food science and technology international vol16 issue3

Antioxidant Potential of Chestnut (Castanea sativa L.)

and Almond (Prunus dulcis L.) By-products

J.C.M. Barreira,1,2 I.C.F.R. Ferreira,1,* M.B.P.P. Oliveira2 and J.A. Pereira1

1CIMO/Escola Superior Agraria, Instituto Politecnico de Braganca, Campus de Santa ApoloniaPO Box 1172, 5301-855 Braganca, Portugal

2REQUIMTE/Servico de Bromatologia, Faculdade de Farmacia da Universidade do PortoRua Anıbal Cunha, 164, 4099-030 Porto, Portugal

The antioxidant properties of almond green husks (Cvs. Duro Italiano, Ferraduel, Ferranhes, Ferrastar and

Orelha de Mula), chestnut skins and chestnut leaves (Cvs. Aveleira, Boa Ventura, Judia and Longal) wereevaluated through several chemical and biochemical assays in order to provide a novel strategy to stimulatethe application of waste products as new suppliers of useful bioactive compounds, namely antioxidants. All

the assayed by-products revealed good antioxidant properties, with very low EC50 values (lower than380 mg/mL), particularly for lipid peroxidation inhibition (lower than 140mg/mL). The total phenols andflavonoids contents were also determined. The correlation between these bioactive compounds and DPPH

(2,2-diphenyl-1-picrylhydrazyl) radical scavenging activity, reducing power, inhibition of b-carotenebleaching and inhibition of lipid peroxidation in pig brain tissue through formation of thiobarbituricacid reactive substances, was also obtained. Although, all the assayed by-products proved to have ahigh potential of application in new antioxidants formulations, chestnut skins and leaves demonstrated

better results.

Key Words: chestnut, almond, by-products, antioxidant activity, total phenols

INTRODUCTION

The interest in polyphenolic antioxidants has

increased remarkably in the last decade because of

their elevated capacity in scavenging free radicals asso-

ciated with various diseases (Silva et al., 2007). Some

studies indicate that dietary polyphenols have a protec-

tive effect against coronary heart disease (Weisburger,

1999; Engler and Engler, 2006), cancer (Fang et al.,

2002; Nichenametla et al., 2006), neurodegenerative dis-

eases (Lau et al., 2005) and osteoporosis (Weaver and

Cheong, 2005).Chestnut and almond are important sources of phe-

nolic compounds. Particularly chestnut fruits (Ribeiro

et al., 2007), chestnut leaves (Calliste et al., 2005),

almond hulls (Sang et al., 2002; Takeoka and Dao,

2003), almond skins (Sang et al., 2002), almond shells

(Pinelo et al., 2004) and almond fruits (Milbury et al.,

2006) contain those compounds.

Portugal is one of the most important chestnut

producers, with nearly 25% of European production.Tras-os-Montes region represent 75.8% of Portuguese

chestnut crops and 84.9% of chestnut orchards area

(23,338 ha). The best development conditions are

found at altitudes higher than 500m and winter low

temperatures, as in the ‘Terra Fria Transmontana’region (Northeast of Portugal) in which 12,500 ha are

used for chestnut cultivation (Ribeiro et al., 2007).

Almond is also an important product, with 24,522

crops spread trough 36,530 ha. This culture is mainly

located in Algarve and ‘Terra Quente Transmontana’(Cordeiro and Monteiro, 2001; Martins et al., 2003).

Accordingly, it would be very important to perform a

complete characterization of the antioxidant potential of

different by-products originated in these Portuguese

crops or by their industrial applications. Due to themultifunctional characteristics of phytochemicals, the

antioxidant efficacy of a plant extract is best evaluated

based on results obtained by commonly accepted assays,

taking into account different oxidative conditions,

system compositions, and antioxidant mechanisms(Weisburger, 1999).

In the present work, the antioxidant properties of

almond green husks (Cvs. Duro Italiano, Ferraduel,

Ferranhes, Ferrastar and Orelha de Mula) chestnut

skins and chestnut leaves (Cvs. Aveleira, Boa Ventura,Judia and Longal) were evaluated through several

*To whom correspondence should be sent(e-mail: [email protected]).

Food Sci Tech Int 2010;16(3):0209–8� SAGE Publications 2010Los Angeles, London, New Delhi and SingaporeISSN: 1082-0132DOI: 10.1177/1082013209353983

209

chemical and biochemical assays: DPPH (2, 2-diphenyl-1-picrylhydrazyl) radical scavenging activity, reducingpower, inhibition of b-carotene bleaching and thio-barbituric acid reactive substances (TBARS) formationin brain cells. The whole extracts were used sincethey contain different compounds that can act syner-gistically, constituting a benefit in comparison to indi-vidual compounds (Pellegrini et al., 2006; Pereira et al.,2006).

The evaluation of the antioxidant properties stands asan interesting and valuable task, particularly for findingnew sources for natural antioxidants and nutraceuticals,providing a novel strategy to stimulate the application ofthese by-products as new suppliers of useful bioactivecompounds.

MATERIALS AND METHODS

Materials

Standards and Reagents

Standards BHA (2-tert-butyl-4-methoxyphenol),TBHQ (tert-butylhydroquinone), L-ascorbic acid,a-tocopherol, gallic acid and (þ)-catechin were purchasefrom Sigma (St. Louis,MO,USA). 2,2-diphenyl-1-picryl-hydrazyl (DPPH) was obtained from Alfa Aesar (WardHill, MA, USA). All other chemicals were obtained fromSigma Chemical Co. (St. Louis, MO, USA). Methanolwas obtained from Pronalab (Lisbon, Portugal). Waterwas treated in a Mili-Q water purification system (TGIPure Water Systems, USA).

Samples

Chestnut tree leaves and chestnut skins were obtainedfrom four different cultivars (Cvs. Aveleira, BoaVentura, Judia and Longal) and collected from orchardslocated in Vinhais (Tras-os-Montes), in the Northeastside of Portugal. Leaves were collected monthly fromJune to October and used miscellaneously (equalnumber of leaves for each month), and fruits were col-lected in October and November. These samples wereobtained during the crop year of 2006. Almond huskswere obtained from five different cultivars (DuroItaliano, Ferraduel, Ferranhes, Ferrastar and Orelha deMula) and collected in August-September 2006 in orch-ards located in Southwest Tras-os-Montes, NortheastPortugal. Selected plants are not irrigated and no phy-tosanitary treatments were applied.

Chestnut leaves and almond husks were dried at 65 �Cuntil constant weight was achieved and kept at �20 �Cuntil further use. Outer and inner skins were removedfrom chestnuts and submitted to a roasting processconducted at 250 �C in a muffle furnace (ECF 12/22,Lenton Thermal Designs Limited) for 15min, to

mimetize industrial practices. Inner and outer skinswere assayed together maintaining the individual pro-portion found for each variety (outer skins represent ahigher chestnut weight percentage, when compared withinner skins).

For antioxidant compounds extraction, a fine driedpowder (20 mesh) of sample was extracted usingwater, under magnetic stirring (150 rpm) at room tem-perature during 1 h. The extracts were filtered throughWhatman no. 4 paper under reduced pressure, frozen at�80 �C and then lyophilized (Ly-8-FM-ULE, Snijders)at �80 �C to �90 �C under a reduced pressure of �0.045mbar. All the samples were redissolved in water at aconcentration of 50mg/mL, diluted to final concentra-tions and analyzed for their contents in polyphenols andflavonoids, DPPH radical scavenging activity, reducingpower, inhibition of b-carotene bleaching and inhibitionof lipid peroxidation.

Determination of Antioxidants Content

Content of total phenols in the extracts was estimatedby a colorimetric assay based on procedures describedby Singleton and Rossi (1965) with some modifications.Basically, 1mL of sample was mixed with 1mL of Folinand Ciocalteu’s phenol reagent. After 3min, 1mL ofsaturated sodium carbonate solution was added to themixture and adjusted to 10mL with distilled water.The reaction was kept in the dark for 90min, afterwhich the absorbance was read at 725 nm (AnalytikJena 200-2004 spectrophotometer). Gallic acid wasused for constructing the standard curve(0.01�0.4mM, y¼ 2.94848 x�0.09211, R2

¼ 0.99914)and the results were expressed as mg of gallic acid equiv-alents/g of extract (GAEs).

Flavonoid contents in the extracts were determinedby a colorimetric method described by Jia et al. (1999)with some modifications. The extract (250 mL) wasmixed with 1.25mL of distilled water and 75 mL of a5% NaNO2 solution. After 5min, 150 mL of a 10%AlCl3�H2O solution was added. After 6min, 500 mL of1M NaOH and 275 mL of distilled water were added toprepare the mixture. The solution was mixed well andthe absorbance was read at 380, 425 and 510 nm, inorder to compare the results. (þ)-Catechin(0.250�2.500mM) was used to calculate the standardcurves (y¼ 2.4553 x�0.1796, R2

¼ 0.997, at 340 nm,y¼ 0.7376 x�0.0131, R2

¼ 0.997, at 425 nm,y¼ 0.5579 x�0.0494, R2

¼ 0.992, at 510 nm), and theresults were expressed as mg of (þ)-catechin equivalents(CEs) per g of extract.

DPPH Radical-scavenging Activity

Various concentrations of extracts (0.3mL) weremixed with 2.7mL of methanolic solution containingDPPH radicals (6� 10�5mol/L). The mixture was

210 J.C.M. BARREIRA ET AL.

shaken vigorously and left to stand for 60min in thedark (until stable absorbance values were obtained).The reduction of the DPPH radical was determined byreading the absorbance at 517 nm. The radical scaveng-ing activity (RSA) was calculated as a percentage ofDPPH discoloration using the equation: %RSA¼[(ADPPH�AS)/ADPPH]� 100, where AS is the absorbanceof the solution when the sample extract has been addedat a particular level, and ADPPH is the absorbance of theDPPH solution (Barreira et al., 2008). The extractconcentration providing 50% of radicals scavengingactivity (EC50) was calculated from the graph of RSApercentage against extract concentration. BHA anda-tocopherol were used as standards.

Reducing Power

Several concentrations of extracts (2.5mL) weremixed with 2.5mL of 200mmol/L sodium phosphatebuffer and 2.5mL of potassium ferricyanide (1%). Themixture was incubated at 50 �C for 20min. After 2.5mLof trichloroacetic acid (10% w/v) were added, and themixture was centrifuged at 1000 rpm for 8min(Centorion K24OR- 2003 refrigerated centrifuge). Theupper layer (5mL) was mixed with 5mL of deionizedwater and 1mL of ferric chloride (0.1%), and the absor-bance was measured spectrophotometrically at 700 nm(Barreira et al., 2008). The extract concentration provid-ing 0.5 of absorbance (EC50) was calculated from thegraph of absorbance at 700 nm against extract concen-tration. BHA and a-tocopherol were used as standards.

Inhibition of �-carotene Bleaching

The antioxidant activity of aqueous extracts was eval-uated by the b-carotene linoleate model system. A solu-tion of b-carotene was prepared by dissolving 2mg ofb-carotene in 10mL of chloroform. Two mL of thissolution were pipetted into a 100mL round-bottomflask. After the removal of the chloroform at 40 �Cunder vacuum, 40mg of linoleic acid, 400mg of Tween80 emulsifier and 100mL of distilled water were addedto the flask with vigorous shaking. Aliquots (4.8mL) ofthis emulsion were transferred into different test tubescontaining 0.2mL of different concentrations of chest-nut extracts. The tubes were shaken and incubated at50 �C in a water bath. As soon as the emulsion wasadded to each tube, the zero time absorbance was mea-sured at 470 nm. Absorbance readings were thenrecorded at 20-min intervals until the control samplehad changed color. A blank, devoid of b-carotene, wasprepared for background subtraction. Lipid peroxida-tion (LPO) inhibition was calculated using the followingequation: LPO inhibition¼ (b-carotene content after 2 hof assay/initial b-carotene content)� 100 (Barreira et al.,2008). The extract concentration providing 50% antiox-idant activity (EC50) was calculated from the graph of

antioxidant activity percentage against extract concen-tration. TBHQ was used as standard.

Inhibition of Lipid Peroxidation Using ThiobarbituricAcid Reactive Substances

Brains were obtained from pig (Sus scrofa) of bodyweight �150 kg, dissected and homogenized with aPolytron in ice-cold Tris-HCl buffer (20mM, pH 7.4)to produce a 1:2 (w/v) brain tissue homogenate whichwas centrifuged at 3000 g for 10min. An aliquot(0.1mL) of the supernatant was incubated with theextracts (0.2mL) in the presence of FeSO4 (10 mM,0.1mL) and ascorbic acid (0.1mM, 0.1mL) at 37 �Cfor 1 h. The reaction was stopped by the addition oftrichloroacetic acid (28% w/v, 0.5mL), followed bythiobarbituric acid (TBA, 2 %, w/v, 0.38mL), and themixture was then heated at 80 �C for 20min. After cen-trifugation at 3000 g for 10min to remove the precipi-tated protein, the color intensity of the TBARS in thesupernatant was measured by its absorbance at 532 nm.The inhibition ratio (%) was calculated using the follow-ing formula: Inhibition ratio (%)¼ [(A�B)/A]� 100%,where A and B were the absorbance of the control andthe compound solution, respectively (Barreira et al.,2008). The extract concentration providing 50% lipidperoxidation inhibition (EC50) was calculated from thegraph of antioxidant activity percentage against extractconcentration. BHA was used as standard.

Statistical Analysis

For all the experiments three samples were analyzedand all the assays were carried out in triplicate. Theresults are expressed as mean values and standarderror or standard deviation (SD). The differences bet-ween the different extracts were analyzed using one-wayanalysis of variance (ANOVA) followed by Tukey’shonestly significant difference post hoc test witha¼ 0.05, coupled with Welch’s statistic. The regressionanalysis between total phenols or flavonoid contents,and EC50 values for antioxidant activity used the samestatistical package. These treatments were carried outusing SPSS v. 16.0 program.

RESULTS AND DISCUSSION

Table 1 presents extraction yields (expressed as w/wpercentages), total phenols and flavonoids content (mg/gof extract) obtained for chestnut and almondby-products. The results are presented for each singlevariety in order to analyze possible differences.However, and regarding the aim of this work, the resultsobtained for each by-product, as presented in the bottomof the table, are the most significant, once it would bedifficult to obtain supplies of these by-products selected

Antioxidant Potential of Chestnut and Almond By-products 211

by variety. Among all of the extracts analyzed, an inter-esting content of total phenols (from 228 to 859mg/g)was detected with mean values of 592mg/g for almondhusk, 413mg/g for chestnut leaf and 710mg/g for chest-nut skins. The marked differences of the results obtainedfor Longal leaf when compared with our previous study(Barreira et al., 2008) can be explained on the basis ofthree different factors. First, the leaves used in our pre-vious work presented a higher ripeness state, second,they were utilized in fresh (a drying step was not con-ducted), and finally the extraction procedure was con-ducted at water boiling temperature. These resultsrevealed the high potential of the assayed by-productsas new sources of antioxidant compounds. Extractionyields were generally low, but their bioactivity indicatesthat the extraction procedure was effective, consideringthat the objective was to achieve a clean extract. Despitethis consideration, not all cases revealed a relationshipbetween extracted mass and total phenols content.Actually, extracts obtained with chestnut skins provedto be the most uncontaminated, promoting it as themore adequate by-products, considering the posteriorpurifying processes. Likewise, this observation couldprobably be explained by a higher amount of otherpolar compounds in chestnut leaves and almond husks.

Several biochemical assays were used to screen theantioxidant properties: inhibition of b-carotene bleach-ing (by neutralizing the linoleate-free radical and otherfree radicals formed in the system which attack thehighly unsaturated b-carotene models), inhibition oflipid peroxidation in brain tissue (measured by thecolor intensity of MDA-TBA complex), scavengingactivity on DPPH radicals (measuring the decrease inDPPH radical absorption after exposure to radical scav-engers) and reducing power (measuring the conversion

of a Fe3þ/ferricyanide complex to the ferrous form). Theassays were carried out using whole extracts instead ofindividual compounds, once additive and synergisticeffects of phytochemicals in fruits and vegetables areresponsible for their potent bioactive properties andthe benefit of a diet rich in fruits and vegetables is attrib-uted to the complex mixture of phytochemicals presentin whole foods (Liu, 2003). This enhances the advan-tages of natural phytochemicals over single antioxidantswhen they are used to achieve health benefits.Antioxidant activity increased with the concentration,being obtained very good results even at low extractconcentrations, especially for TBARS assay.

The linoleic acid free radical attacks the highly unsat-urated b-carotene model. The presence of different anti-oxidants can hinder the extent of b-carotene-bleachingby neutralizing the linoleate free radical and other freeradicals formed in the system (Jayaprakasha et al.,2001). Hence, the absorbance diminishes fast in sampleswithout antioxidant, whereas in the presence of an anti-oxidant, they maintain their color, and thus absorbance,for a longer time (Figure 1). Bleaching inhibition in thepresence of different extracts increased with concentra-tion and proved to be very good. At 500 mg/mL, all theextracts presented inhibition percentages superior to65%, except in the cases of Orelha de Mula husk, avery good result once that the antioxidant activity ofTBHQ standard reached 82.2% only at 2mg/mL. It isexpectable that the antioxidative components in thechestnut extracts reduce the extent of b-carotenedestruction by neutralizing the linoleate free radicaland other free radicals formed in the system. Itbecame clear that chestnut derived by-products revealedhigher efficiency in this antioxidant activity biochemicalassay when compared with almond by-products.

Table 1. Extraction yields, content of total phenols and flavonoids in the extracts of chestnut andalmond by-products.

Cultivar Extraction yield (%) Total phenols (mg/g) Flavonoids (mg/g)

Almond husk (AH) Duro Italiano 17.65±1.02 c 777.21±18.78 b 237.20±2.52 bFerraduel 14.14±0.60 c 304.79±22.06 e 70.48±3.61 eFerranhes 27.49±2.11 a 378.70±9.42 d 130.68±5.91 cFerrastar 22.58±1.18 b 859.07±74.50 a 284.61±12.06 aOrelha de Mula 22.81±1.55 b 639.75±33.91 c 116.88±19.49 d

Chestnut leaf (CL) Aveleira 17.67±0.94 a 468.34±25.47 b 84.68±3.72 bBoa Ventura 15.62±0.93 bc 432.16±37.59 c 83.09±6.82 bJudia 17.08±0.62 ab 228.37±13.99 d 73.31±4.89 cLongal 13.73±0.49 c 522.98±23.82 a 90.39±5.57 a

Chesnut skin (CS) Aveleira 7.17±0.29 b 533.81±30.90 c 49.92±1.93 dBoa Ventura 6.43±0.32 b 805.74±74.31 a 146.08±4.19 aJudia 12.59±0.84 a 757.95±67.51 b 98.10±6.62 bLongal 6.47±0.43 b 742.33±37.46 b 72.27±3.78 c

AH �x 20.93±4.91 a 591.90±221.39 b 167.97±80.88 aCL �x 16.02±1.72 b 412.96±114.91 c 82.87±8.13 bCS �x 8.16±2.72 c 709.96±118.38 a 91.59±36.21 b

Mean values followed by different letter within a column are significant different (p<0.05).


Inhibition of lipid peroxidation was evaluated usingTBARS. When oxidation processes occur, a pinkishsolution is formed. If antioxidant compounds are pre-sent in the system, the formation of the substancesresponsible for the coloration is prevented. As it canbe easily understood after Figure 2 observation, thecapacity of inhibition of lipid peroxidation is propor-tional to the extract concentration. This methodrevealed very high inhibition percentages at extremelylow concentrations. All extracts showed inhibition per-centages superior to 60% at concentrations of 100 mg/mL, except for Ferraduel husk and Judia leaf. Generally,chestnut skins and almond husks extracts proved to bebetter inhibitors in this model.The RSA values were expressed as the ratio percent-

age of sample absorbance decrease and the absorbance

of DPPH solution in the absence of extract at 517 nm.The scavenging effects of all extracts on DPPH radicalsincreased with the concentration increase and wereremarkably good (Figure 3), with RSA percentagessuperior to 90% at 500 mg/mL for almost all the extracts,except for Aveleira and Judia leaves and Ferraduel andFerranhes husks, again better than the scavengingeffects of some usual standards like BHA (96% at3.6mg/mL) and a-tocopherol (95% at 8.6mg/mL).

Like in the other assays previously referred, the reduc-ing power increased with concentration, and the valuesobtained for all the extracts were very good (Figure 4).At 250 mg/mL, the absorbance values were higher than0.5 for all extracts, with the exception of Judia leaf andFerraduel and Orelha de Mula husks, proving once

Almond husk

0

20

40

60

80

100

0

Concentration (µg/mL)

b-ca

rote

ne b

leac

hing

Inhi

bitio

n (%

)b-

caro

tene

ble

achi

ngIn

hibi

tion

(%)

b-ca

rote

ne b

leac

hing

Inhi

bitio

n (%

)

Chestnut leaf

0

20

40

60

80

100


Chestnut skin

0

20

40

60

80

100


200 400 600 800 1000

0 200 400 600 800 1000

0 200 400 600 800 1000

Figure 1. Inhibition of b-carotene bleaching as afunction of extracts concentration. Almond husk: (4)Ferraduel, (^) Duro Italiano, (œ) Ferranhes, (g)Ferrastar, (f) Orelha de Mula. Chestnut leaf: (4)Aveleira, (^) Boa Ventura, (œ) Judia, (g) Longal.

Almond husk

0

20

40

60

80

100


TB

AR

S in

hibi

tion

(%)

Chestnut leaf

0

20

40

60

80

100


TB

AR

S in

hibi

tion

(%)

Chestnut leaf

0

20

40

60

80

100


TB

AR

S in

hibi

tion

(%)

0 200 400 600 800 1000

0 200 400 600 800 1000

0 200 400 600 800 1000

Figure 2. Lipid peroxidation (LPO) inhibition as afunction of extracts concentration. (4) Ferraduel, (^)Duro Italiano, (œ) Ferranhes, (g) Ferrastar, (f)Orelha de Mula. Chestnut leaf: (4) Aveleira, (^) BoaVentura, (œ) Judia, (g) Longal.


more to have much more high antioxidant activitythan some common standards (reducing powers ofBHA at 3.6mg/mL and a-tocopherol at 8.6mg/mLwere only 0.12 and 0.13, respectively).The extractsobtained with chestnut skins revealed better reducingproperties. This difference could be explained by thepresence of high amounts of reductones, which havebeen associated with antioxidant action due to breakingthe free radical chain by donating a hydrogen atom(Shimada et al., 1992).

In overall, chestnut skins revealed better antioxidantproperties (significantly lower EC50 values, p< 0.05;Table 2). The EC50 values obtained for these extractswere excellent (less than 110 mg/mL, average value), par-ticularly for LPO inhibition (less than 40 mg/mL, averagevalue). However, chestnut leaves (less than 220 mg/mL inaverage, for all assays) and almond husks (less than

260 mg/mL in average, for all assays) also revealed verygood antioxidant activity.

The obtained results are generally in agreement withthe total phenol and flavonoid contents determined foreach sample as shown in Table 1. The EC50 valuesobtained for lipid peroxidation inhibition were betterthan for reducing power, scavenging effects on DPPHradicals and b-carotene bleaching inhibition caused bylinoleate free radical, which were similar.

Other tree nuts had demonstrate their potential anti-oxidant activity namely walnuts (Anderson et al., 2001;Fukuda et al., 2004) and hazelnuts (Sivakumar andBacchetta, 2005; Alasalvar et al., 2006). Nevertheless,those studies were carried out with extracts from thefruits.

In previous works (Barreira et al., 2008; Sousa et al.,2008; Barros et al., 2007) we observed a significantly

Almond husk

0

20

40

60

80

100


DP

PH

sca

veng

ing

activ

ity (

%)

Chestnut leaf

0

20

40

60

80

100


DP

PH

sca

veng

ing

activ

ity (

%)

Chestnut skin

0

20

40

60

80

100


DP

PH

sca

veng

ing

activ

ity (

%)

0 200 400 600 800 1000

0 200 400 600 800 1000

0 200 400 600 800 1000

Figure 3. Radical scavenging activity (RSA) as afunction of extracts concentration. (4) Ferraduel, (^)Duro Italiano, (œ) Ferranhes, (g) Ferrastar, (f)Orelha de Mula. Chestnut leaf: (4) Aveleira, (^) BoaVentura, (œ) Judia, (g) Longal.

