr. Gastroenterol. Nutr. 20, 333338.

Malin, M., Suomalainen, H., Saxelin, M., Isolauri, E., 1996.Promotion of IgA immune response in patients withCrohns disease by oral bacteriotherapy with LactobacillusGG. Ann. Nutr. Metab. 40, 137145.

Marteau, P., Pochart, P., Bouhnik, Y., Rambaud, J.C., 1993.Fate and effects of some transiting micro-organisms in thehuman gastrointestinal tract. World Rev. Nutr. Diet. 74,121.

Marteau, P., Gerhardt, M.F., Myara, A., Bouvier, E., Trivin,F., Rambaud, J.C., 1995. Metabolism of bile salts byalimentary bacteria during transit in the human smallintestine. Microb. Ecol. Health Dis. 8, 151157.

Marteau, P., Pochart, P., Flourie, B., Pellier, P., Santos, L.,Desjeux, J.F., Rambaud, J.C., 1990. Effect of chronicingestion of a fermented dairy product containing Lacto-bacillus acidophilus and Bifidobacterium bifidum onmetabolic activities of the colonic flora in humans. Am. J.Clin. Nutr. 52, 685688.

Marteau, P., Vaerman, J.P., Dehennin, J.P., Bord, S., Bras-sart, D., Pochart, P., Desjeux, J.F., Rambaud, J.C., 1997.Effects of intrajejunal perfusion and chronic ingestion ofLactobacillus johnsonii strain La1 on serum concentrationsand jejunal secretions of immunoglobulins and serumproteins in healthy humans. Gastroenterol. Clin. Biol. 21,293298.

Matteuzzi, D., Crociani, F., Brigidi, P., 1983. Antimicrobialsusceptibility of Bifidobacterium. Ann. Microbiol. (Paris)134A, 339349.

Mattila-Sandholm, T., Kauppila, T. (Eds.), 1998. FunctionalFood Research in Europe 3rd Workshop, FAIR CT96-1028, PROBDEMO, VTT Symposium 187. Haikko, Fin-land, p. 125.

Mattila-Sandholm, T., Salminen, S., 1998. Up-to-date on pro-biotics in Europe. Gastroenterol. Int. 11, 816.

Mattila-Sandholm, T., Matto, J., Saarela, M., 1999. Lacticacid bacteria with health claims interference and inter-actions with gastrointestinal flora. Int. Dairy J. 9, 2535.

M. Saarela et al. : Journal of Biotechnology 84 (2000) 197215 213

McFarland, L.V., Surawicz, C.M., Greenberg, R.N., 1994. Arandomised placebo-controlled trial of Saccharomyces bou-lardii in combination with standard antibiotics forClostridium difficile disease. JAMA 271, 19131918.

Michetti, P., Dorta, G., Wiesel, P.H., Brassart, D., Verdu, E.,Herranz, M., Felley, C., Porta, N., Rouvet, M., Blum,A.L., Corthesy-Theulaz, I., 1999. Effect of whey-basedculture supernatant of Lactobacillus acidophilus( johnsonii ) La1 on Helicobacter pylori infection in hu-mans. Digestion 60, 203209.

Miller, L.G., Finegold, S.M., 1967. Antimicrobial sensitivity ofBifidobacterium (Lactobacillus bifidus). J. Bacteriol. 93,125130.

Misra, A., Kulia, R., 1994. Use of Bifidobacterium bifidum forthe manufacturing of bio-yoghurt and fruit bio-yoghurt.Indian J. Dairy Sci. 47, 192197.

Mitsuoka, T., 1982. Recent trends in research on intestinalflora. Bifidobacteria Microflora 1, 324.

Mogensen, G., Friis, M., 1997. L. casei 431 a strategicprobiotic strain from Chr. Hansen A:S. The World ofIngredients. October, 4142.

Morotomi, M., Mutai, M., 1986. In vitro binding of potentmutagenic pyrolyzates to intestinal bacteria. J. Natl. Can-cer Inst. 77, 195201.