Almond husk

0

1

2

3

4


Red

ucin

g po

wer

(A

bs 7

00nm

)

Chestnut leaf

0

1

2

3

4


Red

ucin

g po

wer

(A

bs 7

00nm

)Chestnut skin

0

1

2

3

4


Red

ucin

g po

wer

(A

bs 7

00nm

)

0 200 400 600 800 1000

0 200 400 600 800 1000

0 200 400 600 800 1000

Figure 4. Reducing power as a function of extractsconcentration. (4) Ferraduel, (^) Duro Italiano, (œ)Ferranhes, (g) Ferrastar, (f) Orelha de Mula.Chestnut leaf: (4) Aveleira, (^) Boa Ventura, (œ)Judia, (g) Longal.


negative linear correlation between the total phenolscontent and EC50 antioxidant activity values. This neg-ative linear correlation proves that the samples withhighest total phenols content show lower EC50 values,confirming that phenols are likely to contribute to theantioxidant activity of the extracts, as it has beenreported in other species (Velioglu et al., 1998). The fla-vonoids contents were also correlated with EC50 scav-enging capacity values with similar correlationcoefficients values. Furthermore, approximately half ofthe results showed statistical significance, as it can beseen in Table 3. This may represent an important toolto predict this kind of bioactivity just by quantifyingphenols.In conclusion, all the assayed by-products revealed

good antioxidant properties, with very low EC50

values, particularly for lipid peroxidation inhibition,and might provide a novel strategy to stimulate theapplication of waste products as new suppliers ofuseful bioactive compounds, particularly antioxidants.This represents an additional advantage since almondand chestnut are important products, with high eco-nomic value, which originate high amounts of the stud-ied by-products.

ACKNOWLEDGMENTS

The authors are grateful to Foundation forScience and Technology (Portugal) for financial supportto J.C.M. Barreira (SFRH/BD/29060/2006) andINTERREG IIIA project PIREFI.

Table 2. EC50 values obtained in the antioxidant assays for chestnut and almond by-products andcorresponding coefficients of variation (%, in parenthesis).

Cultivar

EC50 values vs treatment* (mg/mL)

Bleaching inhibition LPO inhibition RSA Reducing power

Almond husk (AH) Duro Italiano 227.37 c (18.44) 29.20 d (2.65) 175.03 c (11.42) 206.96 c (20.63)Ferraduel 284.91 a (17.52) 103.52 a (6.78) 216.37 a (14.15) 376.30 a (27.67)Ferranhes 250.23 b (18.83) 39.95 c (3.63) 209.22 a (14.61) 218.11 c (21.06)Ferrastar 211.37 d (9.25) 28.11 d (1.15) 176.82 c (12.34) 169.85 d (4.53)Orelha de Mula 276.77 a (10.53) 74.15 b (3.61) 190.33 b (4.53) 306.46 b (22.13)

Chestnut leaf (CL) Aveleira 99.47 b (5.33) 78.32 b (6.01) 182.97 b (8.23) 210.09 b (18.92)Boa Ventura 99.09 b (5.37) 71.54 c (5.86) 161.34 c (9.08) 215.62 b (8.87)Judia 160.04 a (15.17) 133.52 a (5.60) 367.06 a (27.89) 267.00 a (26.54)Longal 64.14 c (3.76) 69.04 c (3.53) 129.91 d (5.02) 152.38 c (2.39)

Chesnut Skin (CS) Aveleira 151.27 a (15.55) 49.07 a (4.83) 159.99 a (15.37) 117.58 a (12.71)Boa Ventura 74.62 d (8.92) 27.29 d (0.48) 82.41 c (5.52) 79.25 d (6.39)Judia 86.07 c (7.16) 30.47 c (2.05) 86.52 c (7.77) 104.61 b (8.22)Longal 120.84 b (7.84) 34.53 b (3.21) 108.87 b (6.73) 94.55 c (6.31)

AH �x 250.13 a (32.03) 54.98 b (29.82) 193.56 a (20.52) 255.53 a (78.19)CL �x 105.68 b (35.71) 88.10 a (27.08) 210.32 a (94.11) 211.27 b (44.08)CS �x 108.20 b (31.97) 35.34 c (8.90) 109.45 b (32.44) 99.00 c (16.54)

*Mean values followed by different letter within a column are significant different (p<0.05).

Table 3. Correlations established between total phenols and flavonoids with antioxidant activity EC50 values.

Total phenols Flavonoids

Equation R2 F Sign. Equation R2 F Sign.

Almond husk Bleaching inhibition y¼�0.0001 xþ 0.3086 0.584 4.218 n.s. y¼�0.0003 xþ 0.3073 0.937 44.610 **LPO inhibition y¼�0.0001 xþ 0.1096 0.463 2.590 n.s. y¼�0.0003 xþ 0.1080 0.733 18.238 n.s.RSA y¼�0.0001 xþ 0.2386 0.976 120.893 ** y¼�0.0001 xþ 0.2245 0.774 10.269 *Reducing Power y¼�0.0002 xþ 0.3964 0.473 2.6886 n.s. y¼�0.0008 xþ 0.3942 0.769 9.979 n.s.

Chestnut leaf Bleaching inhibition y¼�0.0003 xþ 0.2312 0.962 50.278 * y¼�0.0056 xþ 0.5686 0.990 208.436 *LPO inhibition y¼�0.0002 xþ 0.1825 0.927 25.419 * y¼�0.0040 xþ 0.4162 0.848 11.133 n.s.RSA y¼�0.0008 xþ 0.5452 0.955 42.044 * y¼�0.0143 xþ 1.3957 0.905 19.055 *Reducing Power y¼�0.0003 xþ 0.3507 0.857 12.020 n.s. y¼�0.065 xþ 0.7466 0.957 44.141 *

Chestnut skin Bleaching inhibition y¼�0.0003 xþ 0.2949 0.830 9.738 n.s. y¼�0.0008 xþ 0.1800 0.866 12.890 n.s.LPO inhibition y¼�0.0001 xþ 0.0916 0.984 121.371 ** y¼�0.0002 xþ 0.0537 0.742 5.736 n.s.RSA y¼�0.0003 xþ 0.3150 0.958 45.713 * y¼�0.0007 xþ 0.1759 0.708 4.851 n.s.Reducing Power y¼�0.0001 xþ 0.1811 0.741 5.731 n.s. y¼�0.0003 xþ 0.1299 0.741 5.727 n.s.

* p�0.05, ** p� 0.01, *** p� 0.001, s.n., not significant correlation.


REFERENCES

Alasalvar C., Karamaca M., Amarowicz R. and Shahidi F. (2006).Antioxidant and antiradical activities in extracts of hazelnutkernel (Corylus avellana L.) and hazelnut green leafy cover.Journal of Agriculture and Food Chemistry 54: 4826�4832.

Anderson K.J., Teuber S.S., Gobeille A., Cremin P., Waterhouse A.L.and Steinberg F.M. (2001). Walnut polyphenolics inhibit in vitrohuman plasma and LDL oxidation. The Journal of Nutrition 131:2837�2842.

Barreira J.C.M., Ferreira I.C.F.R., Oliveira M.B.P.P. and Pereira J.A.(2008). Antioxidant activity of the extracts from chestnut flower,leaf, skins and fruit. Food Chemistry 107: 1106�1113.

Barros L., Baptista P. and Ferreira I.C.F.R. (2007). Effect of Lactariuspiperatus fruiting body maturity stage on antioxidant activitymeasured by several biochemical assays. Food and ChemicalToxicology 45: 1731�1737.

Calliste C.A., Trouillas P., Allais D.P. and Duroux J.L. (2005).Castanea sativa Mill. leaves as new sources of natural antioxi-dant: an electronic spin resonance study. Journal of Agricultureand Food Chemistry 53: 282�288.

Cordeiro V. and Monteiro A. (2001). Almond growing in Tras-os-Montes region (Portugal). Acta Horticulturae 591: 161�165.

Engler M.B. and Engler M.M. (2006). The emerging role of flavonoid-rich cocoa and chocolate in cardiovascular health and disease.Nutrition Reviews 64: 109�118.

Fang Y.Z., Yang S. and Wu G. (2002). Free radicals, antioxidants,and nutrition. Nutrition 18: 872�879.

Fukuda T., Ito H. and Yoshida T. (2004). Effect of the walnut poly-phenol fraction on oxidative stress in type 2 diabetes mice.BioFactors 21: 251�253.

Jayaprakasha G.K., Singh R.P. and Sakariah K.K. (2001).Antioxidant activity of grape seed (Vitis vinifera) extracts on per-oxidation models in vitro. Food Chemistry 73: 285�290.

Jia Z., Tang M. and Wu J. (1999). The determination of flavonoidcontents in mulberry and their scavenging effects on superoxideradicals. Food Chemistry 64: 555�559.

Lau F.C., Shukitt-Hale B.J. and Joseph A. (2005). The beneficialeffects of fruit polyphenols on brain aging. Neurobiology ofAging 26: 128�132.

Liu R.H. (2003). Health benefits of fruits and vegetables are fromadditive and synergistic combination of phytochemicals.American Journal of Clinical Nutrition 78: 517 S�520 S-9.

Martins M., Tenreiro R. and Oliveira M.M. (2003). Genetic related-ness of Portuguese almond cultivars assessed by RAPD and ISSRmarkers. Plant Cell Reports 22: 71�78.

Milbury P.E., Chen C.Y., Dolnikowski G.G. and Blumberg J.B.(2006). Determination of flavonoids and phenolics and their dis-tribuition in almonds. Journal of Agriculture and Food Chemistry54: 5027�5033.

Nichenametla S.N., Taruscio T.G., Barney D.L. and Exon J.H. (2006).A review of the effects and mechanisms of polyphenolics incancer. Critical Review in Food Science and Nutrition 46:161�183.

Pellegrini N., Serafini M., Salvatore S., Del Rio D., Bianchi M. andBrighenti F. (2006). Total antioxidant capacity of spices, driedfruits, nuts, pulses, cereals and sweets consumed in Italy assessedby three different in vitro assays. Molecular Nutrition and FoodResearch 50: 1030�1038.

Pereira J.A., Pereira A.P.G., Ferreira I.C.F.R., Valentao P., AndradeP.B., Seabra R., Estevinho L. and Bento A. (2006). Table olivesfrom Portugal: phenolic compounds, antioxidant potential andantimicrobial activity. Journal of Agriculture and Food Chemistry54: 8425�8431.

Pinelo M., Rubilar M., Jerez M., Sineiro J. and Nunez M.J. (2004).Extraction of antioxidant phenolics from almond hulls (Prunusamygdalus) and pine sawdust (Pinus pinaster). Food Chemistry 85:267�273.

Ribeiro B., Rangel J., Valentao P., Andrade P.B., Pereira J.A., BolkeH. and Seabra R.M. (2007). Organic acids in two Portuguesechestnut (Castanea sativa Miller) varieties. Food Chemistry 100:504�508.

Sang S., Lapsley K., Jeong W.S., Lachance P.A., Ho C.H. and RosenR.T. (2002). Antioxidative phenolic compounds isolated fromalmond skins (Prunus amygdalus Batsch). Journal of Agricultureand Food Chemistry 50: 2459�2463.

Shimada K., Fujikawa K., Yahara K. and Nakamura T. (1992).Antioxidative properties of xanthan on the autoxidation of soy-bean oil in cyclodextrin emulsion. Journal of Agriculture and FoodChemistry 40: 945�948.

Silva E.M., Souza J.N.S., Rogez H., Rees J.F. and Larondelle Y.(2007). Antioxidant activities and polyphenolic contents of fif-teen selected plant species from the Amazonian region. FoodChemistry 101: 1012�1018.

Singleton V.L. and Rossi J.A. Jr (1965). Colorimetric oftotal phenolics with phosphomolybdic-phosphotungstic acidreagents. American Journal of Enology and Viticulture 16:144�158.

Sivakumar G. and Bacchetta L. (2005). Determination of natural vita-min E from Italian hazelnut leaves. Chemistry of NaturalCompounds 41: 654�656.

Sousa A., Ferreira I.C.F.R., Barros L., Bento A. and Pereira J.A.(2008). Effect of solvent and extraction temperatures on the anti-oxidant potential of traditional stoned table olives ‘‘alcaparras’’.LWT- Food Science and Technology 41: 739�745.

Takeoka G.R. and Dao L.T. (2003). Antioxidant constituents ofalmond (Prunus dulcis (Mill.) D.A. Webb) hulls. Journal ofAgriculture and Food Chemistry 51: 496�501.

Velioglu Y.S., Mazza G., Gao L. and Oomah B.D. (1998).Antioxidant activity and total phenolics in selected fruits, vege-tables, and grain products. Journal of Agriculture and FoodChemistry 46: 4113�4117.

Weaver C.M. and Cheong J.M. (2005). Soy isoflavones and bonehealth: the relationship is still unclear. Journal of Nutrition 135:1243�1247.

Weisburger J. (1999). Mechanisms of action of antioxidants as exem-plified in vegetables, tomatoes and tea. Food and ChemicalToxicology 37: 943�948.


http://fst.sagepub.com/International

Food Science and Technology

http://fst.sagepub.com/content/16/3/277The online version of this article can be found at:

DOI: 10.1177/1082013209353830

2010 16: 277 originally published online 12 August 2010Food Science and Technology InternationalC. Chotimarkorn, N. Silalai and N. Chaitanawisuit

Post-mortem Changes in Farmed Spotted Babylon Snail (Babylonia areolata) During Iced Storage

Published by:

http://www.sagepublications.com

On behalf of:

Consejo Superior de Investigaciones Científicas (Spanish Council for Scientific Research)

can be found at:Food Science and Technology InternationalAdditional services and information for

http://fst.sagepub.com/cgi/alertsEmail Alerts:

http://fst.sagepub.com/subscriptionsSubscriptions:

http://www.sagepub.com/journalsReprints.navReprints:

http://www.sagepub.com/journalsPermissions.navPermissions:

http://fst.sagepub.com/content/16/3/277.refs.htmlCitations:

at HINARI on February 22, 2011fst.sagepub.comDownloaded from

http://fst.sagepub.com/

http://fst.sagepub.com/content/16/3/277


http://www.iata.csic.es/

http://fst.sagepub.com/cgi/alerts

http://fst.sagepub.com/subscriptions

http://www.sagepub.com/journalsReprints.nav

http://www.sagepub.com/journalsPermissions.nav

http://fst.sagepub.com/content/16/3/277.refs.html


Post-mortem Changes in Farmed Spotted Babylon Snail

(Babylonia areolata) During Iced Storage

C. Chotimarkorn,1,* N. Silalai2 and N. Chaitanawisuit3

1Faculty of Science and Industrial Technology, Prince of Songkla University, Surat Thani Campus,Muang Surat Thani 84100, Thailand

2Department of Food and Nutritional Sciences, University College Cork, Cork, Ireland3Aquatic Resources Research Institute, Chulalongkorn University, Bangkok 10330, Thailand

Post-mortem changes in farmed spotted Babylon snail stored in ice for 7 days were evaluated usingnucleotide degradation products, K-value, total volatile base nitrogen (TVB-N), trimethylamine nitrogen(TMA-N), organic acid, free amino acids and biogenic amines. During a 7-day ice storage, K-value,TVB-N, TMA-N and organic acids contents increased with increasing storage time (p< 0.05). Changes

in free amino acid, such as aspartic acid, serine, glutamic acid, glycine, arginine and proline were observedthroughout the ice storage (p< 0.05), while total free amino acids were found to decrease significantly(p< 0.05). Biogenic amines found in snail muscle during ice storage were tyramine, putrescine, cadaverine,

agmatine and histamine. Bacteria counts of snail muscle exceeded 7 logCFU/g, which was considered asthe limit for acceptability after 7 days of iced storage. This result initiates the use of ice storage as apreliminary treatment for snails transported from farms.

Key Words: ice storage, spotted Babylon snail, biogenic amines, free amino acids

INTRODUCTION

Seafood products have attracted considerable atten-

tion as sources of nutrients in the human diet. Post-

mortem changes in fish or shellfish tissue depend upon

the endogenous enzymes, microbial contamination and

post-catch handling (Iwamoto et al., 1987). The rate of

quality loss depends on fish or shellfish species and stor-

age condition (Olafsdottir et al., 1997). To decrease the

processes involved in quality loss, fish and shellfish

should be refrigerated immediately. Freshness makes a

major contribution to the quality of fish and fishery

products. Freshness and technological functionality of

fish and shellfish muscle are rapidly lost if stored at

high temperature. Flaked ice has been used as a tradi-

tional system to cool fresh seafood products to a final

temperature slightly above 0 �C to maintain freshness

during storage or transportation (Heen, 1982). The

effects of ice storage on the quality of several fish and

shellfish species has been studied by several authors

(Mazorra-Manzano et al., 2000; Scherer et al., 2005;

Kilinc et al., 2007; Marquez-Rıos et al., 2007). Duringstorage of fish or shellfish in ice, significant deteriorationlosses of nutritional value have been detected due to theeffects of a variety of degradation mechanisms (Ashieet al., 1996).

Spotted Babylon snails are marine invertebrates thatare important products in Thailand, due to a growingdemand and an expanding domestic market for seafood(Chaitanawisuit et al., 2002). Due to its relatively newimportance as a marketable product and the lack ofinformation regarding post-mortem changes, a studyof these changes in snails during ice storage is required.Basic knowledge of post-mortem biochemical changesduring ice storage of spotted Babylon snails wouldgreatly benefit its utilization and processing for humanfood as well as improve commercial practice. The objec-tive of this study was to investigate post-mortem andquality changes in spotted Babylon snail cultured inThailand during ice storage.


Materials

Snail Samples

Live spotted Babylon snails (Babylonia areolata),with an average weight of 8�10 g, were purchasedfrom a farm in Petchburie Province, Thailand, in

*To whom correspondence should be sent(e-mail: [email protected]).Received 26 February 2009; revised 30 April 2009.


277 at HINARI on February 22, 2011fst.sagepub.comDownloaded from


January 2008. Live snails with shells were stored in anisothermic box containing sea water, and transported tothe Faculty of Science and Industrial Technology, Princeof Songkla University, Surat Thani Campus within 6 h.Whole snails with shells were immediately washed withtap water and kept in flaked ice at a 1 : 2 (w/w) ratio ofsnail-to-ice for 30min until their death. They were ran-domly divided into 5.0 kg lots and packed in ice flakes,(snail-to-ice ratio 1 : 2, w/w) then placed into polystyreneboxes and stored in a cool room at 2�4 �C. To maintainthe ice flake content, new ice replaced the melted iceevery 24 h. During storage, 50 snails were randomlypicked out for analysis daily. All analyses were per-formed in triplicate.

Chemicals

Standards for ATP-related compounds (ATP, adeno-sine triphosphate), amino acids and biogenic amineswere purchased from Sigma�Aldrich Chemical Co.Ltd (St. Louis, MO, USA). These compounds were asfollows: ATP-related compounds (adenosine 50-tripho-sphate disodium salt, adenosine 50-diphosphate sodiumsalt, adenosine 50-monophosphate sodium salt, inosine50-monophosphate disodium salt, inosine and hypoxan-thine), amino acid standards (aspartic acid (asp), serine(ser), glutamic acid (glu), glycine (gly), histidine (his),arginine (arg), threonine (thr), alanine (ala), proline(pro), cysteine (cys), tyrosine (tyr), valine (val), methio-nine (met), lysine (lys), isoleucine (ile), leucine (leu) andphenylalanine (phe)), biogenic amines standards (tyra-mine hydrochloride, putrescine dihydrochloride, cadav-erine dihydrochloride, agmatine sulfate and histaminedihydrochloride). High performance liquid chromatog-raphy (HPLC)-grade organic solvents were purchasedfrom the British Drug House (BDH) (Poole, UK) orMerck (Darmstadt, Germany). Other chemicals andorganic solvents used in this study were of analyticalgrade without further purification.

Methods

Determination of ATP-related Compounds

Nucleotide extracts were prepared and concentrationsdetermined according to the method of Veciana-Nogueset al. (1997) using HPLC with some modification. Inbrief, 10.0 g of snail meat was homogenized with15mL of 0.6N HClO4 at 0

�C for 1min with a homog-enizer. The homogenate was centrifuged at 1500� g for10min and 10.0mL of supernatant was neutralized topH 6.5 with 0.1N KOH and left as such for 30min at4 �C. KClO4 was removed by filtration through a 0.2mmcellulose acetate membrane syringe filter and storedat �80 �C until analysis. Reversed phase (RP)-HPLCwas performed using an Agilent 1100 series (Palo Alto,CA, USA) equipped with a Hypersil ODS column

(4.0� 250mm2, 5 mm, Agilent Technologies, Palo Alto,CA, USA) and UV�Vis detector (model G1379A) at254 nm. The mobile phase used was 0.1M phosphatebuffer pH 7.0 (0.04M KH2PO4 and 0.06M K2HPO4)with a flow rate 0.75mL/min. The content of ATP-related compounds were identified by their relativeretention times and quantified based upon their peakareas using standard curves for ATP, adenosine diphos-phate (ADP), adenosine monophosphate (AMP), aden-osine (AdR), inosine monophosphate (IMP), inosine(HxR) and hypoxanthine (Hx). K-value was calculatedin accordance with the method of Saito et al. (1959)using this formula; K-value (%)¼ ([HxR]þ [Hx])/([ATP]þ [ADP]þ [AMP]þ [IMP]þ [HxR]þ [Hx])� 100.

Determination of Total Volatile Base Nitrogen

Total volatile base nitrogen (TVB-N) was measured inaccordance with the method of Woyewoda et al. (1986)using distillation and titration. In brief, 10 g of thesample was homogenized with 300mL of distilledwater. This homogenate was transferred to a 1000mLround bottom distillation flask containing 2.0 g of MgO.The distillation flask was connected to a vertical distil-lation apparatus and heated for 25min. The condensatewas received in a flask containing 2.0% of boric acidsolution and titrated back to the original color using a0.05N H2SO4 standard solution. TVB-N was expressedas mgN/100 g sample.

Determination of Trimethylamine Nitrogen

Trimethylamine nitrogen (TMA-N) was measured inaccordance with the method of Woyewoda et al. (1986)using a spectrophotometer. In brief, 50 g of sample washomogenized in 100mL of 7.5% trichloroacetic acid(TCA) solution and centrifuged at 4 �C for 10min at2000� g. The supernatant was filtered throughWhatman No. 4 filter paper. Then 1.0mL of 10% form-aldehyde, 10.0mL of toluene and 3.0mL of 25% KOHwere added to 2mL of extract. The mixture was vigor-ously agitated for 10min, and then 7.0mL of the super-natant was transferred to a tube containing 0.5 ganhydrous Na2SO4. An aliquot (5mL) of the clarifiedsolution was added to 5.0mL of picric acid to developthe color. The absorbance was measured at 410 nm byusing UV�Vis spectrophotometer (model LambdaEZ201 UV�Vis spectrophotometer, Perkin Elmer,Waltham, MA, USA). TMA-N contents were calculatedfrom a standard curve and expressed as mgN/100 gsample.

Determination of Organic Acids

Organic acids content were extracted according tothe method of Wongso and Yamanaka (1996) and

278 C. CHOTIMARKORN ET AL.



determined by HPLC in accordance with the method ofAguiar et al. (2005). In brief, 5 g of sample was homog-enized with 15.0mL of 6% HClO4 at 0 �C for 1min.This suspension was centrifuged at 1500� g for10min, and then the supernatant was neutralized withKOH and made up to 25.0mL with distilled water. Itwas then filtered through a 0.2-mm cellulose acetatemembrane syringe filter and stored at �80 �C until ana-lyzed. HPLC was performed using a Shimadzu 6A(Kyoto, Japan) equipped with an ionic exchangeAminex HPX-87H column (7.8� 300mm2, 5 mm, Bio-Rad Laboratory (Hemel Hemstead, UK), and aUV�vis detector (model SPD-10AV, Shimadzu,Kyoto, Japan) at 210 nm. The mobile phase used was0.005M of H2SO4 (pH 2.1) at a flow rate 0.75mL/min.The amount of succinic, malic and lactic acid was quan-tified, based upon the peak area using the appropriatestandard curves.

Determination of Biogenic Amines

Biogenic amines were determined in accordance withthe method of Ozogul et al. (2002) with some modifica-tion. In brief, 10 g of the sample was homogenized with50mL of 1% TCA at 0 �C for 1min and filtered throughWhatman No. 1 filter paper. For derivatization, 2.0mLof supernatant was added to 1.0mL of 2.0M NaOH,followed by 1.0mL of 2% benzoyl chloride andshaken with a vortex mixer for 1min. This mixturewas left at 25 �C for 30min and then 2.0mL of saturatedNaCl was added. Amines were extracted twice with2.0mL of diethyl ether and the pooled ether extractevaporated under a stream of N2 gas. The residue wasdissolved with acetonitrile before analysis. HPLC wasperformed using a Shimadzu 6A (Kyoto, Japan)equipped with a Spherisorb 5 Si C18 column(4.6� 250mm2) (Phenomenex, Macclesfield, UK) anda diode array detector (model SPD-M20A, Shimadzu,Kyoto, Japan) at 254 nm. The mobile phase used was agradient solution of acetonitrile and water at a flow rate1mL/min. Biogenic amines content was quantifiedbased upon peak area using appropriate standardcurves.

Determination of Free Amino Acid

Amino acids were determined in accordance with themethod of Aquino et al. (2008) using o-phthaldehyde(OPA) and 2-mercaptoethanol derivatization combinedby HPLC with fluorescence detection. HPLC was per-formed using a Shimadzu 6A (Kyoto, Japan) equippedwith a fluorescence detector (model, RF-10AXL,Shimadzu, Kyoto, Japan). The excitation and emissionwavelengths were set at 340 and 440 nm for measuringamino acid derivatives, with the exception that the cysand pro derivatives were monitored at excitation andemission wavelengths of 425 and 450 nm. In brief,

10 g of sample was homogenized with 15mL of 0.6NHClO4 at 0 �C for 1min. The homogenate was centri-fuged at 1500� g for 10min, 10.0mL of supernatantwas neutralized to pH 6.5 with 0.1N KOH and thenleft for 30min at 4�C. KClO4 was removed by filteringthrough a 0.2 mm cellulose acetate membrane syringefilter and the filtrate stored at �80 �C until analyzed.For derivatization of the amino acids, 1.0mL of super-natant was mixed with 2.0mL of derivatizing solution.This solution was mixed and allowed reacting at roomtemperature for 1min; then an aliquot was withdrawnand injected into the HPLC.

Microbiological Analysis

The viable mesophilic and psychrophilic bacterialcounts were performed by the pour plate method usingplate count agar (Difco

TM

, 0479-17), according tothe method of Harrigan and McCance (1976).Approximately 10 g of snail muscle was dissected asep-tically from the iced whole shell snail specimens, mixedwith 90.0mL of 0.1% peptone water (Difco

TM

, 0118-17-0)and homogenized in a stomacher for 1.0min. Serial dilu-tions from the microbial extracts were prepared in 0.1%peptone water. The plates were incubated at 30 �C for48 h and 5 �C for 72 h for mesophilic counts and psychro-philic counts, respectively.


The experimental data were evaluated by a one-wayANOVA followed by Tukey’s honestly significant dif-ference (HSD) were performed using SPSS 11.0 (SPSSInc., Chicago, IL, USA) to analyze and compare thedata. Results were represented for mean±SD (n¼ 3)and p-values <0.05 were regarded as statisticalsignificance.


The nucleotide breakdown patterns of snail meatstored in ice is shown in Figure 1. At the beginning ofice storage, ATP, ADP, AMP and IMP contents werefound to be 4.8±0.3, 3.8±0.2, 2.7±0.2 and1.4±0.1 mmole/g, respectively. While AdR, HxR andHx contents were found to be very low at 0.5±0.02,0.2±0.02 and 0.4±0.01mmole/g, respectively. Duringiced storage, a small increase in ATP content wasobserved in snail meat after day 1 of iced storage(p� 0.05). The ATP content increased a short timeafter the snails died because the nucleotide was regener-ated by the degradation of phosphagen prior to thedestruction of ATP (Iwamoto et al., 1988). Similarly,results were observed in several marine invertebratesstored in ice or chilled conditions as reported by differ-ent authors (Watanabe et al., 1992a; Massa et al., 2002).

Post-mortem Changes in Farmed Spotted Babylon Snail 279



A marked decrease in ATP content was observed fromdays 1 to 5 (p< 0.05) after which it remained stable untilday 7 of iced storage (0.3±0.02 mmole/g sample). Thephenomenon that ATP remained stable was alsoobserved in scallops stored at 4�6 �C (Massa et al.,2002). However, the reason is still obscure; therefore,further study is required to elucidate the ATP enzymaticdegradation pathway in snail muscle during storage.

The fact that ATP degradation proceeded with timewas also reported by Saito et al. (1959). Similar resultsregarding the rapid ATP degradation in several marineinvertebrates during ice or chilled storage has beenreported by different authors (Watanabe et al., 1992a,b; Wongso and Yamanaka, 1996, 1998; Yamanaka andShimada, 1996; Shimada et al., 1998).

ADP content increased until day 3 (5.1±0.2mmole/gsample) and then decreased gradually until the end ofstorage.

AMP has been reported to be a useful index of tastefor fish and shellfish since AMP enhanced the umamitaste in abalone, crab and scallop (Konosu et al., 1988).In spotted Babylon snails, a small decrease in AMP con-tent was observed in snail meat after day 1 of iced stor-age (p� 0.05). After that, AMP content increased to day5 of iced storage (p< 0.05), but thereafter, a significantdecrease in AMP content was found between days 5 and7 (p< 0.05, Figure 1). The degradation pathway of AMPvaries with the fish and invertebrate species. Severalresearchers have reported that AMP decomposes viaIMP in fish (Mazorra-Manzano et al., 2000) and crus-taceans, such as prawns (Matsumoto and Yamanaka,1990), but via AdR in mollusks (Suweja et al., 1989).

IMP provides the characteristic sweetness (umamiflavor) of fresh fish muscle (Kawai et al., 1992;Chruch, 1998). No changes in IMP and AdR contentswere found during the 7-day iced storage (p� 0.05). Thisresult indicates that AMP was metabolized throughboth the IMP and AdR pathways. Our result is in agree-ment with several researchers who reported that AMP in

some marine invertebrates, such as disk abalone(Watanabe et al., 1992a,b), noble scallop (Wongso andYamanaka, 1996), Japanese baking scallop (Wongsoand Yamanaka, 1998) and lobster (Shimada et al.,1998), is decomposed via both IMP and AdR duringiced or chilled storage.

Hypoxanthine accumulation in fish and marine inver-tebrates’ muscles reflects the initial phase of autolyticdeterioration as well as bacteria spoilage (Woyewodaet al., 1986). In this study, the Hx content of snailmuscle increased as the storage time increased. Theincrease in Hx was relatively small during the first 4days of iced storage, but thereafter it increased rapidlyup to 7 days (3.7±0.3 mmole/g sample, p< 0.05).

The K-value has been proposed as an index for assess-ing ATP degradation as a measure of the loss of fresh-ness during the storage of fish and marine invertebrates(Woyewoda et al., 1986). The changes in the K-value ofsnail meat during iced storage in this study are shown inFigure 2. The K-value of the snail meat samplesincreased as the storage time increased (p< 0.05). Atthe beginning of iced storage, the initial K-value was2.1±0.1%. This value is similar to that given in a pre-vious report where the initial value of K in fish andmarine invertebrates’ muscles immediately after capturewas <10% (Sikorkski et al., 1990). In general, the upperlimit for K-value before the rejection of freshwater andmarine fish usually exceeded 60% (Ehira and Uchiyama,1987). Thus, the increase in K-value with increasing stor-age time could be used to indicate the deterioration andunacceptability of snail meat during storage.

The K-value increased sharply during the 7 days oficed storage, reaching a K-value of 45.1±2.9%. Theincrease in the K-value indicated the progressive deteri-oration and increased unacceptability of the snail meatduring iced storage.

The amount of TVB-N in fish or marine products hasbeen used as an indicator of bacterial spoilage duringstorage. The pattern of change in TVB-N content insnail meat during the 7-day storage period is shown inFigure 3. TVB-N content increased with increasing

0

1

2

3

4

5

6

7

8

0 2 4 6 8

Storage time (days)

Con

tent

s (m

mol

e/g

sam

ple)

Figure 1. Changes in ATP-related compound(mmole/g sample) in snail muscle during ice storage.Values are mean±SD (n¼3). (�) ATP; (*) ADP;(m) AMP; (i) IMP; (#) AdR; («) Ino; and (¨) Hx.

0

10

20

30

40

50

60

0 2 4 6 8Storage time (days)

K-v

alue

(%

)

Figure 2. Changes in K-value (%) in snail muscleduring ice storage. Values are mean±SD (n¼ 3).




storage time (p< 0.05). The TVB-N content of thesample was 0.9±0.1mg/100 g at the beginning of stor-age. This was probably due to the endogenous produc-tion of ammonia from the enzymatic degradation ofprotein, amino acids and nucleotides immediately post-mortem. TVB-N content increased rapidly from days 4to 7 of ice storage time. At the end of iced storage, themean TVB-N content in samples was 33.6±2.6mg/100 g. The limit for TVB-N content in edible fish andfishery products has been reported to be 30.0mg/100 g(Woyewoda et al., 1986).The levels of TMA-N produced in the muscle during

iced storage could be used to indicate the quality of fishand shellfish. TMA-N is considered a valuable tool inthe evaluation of the quality of fish stored in ice becauseof its rapid accumulation in muscle under refrigeratedconditions (Kryzmien and Elias, 1990). In this study, theinitial mean TMA-N content of snail meat was0.1±0.01mg/100 g. TMA-N content of the sampleincreased slightly during the first 2 days of iced storage(0.14±0.03mg/100 g, p� 0.05). After this time, amarked increase in TMA-N was observed until day 7of iced storage (5.7±0.6mg/100 g, p< 0.05). In general,the upper limit for TMA-N before consumer rejection offish is usually 5.0�10.0mg/100 g (Sikorkski et al., 1990).Thus, the increase in TVB-N and TMA-N contents withincreasing storage time could be used to indicate thedeterioration and unacceptability of snail meat at theend of storage.Succinic, malic and lactic acids are the final products

of glycolysis in many fish and shellfish. Changes in thecontent of these acids in snail meat during iced storageare shown in Figure 4. At the beginning of iced storage,the mean succinic acid content was 3.9±0.9mg/100 g,while the malic and lactic acid contents were 0.7±0.03and 0.4±0.03mg/100 g, respectively. The succinic acidcontent of snail meat increased markedly (p< 0.05) in

the first 3 days of storage, but then decreased signifi-cantly until the end of the 7-day storage period(p< 0.05). This was possibly due to the degradation ofsuccinic acid by bacteria. The changes in succinic acidcontent during iced storage suggested that succinic acidmight be a temporarily terminal product of glycolysis insnail meat during anaerobic metabolism. Malic andlactic acid contents increased slightly during the 7-daystorage period. At the end of storage, succinic, malic andlactic acids contents were 31.7±1.9, 25.5±2.5 and13.8±1.5mg/100 g, respectively. Lactic and malic acidscontribute to the pH level, which is one of the indices forfreshness of fish and fisheries products. Wongso andYamanaka (1996) also found an increase in lactic acidcontent during chilled storage of the noble scallop. Theaccumulation of organic acids in snail meat could beattributable to glycolysis, which was enhanced byextended storage.

Free amino acids (FAAs), which are one of the non-protein nitrogen sources in fish and marine inverte-brates, are responsible for their flavor and taste.Yamanaka and Shimada (1996) reported that alanine,glycine and glutamic acid showed a sweet taste andumami. At the beginning of iced storage, the meantotal amount of FAAs was 1144.5±67.8mg/100 g,decreasing to 766.2±43.1mg/100 g after 7 days of icedstorage (Table 1). During storage, decreases in asparticacid, glutamic acid, glycine, arginine and proline con-tents were observed (p< 0.05), whereas alanine contentincreased (p< 0.05). The total amount of FAAsincreased after storage for 1 day and then decreasedcontinuously. The cause for the increment of the totalamount of FAAs was probably the activity of endoge-nous proteolytic enzymes, while the decrease in FAAs

0

10

20

30

40

50

60

70

80

0 2 4 6 8

Storage time (days)

Con

tent

(m

g/10

0g

sam

ple)

Figure 4. Changes in succinic, malic and lactic acids(mg/100 g sample) in snail muscle during ice storage.Values are mean±SD (n¼ 3). (�) Succinic acid;(*) malic acid; and (n) lactic acid.

0

5

10

15

20

25

30

35

40

0 2 4 6 8Storage time (days)

mg/

100

g sa

mpl

e

Figure 3. Changes in TVB-N (mg/100 g sample) andTMA-N (mg/100 g sample) in snail muscle during icestorage. Values are mean±SD (n¼ 3). (�) TVB-N and(*) TMA-N.




was caused by the breakdown of FAAs as a result ofbacterial enzymes.

Biogenic amines are generated by microbial decarbox-ylation of amino acids in food products. The adverseeffects on human health and food quality of biogenicamines, such as histamine, putrescine, cadaverine, tyra-mine and agmatine make it important to monitor theirlevels in fish and shellfish products (Chu and Bjeldanes,1981; Stratton et al., 1991). Tyramine, putrescine, cadav-erine, agmatine and histamine were not found in snailmeat at the beginning of storage (Table 2). These bio-genic amines increased with more storage time. A signif-icant rise in biogenic amine contents was observedthroughout the 7-day storage (p< 0.05). In this study,tyramine and histamine levels increased more slowlythan those of putrescine, agmatine and cadaverine andwere not detected until days 5 and 6, respectively. Ourresults are in agreement with those of Marks (Rupp) and

Anderson (2005), who reported that histamine is notalways found in spoiled fish or shellfish and that putres-cine as well as cadaverine may be a better indicator fortheir decomposition. Putrescine, agmatine and cadaver-ine were the most useful indicators for freshness amongbiogenic amines since these amines increased in propor-tion to the storage time.

The initial quality of the whole shell snails used in thisstudy was good, indicated by a low initial number bacte-ria (<103CFU/g) before the snails were stored. Initialpsychrophilic and mesophilic viable counts of snailmuscle were 2.03±0.17 and 2.00±0.12 logCFU/g.Total viable psychrophilic and mesophilic bacterialcounts increased throughout the storage period(Table 3). The storage of snail muscle in ice delayedmicrobial growth and lag phase of microbial growth atleast 48h was observed. After that, a significant increasein psychrophilic and mesophilic viable counts was

Table 1. FAAs in snail muscle during ice storage.

FAA

FAA content (mg/100 g sample)

Storage time (days)

0 1 2 3 4 5 6 7

Aspartic acid 2.8±0.3 c 2.4±0.2 bc 2.0±0.3 ab 2.2±0.3 abc 2.1±0.2 abc 1.9±0.2 ab 1.7±0.3 ab 1.5±0.2 aSerine 19.9±1.6 abc 1.2±1.9 bc 22.3±1.6 c 19.5±2.1 abc 18.1±1.3 abc 8.0±1.2 abc 17.4±2.0 ab 16.6±0.7 aGlutamic acid 144.4±12.2 d 132.4±14.3 d 122.1±10.3 cd 102.4±9.9 bc 96.5±8.7 bc 81.2±9.8 ab 74.3±6.6 ab 62.1±5.5 aGlycine 22.2±0.8 bc 24.3±0.7 cd 26.1±0.7 d 26.4±0.7 d 22.9±0.6 c 20.1±1.3 ab 19.3±1.4 a 17.4±1.1 aHistidine 24.8±1.1 24.5±1.7 24.3±1.4 23.7±2.1 24.3±1.4 23.6±2.5 23.1±2.4 24.9±2.3Arginine 599.9±67.8 d 662.0±43.2 d 478.9±34.2 c 455.8±39.8 bc 412.6±32.2 abc 389.7±27.6 abc 354.2±22.3 ab 322.1±18.9 aThreonine 27.7±1.9 27.0±2.1 28.3±2.3 27.6±1.9 28.2±2.5 26.5±3.2 26.9±2.8 27.5±1.9Alanine 132.1±11.1 144.3±12.2 149.9±13.2 156.7±11.7 159.8±13.1 155.3±14.2 150.9±13.3 145.9±15.6Proline 29.9±3.1 b 28.6±2.1 b 27.8±1.6 b 27.5±2.3 b 25.4±1.9 ab 24.6±1.1 ab 21.2±2.5 a 20.1±2.4 aCystine 29.9±3.1 b 28.6±2.1 b 27.8±1.6 b 27.5±2.3 b 25.4±1.9 ab 24.6±1.1 ab 21.2±2.5 a 20.1±2.4 aTyrosine 18.9±1.3 18.5±1.5 18.8±1.6 19.0±2.3 18.1±2.7 18.6±2.6 17.9±2.7 18.0±2.4Valine 17.8±1.5 16.3±1.1 17.3±2.2 16.7±2.4 17.0±2.1 7.5±2.4 16.5±1.5 15.3±1.1Methionine 16.7±1.3 16.0±1.5 17.0±1.7 16.6±2.1 16.9±2.1 17.3±1.9 16.9±2.3 17.7±2.1Lysine 31.2±2.7 2.3±1.9 30.3±2.8 31.8±1.8 30.7±2.6 31.4±3.1 32.4±2.9 32.2±2.8Isoleucine 15.4±2.1 14.6±2.3 15.9±3.2 16.2±2.1 15.4±1.7 14.5±2.3 12.3±1.4 14.4±2.3Leucine 12.1±1.5 3.2±2.1 13.1±1.2 11.7±2.3 11.5±1.9 12.9±2.2 13.7±2.4 12.4±1.8

Values are mean±SD (n¼3). Means followed by the different letter within the same row are significantly different in the same row (p� 0.05).

Table 2. Biogenic amines in snail muscle during ice storage.

Storagetime (days)

Biogenic amines (mg/100 g sample)

Tyramine Putrescine Cadaverine Agmatine Histamine

0 0 a 0 a 0 a 0 a 0 a1 0 a 0.22±0.09 a 0 a 0.77±0.05 a 0 a2 0 a 0.31±0.08 ab 0.21±0.12 2.33±0.21 ab 0 a3 0 a 1.22±0.30 bc 0.34±0.02 a 6.98±0.52 b 0 a4 0 a 1.44±0.11 c 0.86±0.04 b 12.35±1.36 c 0 a5 0.11±0.02 a 3.32±0.24 d 1.31±0.12 c 19.44±2.55 d 0 a6 1.14±0.04 b 8.75±0.56 e 2.44±0.24 d 24.56±3.67 d 0.22±0.08 b7 1.43±0.21 c 11.66±1.32 f 6.66±0.47 e 28.87±3.31 e 0.29±0.03

Values are mean±SD (n¼3). Means followed by the different letter within the same column are significantly different in the same column (p�0.05).




observed from days 3 to 7 (p< 0.05). At the end of stor-age, psychrophilic and mesophilic viable counts were7.83±0.43 and 7.09±0.45 logCFU/g. The results, withregard to the increasing bacteria content during iced orchilled storage in several fish, had been reported by dif-ferent authors (Rodrıguez et al., 2004; Ozogul et al., 2008;Reza et al., 2008). In general, the upper limit for meso-philic viable counts before the rejection of freshwater andmarine fish usually exceeded 7 logCFU/g (InternationalCommission on Microbiological Specifications for Foods(ICMSF), 1978). Thus, the increase in psychrophilic andmesophilic viable counts with increasing storage timecould be used to indicate the deterioration and unaccept-ability of snail meat at the end of storage.

CONCLUSION

In this study, the effects of flake ice pretreatment onthe quality of spotted Babylon snails were evaluated.According to the results of post-mortem changesduring 7 days of storage in ice, nucleotide determinationcan be used to evaluate the loss of freshness during stor-age. An increase in K-value, TVB-N, TMA-N, organicacids and biogenic amines and changes in chemicalproperties of snail meat stored in chilled conditions,illustrated the susceptibility of post-mortem changesduring storage. Mesophilic counts of spotted Babylonsnails stored in ice exceeded 7 logCFU/g after 7 daysof storage, which is considered the maximum level foracceptability of marine products. On the basis of thisinformation, it is highly recommended that the commer-cial handling method of spotted Babylon snails be mod-ified to extend the shelf life of this product.

ACKNOWLEDGMENTS

This study was supported partially by a grant-in-aidfrom the National Research Council of Thailand to

Chatchawan Chotimarkorn under the project,‘Preservation of whole shell and streamed meat of themarketable sizes of spotted Babylon snail (B. areolata)using freezing and chilling for commercial practices /2550.’

REFERENCES

Aguiar A., Nascimento R.A.A., Ferretti L.P. and Goncalves A.R.(2005). Determination of organic acids and ethanol in commer-cial vinegars. Brazilian Journal of Food Technology 5: 51�56.

Aquino F.W.B., Boso L.M., Cardoso D.R. and Franco D.W. (2008).Amino acids profile of sugar cane spirit (cachaca), rum, andwhisky. Food Chemistry 108: 784�793.

Ashie I., Smith J. and Simpson B. (1996). Spoilage and shelf-life exten-sion of fresh fish and shellfish. Critical Reviews in Food Scienceand Nutrition 36: 87�121.

Chaitanawisuit N., Kritsanapuntu S. and Natsukari Y. (2002).Economic analysis of a pilot commercial production for spottedBabylon, Babylonia areolata Link, 1807 of marketable sizes usinga flow-through culture system in Thailand. Aquaculture Research15: 1265�1272.

Chruch N. (1998). MAP fish and crustaceans-sensory enhancement.Food Science and Technology Today 12: 73�83.

Chu C.H. and Bjeldanes L.F. (1981). Effects of diamines, polyaminesand tuna fish extract on the binding of histamine to mucinin vitro. Journal of Food Science 47: 7980�7988.

Ehira S. and Uchiyama H. (1987). Biochemical changes in relation tofreshness. In: Kramer D.E. and Liston J. (eds), Seafood QualityDetermination, New York: Elsevier, pp. 85�204.

Harrigan W.F. and McCance M.E. (1976). Biochemical test for bac-teria. In: Harrigan W.F. (ed.), Laboratory Methods in Food andDairy Microbiology, London: Academic Press Inc., pp. 15�22.

Heen E. (1982). Development in chilling and freezing of fish.International Journal of Refrigeration 5: 45�49.

ICMSF (1978). Their significance and methods of enumeration.In: Microorganism in Foods, Canada: Toronto: University ofToronto Press, pp. 50�188.

Iwamoto M., Yamanaka H., Abe H., Ushio H., Watanabe S. andHashimoto K. (1988). ATP and creatine phosphate breakdownin spiked plaiced muscle during storage and activity of someenzyme involved. Journal of Food Science 53: 1662�1665.

Iwamoto M., Yamanaka H., Watabe S. and Hashimoto K. (1987).Effects of storage temperature on rigor-mortis and ATP degra-dation in plaice Paralichthys olivaceus muscle. Journal of FoodScience 52: 1541�1547.

Kawai M., Okiyama A. and Ueda Y. (1992). Taste enhancementsbetween various amino acids and IMP. Chemicals Senses 27:739�745.

Kilinc B., Cakli S., Cadum A., Dincer T. and Tolasa S. (2007).Comparison of effects of slurry ice and flake ice parameters onthe quality of aquacultured sea bream (Sparus aurata) and seabass (Dicentrarchus labrax) stored at 4 �C. Food Chemistry 104:1611�1617.

Konosu S., Watanabe K., Koriyama T., Shirai T. and Yamaguchi K.(1988). Extractive component of scallop and identification of itstaste-active component by omission test. Journal of the JapanSociety of Food Science and Technology 35: 252�258.