Myllarinen, P., Forssell, P., von Wright, A., Alander, M.,Mattila-Sandholm, T., 1998. The use of starch as a capsu-lation material for lactic acid bacteria. In: Mattila-Sand-holm, T., Kauppila, T. (Eds.), Functional Food Researchin Europe, 3rd Workshop, FAIR CT96-1028, PROB-DEMO, VTT Symposium 187. Haikko, Finland, p. 91.

Nicas, T.I., Cole, C.T., Preston, D.A., Schabel, A.A., Nagara-jan, R., 1989. Activity of glycopeptides against van-comycin-resistant gram-positive bacteria. Antimicrob.Agents Chemother. 33, 14771481.

Nielsen, O.H., Jorgensen, S., Pedersen, K., Justesen, T., 1994.Microbiological evaluation of jejunal aspirates and faecalsamples after oral administration of bifidobacteria andlactic acid bacteria. J. Appl. Bacteriol. 76, 469474.

Niku-Paavola, M.-L., Latva-Kala, K., Laitila, A., Mattila-Sandholm, T., Haikara, A., 1999. New types of antimicro-bial compounds produced by Lactobacillus plantarum. J.Appl. Microbiol. 86, 2935.

Oberman, H., Libudzisz, Z., 1998. Fermented milks. In:Wood, B.J.B. (Ed.), Microbiology of Fermented Foods.Blackie Academic & Professional, London, pp. 308350.

Okawa, T., Niibe, H., Arai, T., Sekiba, K., Noda, K.,Takeuchi, S., Hashimoto, S., Ogawa, N., 1993. Effect ofLC9018 combined with radiation therapy on carcinoma ofthe uterine cervix. A phase III, multicenter, randomized,controlled study. Cancer 72, 19491954.

Okongi, S., Ono, J., Kudo, T., Hiramatsu, A., Teraguchi, S.,1984. Jap. Patent cJ59085288-A.

Orrhage, K., Sillerstrom, E., Gustafsson, J.A., Nord, C.E.,Rafter, J., 1994. Binding of mutagenic heterocyclic aminesby intestinal and lactic acid bacteria. Mut. Res. 311, 239248.

Ouwehand, A.C., Salminen, S.J., 1998. The health effects ofcultured milk products with viable and non-viable bacteria.Int. Dairy J. 8, 749758.

Ouwehand, A.C., Isolauri, E., Kirjavainen, P.V., Salminen,S.J., 1999. Adhesion of four Bifidobacterium strains tohuman intestinal mucus from subjects in different agegroups. FEMS Microbiol. Lett. 172, 6164.

Pedrosa, M.C., Golner, B.B., Goldin, B.R., Barakat, S., Dal-lal, G.E., Russell, R.M., 1995. Survival of yougurt-contain-ing organisms and Lactobacillus gasseri (ADH) and theireffect on bacterial enzyme activity in the gastrointestinaltract of healthy and hypochlorhydric elderly subjects. Am.J. Clin. Nutr. 61, 353359.

Pelletier, C., Bouley, C., Bourliov, P., Carbon, C., 1996.Evaluation of safety properties of Lactobacillus strains byusing an experimental model of endocarditis in rabbit(abstr.) SOMED Meeting, Paris, France.

Pelto, L., Isolauri, E., Lilius, E.M., Nuutila, J., Salminen, S.,1998. Probiotic bacteria down-regulate the milk-inducedinflammatory response in milk-hypersensitive subjects buthave an immunostimulatory effect in healthy subjects.Clin. Exp. Allergy. 28, 14741479.

Pettersson, L., Graf, W., Sewelin, U., 1983. Survival of Lacto-bacillus acidophilus NCDO 1748 in human gastrointestinaltract. 2. Ability to pass the stomach and intestine in vivo.In: Hallgren, B. (Ed.), Nutrition and the Intestinal Flora.XV Symp. Swed. Nutr. Found. Almqvist and Wicksell,Uppsala, pp. 127130.

Pochart, P., Marteau, P., Bouhnik, Y., Goderel, I., Bourlioux,P., Rambaud, J.-C., 1992. Survival of bifidobacteria in-gested via fermented milk during their passage through thehuman small intestine: an in vivo study using intestinalperfusion. Am. J. Clin. Nutr. 55, 7880.