Kryzmien M.E. and Elias L. (1990). Feasibility study on the determi-nation of fish freshness by trimethylamine headspace analysis.Journal of Food Science 55: 1228�1233.

Marks (Rupp) H.S. and Anderson C.R. (2005). Determination ofputrescine and cadaverine in seafood (finfish and shellfish) byliquid chromatography using pyrene excimer fluorescence.Journal of Chromatography A 1094: 60�69.

Marquez-Rıos E., Moran-Palacio E.F., Lugo-Sanchez M.E.,Ocano-Higuera V.M. and Pacheco-Aguilar R. (2007).

Table 3. Microbiological changes of snail muscleduring ice storage.

Storagetime (days)

Bacterial counts (log CFU/g sample)

Psychrophilic Mesophilic

0 2.03±0.17 a 2.00±0.12 a1 2.16±0.12 a 2.06±0.13 a2 2.34±0.15 a 2.13±0.14 a3 3.53±0.22 b 3.09±0.23 b4 3.98±0.24 bc 3.12±0.32 b5 4.56±0.36 c 4.16±0.24 c6 6.08±0.43 d 5.76±0.34 d7 7.83±0.43 e 7.09±0.45 e

Values are mean±SD (n¼ 3). Means followed by the different letter within thesame column are significantly different in the same column (p�0.05).




Postmortem biochemical behavior of giant squid (Dosidicusgigas) mantle muscle stored in ice and its relation with qualityparameters. Journal of Food Science 72: C356�C362.

Massa A.E., Paredi M.E. and Crupkin M. (2002). Nucleotide catabo-lism in cod stored adductor muscle of scallop (ZygochlamysPatagonica). Journal of Food Biochemistry 26: 295�305.

Matsumoto M. and Yamanaka H. (1990). Post-mortem biochemicalchanges in muscle of kuruma prawn during storage and evalua-tion of freshness. Nippon Suisan Gakkaishi 56: 1145�1149.

Mazorra-Manzano M.A., Pacheco-Aguilar R., Dıaz-Rojas E.I. andLugo-Sanchez M.E. (2000). Postmortem changes in black skip-jack muscle during storage in ice. Journal of Food Science 65:774�779.

Olafsdottir G., Martinsdottir E., Oehlenschlager J., Dalgard P., JensenB., Underland I., Mackie I., Henehan G., Nielsen J. and NilsenH. (1997). Methods to evaluate fish freshness in research andindustry. Trend in Food Science and Technology 8: 285�265.

Ozogul F., Ozogul Y. and Kuley E. (2008). Nucleotide degradationand biogenic amine formation of wild white grouper (Epinephelusaeneus) stored in ice and at chill temperature (4 �C). FoodChemistry 108: 933�941.

Ozogul F., Taylor K.D.A. and Ozogul Y. (2002). Biogenic aminesformation in Atlantic herring (Clupea harengus) stored undermodified atmosphere packaging using a rapid HPLC method.International Journal of Food Science and Technology 37:515�522.

Reza Md. S., Bapary M.A.J., Ahasan C.T., slam Md. I. and KamalMd. N. (2008). Shelf life of several marine fish species ofBangladesh during ice storage. International Journal of FoodScience and Technology 43: 1�10.

Rodrıguez O., Losada V., Aubourg S.P. and Velazquez J.B. (2004).Enhanced shelf-life of chilled European hake (Merluccius merluc-cius) stored in slurry ice as determined by sensory analysis andassessment of microbiological activity. Food ResearchInternational 37: 749�757.

Saito T., Arai K. and Matuyoshi M. (1959). A new method for esti-mating the freshness of fish. Bulletin of the Japanese Society ofScientific Fisheries 24: 749�750.

Scherer R., Augusti P.R., Steffens C., Bochi V.C., HecktheureL.H., Lazzari R., Radunz-Neto J., Pomblum S.C.G. and

Emanuelli T. (2005). Effect of slaughter method on postmor-tem changes of grass carp (Ctenopharyngodon idella) stored inice. Journal of Food Science 70: C348�C353.

Shimada R., Ushio H. and Yamanaka H. (1998). Changes in contentof extractive components in the tail muscle of three species oflobsters during storage. Fisheries Science 64: 979�984.

Sikorkski Z.E., Kolakowska A. and Burt J.R. (1990). Postharvest bio-chemical and microbial changes. In: Sikorski Z.E. (ed.), Seafood:Resources, Nutritional Composition and Preservation, BocaRaton, FL: CRC Press, pp. 55�75.

Stratton J.E., Hutkins R.W. and Taylor S.L. (1991). Biogenic-aminesin cheese and other fermented foods: a review. Journal of FoodProtection 54: 460�470.

Suweja I.K., Hori K., Miyazawa K. and Ito K. (1989). Changes incontent of ATP-related compounds, homarine, and trigonellinein marine invertebrates during ice storage. Nippon SuisanGakkaishi 55: 559�566.

Veciana-Nogues M.T., Izquierdo-Pulido M. and Vidal-Carou M.C.(1997). Determination of ATP related compounds in fresh andcanned tuna fish by HPLC. Food Chemistry 59: 467�472.

Watanabe H., Yamanaka H. and Yamakawa H. (1992a). Seasonalvariation of extractive components in the muscle of disk abalone.Nippon Suisan Gakkaishi 58: 921�925.

Watanabe H., Yamanaka H. and Yamakawa H. (1992b). Post-mortembiochemical changes in the muscle of disk abalone during stor-age. Nippon Suisan Gakkaishi 58: 2081�2088.

Wongso S. and Yamanaka H. (1996). Changes in content of extractivecomponents in the adductor muscle of noble scallop during stor-age. Fisheries Science 62: 815�820.

Wongso S. and Yamanaka H. (1998). Extractive components of theadductor muscle of Japanese baking scallop and changes duringrefrigerated storage. Journal of Food Science 63: 504�508.

Woyewoda A.D., Shaw S.J., Ke P.J. and Burns B.G. (1986).Recommended for assessment of fish quality. Canadian technicalreport of fisheries and aquatic sciences. No. 1448. Nova, Scotia,Canada.

Yamanaka H. and Shimada R. (1996). Post-mortem biochemicalchanges in the muscle of Japanese spiny lobster during storage.Fisheries Science 62: 821�824.




Accumulation of Lignin and Malondialdehyde in Relation to

Quality Changes of Button Mushrooms (Agaricus bisporus)

Stored in Modified Atmosphere

Tianjia Jiang,1 Muhammad Muzammil Jahangir,1 Qiushuang Wang2 andTiejin Ying1,*

1Department of Food Science and Nutrition, Zhejiang University, Hangzhou 310029, PR China2Tea Research Institute, Guangdong Academy of Agricultural Sciences, Guangzhou 510640, PR China

Whole fresh button mushrooms (Agaricus bisporus) were stored in unsealed bags in two types of modified

atmosphere packaging (MAP), namely, active and passive, at 4 �C. The packaging film was 0.04mmlow-density polyethylene and the gas composition of the active MAP was 12% O2, 2.5% CO2 and85.5% N2. Firmness of Mushroom showed a positive correlation with the accumulation of lignin in the

tissue. On the other hand, changes of malondialdehyde content with storage time were proportional to theevolution of browning. The button mushrooms in control treatment developed severe browning at the endof the 15-day storage, while the mushrooms of both active and passive MAP treatments browned slightly.

MAP treatments could not inhibit phenylalanine ammonia lyase activity; however, it can reduce thelignification process by the inhibition of peroxidase (POD) activity and the accumulation of lignin.Correlation between the cinnamyl alcohol dehydrogenase activity and lignin accumulation was notobvious. Our results suggested that an increase in the firmness of mushrooms during senescence may be

a consequence of tissue lignification, a process associated with increase in POD activity. Both active andpassive MAPs were useful for the conservation of tenderness and whiteness in button mushrooms.

Key Words: button mushrooms, Agaricus bisporus, lignin, malondialdehyde, packaging, modifiedatmosphere, texture

INTRODUCTION

Button mushroom (Agaricus bisporus) is one of the

most popular mushrooms, traditionally cultivated in

the world. Consumption and production of this edible

mushroom have grown continuously in the past years.

It is the most extensively cultivated edible mushroom,

comprising 32% of the worldwide production (Chang,

1999). Mushrooms are a good source of vitamin B2,

niacin, folates and many mineral elements (Mattila

et al., 2001). However, button mushrooms only have a

short shelf life of 3�4 days; they lose their commercial

value within a few days, due to senescence, browning,

water loss and microbial attack (Nerya et al., 2006).Lignification has been reported in some fruits and

vegetables. Lignification and the associated enzyme

activities were also found to be related to fruit firmness,

ripening and browning (Ketsa and Atantee, 1998;

Aquino-Bolanos and Mercado-Silva, 2004; Cai et al.,

2006). Lignin is a three-dimensional phenolic structure

derived from free-radical polymerization of p-coumaryl,

coniferyl and sinapyl alcohols within the plant cell wall

(Whetten and Sederoff, 1995). Phenylalanine ammonia

lyase (PAL; EC 4.3.1.5), cinnamyl alcohol dehydroge-

nase (CAD; EC 1.1.1.195) and peroxidase (POD; EC

1.11.1.7) are the three important enzymes involved in

lignin synthesis (Whetten and Sederoff, 1995). PAL,

in catalyzing deamination of phenylalanine to trans-cin-

namate, is believed to be the critical enzyme controlling

the accumulation of lignin in plants (Lewis et al., 1999);

CAD catalyzes the last step of themonolignol pathway,

while POD catalyzes the polymerization of monolignol

to complete the process of lignification (Imberty et al.,

1985). Although they are all parts of the lignification

pathway, the extent to which they may regulate the

pathway in different tissues is still not clear.Modified atmosphere packaging (MAP) has become

increasingly common in recent years and has been

reported to be the most economical and effective

method of extending the shelf life of mushroom (Tano

et al., 1999). Active modification is created by replacing

*To whom correspondence should be sent(e-mail: [email protected]).Received 16 February 2009; revised 8 May 2009.


217

gases in the package with a desired mixture of gases. Onthe other hand, passive modification happens when theproduct is sealed in a package made of a selected film,and a desired atmosphere develops naturally as a conse-quence of product respiration and diffusion of gasesthrough the film (Lee et al., 1996). Modified atmospherein terms of reduced O2 and elevated CO2 can extend thepostharvest life of fruits and vegetables by reducing theirrespiration rate as well as the production of ethylene,minimizing metabolic activity, delaying enzymaticbrowning and retaining visual appearance (Kader,1986). The optimal ranges of O2 and CO2 levels havebeen published for fresh sliced mushroom, and gener-ally, 10�20% O2 and 2.5% CO2 (balanced N2) areassumed as an optimal gas composition to inhibit thegrowth of aerobic bacteria and improve slice appearance(Simon et al., 2005).

Though a lot of work has been done on extending thepostharvest life of Agaricus sp. and Pleurotus sp. mush-rooms using MAP (Villaescusa and Gil, 2003; Areset al., 2007), little is known about the lignin accumula-tion of button mushroom during senescence and its rela-tion to firmness and browning in either active or passiveMAP treatment. Physiological and biochemical changesof button mushrooms with different treatments wereinvestigated during the storage, in order to betterunderstand the changes in lignin content of buttonmushrooms and its relation to key enzymic activities inlignification as well as its effect on tissue firmness.


Sample Preparation and Storage Conditions

Button mushrooms (A. bisporus) used in this studywere harvested in July from a local farm in Hangzhou,China. Mushrooms were picked from the same flowerand from the same area of the shed so as to reducepossible variations caused by cultivation and environ-mental conditions. The mushrooms were transportedto the laboratory within 1 h after picking. In the labora-tory, the mushrooms were screened for uniform size andabsence of mechanical damage.

Mushrooms (65±5 g) were selected at random andput in 18� 20 cm2 bags of low-density polyethylene(PE with thickness 0.04mm). The PE gas transmissionrates were 1078� 10�18mol/m s Pa for O2, 4134�10�18mol/m s Pa for CO2 (both at 20 �C and 100%RH) and (2.8�6.5)� 10�5 g/m2 s for H2O (at 37 �C and90% RH). Then, the mushrooms were divided into threetreatments: (1) packaged in PE bags with a modifiedatmosphere of 12% O2, 2.5% CO2 and 85.5% N2, gen-erated by a gas mixture (MAP Mix9000, PBI, Denmark)

and heat-sealed (active modified atmosphere, AM);(2) packaged in PE bags with normal air and heat-sealed (passive modified atmosphere, PM); (3) openstorage with unsealed PE bags as control. The storagetemperature was 4±1 �C with a relative humidity ofabout 75%. The mushroom bags were taken out after0, 3, 6, 9, 12 and 15 days of storage to determine thepackage atmosphere composition, browning, firmness,the contents of MDA, polyphenol, lignin and the activ-ity of PAL, POD and CAD.

Methods

Package Atmosphere Composition

O2 and CO2 concentrations in the packages wereevaluated using a SCY-2A O2 and CO2 analyzer(Xinrui Instrument Co., Shanghai, China). Gas sampleswere taken from the bags with a 20mL syringe.

Weight Loss

Weight loss was determined by weighing the wholemushroom before and after the storage period. Weightloss was expressed as the percentage of loss of weightwith respect to initial weight.

Color

The surface color of mushroom caps was measuredwith a WSC-S Colorimeter (Shanghai PrecisionInstrument Co. Ltd., Shanghai, China). To analyze theL* (light/dark), a* (red/green) and b* (yellow/blue)values, each mushroom was measured at three equidis-tant points of the cap and compared to the ideal mush-room color values of L*¼ 97, a*¼�2 and b*¼ 0 using�E as described by the following equation (Ajlouni,1991):

�E ¼ ðL� 97Þ2 þ ða� ð�2ÞÞ2 þ b2� �1=2

where �E indicates the degree of overall color change incomparison to color values of an ideal mushroom.

Firmness Evaluation

A penetration test was performed on the mushroomcap with a TA.XT2i texture analyzer (Stable MicroSystems, UK), using a 5mm diameter cylindricalprobe. Samples were penetrated to a depth of 5mm.The speed of the probe was 2.0mm/s during the pretestand 2.0mm/s during penetration. Force and time datawere recorded with Texture Expert (version 1.0) fromStable Micro Systems. From the force versus timecurves, firmness was defined as the maximum force.

218 T. JIANG ET AL.

Measurements were performed in duplicate on threemushroom caps for each sample.

Malondialdehyde Content

The level of lipid peroxidation products in mushroomcaps was determined as expressed in the MDA content(Health and Pacontroler, 1968). One gram of mushroomwas ground in 0.25% 2-thiobarbituric acid (TBA) in10% trichloroacetic acid (TCA) using a mortar andpestle. After heating at 95 �C for 30min, the mixturewas cooled quickly in an ice bath and centrifuged at10 000� g for 10min. The absorbance of the superna-tant was read at 532 nm and corrected for unspecificturbidity by substracting the absorbance of the sameat 600 nm. The blank was 0.25% TBA in 10% TCA.The amount of MDA was calculated with an extinctioncoefficient of 155mM/cm and expressed as nanomolesper gram of fresh weight.

Polyphenol Content

Quantification of the total soluble phenolic com-pounds was carried out using the method proposed bySingleton and Rossi (1965). Five grams of mushroomcap were homogenized with 20mL of 80% ethanol for24 h; the homogenized mix was then filtered through twolayers of cheesecloth, and the filtered liquid was centri-fuged at 10 000� g for 15min. One milliliter of the super-natant liquid was mixed with 1mL of Folin�Ciocalteureagent and 10mL of sodium carbonate (7%). This wasincreased to 25mL with distilled water and left to settlefor 1 h. The absorbance was then read at 750nm. Astandard curve of gallic acid was used for quantification.

Lignin Content

To measure the lignin content, mushroom samplesfrom each treatment were frozen in liquid nitrogen andstored at �20 �C until analysis. Lignin was extractedand measured by the method of Bruce and West(1989). Three grams of frozen tissue powder of mush-room cap were homogenized in 10mL of 99.5% (v/v)ethanol and centrifuged at 20 000� g for 20min. Thepellet was dried at 60 �C for 24 h. Fifty milligrams ofdried residue were placed in a screw-cap tube, and then5mL of 2M HCl and 0.5mL of thioglycolic acid wereadded. The sample was heated at 100 �C for 6 h, thencooled and centrifuged at 20 000� g for 20min at roomtemperature. The supernatant was then discarded andthe pellet was washed with distilled water. The resultingpellet was resuspended in 5mL of 1M NaOH sealedwith parafilm. The solution was agitated gently at25 �C for 18 h, and then centrifuged at 20 000� g for20min, and the supernatant then transferred to a test

tube. One milliliter of concentrated HCl was added tothe test tube and the lignin thioglycolic acid was allowedto precipitate at 4 �C for 4 h. Following centrifugation at10 000� g for 10min, the pellet was dissolved in 1mL of1M NaOH. The absorbance was measured against aNaOH blank at 280 nm, and data were expressed on afresh weight basis.

Assay of Enzyme Activity

PAL was extracted from 5 g of frozen tissue powder ofmushroom cap using 10mL of sodium borate buffer(200mM, pH 8.8) as described by Koukol and Conn(1961) and assayed by measuring the absorbance oftrans-cinnamic acid at 290 nm. The reaction mixture(3mL) which contained 0.8mL of supernatant and50mM of L-phenylalanine in sodium borate buffer(200mM, pH 8.8) was incubated at 37 �C for 90min,and the reaction was terminated by ice water. One unitof PAL activity was defined as the amount of enzymethat caused the increase in absorbance at 290 nm of 0.01in 1 h under the specified conditions, and data wereexpressed on a protein basis.

For CAD analysis, 5 g of frozen tissue powder ofmushroom cap was extracted using 10mL of Tris :HClbuffer (200mM, pH 7.5) as described by Wyrambik andGrisebach (1975). The reaction mixture which contained100mM of Tris :HCl (pH 8.8), 20mM of coniferyl alco-hol, 5mM of NADPþ and 50 mL of extract was incu-bated at 30 �C for 15min. One unit of CAD activity wasdefined as the amount of enzyme that caused the changein absorbance at 340 nm of 0.01/min under the specifiedconditions, and data were expressed on a protein basis.

POD was extracted from 5 g of frozen tissue powder ofmushroom cap with 10mL of sodium phosphate buffer(200mM, pH 6.0) as described by Moerschbacher et al.(1988). In this study, POD activities were measuredspectrophotometrically using substrate guaiacol. Thereaction mixture for the determination of POD activityconsisted of 50mM of sodium phosphate buffer (pH6.0), 5mM of guaiacol, 5mM of H2O2 and 50 mL oftissue extract. One unit of POD activity was defined asthe amount of enzyme that caused the change in absor-bance at 470 nm of 0.01/min under the specified condi-tions, and data were expressed on a protein basis.


There were three replications per treatment and eval-uation period. All data represent the mean of three rep-licates. Analysis of variance (ANOVA), followed byDuncan’s multiple range test (DPS version 6.55) at the5% level. Least significant differences (LSDs, p¼ 0.05)were calculated for mean separation.

Accumulation of Lignin and Malondialdehyde 219


Package Atmosphere Composition

Atmosphere composition of control package was con-

stant and equal to atmospheric composition; therefore,

button mushrooms in this packaging were stored under

air during the entire tested period.In the active package, O2 concentration decreased

during the first 6 days, reaching a concentration of

3.8%, and after about 9 days, it increased to 13.4% at

the end of the storage time. On the contrary, CO2 con-

centration increased from the initial 2.5% to 4.9%

during the first 6 days and decreased to 2.7% at the

15th day. On the other hand, in the passive package,

O2 concentration decreased sharply in the first 3 days,

from 21% to 6.2% (Figure 1). This considerable

decrease in O2 concentration could be explained by the

high respiration rate of button mushrooms right after

harvest; similar results had been reported by Ares

et al. (2007) and Parentelli et al. (2007) for shiitake

mushrooms. Contrary to the O2 concentration, CO2

concentration increased sharply to 7.4% within 3 days;

the peak on the 3rd day is also believed to be the result

of a sharp rise in postharvest respiration that occurswith fruits and vegetables (Eskin et al., 1971).

Weight Loss

Weight loss in mushrooms is mainly caused by watertranspiration rate and CO2 loss during respiration (Royet al., 1995). The percent weight loss increased with theduration of storage; however, there were no significantdifferences between the active and passive MAP treat-ments (Figure 2). Their weight losses were below 1%during storage. The low water vapor transmission rateof PE films, combined with the high transpiration rate ofmushrooms, developed a nearly saturated condition inthe packages, which was responsible for the small weightloss (Antmann et al., 2008).

The button mushrooms stored under control packagehad a weight loss of more than 4% at the end of thestorage, suggesting that dehydration is an importantprocess in the loss of mushroom quality during posthar-vest storage. This could be attributed to the fact thatmushrooms are only protected by a thin epidermalstructure, which does not prevent a quick superficialdehydration (Singer, 1986).

Color and MDA Content

There were significant differences (p< 0.05) in white-ness between the control and MAP-treated mushrooms(Table 1). The L-value of control decreased sharply afterthe first 3 days, and it was 83.5 at the 6th day and 77.1 atthe 12th day; the last value may not be considered ascommercially acceptable if a L-value of 80 for whole-salers (Lopez-Briones et al., 1992) was taken intoaccount. In the active and passive MAP treatments,L-value was still acceptable after 12 days, when theirL-value was 87.5 and 86.4 at the 12th day, respectively.

(a)

(b)

0

5

10

15

20

25

0 3 6 9 12 15 18

O2

conc

ent

ratio

n (%

v/v

)

0

2

4

6

8

10

0 3 6 9 12 15 18Storage time (days)

Storage time (days)

CO

2 co

ncen

tratio

n (%

v/v

)

Figure 1. Concentration of O2 (a) and CO2 (b) inpackaged button mushrooms (A. bisporus) storedunder different treatments at 4 �C. Data representthe mean±SD (n¼ 3). (�) Active MA; (#) PassiveMA; (m) Control.

0

1

2

3

4

5

0

Storage time (days)

Wei

ght l

oss

(%)

3 6 9 12 15 18

Figure 2. Weight loss of button mushrooms (A. bis-porus) stored under different treatments at 4 �C. Datarepresent the mean±SD (n¼ 3). (�) Active MA;(#) Passive MA; (m) Control.

220 T. JIANG ET AL.

The button mushrooms began to brown after the 15thday. Similar results were obtained by Simon et al.(2005), who reported that the L-value of MAP freshsliced mushrooms was between 86 and 88 after13 days. Interestingly, we found that mushrooms of pas-sive MAP exhibited even higher L-value than those ofactive MAP treatment at the same storage time. At theend of the storage, the button mushrooms of MAP treat-ment browned slightly, but they also had commercialvalue and edibility. Compared with control, both activeand passive MAP treatments significantly (p< 0.05)inhibited the browning of the button mushrooms.The level of lipid peroxidation was measured in terms

of MDA content (Figure 3). Throughout the storage of15 days, MDA content of the control package increasedfrom the initial value of 11.45�72.66 nmol/g, with theincrease being sharp after the 6th day. While for theMAP treatment, MDA content only reached 21.74 and26.77 nmol/g at the end of storage, respectively.Significant differences (p< 0.05) in MDA content ofbutton mushrooms were observed between control andMAP treatment. Active MAP treatment showed thelowest MDA content during the whole storage.Meanwhile, there was a correlation between the

browning of the button mushrooms and the MDA con-tent. MAP treatment could inhibit the accumulation ofMDA in button mushrooms, and thus attributed to inhi-bition of browning. Although this does not necessarilymean that lipid peroxidation is directly responsible forthe development of browning, this indicated that bothprocesses occurred simultaneously.

Polyphenol Content

Figure 4 shows the variation in polyphenol content ofbutton mushrooms stored at 4 �C under three treat-ments. There were no significant differences (p> 0.05)between the MAP treatments. However, lower polyphe-nol content was found for the mushroom in controltreatment as compared to the MAP treatment, whichmight be explained for the participation of polyphenolin the browning synthesis during the storage. But, it is

not known if the reduced PAL activity in button mush-

rooms found in this study, as has been observed in

potato tuber disks (Zucker, 1968) and iceberg lettuce

midrib tissue (Ritenouv and Saltveit, 1996), which in

turn decreased the phenolic content.

Table 1. Lightness value (L*) and �E of button mushrooms (A. bisporus) stored under active MAP, passive MAPand control treatment at 4 �C for 15 days.

Storage (days)

Active MAP Passive MAP Control

L* �E L* �E L* �E

0 94.0±1.3 a 13.8±1.1 i 94.1±1.0 a 13.7±0.8 i 94.0±1.1 a 13.8±0.7 i3 88.8±2.1 cd 17.8±1.8 gh 90.5±2.2 b 17.9±2.1 gh 86.1±2.6 ef 20.9±2.1 f6 86.3±1.7 ef 22.3±1.3 f 89.8±2.5 bc 18.6±1.8 g 83.5±2.4 g 24.6±1.6 e9 86.3±1.5 ef 22.8±1.2 f 91.8±1.3 b 16.9±1.0 h 81.4±1.6 gh 27.2±1.1 cd12 87.5±2.2 de 21.7±1.7 f 86.4±2.8 ef 26.2±2.3 d 77.1±3.0 h 31.0±2.2 b15 79.2±3.1 h 28.4±2.3 c 83.5±2.4 g 26.5±1.6 d 74.0±2.1 i 37.1±1.4 a

Values within a column followed by the different letter indicate that mean values are significantly different by Duncan’s multiple-range test (p<0.05), LSD0.05¼6.63.

0

0.2

0.4

0.6

0.8

1

1.2

0

Storage time (days)

Pol

yphe

nol c

onte

nts

(mg/

g)

3 6 9 12 15 18

Figure 4. Polyphenol contents in button mushrooms(A. bisporus) stored under different treatments at 4 �C.Data represent the mean±SD (n¼3). (�) Active MA;(#) Passive MA; (m) Control.

0

10

20

30

40

50

60

70

80

0

Storage time (days)

MD

A c

onte

nts

(nm

ol/g

)

3 6 9 12 15 18

Figure 3. MDA contents in button mushrooms(A. bisporus) stored under different treatmentsat 4 �C. Data represent the mean±SD (n¼ 3).(�) Active MA; (#) Passive MA; (m) Control.


Firmness and Lignification

Texture is an important quality parameter for fresh

mushrooms (Lopez-Briones et al., 1992). As shown in

Figure 5a, the firmness of button mushrooms in control

package had a tendency to increase, with a 22.2%

increase over the 12-day storage period at 4 �C; similar

results had been reported by Parentelli et al. (2007) and

Antmann et al. (2008) for shiitake mushrooms. On the

other hand, a slight reduction in the firmness of mush-

rooms in both active and passive MAP treatments was

found with storage time. Significant differences

(p< 0.05) were found between the MAP and control

treatments in terms of firmness value.Lignin is a complex polymer of phenylpropanoid,

mainly deposited in cell walls (Whetten and Sederoff,

1995). Lignin contents of button mushrooms in both

active and passive MAP treatments demonstrated a

slight decrease during the storage, and this tendency

was more apparent in the passive MAP treatment

(Figure 5b). This is a sharp contrast to the control treat-

ment in which the lignin content increased by 23.8%

during the same period.

The results indicated that MAP treatment could inhi-bit the formation of lignin, and there was a positivecorrelation between firmness and lignin increases.Atalla and Agarwal (1985) have shown that the aro-matic rings of lignin are often oriented within theplane of the cell wall. Lignin imparts rigidity to cellwalls (Hu et al., 1999), providing a close connectionbetween the lignin content and tissue firmness. This sup-ports the role of lignin to help maintain cell wall struc-ture. In general, tissue firmness is mainly a combinedresult of the mechanical property of cell wall and theturgidity of cells. Since in MAP treatment, atmospheressaturated with water vapor reduced water loss and con-tributed to cell turgidity, it is apparent that the changesin cell wall strength as affected by lignification may beeven more significant than that reflected by the index oftissue firmness.

Lignification Enzyme Activities

Lignification of plant tissues proceeds through PALand CAD activities, and monolignols synthesized bythese actions are polymerized to form macromoleculesof lignin along with the participation of POD (Boerjanet al., 2003). PAL activities in MAP treatment mush-rooms were significantly higher (p< 0.05) than controltreatment mushrooms (Figure 6a). This is a quite unex-pected result in the context of reduced lignification levelsin MAPs. Increase in PAL activity has been associatedwith the activation of reactions involved in biosynthesisof monolignols in potatoes after gamma irradiation,resulting in tissue lignification (Ramamurthy et al.,2000). However, in studies on cherimoya fruit, increasesin PAL activity do not necessarily result in an increase inlignin (Assis et al., 2001). It is well known that PAL isthe key enzyme catalyzing the first step in the plantphenylpropanoid biosynthetic pathway. Higher PALactivity in MAP treatment mushrooms could explainthe higher levels of total polyphenol found under thistreatment. Furthermore, total polyphenol was relativelyhigher throughout the storage time for MAP treatment(Figure 4), which in turns agrees with the higher PALactivity observed during most of the storage time.

Antisense POD plants such as poplars have beenshown to have a 10�20% reduction in the lignin contentof tissue, providing evidence for the direct involvementof POD in polymerization of monolignols (Ipelcl et al.,1999). As shown in Figure 6b, passive MAP treatmentexhibited the lowest POD activity during the storagetime, while the POD activity in control mushroomswas much higher than that in both active and passiveMAP treatments. Meanwhile, the lowest L-value wasfound in the control at the same time. Such behaviorcoincides with the findings of Howard and Griffin(1993), who reported an increase in the activity of thisenzyme in carrot sticks along with an increase in ligninsassociated with color change. Greater activity of this

0

5

10

15

20

25

Storage time (days)

Firm

ness

(N

)

0

0.5

1

1.5

2

2.5

3

3.5

4

Storage time (days)

Lign

in c

onte

nt (

×10

3A

280/

kg)

(a)

(b)

0 3 6 9 12 15 18

0 3 6 9 12 15 18

Figure 5. Changes in firmness (a) and lignin content(b) in button mushrooms (A. bisporus) stored underdifferent treatments at 4 �C. Data represent themean±SD (n¼3). (�) Active MA; (#) Passive MA;(m) Control.

222 T. JIANG ET AL.

enzyme in the tissue, associated with greater lignin contentand color changes in the tissue, as was also observed inthe case of minimally processed lettuce by Ke and Saltveit(1989), for mangosteen (Garcinia mangostana L.) byKetsa and Atantee (1998) and for sweet potato byUritani (1999), indicated the common involvement ofthe enzyme in lignification and browning process besidesphenoloxidase (PPO), which plays an important role inbrowning of damaged fruits and vegetables (Espin et al.,1998).

The role of CAD in lignification in literature is lessclear, with lignin composition sometimes being affectedand sometimes not in CAD antisense transgenic plants(Boudet, 2000). As shown in Figure 6c, the CAD activityin mushrooms of all treatments increased, peaking at the6th day and decreased thereafter. At the end of the stor-age, the order of CAD activity of different treatmentsis as follows: control> active MAP> passive MAP.The correlation between the CAD activity and ligninaccumulation was not obvious in this study.

CONCLUSION

Our data showed relationships between lignin accu-mulation and firmness development, as well as associa-tion between MDA content, POD activity and colorchanges in button mushrooms during storage. Theincrease of firmness in postharvest button mushroomsis mainly associated with the process of lignin accumu-lation in which a POD is involved. Detrimental colorchanges of mushrooms were related to MDA formation,packaging conditions, and consequently, to changes ininternal atmosphere composition. MAP treatment canretard the senescence of button mushrooms by the inhi-bition of POD activity and the accumulation of ligninand MDA. Results from this study also suggest that thehigh relative humidity inside the packages mightincrease mushroom deterioration due to microorganismgrowth. These results suggest that the use of moistureabsorbers could extend button mushroom postharveststorage.

ACKNOWLEDGMENT

The authors are grateful to National Key TechnologyR and D Program of China for providing financial sup-port (Program no. 2006BAD22B01).

REFERENCES

Ajlouni S.O. (1991). Quality characteristics of two hybrids of the culti-vated mushrooms (Agaricus bisporus) and the improvement oftheir shelf life using stripe trimming and gamma irradiation. PhDThesis, The Pennsylvania State University, University Park, PA.

Antmann G., Ares G., Lema P. and Lareo C. (2008). Influence ofmodified atmosphere packaging on sensory quality of shiitakemushrooms. Postharvest Biology and Technology 49: 164�170.

Aquino-Bolanos E.N. and Mercado-Silva E. (2004). Effects of poly-phenol oxidase and peroxidase activity, phenolics and lignin con-tent on the browning of cut jicama. Postharvest Biology andTechnology 33: 275�283.

Ares G., Lareo C. and Lema P. (2007). Modified atmosphere packag-ing for postharvest storage of mushrooms. A review. FreshProduce 1: 32�40.

Assis J.S., Maldonado R., Munoz T., Escribano M.I. and Merodio C.(2001). Effect of high carbon dioxide concentration on PAL

0

10

20

30

40

50

60

Storage time (days)

PA

L ac

tivity

(U

/mg

prot

ein)

0

10

20

30

40

50

60

Storage time (days)

PO

D a

ctiv

ity (

U/m

g pr

otei

n)

0

10

20

30

40

50

Storage time (days)

CA

D a

ctiv

ity (

U/m

g p

rote

in)

(a)

(b)

(c)

0 3 6 9 12 15 18

0 3 6 9 12 15 18

0 3 6 9 12 15 18

Figure 6. Changes in PAL (a); POD (b); and CAD(c) activities of button mushrooms (A. bisporus)stored under different treatments at 4 �C. Data repre-sent the mean±SD (n¼ 3). (�) Active MA; (#) PassiveMA; (m) Control.


activity and phenolic contents in ripening cherimoya fruit.Postharvest Biology and Technology 23: 33�39.

Atalla R.H. and Agarwal U.P. (1985). Raman microprobe evidencefor lignin orientation in cell walls of native woody tissue. Science227: 636�638.

Boerjan W., Ralph J. and Baucher M. (2003). Lignin biosynthesis.Annual Review of Plant Biology 54: 519�546.

Boudet A.M. (2000). Lignins and lignification: selected issues. PlantPhysiology and Biochemistry 38: 81�96.

Bruce R.J. and West C.A. (1989). Elicitation of lignin biosynthesis andisoperoxidase activity by pectic fragments in suspension culturesof castor bean. Plant Physiology 91: 889�897.

Cai C., Xu C.J., Li X., Ferguson I. and Chen K.S. (2006).Accumulation of lignin in relation to change in activities of lig-nification enzymes in loquat fruit Fesh after harvest. PostharvestBiology and Technology 40: 163�169.

Chang S.T. (1999). World production of cultivated edible and medic-inal mushrooms in 1997 with emphasis on Lentinus edodes (Berk.)Sing. in China. International Journal of Medicinal Mushrooms 1:291�300.

Eskin N.A.M., Henderson H.M. and Townsend R.J. (1971).Biochemistry of Foods. New York: Academic Press, p. 33.

Espin J.C., Jolivet S. and Wichers H. (1998). Inhibition of mushroompolyphenol oxidase by agaritine. Journal of Agricultural and FoodChemistry 46: 2976�2980.

Health R.T. and Pacontroler L. (1968). Photoperoxidation in isolatedchloroplasts I. Kinetics and stoichiometry of fatty acid peroxida-tion. Archives of Biochemistry and Biophysics 125: 189�198.

Howard L.R. and Griffin L.E. (1993). Lignin formation and surfacediscolouration of minimally processed carrot sticks. Journal ofFood Science 58: 1065�1067.

Hu W.J., Harding S.A., Lung J., Popko J.L., Ralph J., Stokke D.D.,Tsai C.J. and Chiang V.L. (1999). Repression of lignin biosyn-thesis promotes cellulose accumulation and growth in transgenictrees. Nature Biotechnology 17: 808�812.

Imberty A., Goldberg R. and Catesson A.M. (1985). Isolation andcharacterization of Populus isoperoxidases involved in the laststep of lignin formation. Planta 164: 221�226.

Ipelcl Z., Ogras T. and Altinkut A. (1999). Reduced leaf peroxidaseactivities associated with reduced lignin content in transgenicpoplar. Plant Biotechnology 16: 381�387.

Kader A.A. (1986). Biochemical and physiological basis for effects ofcontrolled atmospheres. Food Technology 40: 99�104.

Ke D. and Saltveit M.E. (1989). Wound-induced ethylene production.Plant Physiology 76: 412�418.

Ketsa S. and Atantee S. (1998). Phenolics, lignin, peroxidase activityand increased firmness of damaged pericarp of mangosteen fruitafter impact. Postharvest Biology and Technology 14: 117�124.

Koukol J. and Conn E.E. (1961). The metabolism of aromatic com-pounds in higher plants. IV. Purification and properties of thephenylalanine deaminase of Hordeum vulgare. The Journal ofBiological Chemistry 236: 2692�2698.

Lee L., Arul J., Lencki R. and Castaigne F. (1996). A review on mod-ified atmosphere packaging and preservation of fresh fruits andvegetables: physiological basis and practical aspects. PackagingTechnology and Science 9: 1�17.

Lewis N.G., Davin L.B. and Sarkanen S. (1999). The nature andfunction of lignins. In: Barton D.H.R., Nakanishi K. andMeth-Cohn O. (eds). Comprehensive Natural ProductsChemistry, Vol. 3, Oxford, UK: Elsevier Science, pp. 617�745.

Lopez-Briones G.L., Varoguaux P., Chambroy Y., Bouquant J.,Bureau G. and Pascat B. (1992). Storage of common mushroomunder controlled atmospheres. International Journal of FoodScience and Technology 27: 493�505.

Mattila P., Konko K., Eurola M., Pihlava J.M., Astola J., VahteristoL., Hietaniemi V., Kumpulainen J., Valtonen M. and Piironen V.(2001). Contents of vitamins, mineral elements, and some pheno-lic compounds in cultivated mushrooms. Journal of Agriculturaland Food Chemistry 49: 2343�2348.

Moerschbacher B.M., Noll U.M., Flott B.E. and Reisener H.J. (1988).Lignin biosynthetic enzyme in stem rust infected resistant andsusceptible near-isogenic wheat lines. Physiological andMolecular Plant Pathology 33: 33�46.

Nerya O., Ben-Arie R., Luzzatto T., Musa R., Khativ S. and Vaya J.(2006). Prevention of Agaricus bisporus postharvest browningwith tyrosinase inhibitors. Postharvest Biology and Technology39: 272�277.

Parentelli C., Ares G., Corona M., Lareo C., Gambaro A., Soubes M.and Lema P. (2007). Sensory and microbiological quality of shii-take mushrooms in modified atmosphere packages. Journal of theScience of Food and Agriculture 87: 1645�1652.

Ramamurthy M.S., Ussuf K.K., Nair P.M. and Thomas P. (2000).Lignin biosynthesis during wound healing of potato tubers inresponse to gamma irradiation. Postharvest Biology andTechnology 18: 267�272.

Ritenouv M.A. and Saltveit M.E. (1996). Identification of a phenylal-anine ammonia lyase inactivating factor in harvested iceberg let-tuce (Lactuca sativa L.). Physiologia Plantarum 97: 327�331.

Roy S., Anantheswaran R.C. and Beelman R.B. (1995). Fresh mush-room quality as affected by modified atmosphere packaging.Journal of Food Science 60: 334�340.

Simon A., Gonzalez-Fandos E. and Tobar V. (2005). The sensory andmicrobiological quality of fresh sliced mushroom (Agaricus bis-porus L.) packaged in modified atmospheres. InternationalJournal of Food Science and Technology 40: 943�952.

Singer R. (1986). The Agaricales in Modern Taxonomy, 4th edn.Koeningstein, Germany: Koeltz Scientific Books.

Singleton V.L. and Rossi Jr J.A. (1965). Colourimetry of totalphenolics with phosphomolybdic�phosphotungstic acidreagents. American Journal of Enology and Viticulture 16:144�158.

Tano K., Arul J., Doyon G. and Castaigne F. (1999). Atmosphericcomposition and quality of fresh mushrooms in modified atmo-sphere packages as affected by storage temperature abuse.Journal of Food Science 64: 1077�1079.

Uritani I. (1999). Biochemistry on postharvest metabolism and dete-rioration of some tropical tuberous crops. Botanical Bulletin ofAcademia Sinica Taipei 40: 177�183.

Villaescusa R. and Gil M.I. (2003). Quality improvement of Pleurotusmushrooms by modified atmosphere packaging and moistureabsorbers. Postharvest Biology and Technology 28: 169�179.

Whetten R. and Sederoff R. (1995). Lignin biosynthesis. Plant Cell 7:1001�1013.

Wyrambik D. and Grisebach H. (1975). Purification and properties ofisoenzymes of cinnamyl-alcohol dehydrogenase from soybean-cell-suspension cultures. European Journal of Biochemistry 59:9�15.

Zucker M. (1968). Sequential induction of phenylalanine ammonia-lyase and a lyase-inactivating system in potato tuber disks.Plant Physiology 43: 365�374.

224 T. JIANG ET AL.




DOI: 10.1177/1082013209353986

2010 16: 225 originally published online 12 August 2010Food Science and Technology InternationalX.H. Li, Y.G. Xing, W.L. Li, Y.H. Jiang and Y.L. Ding

Antibacterial and Physical Properties of Poly(vinyl chloride)-based Film Coated with ZnO Nanoparticles

Published by:


On behalf of:



















Antibacterial and Physical Properties of Poly(vinyl

chloride)-based Film Coated with ZnO Nanoparticles

X.H. Li,1 Y.G. Xing,1,* W.L. Li,1 Y.H. Jiang2 and Y.L. Ding2

1Key Laboratory of Food Nutrition and Safety (Tianjin University of Science and Technology),Ministry of Education, Tianjin 300457, P.R. China

2Institute of Particle Science and Engineering, University of Leeds, Leeds, LS29JT, UK

Nanoparticles of ZnO and their application in coating systems have attracted a great deal of attention inrecent years because of its multifunction property, especially antibacterial activity. In this study, antibac-terial and physical properties of poly(vinyl chloride) (PVC) based film coated with ZnO nanoparticles wereinvestigated. It was found that the antibacterial action should be attributed to the killing effect property of

ZnO nanoparticles. The ZnO-coated films treated by shaking for 10 h exhibited a similar high antibacterialactivity against Escherichia coli and Staphylococcus aureus as the untreated ZnO-coated films. This resultindicated that the ZnO nanoparticles adhered very well to the plastic film. The antibacterial activity of the

ZnO-coated film to inactivate E. coli or S. aureus was improved by UV irradiation. The analysis of physicalproperties of the ZnO-coated films revealed that the nano-ZnO particles showed less effects on the tensilestrength and elongation at break of the film. The ultraviolet (UV) light fastness of the ZnO-coated PVC

film was improved, which may be attributed to the absorption of ZnO nanoparticles against UV light.Water vapor transmission of the ZnO-coated film decreased from 128 to 85 g/m2

� 24 h, whereas the thick-ness of film increased from 6.0mm with increasing the amount of nano-ZnO particles coated from 0 to

187.5 mg/cm2. This research revealed that the PVC film coated with nano-ZnO particles has a good poten-tial to be used as an active coating system for food packaging.

Key Words: PVC film, ZnO nanoparticles, active packaging, physical property

INTRODUCTION

Compared to the instability of organic antibacterialmaterials, the inorganic antibacterial agents exhibit agood stability to withstand harsh process conditionssuch as high pressures or temperatures (Sawai, 2003).Therefore, the inorganic bacteriostatic agents such aszinc oxide (ZnO) nanoparticles have attracted a greatdeal of attention over the past decade (Fu et al., 2005;Zhang et al., 2007). The antibacterial property of ZnOparticles has been investigated by many researchers(Yamamoto et al., 1998; Stoimenov et al., 2002; Xuand Xie, 2003). The antibacterial activity of ZnO nano-particles is greater than that of microparticles which alsodepends on the concentration and surface area, whereasthe particle shape and crystalline structure have lesseffect (Sawai et al., 1996a; Yamamoto et al., 1998;Yamamoto, 2001; Zhang et al., 2007). They observed

that antibacterial activity increased with the reduction

in particle size. On the other hand, Liu and Yang (2003)

have reported that the photocatalytic inactivation of

ZnO with 365 nm ultraviolet (UV) light inactivated

Gram-negative Escherichia coli and Gram-positive

Lactobacillus helveticus. Almost all the initial E. coli

cells (108CFU/mL) were inactivated in 40min in the

presence of 2 g/L ZnO. As the results reported by

Zhang et al. (2008), ZnO nanofluids, which were

stored for 120 days under the light, had the best anti-

bacterial behavior against E. coli bacteria. The best anti-

bacterial activity was believed to be caused by the

photocatalytic properties of ZnO particles. So, the

results above show that ZnO nanoparticles as potential

antibacterial agents can be used in many industries such

as food industry.Microbial growth on food surfaces is one of the most

important problems for its storage. However, direct

application of antibacterial substances onto the surface

of foods has been limited due to the fact that some of the

active agents may diffuse rapidly into the food mass

(Pranoto et al., 2005). Thus, there is an urgent need

for alternative technologies to inhibit the changes in

the quality of food during storage. In recent years,

many researchers have focused on investigating active

packaging which can offer functions such as regulation





of moisture and/or provision of antimicrobial activity(Ouattara et al., 2000). The use of active packagingfilms could be more efficient for the storage of foodbecause the active agents can transfer slowly from thepackaging material to the surface of the product. So far,the application of biodegradable films has been seriouslylimited due to their weak mechanical and poor barrierproperties (Li et al., 2009a). Hence, the commonly usedfood packing materials are still natural polymers andsynthetic polymers (Petersen et al., 1999). Plastic filmsare anticipated to continue to play an important role infood distribution because of their better physical prop-erties and barrier properties, higher strength and elon-gation, lower cost and lightness (Arvanitoyannis et al.,1997b; Psomiadou, 1997). Polymers, such as poly(vinylchloride) (PVC) and polyethylene, are the most com-monly used polymeric materials for packaging applica-tions. Therefore, in order to control undesirablemicroorganisms on food surfaces, the methods thatcoat antimicrobial agents onto the surface of the plasticfilm and incorporate antimicrobial agents into the poly-mer can be used (Appendini and Hotchkiss, 2002).

Several researchers have reported the effect of nano-particles on the properties of films. Li et al. (2009a) havereported that the polyurethane (PU) films doped withZnO nanoparticles exhibited excellent antibacterialactivity, especially for E. coli. The antifungal activityof TiO2-coated plastic film against Penicillium expansumhas been investigated by Maneerat and Hayata (2006).Moreover, the antimicrobial property of TiO2-coatedpackaging film against E. coli in vitro and in actualtests under two kinds of artificial light has also beenshown by Chawengkijwanich and Hayata (2008).According to Shanmugam et al. (2006), the Ag and Aunanoparticle-embedded composite film exhibits antimi-crobial activity against Gram-negative E. coli. Thenanopackaging material, which was synthesized byblending polyethylene with nanopowder (nano-Ag,kaolin, anatase TiO2 and rutile TiO2), could be appliedto improve the preservation quality of Chinese jujube(Li et al., 2009a). On the other hand, the effects of mate-rials, such as nanoparticles and starches coated on orincorporated into plastic films, on the mechanical andphysical benefits have been reported by many research-ers (Krochta and Johnston, 1997; Pranoto et al., 2005).Li et al. (2009a) reported that the Young’s modulus andtensile strength of the PU films improved significantlyby incorporating ZnO nanoparticles up to 2.0wt% andthe abrasion resistance of the PU coatings improved aswell. The mechanical properties of biodegradable filmsmade from low-density polyethylene (LDPE), wheatstarch, soluble starch and ethylene acrylic acid (EAA)or polycaprolactone (PCL) were studied byArvanitoyannis et al. (1997b,c) and Psomiadou et al.(1997). Moreover, the physical properties and micro-structure of nanopacking were also studied by Li et al.(2009a).