Pool-Zobel, B.L., Bertram, B., Knoll, M., Lamertz, R., Neu-decker, C., Schillinger, U., Schmezer, P., Holzapfel, W.H.,1993. Antigenotoxic properties of lactic acid bacteria invivo in the gastrointestinal tract of rats. Nutr. Cancer 20,271281.

Pool-Zobel, B.L., Neudecker, C., Domizlaff, I., Ji, S.,Schillinger, U., Rumney, C., Moretti, M., Vilarini, I.,Scassellati-Sforzolini, R., Rowland, I., 1996. Lactobacillus-and Bifidobacterium-mediated antigenotoxicity in the colonof rats. Nutr. Cancer 26, 365380.

Rautanen, T., Isolauri, E., Salo, E., Vesikari, T., 1998. Man-agement of acute diarrhoea with low osmolarity oral rehy-dration solution and Lactobacillus strain GG. Arch. Dis.Child 79, 157160.

Rautio, M., Jousimies-Somer, H., Kauma, H., Pietarinen, I.,Saxelin, M., Tynkkynen, S., Koskela, M., 1999. Liverabscess due to a Lactobacillus rhamnosus strain indistin-guishable from L. rhamnosus strain GG. Clin. Infect. Dis.28, 11601161.

Raza, S., Graham, S.M., Allen, S.J., Sultana, S., Cuevas, L.,Hart, C.A., 1995. Lactobacillus GG promotes recoveryfrom acute nonbloody diarrhea in Pakistan. Pediatr. Infect.Dis. J. 14, 107111.

M. Saarela et al. : Journal of Biotechnology 84 (2000) 197215214

Reddy, M.S., 1989. Enhancement of L. acidophilus viability inyoghurt and other cultured products. US Patent 4.797.289.

Reddy, B.S., Rivenson, A., 1993. Inhibitory effect ofBifidobacterium longum on colon, mammary, and livercarcinogenesis induced by 2-amino-3-methylimidazo[4,5-f]quinoline, a food mutagen. Cancer Res. 53, 39143918.

Rinckel, L.A., Savage, D.C., 1990. Characterization of plas-mids and plasmid-borne macrolide resistance from Lacto-bacillus sp. strain 100-33. Plasmid 23, 119125.

Roberfroid, M.B., 1998. Prebiotics and synbiotics: conceptsand nutritional properties. Br. J. Nutr. 80, S197S202.

Robredo, B., Singh, K.V., Baquero, R., Murray, B.E., Torres,C., 2000. Vancomycin-resistant enterococci isolated fromanimals and food. Int. J. Food Microbiol. 54, 197204.

Ruseler-van Embden, J.G.H., van Lieshout, L.M.C., Marteau,P., 1995. No degradation of intestinal mucus glycoproteinsby Lactobacillus casei strain GG. Microecol. Ther. 25,304309.

Rybka, S., Fleet, G., 1997. Populations of L. delbrueckii sspbulgaricus, S. thermophilus, L. acidophilus and Bifidobac-terium species in Australian yoghurts. Food Austr. 49,471475.

Saavedra, J.M., Bauman, N.A, Oung, I., Perman, J.A.,Yolken, R.H., 1994. Feeding of Bifidobacterium bifidumand Streptococcus thermophilus to infants in hospital forprevention of diarrhea and shedding of rotavirus. Lancet344, 10461049.

Salminen, S., Isolauri, E., Salminen, E., 1996a. Probiotics andstabilisation of the gut mucosal barrier. Asia Pacific J.Clin. Nutr. 5, 5356.

Salminen, S., Ouwehand, A.C., Isolauri, E., 1998a. Clinicalapplications of probiotic bacteria. Int. Dairy J. 8, 563572.

Salminen, S., Ouwehand, A., Benno, Y., Lee, Y.K., 1999.Probiotics: how should they be defined? Trends Food Sci.Technol. 10, 107110.