It seems to be very important for the investigation ofthe possibility of producing antimicrobial film by coat-ing with ZnO nanoparticles. The objective of this studywas to explore the feasibility of producing antibacterialPVC film by coating with ZnO nanoparticles.Antibacterial activity against the food pathogenic bac-teria E. coli and Staphylococcus aureus and the physicalproperty of the PVC film coated with ZnO nanoparticlesunder different conditions were investigated.


Materials

Dry zinc oxide nanoparticles from NanophaseTechnologies (USA) were used in this study. The pri-mary sizes of the nanoparticles given by the manufac-turers were 200�400 nm. Polyethylene glycol 400 (PEG400) from Fluka was used as a dispersant. E. coli and S.aureus were provided by the Department of BiologicalScience of Tianjin Univeristy of Science andTechnology. Luria-Bertani (LB) medium used for grow-ing and maintaining bacteria was purchased fromSigma�Aldrich (UK).

Methods

Preparation of Nano-ZnO Coating

Suspension of ZnO nanoparticles was prepared by themethod reported by Zhang et al. (2007). In order toimprove the stability of the suspension, PEG 400 as adispersant was used and the amount was 10% of theamount of ZnO nanoparticles added. The mixture con-taining a preset amount of dried ZnO nanoparticles anddispersant was mixed with distilled water in a glassbeaker with the aid of a magnetic stirrer. The beakerwas placed in an ultrasonicator for 30min (Clifton,UK) until the particles were dispersed in water. Aftersonication, the obtained suspension was milled using aplanetary Dyno-mill (Willy A. Bachofen, Switzerland)for 1 h at room temperature. Then, the ZnO nanoparti-cle coating was prepared, which had a concentration of5.0 g/L.

Preparation of ZnO-coated PVC Film

ZnO nanofluid (1.5 or 3.0mL) with a concentration of5.0 g/L was coated on the surface of the PVC film(80� 100mm2) which was spread on a plastic support(AFA-II, XD Env. Eng. Tech. Inc., China). After airing,the ZnO-coated PVC film on the carrier support wasroasted and pressed at 100 �C for 10min. Finally, theZnO-coated PVC film was peeled from the substratecarefully when the carrier support was cooled to roomtemperature (Li et al., 2009a). The final concentrations

226 X.H. LI ET AL.



of nano-ZnO on the dried films were 93.75 and187.5mg/cm2, respectively.

Morphological Observation

Morphology was characterized by scanning electronmicroscopy (SEM). The samples of coated and uncoatedfilms were placed on the SEM stubs using a two-sidedadhesive tape and then, SEM analyses using a LEOGemini 1530 field emission SEM at an acceleratingvoltage of 5 kV after Pt sputtering were performed.

Antibacterial Activity Against E. coli and S. aureus

In order to study how well the ZnO nanoparticlesadhere to the film, the antibacterial activities of ZnO-coated PVC films (80� 100mm2) before and after shak-ing were investigated. The sample film in a 250-mL flaskcontaining 100mL of distilled H2O was shook in a con-stant temperature shaker (Changzhou GuohuaElectrical Appliance Co. Ltd., Jiangsu, China) at120 rpm and 37�C for 10 h. Effects of the sample filmsbefore and after shaking on the growth of E. coli andS. aureus were investigated by the method reported byZhang et al. (2007). The sample film was added to a250mL flask containing 99mL of nutrient broth and1mL of the 24 h grown bacterial culture with an approx-imate concentration of 106�107CFU/mL. The flaskswith the same nutrient broth and grown bacterial culturewith blank film and without film were used as the con-trol and the negative control, respectively. Mediagrowths in flasks were incubated in a constant temper-ature shaker at 120 rpm and 37�C. The growth curveswere determined by measuring the time evolution of theoptical density (OD) of the sample. The experimentswere done with a spectrophotometer (Thermo ElectronCorporation, USA) at 570 nm for S. aureus and at600 nm for E. coli, respectively. Moreover, the inhibitionrate (%) can be calculated according to ([A]i� [A]t)/[A]i,where [A]i is the optical absorption of untreated bacteriaat 8 h for E. coli and 12 h for S. aureus and [A]t is theoptical absorption of treated bacteria (Yao et al., 2007).Photocatalytic inactivation of ZnO nanoparticles

coated on the surface of PVC film against E. coli or S.aureus was performed using the antibacterial drop testmethod with some modifications as described by Yaoet al. (2007). Samples (100mL) of the different bacterialstrains diluted with sterilized distilled H2O were addedonto the surface of each ZnO-coated film. One uncoatedfilm was used as a negative control along with a darkreaction. The bacterial solutions were then irradiatedwith UV A (365 nm) light at room temperature for120min. Then, the treatments were washed from thefilm surfaces with 1mL of 0.05% Tween-20 solution inthe sterilized dish, respectively. Then, 1mL of each bac-terial suspension was pipetted into sterilized plates con-taining 9mL of LB medium and then, the plates were

incubated at 37�C for 16 h. The solutions were read witha spectrophotometer at 570 nm for S. aureus and at600 nm for E. coli, respectively. The optical absorptions[A] were obtained. The inhibition rate (%) can be calcu-lated according to ([A]i� [A]t)/[A]i, where [A]i is the opti-cal absorption of untreated bacteria (Yao et al., 2007).

Elongation at Break (E) and tensile strength (TS)

In order to test the effect of ZnO nanoparticles on thephysical property of PVC film, the elongation at break(E) and tensile strength (TS) of the sample filmuntreated and treated by UV-ray were measured.Various PVC-based films laid on a support were irradi-ated for 6 h under a UV lamp (20W, 365 nm) and thedistance between the support and the UV lamp wasabout 10 cm (Li et al., 2009a). The samples were cutinto a dumbbell shape. Elongation at break (E) and ten-sile strength (TS) of the films were examined using theuniversal material testing machine (New Sansi MaterialDetection Ltd., Shenzhen, China). The films were heldparallel with an initial grip separation of 5 cm and thenpulled apart at a head speed of 200mm/min. Percent Ewas calculated based on the length extended and originallength of the films. Tensile strength was calculated bydividing the maximum force at break (read frommachine) by cross-sectional area of film (Pranotoet al., 2005).

Thickness and Water Vapor Transmission

The thickness of film was measured at several pointswith a hand micrometer (Liuling Instrument, Shanghai,China). WVT of the samples was determined gravimet-rically using the method reported by Pranoto et al.(2005). A cup was covered with the sample film inwhich silica gel was used as a desiccant and placed ina controlled desiccator. The relative humidity was keptat 90±2%, and temperature was kept at 38±0.6 �Cinside the desiccator and checked periodically. Theweight of the cup was measured at 24 h intervalswithin 72 h. The WVT value was expressed ing/m2�24 h. The WVT was calculated according to the

following equation: WVT¼ 24�m/(A�t), where WVTis the water vapor transmission (g/m2

�24 h), �m theincreased weight in the internal time, A the area of per-meation and t is the internal time when the weight of cupincreased constantly.


The tests were carried out in triplicate and experimen-tal dates were analyzed using SPSS 13.0 software (SPSSInc.). The one-way analysis of variance (ANOVA) pro-cedure followed by least significant difference (LSD) testwas used to determine the significant difference(p< 0.05) between the treatment means.

Antibacterial and Physical Properties of Poly(vinyl chloride)-based Film 227





ZnO nanoparticles were dispersed uniformly, but little

agglomerates exist on the film surface (Figure 1, right).

The ZnO particles coated on the surface of film were

about 200 nm and a few particles were smaller than

100 nm (Zhang et al., 2008). Few particles were found

on the surface of the blank film (Figure 1, left). As Xu

and Xie (2003) reported, with decreasing the size of the

ZnO particles, especially getting to nanoscale, antibac-

terial activity of ZnO particles increased sharply (Zhang

et al., 2007). Arvanitoyannis et al. (1997b) have investi-

gated the morphology of LDPE/EAA/Wheat starch

blends after fracture. The SEM observations showed

that the starch particles were occluded and interspersed

within the LDPE/EAA matrix. The formations of metal-

lic silver and gold nanoparticles in composites were stu-

died by Shanmugam et al. (2006). TEM images showed

that Ag and Au nanoparticles were embedded into the

organic�inorganic composite film. According to Li et al.

(2009a), it appeared that the nanoparticles (Ag, TiO2)

were uniformly distributed in the nanopacking film withan irregular shape. The result showed that the nanopar-ticles tended to give better barrier and mechanical prop-erties to the nanopacking film (Li et al., 2009a). TheSEM image showed that the ZnO nanoparticles were uni-formly coated on the PVC film with an irregular shape.Therefore, it is important to investigate howwell the ZnOnanoparticles adhered to the film and the effect of thenanoparticles on the antimicrobial and physical proper-ties of ZnO-coated film.

Effect of ZnO-coated Films Against E. coli andS. aureus

Effects of the sample films before and after shakingfor 10 h on the growth of Gram-negative bacteria E. colior Gram-positive bacteria S. aureus were investigated.The value of the OD rose with an increase in the numberof bacteria due to more light being absorbed. Comparedwith Figures 2a and 2b for E. coli, the increasing trendsof the growth curves were quite similar betweenuntreated and treated by shaking for 10 h for the sameamount of nanoparticles coated. The similar result wasobtained by comparing Figures 3a and 3b for S. aureus.

Figure 1. SEM images of the blank PVC film (left)and ZnO-coated PVC film (right).

0

0.5

1

1.5

2

2.5

3

3.5

4

0

Time (h)

OD

600

nm

(A

)

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

Time (h)

OD

600

nm (

A)

(b)

(a)

2 4 6 8 10

0 2 4 6 8 10

Figure 2. Growth curves of E. coli in LB mediuminoculated with 106�107 CFU/mL of bacteria. (#) neg-ative control; (*) blank film; (¨) film coated with93.75 mg/cm2 ZnO nanoparticles; and (i) filmcoated with 187.5 mg/cm2 ZnO nanoparticles.

228 X.H. LI ET AL.



No significant difference (p< 0.05) in the inhibitionratios of the ZnO-coated film for E. coli or S. aureuswas observed after the treatment of shaking (Table 1).It was reasonable to get the result that the ZnO nano-particles adhered very well to the plastic film. On theother hand, the growth curves of E. coli and S. aureusin LB medium were affected significantly (p< 0.05) bythe presence of nano-ZnO-coated films in comparisonwith the blank film. A small difference in the growthcurves of E. coli or S. aureus between the negative

control and the blank film control can be seen. Thisresult indicated that the antimicrobial activity of ZnO-coated film should be attributed to the nano-ZnO par-ticles. The antibacterial activity against E. coli improvedwith increasing the amount of nanoparticles coated(Figure 2). However, this increasing trend changedslower for S. aureus. The antibacterial behavior ofZnO-coated PVC film under the UV light for 3 h wasalso investigated. With the presence of the same amountof ZnO nanoparticles, the inhibition ratios of the ZnO-coated film samples irradiated by UV light improvedsignificantly (p< 0.05), 29.8% for E. coli and 26.0%for S. aureus, respectively, compared to that of theuntreated ZnO-coated film (Figure 4). This result dem-onstrated that the antibacterial activity of the ZnO-coated film to inactivate E. coli and S. aureus wasimproved by UV irradiation.

The results above indicated that the antimicrobialactivity of ZnO-coated film should be attributed to thenano-ZnO particles (Li et al., 2009a,b). The result alsoagreed with Sawai et al. (1995a, b, 1996a, b) and Nairet al. (2009) when reported ZnO particles were effectivefor inhibiting both Gram-positive and Gram-negativebacteria. While the mechanisms of the antibacterialactivity of ZnO nanoparticles have not yet been clearlyelucidated, Nair et al. (2009) suggested mechanismsincluded the role of reactive oxygen species (ROS)such as O�2 , hydrogen peroxide (H2O2), singlet oxygen(1O2) and hydroxyl radical (�OH) generated on the sur-face of the particles (Sawai et al., 1996b, 1997, 1998;Li et al., 2009a), zincion release (Yang and Xie, 2006),membrane dysfunction (Zhang et al., 2007, 2008) andnanoparticle internalization (Brayner et al., 2006). Anexcellent study by Sawai et al. (1996b, 1998) reportedthat ROS concentrations increased with the ZnO con-tent of slurries. These active oxygen species are toxic tothe bacterial cell because they are very reactive and pow-erful oxidizing agents (Sawai et al., 1998). H2O2 was alsodetected by spectrophotofluorometry by Yamamotoet al. (2004). The study conducted by Zhang et al.(2008) showed that the antibacterial activities in the

0

0.5

1

1.5

2

2.5

3

3.5

4

0Time (h)

OD

570

nm

(A

)

0

0.5

1

1.5

2

2.5

3

3.5

Time (h)

OD

570

nm

(A

)

(a)

(b)

2 4 6 8 10 12 14

0 2 4 6 8 10 12 14

Figure 3. Growth curves of S. aureus in LB mediuminoculated with 106�107 CFU/mL of bacteria. (#) neg-ative control; (*) blank film; (¨) film coated with93.75 mg/cm2 ZnO nanoparticles; and (i) filmcoated with 187.5 mg/cm2 ZnO nanoparticles.

0

20

40

60

80

100

Blank film+UV

Inhi

bitio

n ra

te (

%)

ZnO-coated film

ZnO-coated film+UV

Figure 4. Inhibition rates of PVC-based film with orwithout UV («, the blank film with UV; #, the ZnO-coated film (93.75 mg/cm2) without UV; «, the ZnO-coated film (93.75 mg/cm2) with UV irradiation).

Table 1. Comparisons of inhibition rates ofZnO-coated PVC film with different treatments against

E. coli and S. aureus.

ZnO-coated films(mg/cm2) E. coli S. aureus

93.75 74.52±1.07 a 91.20±0.62 a187.5 85.02±0.79 b 95.91±0.46 b93.75 (shaking) 75.40±0.81 a 91.31±0.38 a187.5 (shaking) 84.17±0.23 b 96.54±0.71 b

Values are mean±SD. Means in same column with different letters are sig-nificantly different (p¼0.05).




ZnO nanofluids with longer storage period had a higherefficiency. With longer storage period, there is morechance to produce more amount of active oxygen,such as H2O2 (Zhang et al., 2008). The ZnO nanofluidsstored for 120 days under the light had a better antibac-terial behavior against E. coli bacteria than that underthe dark (Zhang et al., 2008). The increasing inhibitionrates of ZnO-coated PVC film irradiated by UV lightwas presumably due to the photocatalytic reactions ofthe ZnO particles (Liu and Yang, 2003; Zhang et al.,2008). When the ZnO-coated film was irradiated byUV, there is more chance to have certain photocatalyticreactions. The production of the active oxygen speciescan be promoted in the ZnO sample kept under the UVlight (Yang et al., 2004; Zhang et al., 2008). On the otherhand, under the UV illuminated process, ZnO is acti-vated to generate a conduction electron and a hole(ZnO-hþ). This kind of hole can activate oxygen in airand give active oxygen with very strong chemical activity(Xu and Xie, 2003). According to Liu and Yang (2003),the creation of the electron�hole pair is very importantfor the oxidation process. The water molecule can besplit into OH and Hþ with the hole. The externally sup-plied or dissolved oxygen molecules are transformed to asuperoxide radical anion (�O�2 ) which reacts with Hþ togenerate a hydrogen peroxide radical (�HO�2 ). On sub-sequent collisions with an electron and a hydrogen ion, amolecule of H2O2 is produced. On the other hand, thepresence of ZnO nanoparticles leads to damages in themembrane wall of E. coli (Zhang et al., 2007, 2008).Such damages may be partly due to direct interactionsbetween ZnO nanoparticles and bacterial membranesurfaces, which were also suggested by Stoimenovet al. (2002) and Nair et al. (2009).

Elongation at Break and Tensile Strength

Both properties of film, elongation at break and tensilestrength are important characteristics for packagingmaterial. They are two measures of stretchability priorto breakage and film strength (Krochta and Johnston,1997; Pranoto et al., 2005). In order to investigatethe effect of ZnO nanoparticles on the mechanical prop-erty of the PVC film, the elongation at break and ten-sile strength values of sample films were measured(Figure 5). Coating of nano-ZnO particles on the surfaceof film had a negligible effect on the elongation at breakand tensile strength values of the PVC film. The resultobtained was different from that reported by Li et al.(2009a). In their investigation, the ZnO nanoparticlesincorporated into the PU chains limited the movingscale of chain segments and generated an interactiveforce against the chains. This fact induced that the tensilestrength increases first and then decreases with anincrease in ZnO content, and the elongation at rupturevaries inversely. The property of materials incorporatedinto the film has an important effect on the mechanical

properties. As reported by Psomiadou et al. (1997), thepresence of starch at contents >30% had an adverseeffect on the mechanical properties of LDPE�starchblends. In this article, the ZnO nanoparticles were notfilled into the interstice of PVC chains but were coatedon the surface of the film. So, the interaction forcebetween the nanoparticles and the PVC chains may notbe generated. Therefore, the ZnO particles coated on thefilm had less effect on the mechanical properties of PVC-based film. Figure 5 shows the down trend of elongationat break and tensile strength of the films by the irradia-tion. While tensile strength (Figure 5a) and elongation atbreak (Figure 5b) of the ZnO-coated films decreases asthe pure films do, they are still larger than that of the purePVC films. It was confirmed that the UV light fastness ofthe PU composite films was improved, which may beattributed to the absorption and dispersion of ZnO nano-particles against UV light (Xu and Xie, 2003; Li et al.,2007, 2009a). The longitudinal strength of nanopacking,which was synthesized by blending polyethylene withnano-Ag, kaolin, anatase TiO2 and rutile TiO2 was1.24-fold higher than that of the control. These resultsindicated that the novel material possessed bettermechanical properties (Li et al., 2009a).

Thickness and WVT

WVT of film has a great influence on the shelf life offood, which is a measure of ease of the moisture to

0

5

10

15

20

25

30

0

Time (h)

Ten

sile

str

engt

h (M

Pa)

0

50

100

150

200

250

Time (h)E

long

atio

n at

bre

ak(%

)

(a)

(b)

3 6

0 3 6

Figure 5. Tensile strength (a) and elongation atbreak (b) related to the irradiation time. (*) theblank film; (¨) the ZnO-coated film (93.75mg/cm2);and (i) ZnO-coated film (187.5mg/cm2).

230 X.H. LI ET AL.



penetrate and pass through a material (Li et al., 2006).The ability to control the loss of water molecule fromthe product for the film is an important characteristicthat affects the quality of the food product (Wu et al.,2001). As shown in Table 2, addition of ZnO nanopar-ticles on the film decreased WVT from 128 to85 g/m2

�24 h and the thickness of ZnO-coated filmincreased from 6.0mm. The WVT value of PVC filmscoated with ZnO nanoparticles is lower than that ofthe neat film, and the WVT decreased significantly(p< 0.05) with increasing the amount of nanoparticles.It is probably due to that the ZnO nanoparticles coatedon one side of film prevent a part of the moisture frompassing through the PVC film. The WVT value shouldbe taken into account for the packaging film or othercoating materials when it applies to a moist food prod-uct. In general, the WVT value decreased with increasingthe amount of ZnO nanoparticles. Therefore, thisZnO-coated film could keep more water molecules inthe packaging system and thereby delay the shelf lifeof some kinds of foods, such as fruits and vegetables.WVT of film was related to the materials coated onto orincorporated into the plastic films. As reported byArvanitoyannis et al. (1997b), an increase in starchwith the strong hydrophilic character and/or PCL con-tent was directly proportional to WVTR of the plasticfilm. It was well known that the main disadvantage ofedible films has been their high WVTR. For this reasonit was important either to use a high/multilayer or ablend with a hydrophobic polymer (Arvanitoyanniset al., 1997a). The transmission rates of relative humid-ity and oxygen of the nanopacking material, which wassynthesized by blending polyethylene with nanoparticlesincluding nano-Ag, kaolin, anatase TiO2 and rutileTiO2, were decreased by 28.0% and 2.1% comparedwith those of normal packing. These results indicatedthat the novel material possessed a better barrier prop-erty (Li et al., 2009a).

CONCLUSIONS

SEM image of ZnO-coated PVC film clearly indicatedthat the ZnO nanoparticles were uniformly coated onthe PVC film with an irregular shape. On the aspect of

antibacterial function, no significant difference(p< 0.05) was observed for the inhibition ratios of theZnO-coated film against E. coli or S. aureus after shak-ing for 10 h. This result indicated that the ZnO nanopar-ticles adhered very well to the plastic film. Theantibacterial activity of the ZnO-coated film to inacti-vate E. coli or S. aureus was improved by 29.8% and26.0% for E. coli and S. aureus, respectively, after irra-diating by UV light. Coating of ZnO nanoparticlesenhanced the UV light fastness of the PVC film buthad less effects on the TS and elongation at break ofthe ZnO-coated film. The WVT and the thickness ofZnO�PVC film were significantly affected (p< 0.05) bythe addition of ZnO nanoparticles. WVT of the ZnO-coated film decreased from 128 to 85 g/m2

� 24 h, whereasthe thickness of film increased from 6.0 mm with increas-ing the amount of nano-ZnO particles to 187.5 mg/cm2.This research indicated that PVC films coated withnano-ZnO particles have a good potential to be usedfor food packaging.

ACKNOWLEDGMENTS

This study is supported by the Outstanding DoctorateDissertation Foster Foundation of Tianjin University ofScience and Technology (B201003) and by the 11thFive-Year Key Technologies R&D Program of China(2006BAD22B03). This research is also supported bythe Nanomanufacturing Institute of the University ofLeeds.

REFERENCES

Appendini P. and Hotchkiss J.H. (2002). Review of antimicrobial foodpackaging. Innovative Food Science and Emerging Technologies 3:113�126.

Arvanitoyannis I., Kolokuris I., Nakayama A. and Aiba S. (1997a).Preparation and study of novel biodegradable blends based ongelatinized starch and 1,4-trans-polyisoprene (gutta percha) forfood packaging or biomedical applications. CarbohydratePolymers 34: 291�302.

Arvanitoyannis I., Psomiadou E., Biliaderis C.G., Ogawa H.,Kawasaki N. and Nakayama A. (1997b). Biodegradable filmsmade from low density polyethylene (LDPE), ethylene acrylicacid (EAA), polycaprolactone (PCL) and wheat starch for foodpackaging applications: Part 3. Starch/Starke 49: 306�322.

Arvanitoyannis I., Psomiadou E., Ogawa H. and Kawasaki N.(1997c). Biodegradable films made from low density polyethylene(LDPE) and rice starch for food packaging Uses; Part 1.Carbohydrate Polymers 36: 89�104.

Brayner R., Ferrari-Iliou R., Brivois N., Djediat S., Benedetti M.F.and Fievet F. (2006). Toxicological impact studies based onEscherichia coli bacteria in ultrafine ZnO nanoparticles colloidalmedium. Nano Letters 6: 866�870.

Chawengkijwanich C. and Hayata Y. (2008). Development of TiO2

powder-coated food packaging film and its ability to inactivateEscherichia coli in vitro and in actual tests. International Journalof Food Microbiology 123: 288�292.

Table 2. WVT and thickness of PVC film with differentamounts of ZnO nanoparticles.

ZnO-coated films(mg/cm2) WVT (g/m2

�24 h) Thickness (�m)

0 (control) 128±1.04 c 48.77±0.06 a93.75 107±0.52 b 50.30±0.10 b187.5 85±0.66 a 53.77±0.10 c





Fu G., Vary P.S. and Lin C.T. (2005). Anatase TiO2 nanocompositesfor antimicrobial coating. Journal of Physical Chemistry B 109:8889�8898.

Krochta J.M. and Johnston C.D.M. (1997). Edible and biodegradablepolymer films: challenges and opportunities. Food Technology 51:61�74.

Li J.H., Hong R.Y., Li M.Y., Li H.Z., Zheng Y. and Ding J. (2009a).Effects of ZnO nanoparticles on the mechanical and antibacterialproperties of polyurethane coatings. Progress in Organic Coatings64: 504�509.

Li H., Li B., Kennedy J.F., Peng J.L., Yie X. and Xie B.J. (2006).Preparation and performance evaluation of glucomannan�chito-san�nisin ternary antimicrobial blend film. CarbohydratePolymers 65: 488�494.

Li F., Wang L., Sheng J., Xin Z., Zhao L., Xiao H., Zheng Y. andHu Q. (2009b). Effect of nano-packing on preservation quality ofChinese jujube (Ziziphus jujuba Mill. var. inermis (Bunge) Rehd).Food Chemistry 114: 547�552.

Li Y.Q., Yang Y. and Fu S.Y. (2007). Photo-stabilization properties oftransparent inorganic UV-filter/epoxy nanocomposites.Composites Science and Technology 67: 3465�3471.

Liu H.L. and Yang T.C.K. (2003). Photocatalytic inactivation ofEscherichia coli and Lactobacillus helveticus by ZnO and TiO2

activated with ultraviolet light. Process Biochemistry 39:475�481.

Maneerat C. and Hayata Y. (2006). Antifungal activity of TiO2 photo-catalysis against Penicillium expansum in vitro and in fruit tests.International Journal of Food Microbiology 107: 99�103.

Nair S., Sasidharan A., Rani V.V.D., Menon D., Nair S., Manzoor K.and Raina S. (2009). Role of size scale of ZnO nanoparticles andmicroparticles on toxicity toward bacteria and osteoblast cancercells. Journal of Materials Science: Materials in Medicine 20:235�241.

Ouattara B., Simard R.E., Piette G., Begin A. and Holley R.A. (2000).Diffusion of acetic and propionic acids from chitosan-based anti-microbial packaging films. Journal of Food Science 65: 768�773.

Petersen K., Nielsen P.V., Bertelsen G., Lawther M., Olsen M.B.,Nilssonk N.H. and Mortensen G. (1999). Potential of biobasedmaterials for food packaging. Trends in Food Science andTechnology 10: 52�68.

Pranoto Y., Salokhe V.M. and Rakshit S.K. (2005). Physical andantibacterial properties of alginate-based edible film incorpo-rated with garlic oil. Food Research International 38: 267�272.

Psomiadou E., Arvanitoyannis I., Biliaderis C.G., Ogawa H. andKawasaki N. (1997). Biodegradable films made from low densitypolyethylene (LDPE), wheat starch and soluble starch for foodpackaging applications: Part 2. Carbohydrate Polymers 33:227�242.

Sawai J. (2003). Quantitative evaluation of antibacterial activities ofmetallic oxide powders (ZnO, MgO, CaO) by conductimetricassay. Journal of Microbiological Methods 54: 177�182.

Sawai J., Igarashi H., Hashimoto A., Kokugan T. and Shimizu M.(1995a). Evaluation of growth inhibitory effect of ceramicspowder slurry on bacteria by conductance method. Journal ofChemical Engineering of Japan 28: 288�293.

Sawai J., Igarashi H., Hashimoto A., Kokugan T. and Shimizu M.(1996a). Effect of particle size and heating temperature of cera-mic powders on antibacterial activity of their slurries. Journal ofChemical Engineering of Japan 29: 251�256.

Sawai J., Kawada E.I., Kanou F., Igarashi H., Hashimoto A.,Kokugan T. and Shimizu M. (1996b). Detection of activeoxygen generated from ceramic powders having antibacterialactivity. Journal of Chemical Engineering of Japan 29: 627�633.

Sawai J., Kojima H., Igarashi H., Hashimoto A., Shoji S.,Takehara A., Sawaki T., Kokugan T. and Shimizu M. (1997).Escherichia coli damage by ceramic powder slurries. Journal ofChemical Engineering of Japan 30: 1034�1039.

Sawai J., Saito I., Kanou F., Igarashi H., Hashimoto A., Kokugan T.and Shimizu M. (1995b). Mutagenicity test of ceramic powderwhich have growth inhibitory effect on bacteria. Journal ofChemical Engineering of Japan 28: 352�354.

Sawai J., Shouji S., Igarashi H., Hashimoto A., Kokugan T.,Shimizu M. and Kojima H. (1998). Hydrogen peroxide as anantibacterial factor in zinc oxide powder slurry. Journal ofFermentation and Bioengineering 86: 521�522.

Shanmugam S., Viswanathan B. and Varadarajan T.K. (2006). Anovel single step chemical route for noble metal nanoparticlesembedded organic�inorganic composite films. MaterialsChemistry and Physics 95: 51�55.

Stoimenov P.K., Klinger R.L., Marchin G.L. and Klabunde K.J.(2002). Metal oxide nanoparticles as bactericidal agents.Langmuir 18: 6679�6686.

Wu Y., Weller C.L., Hamouz F., Cuppett S. and Schnepf M. (2001).Moisture loss and lipid oxidation for precooked ground-beef pat-ties packaged in edible starch-alginate-based composite films.Journal of Food Science 66: 486�493.

Xu T. and Xie C.S. (2003). Tetrapod-like nano-particle ZnO/acrylicresin composite and its multi-function property. Progress inOrganic Coatings 46: 297�301.

Yamamoto O. (2001). Influence of particle size on the antibacterialactivity of zinc oxide. International Journal of Inorganic Materials3: 643�646.

Yamamoto O., Hotta M., Sawai J., Sasamoto T. and Kojima H.(1998). Influence of powder characteristic of ZnO on antibacter-ial activity � effect of specific surface area. Journal of the CeramicSociety of Japan 106: 1007�1011.

Yamamoto O., Komatsu M., Sawai J. and Nakaquwa Z.E. (2004).Effect of lattice constant of zinc oxide on antibacterial charac-teristics. Journal of Materials Science: Material in Medicine 15:847�851.

Yang J.L., An S.J., Park W.I., Yi G.C. and Choi W. (2004).Photocatalysis using ZnO thin films and nanoneedles grown bymetal�organic chemical vapor deposition. Advanced Materials16: 1661�1664.

Yang Z. and Xie C. (2006). Zn2þ release from zinc and zinc oxideparticles in simulated uterine solution. Colloids and Surface B:Biointerfaces 47: 140�145.

Yao K.S., Wang D.Y., Ho W.Y., Yan J.J. and Tzeng K.C. (2007).Photocatalytic bactericidal effect of TiO2 thin film on plant path-ogens. Surface and Coatings Technology 201: 6886�6888.

Zhang L.L., Ding Y.L., Povey M. and York D. (2008). ZnO nano-fluids � a potential antibacterial agent antibacterial. Progess inNatural Science 18: 939�944.

Zhang L.L., Jiang Y.H., Ding Y.L., Povey M. and York D. (2007).Investigation into the antibacterial behaviour of suspensions ofZnO nanoparticles (ZnO nanofluids). Journal of NanoparticleResearch 9: 479�489.

232 X.H. LI ET AL.






DOI: 10.1177/1082013209353986

2010 16: 225 originally published online 12 August 2010Food Science and Technology InternationalX.H. Li, Y.G. Xing, W.L. Li, Y.H. Jiang and Y.L. Ding

Antibacterial and Physical Properties of Poly(vinyl chloride)-based Film Coated with ZnO Nanoparticles

Published by:


On behalf of:



















Antibacterial and Physical Properties of Poly(vinyl

chloride)-based Film Coated with ZnO Nanoparticles

X.H. Li,1 Y.G. Xing,1,* W.L. Li,1 Y.H. Jiang2 and Y.L. Ding2

1Key Laboratory of Food Nutrition and Safety (Tianjin University of Science and Technology),Ministry of Education, Tianjin 300457, P.R. China

2Institute of Particle Science and Engineering, University of Leeds, Leeds, LS29JT, UK

Nanoparticles of ZnO and their application in coating systems have attracted a great deal of attention inrecent years because of its multifunction property, especially antibacterial activity. In this study, antibac-terial and physical properties of poly(vinyl chloride) (PVC) based film coated with ZnO nanoparticles wereinvestigated. It was found that the antibacterial action should be attributed to the killing effect property of

ZnO nanoparticles. The ZnO-coated films treated by shaking for 10 h exhibited a similar high antibacterialactivity against Escherichia coli and Staphylococcus aureus as the untreated ZnO-coated films. This resultindicated that the ZnO nanoparticles adhered very well to the plastic film. The antibacterial activity of the

ZnO-coated film to inactivate E. coli or S. aureus was improved by UV irradiation. The analysis of physicalproperties of the ZnO-coated films revealed that the nano-ZnO particles showed less effects on the tensilestrength and elongation at break of the film. The ultraviolet (UV) light fastness of the ZnO-coated PVC

film was improved, which may be attributed to the absorption of ZnO nanoparticles against UV light.Water vapor transmission of the ZnO-coated film decreased from 128 to 85 g/m2

� 24 h, whereas the thick-ness of film increased from 6.0mm with increasing the amount of nano-ZnO particles coated from 0 to

187.5 mg/cm2. This research revealed that the PVC film coated with nano-ZnO particles has a good poten-tial to be used as an active coating system for food packaging.

Key Words: PVC film, ZnO nanoparticles, active packaging, physical property

INTRODUCTION

Compared to the instability of organic antibacterialmaterials, the inorganic antibacterial agents exhibit agood stability to withstand harsh process conditionssuch as high pressures or temperatures (Sawai, 2003).Therefore, the inorganic bacteriostatic agents such aszinc oxide (ZnO) nanoparticles have attracted a greatdeal of attention over the past decade (Fu et al., 2005;Zhang et al., 2007). The antibacterial property of ZnOparticles has been investigated by many researchers(Yamamoto et al., 1998; Stoimenov et al., 2002; Xuand Xie, 2003). The antibacterial activity of ZnO nano-particles is greater than that of microparticles which alsodepends on the concentration and surface area, whereasthe particle shape and crystalline structure have lesseffect (Sawai et al., 1996a; Yamamoto et al., 1998;Yamamoto, 2001; Zhang et al., 2007). They observed

that antibacterial activity increased with the reduction

in particle size. On the other hand, Liu and Yang (2003)

have reported that the photocatalytic inactivation of

ZnO with 365 nm ultraviolet (UV) light inactivated

Gram-negative Escherichia coli and Gram-positive

Lactobacillus helveticus. Almost all the initial E. coli

cells (108CFU/mL) were inactivated in 40min in the

presence of 2 g/L ZnO. As the results reported by

Zhang et al. (2008), ZnO nanofluids, which were

stored for 120 days under the light, had the best anti-

bacterial behavior against E. coli bacteria. The best anti-

bacterial activity was believed to be caused by the

photocatalytic properties of ZnO particles. So, the

results above show that ZnO nanoparticles as potential

antibacterial agents can be used in many industries such

as food industry.Microbial growth on food surfaces is one of the most

important problems for its storage. However, direct

application of antibacterial substances onto the surface

of foods has been limited due to the fact that some of the

active agents may diffuse rapidly into the food mass

(Pranoto et al., 2005). Thus, there is an urgent need

for alternative technologies to inhibit the changes in

the quality of food during storage. In recent years,

many researchers have focused on investigating active

packaging which can offer functions such as regulation





of moisture and/or provision of antimicrobial activity(Ouattara et al., 2000). The use of active packagingfilms could be more efficient for the storage of foodbecause the active agents can transfer slowly from thepackaging material to the surface of the product. So far,the application of biodegradable films has been seriouslylimited due to their weak mechanical and poor barrierproperties (Li et al., 2009a). Hence, the commonly usedfood packing materials are still natural polymers andsynthetic polymers (Petersen et al., 1999). Plastic filmsare anticipated to continue to play an important role infood distribution because of their better physical prop-erties and barrier properties, higher strength and elon-gation, lower cost and lightness (Arvanitoyannis et al.,1997b; Psomiadou, 1997). Polymers, such as poly(vinylchloride) (PVC) and polyethylene, are the most com-monly used polymeric materials for packaging applica-tions. Therefore, in order to control undesirablemicroorganisms on food surfaces, the methods thatcoat antimicrobial agents onto the surface of the plasticfilm and incorporate antimicrobial agents into the poly-mer can be used (Appendini and Hotchkiss, 2002).

Several researchers have reported the effect of nano-particles on the properties of films. Li et al. (2009a) havereported that the polyurethane (PU) films doped withZnO nanoparticles exhibited excellent antibacterialactivity, especially for E. coli. The antifungal activityof TiO2-coated plastic film against Penicillium expansumhas been investigated by Maneerat and Hayata (2006).Moreover, the antimicrobial property of TiO2-coatedpackaging film against E. coli in vitro and in actualtests under two kinds of artificial light has also beenshown by Chawengkijwanich and Hayata (2008).According to Shanmugam et al. (2006), the Ag and Aunanoparticle-embedded composite film exhibits antimi-crobial activity against Gram-negative E. coli. Thenanopackaging material, which was synthesized byblending polyethylene with nanopowder (nano-Ag,kaolin, anatase TiO2 and rutile TiO2), could be appliedto improve the preservation quality of Chinese jujube(Li et al., 2009a). On the other hand, the effects of mate-rials, such as nanoparticles and starches coated on orincorporated into plastic films, on the mechanical andphysical benefits have been reported by many research-ers (Krochta and Johnston, 1997; Pranoto et al., 2005).Li et al. (2009a) reported that the Young’s modulus andtensile strength of the PU films improved significantlyby incorporating ZnO nanoparticles up to 2.0wt% andthe abrasion resistance of the PU coatings improved aswell. The mechanical properties of biodegradable filmsmade from low-density polyethylene (LDPE), wheatstarch, soluble starch and ethylene acrylic acid (EAA)or polycaprolactone (PCL) were studied byArvanitoyannis et al. (1997b,c) and Psomiadou et al.(1997). Moreover, the physical properties and micro-structure of nanopacking were also studied by Li et al.(2009a).

It seems to be very important for the investigation ofthe possibility of producing antimicrobial film by coat-ing with ZnO nanoparticles. The objective of this studywas to explore the feasibility of producing antibacterialPVC film by coating with ZnO nanoparticles.Antibacterial activity against the food pathogenic bac-teria E. coli and Staphylococcus aureus and the physicalproperty of the PVC film coated with ZnO nanoparticlesunder different conditions were investigated.


Materials

Dry zinc oxide nanoparticles from NanophaseTechnologies (USA) were used in this study. The pri-mary sizes of the nanoparticles given by the manufac-turers were 200�400 nm. Polyethylene glycol 400 (PEG400) from Fluka was used as a dispersant. E. coli and S.aureus were provided by the Department of BiologicalScience of Tianjin Univeristy of Science andTechnology. Luria-Bertani (LB) medium used for grow-ing and maintaining bacteria was purchased fromSigma�Aldrich (UK).

Methods

Preparation of Nano-ZnO Coating

Suspension of ZnO nanoparticles was prepared by themethod reported by Zhang et al. (2007). In order toimprove the stability of the suspension, PEG 400 as adispersant was used and the amount was 10% of theamount of ZnO nanoparticles added. The mixture con-taining a preset amount of dried ZnO nanoparticles anddispersant was mixed with distilled water in a glassbeaker with the aid of a magnetic stirrer. The beakerwas placed in an ultrasonicator for 30min (Clifton,UK) until the particles were dispersed in water. Aftersonication, the obtained suspension was milled using aplanetary Dyno-mill (Willy A. Bachofen, Switzerland)for 1 h at room temperature. Then, the ZnO nanoparti-cle coating was prepared, which had a concentration of5.0 g/L.

Preparation of ZnO-coated PVC Film

ZnO nanofluid (1.5 or 3.0mL) with a concentration of5.0 g/L was coated on the surface of the PVC film(80� 100mm2) which was spread on a plastic support(AFA-II, XD Env. Eng. Tech. Inc., China). After airing,the ZnO-coated PVC film on the carrier support wasroasted and pressed at 100 �C for 10min. Finally, theZnO-coated PVC film was peeled from the substratecarefully when the carrier support was cooled to roomtemperature (Li et al., 2009a). The final concentrations

226 X.H. LI ET AL.



of nano-ZnO on the dried films were 93.75 and187.5mg/cm2, respectively.


Morphology was characterized by scanning electronmicroscopy (SEM). The samples of coated and uncoatedfilms were placed on the SEM stubs using a two-sidedadhesive tape and then, SEM analyses using a LEOGemini 1530 field emission SEM at an acceleratingvoltage of 5 kV after Pt sputtering were performed.

Antibacterial Activity Against E. coli and S. aureus

In order to study how well the ZnO nanoparticlesadhere to the film, the antibacterial activities of ZnO-coated PVC films (80� 100mm2) before and after shak-ing were investigated. The sample film in a 250-mL flaskcontaining 100mL of distilled H2O was shook in a con-stant temperature shaker (Changzhou GuohuaElectrical Appliance Co. Ltd., Jiangsu, China) at120 rpm and 37�C for 10 h. Effects of the sample filmsbefore and after shaking on the growth of E. coli andS. aureus were investigated by the method reported byZhang et al. (2007). The sample film was added to a250mL flask containing 99mL of nutrient broth and1mL of the 24 h grown bacterial culture with an approx-imate concentration of 106�107CFU/mL. The flaskswith the same nutrient broth and grown bacterial culturewith blank film and without film were used as the con-trol and the negative control, respectively. Mediagrowths in flasks were incubated in a constant temper-ature shaker at 120 rpm and 37�C. The growth curveswere determined by measuring the time evolution of theoptical density (OD) of the sample. The experimentswere done with a spectrophotometer (Thermo ElectronCorporation, USA) at 570 nm for S. aureus and at600 nm for E. coli, respectively. Moreover, the inhibitionrate (%) can be calculated according to ([A]i� [A]t)/[A]i,where [A]i is the optical absorption of untreated bacteriaat 8 h for E. coli and 12 h for S. aureus and [A]t is theoptical absorption of treated bacteria (Yao et al., 2007).Photocatalytic inactivation of ZnO nanoparticles

coated on the surface of PVC film against E. coli or S.aureus was performed using the antibacterial drop testmethod with some modifications as described by Yaoet al. (2007). Samples (100mL) of the different bacterialstrains diluted with sterilized distilled H2O were addedonto the surface of each ZnO-coated film. One uncoatedfilm was used as a negative control along with a darkreaction. The bacterial solutions were then irradiatedwith UV A (365 nm) light at room temperature for120min. Then, the treatments were washed from thefilm surfaces with 1mL of 0.05% Tween-20 solution inthe sterilized dish, respectively. Then, 1mL of each bac-terial suspension was pipetted into sterilized plates con-taining 9mL of LB medium and then, the plates were

incubated at 37�C for 16 h. The solutions were read witha spectrophotometer at 570 nm for S. aureus and at600 nm for E. coli, respectively. The optical absorptions[A] were obtained. The inhibition rate (%) can be calcu-lated according to ([A]i� [A]t)/[A]i, where [A]i is the opti-cal absorption of untreated bacteria (Yao et al., 2007).

Elongation at Break (E) and tensile strength (TS)

In order to test the effect of ZnO nanoparticles on thephysical property of PVC film, the elongation at break(E) and tensile strength (TS) of the sample filmuntreated and treated by UV-ray were measured.Various PVC-based films laid on a support were irradi-ated for 6 h under a UV lamp (20W, 365 nm) and thedistance between the support and the UV lamp wasabout 10 cm (Li et al., 2009a). The samples were cutinto a dumbbell shape. Elongation at break (E) and ten-sile strength (TS) of the films were examined using theuniversal material testing machine (New Sansi MaterialDetection Ltd., Shenzhen, China). The films were heldparallel with an initial grip separation of 5 cm and thenpulled apart at a head speed of 200mm/min. Percent Ewas calculated based on the length extended and originallength of the films. Tensile strength was calculated bydividing the maximum force at break (read frommachine) by cross-sectional area of film (Pranotoet al., 2005).

Thickness and Water Vapor Transmission

The thickness of film was measured at several pointswith a hand micrometer (Liuling Instrument, Shanghai,China). WVT of the samples was determined gravimet-rically using the method reported by Pranoto et al.(2005). A cup was covered with the sample film inwhich silica gel was used as a desiccant and placed ina controlled desiccator. The relative humidity was keptat 90±2%, and temperature was kept at 38±0.6 �Cinside the desiccator and checked periodically. Theweight of the cup was measured at 24 h intervalswithin 72 h. The WVT value was expressed ing/m2�24 h. The WVT was calculated according to the

following equation: WVT¼ 24�m/(A�t), where WVTis the water vapor transmission (g/m2

�24 h), �m theincreased weight in the internal time, A the area of per-meation and t is the internal time when the weight of cupincreased constantly.


The tests were carried out in triplicate and experimen-tal dates were analyzed using SPSS 13.0 software (SPSSInc.). The one-way analysis of variance (ANOVA) pro-cedure followed by least significant difference (LSD) testwas used to determine the significant difference(p< 0.05) between the treatment means.






ZnO nanoparticles were dispersed uniformly, but little

agglomerates exist on the film surface (Figure 1, right).

The ZnO particles coated on the surface of film were

about 200 nm and a few particles were smaller than

100 nm (Zhang et al., 2008). Few particles were found

on the surface of the blank film (Figure 1, left). As Xu

and Xie (2003) reported, with decreasing the size of the

ZnO particles, especially getting to nanoscale, antibac-

terial activity of ZnO particles increased sharply (Zhang

et al., 2007). Arvanitoyannis et al. (1997b) have investi-

gated the morphology of LDPE/EAA/Wheat starch

blends after fracture. The SEM observations showed

that the starch particles were occluded and interspersed

within the LDPE/EAA matrix. The formations of metal-

lic silver and gold nanoparticles in composites were stu-

died by Shanmugam et al. (2006). TEM images showed

that Ag and Au nanoparticles were embedded into the

organic�inorganic composite film. According to Li et al.

(2009a), it appeared that the nanoparticles (Ag, TiO2)

were uniformly distributed in the nanopacking film withan irregular shape. The result showed that the nanopar-ticles tended to give better barrier and mechanical prop-erties to the nanopacking film (Li et al., 2009a). TheSEM image showed that the ZnO nanoparticles were uni-formly coated on the PVC film with an irregular shape.Therefore, it is important to investigate howwell the ZnOnanoparticles adhered to the film and the effect of thenanoparticles on the antimicrobial and physical proper-ties of ZnO-coated film.

Effect of ZnO-coated Films Against E. coli andS. aureus

Effects of the sample films before and after shakingfor 10 h on the growth of Gram-negative bacteria E. colior Gram-positive bacteria S. aureus were investigated.The value of the OD rose with an increase in the numberof bacteria due to more light being absorbed. Comparedwith Figures 2a and 2b for E. coli, the increasing trendsof the growth curves were quite similar betweenuntreated and treated by shaking for 10 h for the sameamount of nanoparticles coated. The similar result wasobtained by comparing Figures 3a and 3b for S. aureus.

Figure 1. SEM images of the blank PVC film (left)and ZnO-coated PVC film (right).

0

0.5

1

1.5

2

2.5

3

3.5

4

0

Time (h)

OD

600

nm

(A

)

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

Time (h)

OD

600

nm (

A)

(b)

(a)

2 4 6 8 10

0 2 4 6 8 10

Figure 2. Growth curves of E. coli in LB mediuminoculated with 106�107 CFU/mL of bacteria. (#) neg-ative control; (*) blank film; (¨) film coated with93.75 mg/cm2 ZnO nanoparticles; and (i) filmcoated with 187.5 mg/cm2 ZnO nanoparticles.

228 X.H. LI ET AL.



No significant difference (p< 0.05) in the inhibitionratios of the ZnO-coated film for E. coli or S. aureuswas observed after the treatment of shaking (Table 1).It was reasonable to get the result that the ZnO nano-particles adhered very well to the plastic film. On theother hand, the growth curves of E. coli and S. aureusin LB medium were affected significantly (p< 0.05) bythe presence of nano-ZnO-coated films in comparisonwith the blank film. A small difference in the growthcurves of E. coli or S. aureus between the negative

control and the blank film control can be seen. Thisresult indicated that the antimicrobial activity of ZnO-coated film should be attributed to the nano-ZnO par-ticles. The antibacterial activity against E. coli improvedwith increasing the amount of nanoparticles coated(Figure 2). However, this increasing trend changedslower for S. aureus. The antibacterial behavior ofZnO-coated PVC film under the UV light for 3 h wasalso investigated. With the presence of the same amountof ZnO nanoparticles, the inhibition ratios of the ZnO-coated film samples irradiated by UV light improvedsignificantly (p< 0.05), 29.8% for E. coli and 26.0%for S. aureus, respectively, compared to that of theuntreated ZnO-coated film (Figure 4). This result dem-onstrated that the antibacterial activity of the ZnO-coated film to inactivate E. coli and S. aureus wasimproved by UV irradiation.