Salminen, S., von Wright, A., Laine, M., Vuopio-Varkila, J.,Korhonen, T., Mattila-Sandholm, T., 1996b. Developmentof selection criteria for probiotic strains to assess theirpotential in functional foods: a Nordic and Europeanapproach. Biosci. Microbiol. 15, 6167.

Salminen, S., von Wright, A., Morelli, L., Marteau, P., Bras-sart, D., de Vos, W.M., Fonden, R., Saxelin, M., Collins,K., Mogensen, G., Birkeland, S.-E., Mattila-Sandholm, T.,1998b. Demonstration of safety of probiotics a review.Int. J. Food Microbiol. 44, 93106.

Samona, A., Robinson, R.K., Marakis, S., 1996. Acid produc-tion by bifidobacteria and yoghurt bacteria during fermen-tation and storage of milk. Food Microbiol. 13, 275280.

Saxelin, M., Grenov, B., Svensson, U., Fonden, R., Reniero,R., Mattila-Sandholm, T., 2000. The technology of probi-otics. Trends Food Sci. Technol. 10, 387392.

Schiffrin, E., Brassart, D., Servin, A.L., Rochat, F., Donnet-Hughes, A., 1997. Immune modulation of blood leukocytesin humans by lactic acid bacteria: criteria for strain selec-tion. Am. J. Clin. Nutr. , S 66, 15S20.

Schiffrin, E.J., Rochat, F., Link-Angler, H., Acschlimann,J.M., Donnet-Hughes, A., 1995. Immuno-modulation of

human blood cells following the ingestion of lactic acidbacteria. J. Dairy Sci. 78, 491497.

Sepp, E., Tamm, E., Torm, S., Lutsar, I., Mikelsaar, M.,Salminen, S., 1995. Impact of a Lactobacillus probiotic onthe faecal microflora in children with shigellosis. Mi-croecol. Ther. 23, 7480.

Shornikova, A.V., Casas, I., Isolauri, E., Mykkanen, H.,Vesikari, T., 1997a. Lactobacillus reuteri as a therapeuticagent in acute diarrhea in young children. J. Pediatr.Gastroenterol. Nutr. 24, 399404.

Shornikova, A.V., Casas, I.A., Mykkanen, H., Vesikari, T.,1997b. Bacteriotherapy with Lactobacillus reuteri in ro-tavirus gastroenteritis. Pediatr. Infect. Dis. 16, 11031107.

Shornikova, A.V., Isolauri, E., Burkanova, L., Lukovnikova,S., Vesikari, T., 1997c. A trial in Karelian republic of oralrehydration and Lactobacillus GG for treatment of acutediarrhoea. Acta Paediatr. 86, 460465.

Siitonen, S., Vapaatalo, H., Salminen, S., Gordin, A., Saxelin,M., Wikberg, R., Kirkkola, A.L., 1990. Effect of Lacto-bacillus GG yoghurt in prevention of antibiotic associateddiarrhea. Ann. Med. 22, 5759.

Silva, M., Jacobus, N.V., Deneke, C., Gorbach, S.L., 1987.Antimicrobial substance from a human Lactobacillusstrain. Antimicrob. Agents Chemother. 31, 12311233.

Singh, J., Riveson, A., Tomita, M., Shimamura, S., Ishibashgi,N., Reddy, B.S., 1997. Bifidobacterium longum, a lactic acidproducing intestinal bacterium, inhibits colon cancer andmodulates the intermediate biomarkers of colon carcino-genesis. Carcinogenesis 18, 833841.

Skytta, E., Haikara, A., Mattila-Sandholm, T., 1992. Produc-tion and characterization of antibacterial compounds pro-duced by Pediococcus damnosus and Pediococcuspentosaceus. J. Appl. Bacteriol. 72, 134142.

Spanhaak, S., Havenaar, R., Schaafsma, G., 1998. The effectof consumption of milk fermented by Lactobacillus caseistrain Shirota on the intestinal microflora and immuneparameters in humans. Eur. J. Clin. Nutr. 52, 899907.

Spork, A.C., 1994. New trends for probiotic cultures. Eur.Dairy Mag. 30763077.

Surawicz, C.M., Elmer, G.W., Speelman, P., McFarland, L.,Chinn, J., van Belle, G., 1989. Prevention of antibiotic-as-sociated diarrhea by Saccharomyces boulardii. A prospec-tive study. Gastroenterology 84, 981988.

Svensson, U., 1999. Industrial perspectives. In: Tannock,G.W. (Ed.), Probiotics: A Critical Review. Horizon Scien-tific Press, Wymondham, pp. 5764.

Swenson, J.M., Facklam, R.R., Thornsberry, C., 1990. An-timicrobial susceptibility of vancomycin-resistant Leu-conostoc, Pediococcus, and Lactobacillus species.Antimicrobiol. Agents Chemother. 34, 543549.

Takagi, A., Matsuzaki, T., Sato, M., Nomoto, K., Morotomi,M., Yokokura, T., 1999. Inhibitory effect of oral adminis-tration of Lactobacillus casei on 3-methycholanthrene-in-duced carcinogenesis in mice. Med. Microbiol. Immunol.(Berl.) 188, 111116.

Tamime, A.Y., Marshall, V.M.E., Robinson, R.K., 1995. Mi-crobiological and technological aspects of milks fermentedby bifidobacteria. J. Dairy Res. 62, 151187.

M. Saarela et al. : Journal of Biotechnology 84 (2000) 197215 215

Tanaka, R., Ohwaki, M.A., 1994. Controlled study on theingestion of Lactobacillus casei fermented milk on theintestinal microflora, its microbiology and immune systemin healthy adults. In: Proceedings of XII Riken Symposiumon Intestinal Flora. Tokyo, Japan, pp. 85104.

Thyagaraja, N., Hosono, A., 1993. Antimutagenicity of lacticacid bacteria from Idly against food-related mutagens. J.Food Prot. 56, 10611066.

Thyagaraja, N., Hosono, A., 1994. Binding properties of lacticacid bacteria from Idly towards food-borne mutagens.Food Chem. Toxic. 32, 805809.

Tuomola, E., 1999. In vitro adhesion of probiotic lactic acidbacteria. Ph.D. thesis, Turku, Finland.

Tuomola, E.M., Ouwehand, A.C., Salminen, S., 1999. Humanileostomy glycoproteins as a model for small intestinalmucus to investigate adhesion of probiotics. Lett. Appl.Microbiol. 28, 159163.

Tuomola, E.M., Salminen, S.J., 1998. Adhesion of some pro-biotic and dairy Lactobacillus strains to Caco-2 cell cul-tures. Int. J. Food Microbiol. 41, 4551.

Tynkkynen, S., Singh, K.V., Varmanen, P., 1998. Vancomycinresistance factor of Lactobacillus rhamnosus GG in relation

to enterococcal vancomycin resistance (6an) genes. Int. J.Food Microbiol. 41, 195204.

Vescovo, M., Morelli, L., Bottazzi, V., 1982. Drug resistanceplasmids in Lactobacillus acidophilus and Lactobacillusreuteri. Appl. Environ. Microbiol. 43, 5056.

Wolf, B.W., Garleb, D.G., Casas, I., 1995. Safety and toler-ance of Lactobacillus reuteri in healthy adult male subjects.Microb. Ecol. Health Dis. 8, 4150.

Wolf, B.W., Wheeler, K.B., Ataya, D.G., Garleb, K.A., 1998.Safety and tolerance of Lactobacillus reuteri supplementa-tion to a population infected with human immunodefi-ciency virus. Food Chem. Toxicol. 36, 10851094.

Young, J., 1998. European market developments in prebiotic-and probiotic containing foodstuffs. Br. J. Nutr. 80, S231S233.

Zhang, X.B., Ohta, Y., 1993. Antimutagenicity of cell frac-tions of microorganisms of potent mutagenic pyrolysates.Mut. Res. 298, 247253.

Zhang, X.B., Ohta, Y., Hosono, A., 1990. Antimutagenicityand binding of lactic acid bacteria from a Chinese cheeseto mutagenic pyrolyzates. J. Dairy Sci. 73, 27022710.

.