The results above indicated that the antimicrobialactivity of ZnO-coated film should be attributed to thenano-ZnO particles (Li et al., 2009a,b). The result alsoagreed with Sawai et al. (1995a, b, 1996a, b) and Nairet al. (2009) when reported ZnO particles were effectivefor inhibiting both Gram-positive and Gram-negativebacteria. While the mechanisms of the antibacterialactivity of ZnO nanoparticles have not yet been clearlyelucidated, Nair et al. (2009) suggested mechanismsincluded the role of reactive oxygen species (ROS)such as O�2 , hydrogen peroxide (H2O2), singlet oxygen(1O2) and hydroxyl radical (�OH) generated on the sur-face of the particles (Sawai et al., 1996b, 1997, 1998;Li et al., 2009a), zincion release (Yang and Xie, 2006),membrane dysfunction (Zhang et al., 2007, 2008) andnanoparticle internalization (Brayner et al., 2006). Anexcellent study by Sawai et al. (1996b, 1998) reportedthat ROS concentrations increased with the ZnO con-tent of slurries. These active oxygen species are toxic tothe bacterial cell because they are very reactive and pow-erful oxidizing agents (Sawai et al., 1998). H2O2 was alsodetected by spectrophotofluorometry by Yamamotoet al. (2004). The study conducted by Zhang et al.(2008) showed that the antibacterial activities in the

0

0.5

1

1.5

2

2.5

3

3.5

4

0Time (h)

OD

570

nm

(A

)

0

0.5

1

1.5

2

2.5

3

3.5

Time (h)

OD

570

nm

(A

)

(a)

(b)

2 4 6 8 10 12 14

0 2 4 6 8 10 12 14

Figure 3. Growth curves of S. aureus in LB mediuminoculated with 106�107 CFU/mL of bacteria. (#) neg-ative control; (*) blank film; (¨) film coated with93.75 mg/cm2 ZnO nanoparticles; and (i) filmcoated with 187.5 mg/cm2 ZnO nanoparticles.

0

20

40

60

80

100

Blank film+UV

Inhi

bitio

n ra

te (

%)

ZnO-coated film

ZnO-coated film+UV

Figure 4. Inhibition rates of PVC-based film with orwithout UV («, the blank film with UV; #, the ZnO-coated film (93.75 mg/cm2) without UV; «, the ZnO-coated film (93.75 mg/cm2) with UV irradiation).

Table 1. Comparisons of inhibition rates ofZnO-coated PVC film with different treatments against

E. coli and S. aureus.

ZnO-coated films(mg/cm2) E. coli S. aureus

93.75 74.52±1.07 a 91.20±0.62 a187.5 85.02±0.79 b 95.91±0.46 b93.75 (shaking) 75.40±0.81 a 91.31±0.38 a187.5 (shaking) 84.17±0.23 b 96.54±0.71 b





ZnO nanofluids with longer storage period had a higherefficiency. With longer storage period, there is morechance to produce more amount of active oxygen,such as H2O2 (Zhang et al., 2008). The ZnO nanofluidsstored for 120 days under the light had a better antibac-terial behavior against E. coli bacteria than that underthe dark (Zhang et al., 2008). The increasing inhibitionrates of ZnO-coated PVC film irradiated by UV lightwas presumably due to the photocatalytic reactions ofthe ZnO particles (Liu and Yang, 2003; Zhang et al.,2008). When the ZnO-coated film was irradiated byUV, there is more chance to have certain photocatalyticreactions. The production of the active oxygen speciescan be promoted in the ZnO sample kept under the UVlight (Yang et al., 2004; Zhang et al., 2008). On the otherhand, under the UV illuminated process, ZnO is acti-vated to generate a conduction electron and a hole(ZnO-hþ). This kind of hole can activate oxygen in airand give active oxygen with very strong chemical activity(Xu and Xie, 2003). According to Liu and Yang (2003),the creation of the electron�hole pair is very importantfor the oxidation process. The water molecule can besplit into OH and Hþ with the hole. The externally sup-plied or dissolved oxygen molecules are transformed to asuperoxide radical anion (�O�2 ) which reacts with Hþ togenerate a hydrogen peroxide radical (�HO�2 ). On sub-sequent collisions with an electron and a hydrogen ion, amolecule of H2O2 is produced. On the other hand, thepresence of ZnO nanoparticles leads to damages in themembrane wall of E. coli (Zhang et al., 2007, 2008).Such damages may be partly due to direct interactionsbetween ZnO nanoparticles and bacterial membranesurfaces, which were also suggested by Stoimenovet al. (2002) and Nair et al. (2009).

Elongation at Break and Tensile Strength

Both properties of film, elongation at break and tensilestrength are important characteristics for packagingmaterial. They are two measures of stretchability priorto breakage and film strength (Krochta and Johnston,1997; Pranoto et al., 2005). In order to investigatethe effect of ZnO nanoparticles on the mechanical prop-erty of the PVC film, the elongation at break and ten-sile strength values of sample films were measured(Figure 5). Coating of nano-ZnO particles on the surfaceof film had a negligible effect on the elongation at breakand tensile strength values of the PVC film. The resultobtained was different from that reported by Li et al.(2009a). In their investigation, the ZnO nanoparticlesincorporated into the PU chains limited the movingscale of chain segments and generated an interactiveforce against the chains. This fact induced that the tensilestrength increases first and then decreases with anincrease in ZnO content, and the elongation at rupturevaries inversely. The property of materials incorporatedinto the film has an important effect on the mechanical

properties. As reported by Psomiadou et al. (1997), thepresence of starch at contents >30% had an adverseeffect on the mechanical properties of LDPE�starchblends. In this article, the ZnO nanoparticles were notfilled into the interstice of PVC chains but were coatedon the surface of the film. So, the interaction forcebetween the nanoparticles and the PVC chains may notbe generated. Therefore, the ZnO particles coated on thefilm had less effect on the mechanical properties of PVC-based film. Figure 5 shows the down trend of elongationat break and tensile strength of the films by the irradia-tion. While tensile strength (Figure 5a) and elongation atbreak (Figure 5b) of the ZnO-coated films decreases asthe pure films do, they are still larger than that of the purePVC films. It was confirmed that the UV light fastness ofthe PU composite films was improved, which may beattributed to the absorption and dispersion of ZnO nano-particles against UV light (Xu and Xie, 2003; Li et al.,2007, 2009a). The longitudinal strength of nanopacking,which was synthesized by blending polyethylene withnano-Ag, kaolin, anatase TiO2 and rutile TiO2 was1.24-fold higher than that of the control. These resultsindicated that the novel material possessed bettermechanical properties (Li et al., 2009a).

Thickness and WVT

WVT of film has a great influence on the shelf life offood, which is a measure of ease of the moisture to

0

5

10

15

20

25

30

0

Time (h)

Ten

sile

str

engt

h (M

Pa)

0

50

100

150

200

250

Time (h)E

long

atio

n at

bre

ak(%

)

(a)

(b)

3 6

0 3 6

Figure 5. Tensile strength (a) and elongation atbreak (b) related to the irradiation time. (*) theblank film; (¨) the ZnO-coated film (93.75mg/cm2);and (i) ZnO-coated film (187.5mg/cm2).

230 X.H. LI ET AL.



penetrate and pass through a material (Li et al., 2006).The ability to control the loss of water molecule fromthe product for the film is an important characteristicthat affects the quality of the food product (Wu et al.,2001). As shown in Table 2, addition of ZnO nanopar-ticles on the film decreased WVT from 128 to85 g/m2

�24 h and the thickness of ZnO-coated filmincreased from 6.0mm. The WVT value of PVC filmscoated with ZnO nanoparticles is lower than that ofthe neat film, and the WVT decreased significantly(p< 0.05) with increasing the amount of nanoparticles.It is probably due to that the ZnO nanoparticles coatedon one side of film prevent a part of the moisture frompassing through the PVC film. The WVT value shouldbe taken into account for the packaging film or othercoating materials when it applies to a moist food prod-uct. In general, the WVT value decreased with increasingthe amount of ZnO nanoparticles. Therefore, thisZnO-coated film could keep more water molecules inthe packaging system and thereby delay the shelf lifeof some kinds of foods, such as fruits and vegetables.WVT of film was related to the materials coated onto orincorporated into the plastic films. As reported byArvanitoyannis et al. (1997b), an increase in starchwith the strong hydrophilic character and/or PCL con-tent was directly proportional to WVTR of the plasticfilm. It was well known that the main disadvantage ofedible films has been their high WVTR. For this reasonit was important either to use a high/multilayer or ablend with a hydrophobic polymer (Arvanitoyanniset al., 1997a). The transmission rates of relative humid-ity and oxygen of the nanopacking material, which wassynthesized by blending polyethylene with nanoparticlesincluding nano-Ag, kaolin, anatase TiO2 and rutileTiO2, were decreased by 28.0% and 2.1% comparedwith those of normal packing. These results indicatedthat the novel material possessed a better barrier prop-erty (Li et al., 2009a).

CONCLUSIONS

SEM image of ZnO-coated PVC film clearly indicatedthat the ZnO nanoparticles were uniformly coated onthe PVC film with an irregular shape. On the aspect of

antibacterial function, no significant difference(p< 0.05) was observed for the inhibition ratios of theZnO-coated film against E. coli or S. aureus after shak-ing for 10 h. This result indicated that the ZnO nanopar-ticles adhered very well to the plastic film. Theantibacterial activity of the ZnO-coated film to inacti-vate E. coli or S. aureus was improved by 29.8% and26.0% for E. coli and S. aureus, respectively, after irra-diating by UV light. Coating of ZnO nanoparticlesenhanced the UV light fastness of the PVC film buthad less effects on the TS and elongation at break ofthe ZnO-coated film. The WVT and the thickness ofZnO�PVC film were significantly affected (p< 0.05) bythe addition of ZnO nanoparticles. WVT of the ZnO-coated film decreased from 128 to 85 g/m2

� 24 h, whereasthe thickness of film increased from 6.0 mm with increas-ing the amount of nano-ZnO particles to 187.5 mg/cm2.This research indicated that PVC films coated withnano-ZnO particles have a good potential to be usedfor food packaging.

ACKNOWLEDGMENTS

This study is supported by the Outstanding DoctorateDissertation Foster Foundation of Tianjin University ofScience and Technology (B201003) and by the 11thFive-Year Key Technologies R&D Program of China(2006BAD22B03). This research is also supported bythe Nanomanufacturing Institute of the University ofLeeds.

REFERENCES

Appendini P. and Hotchkiss J.H. (2002). Review of antimicrobial foodpackaging. Innovative Food Science and Emerging Technologies 3:113�126.

Arvanitoyannis I., Kolokuris I., Nakayama A. and Aiba S. (1997a).Preparation and study of novel biodegradable blends based ongelatinized starch and 1,4-trans-polyisoprene (gutta percha) forfood packaging or biomedical applications. CarbohydratePolymers 34: 291�302.

Arvanitoyannis I., Psomiadou E., Biliaderis C.G., Ogawa H.,Kawasaki N. and Nakayama A. (1997b). Biodegradable filmsmade from low density polyethylene (LDPE), ethylene acrylicacid (EAA), polycaprolactone (PCL) and wheat starch for foodpackaging applications: Part 3. Starch/Starke 49: 306�322.

Arvanitoyannis I., Psomiadou E., Ogawa H. and Kawasaki N.(1997c). Biodegradable films made from low density polyethylene(LDPE) and rice starch for food packaging Uses; Part 1.Carbohydrate Polymers 36: 89�104.

Brayner R., Ferrari-Iliou R., Brivois N., Djediat S., Benedetti M.F.and Fievet F. (2006). Toxicological impact studies based onEscherichia coli bacteria in ultrafine ZnO nanoparticles colloidalmedium. Nano Letters 6: 866�870.

Chawengkijwanich C. and Hayata Y. (2008). Development of TiO2

powder-coated food packaging film and its ability to inactivateEscherichia coli in vitro and in actual tests. International Journalof Food Microbiology 123: 288�292.

Table 2. WVT and thickness of PVC film with differentamounts of ZnO nanoparticles.

ZnO-coated films(mg/cm2) WVT (g/m2

�24 h) Thickness (�m)

0 (control) 128±1.04 c 48.77±0.06 a93.75 107±0.52 b 50.30±0.10 b187.5 85±0.66 a 53.77±0.10 c





Fu G., Vary P.S. and Lin C.T. (2005). Anatase TiO2 nanocompositesfor antimicrobial coating. Journal of Physical Chemistry B 109:8889�8898.

Krochta J.M. and Johnston C.D.M. (1997). Edible and biodegradablepolymer films: challenges and opportunities. Food Technology 51:61�74.

Li J.H., Hong R.Y., Li M.Y., Li H.Z., Zheng Y. and Ding J. (2009a).Effects of ZnO nanoparticles on the mechanical and antibacterialproperties of polyurethane coatings. Progress in Organic Coatings64: 504�509.

Li H., Li B., Kennedy J.F., Peng J.L., Yie X. and Xie B.J. (2006).Preparation and performance evaluation of glucomannan�chito-san�nisin ternary antimicrobial blend film. CarbohydratePolymers 65: 488�494.

Li F., Wang L., Sheng J., Xin Z., Zhao L., Xiao H., Zheng Y. andHu Q. (2009b). Effect of nano-packing on preservation quality ofChinese jujube (Ziziphus jujuba Mill. var. inermis (Bunge) Rehd).Food Chemistry 114: 547�552.

Li Y.Q., Yang Y. and Fu S.Y. (2007). Photo-stabilization properties oftransparent inorganic UV-filter/epoxy nanocomposites.Composites Science and Technology 67: 3465�3471.

Liu H.L. and Yang T.C.K. (2003). Photocatalytic inactivation ofEscherichia coli and Lactobacillus helveticus by ZnO and TiO2

activated with ultraviolet light. Process Biochemistry 39:475�481.

Maneerat C. and Hayata Y. (2006). Antifungal activity of TiO2 photo-catalysis against Penicillium expansum in vitro and in fruit tests.International Journal of Food Microbiology 107: 99�103.

Nair S., Sasidharan A., Rani V.V.D., Menon D., Nair S., Manzoor K.and Raina S. (2009). Role of size scale of ZnO nanoparticles andmicroparticles on toxicity toward bacteria and osteoblast cancercells. Journal of Materials Science: Materials in Medicine 20:235�241.

Ouattara B., Simard R.E., Piette G., Begin A. and Holley R.A. (2000).Diffusion of acetic and propionic acids from chitosan-based anti-microbial packaging films. Journal of Food Science 65: 768�773.

Petersen K., Nielsen P.V., Bertelsen G., Lawther M., Olsen M.B.,Nilssonk N.H. and Mortensen G. (1999). Potential of biobasedmaterials for food packaging. Trends in Food Science andTechnology 10: 52�68.

Pranoto Y., Salokhe V.M. and Rakshit S.K. (2005). Physical andantibacterial properties of alginate-based edible film incorpo-rated with garlic oil. Food Research International 38: 267�272.

Psomiadou E., Arvanitoyannis I., Biliaderis C.G., Ogawa H. andKawasaki N. (1997). Biodegradable films made from low densitypolyethylene (LDPE), wheat starch and soluble starch for foodpackaging applications: Part 2. Carbohydrate Polymers 33:227�242.

Sawai J. (2003). Quantitative evaluation of antibacterial activities ofmetallic oxide powders (ZnO, MgO, CaO) by conductimetricassay. Journal of Microbiological Methods 54: 177�182.

Sawai J., Igarashi H., Hashimoto A., Kokugan T. and Shimizu M.(1995a). Evaluation of growth inhibitory effect of ceramicspowder slurry on bacteria by conductance method. Journal ofChemical Engineering of Japan 28: 288�293.

Sawai J., Igarashi H., Hashimoto A., Kokugan T. and Shimizu M.(1996a). Effect of particle size and heating temperature of cera-mic powders on antibacterial activity of their slurries. Journal ofChemical Engineering of Japan 29: 251�256.

Sawai J., Kawada E.I., Kanou F., Igarashi H., Hashimoto A.,Kokugan T. and Shimizu M. (1996b). Detection of activeoxygen generated from ceramic powders having antibacterialactivity. Journal of Chemical Engineering of Japan 29: 627�633.

Sawai J., Kojima H., Igarashi H., Hashimoto A., Shoji S.,Takehara A., Sawaki T., Kokugan T. and Shimizu M. (1997).Escherichia coli damage by ceramic powder slurries. Journal ofChemical Engineering of Japan 30: 1034�1039.

Sawai J., Saito I., Kanou F., Igarashi H., Hashimoto A., Kokugan T.and Shimizu M. (1995b). Mutagenicity test of ceramic powderwhich have growth inhibitory effect on bacteria. Journal ofChemical Engineering of Japan 28: 352�354.

Sawai J., Shouji S., Igarashi H., Hashimoto A., Kokugan T.,Shimizu M. and Kojima H. (1998). Hydrogen peroxide as anantibacterial factor in zinc oxide powder slurry. Journal ofFermentation and Bioengineering 86: 521�522.

Shanmugam S., Viswanathan B. and Varadarajan T.K. (2006). Anovel single step chemical route for noble metal nanoparticlesembedded organic�inorganic composite films. MaterialsChemistry and Physics 95: 51�55.

Stoimenov P.K., Klinger R.L., Marchin G.L. and Klabunde K.J.(2002). Metal oxide nanoparticles as bactericidal agents.Langmuir 18: 6679�6686.

Wu Y., Weller C.L., Hamouz F., Cuppett S. and Schnepf M. (2001).Moisture loss and lipid oxidation for precooked ground-beef pat-ties packaged in edible starch-alginate-based composite films.Journal of Food Science 66: 486�493.

Xu T. and Xie C.S. (2003). Tetrapod-like nano-particle ZnO/acrylicresin composite and its multi-function property. Progress inOrganic Coatings 46: 297�301.

Yamamoto O. (2001). Influence of particle size on the antibacterialactivity of zinc oxide. International Journal of Inorganic Materials3: 643�646.

Yamamoto O., Hotta M., Sawai J., Sasamoto T. and Kojima H.(1998). Influence of powder characteristic of ZnO on antibacter-ial activity � effect of specific surface area. Journal of the CeramicSociety of Japan 106: 1007�1011.

Yamamoto O., Komatsu M., Sawai J. and Nakaquwa Z.E. (2004).Effect of lattice constant of zinc oxide on antibacterial charac-teristics. Journal of Materials Science: Material in Medicine 15:847�851.

Yang J.L., An S.J., Park W.I., Yi G.C. and Choi W. (2004).Photocatalysis using ZnO thin films and nanoneedles grown bymetal�organic chemical vapor deposition. Advanced Materials16: 1661�1664.

Yang Z. and Xie C. (2006). Zn2þ release from zinc and zinc oxideparticles in simulated uterine solution. Colloids and Surface B:Biointerfaces 47: 140�145.

Yao K.S., Wang D.Y., Ho W.Y., Yan J.J. and Tzeng K.C. (2007).Photocatalytic bactericidal effect of TiO2 thin film on plant path-ogens. Surface and Coatings Technology 201: 6886�6888.

Zhang L.L., Ding Y.L., Povey M. and York D. (2008). ZnO nano-fluids � a potential antibacterial agent antibacterial. Progess inNatural Science 18: 939�944.

Zhang L.L., Jiang Y.H., Ding Y.L., Povey M. and York D. (2007).Investigation into the antibacterial behaviour of suspensions ofZnO nanoparticles (ZnO nanofluids). Journal of NanoparticleResearch 9: 479�489.

232 X.H. LI ET AL.






DOI: 10.1177/1082013209353836

2010 16: 233 originally published online 12 August 2010Food Science and Technology InternationalF.V. Romeo, S. De Luca, A. Piscopo, V. Santisi and M. Poiana

Shelf-life of Almond Pastry Cookies with Different Types of Packaging and Levels of Temperature

Published by:


On behalf of:



















Shelf-life of Almond Pastry Cookies with Different Types of

Packaging and Levels of Temperature

F.V. Romeo,* S. De Luca, A. Piscopo, V. Santisi and M. Poiana

Department of Biotechnologies for Agricultural Food and Environmental Monitoring (BIOMAA),Mediterranean University of Reggio Calabria, Feo di Vito 89060 (RC), Italy

Almond pastries are typical cookies of the south of Italy. Introduction of new packaging for this kind ofcookies requires shelf-life assessments. This study, related to different types of packaging under variousstorage conditions of time and temperature, identifies critical parameters, as color and texture, to trackduring storage studies and to extend the shelf-life. The cookies were packed in three different ways and

stored at two different temperatures. The pastries were separately stored: (1) in polyvinylchloride film;(2) in aluminum foil (ALL); (3) with modified atmosphere (MAP) in plastic vessels sealed into a polyamide/polyethylene film; and (4) in vessels without any polymeric film. The storage temperatures were 20 and

30 �C. Evolution of texture, water activity, dry matter and color was assessed. Texture was evaluated by atexture analyzer with a puncturing test. Indices for hardening were the area under the curve (N�mm) upto 10mm of distance, and the maximum force (N) corresponding to the crust fracture. The best results were

obtained with ALL packaging and MAP condition, and above all, in all the trials a temperature of 30 �Creduced the crust hardness.

Key Words: almond pastries, color, hardening, packaging, shelf-life, texture

INTRODUCTION

Almond pastry cookies are a typical product of the

south of Italy and in some areas they are a very much

requested product. This product normally is flavored by

citrus flavors like orange, citron, bergamot or mandarin,

or pistachio paste is added to it. Storage stability or

the shelf-life of baked products could be defined as

maintenance of the sensory and physical characteristics

associated with freshness (Baixauli et al., 2008). These

almond pastry cookies are a craftsmanlike food, which

suffer a brief shelf-life, particularly due to the ingredi-

ents’ characters and improper packaging applied. The

shelf-life is limited by a qualitative decay due to severe

hardening of the internal paste. This alteration could

be due to both the redistribution of water that leads to

the sugar crystallization and loss of water into the sur-

rounding environment (Farris et al., 2008). In fact, the

almond pastry cookies are high in sugar and low in

moisture. The sucrose recrystallization causes release

of water and subsequent redistribution of moisture.

The water could be absorbed by other components

and lost from cookies. The hypothesis that sucrose

recrystallization is responsible in part for the firming

of soft cookies was demonstrated by Belcourt and

Labuza (2007). In general, cookies have the property

of bending after baking unlike biscuits that break.

This fact could depend on the biscuits’ lower water

activity and moisture content than cookies (Piga et al.,

2005). The maintenance of the fragrance, the standard

of hygiene and the nutritional quality of this kind of

cookies has three main hurdles: the contact with air

oxygen which can improve the lipids oxidization; the

enzymatic activity which contributes to the accelera-

tion of shelf-life; and the contaminant bacteria, molds

and yeasts that represent a real microbiological risk.

Regarding the last case, several authors have found a

great heterogeneity in the total bacteria popula-

tion of almond paste: in fact, genera belonging to

Enterobacteriaceae, Micrococcaceae and Bacillaceae

were isolated (La Rosa et al., 2000). Little research has

been done on cookies, but it could be possible to pre-

vent all the causes of deterioration, to prolong the

nutritional and sensorial quality through a modified

atmosphere (MAP). In fact, the MAP conditions can

protect the color, the taste and the nutritional value

of the packaged food and, as a consequence, protect

their economic value. MAP of food is capable of signif-

icantly extending the shelf-lives of various products

by altering the relative proportions of the surrounding

atmospheric gases (Seiler, 1989; Kotsianis et al., 2002).

*To whom correspondence should be sent(e-mail: [email protected]).Received 18 December 2008; revised 15 May 2009.




The use and application of carbon dioxide and nitro-

gen and the microbiological response to these gases

were investigated by several authors (Sutherland et al.,

1977; Gill and Tan, 1980; Farber, 1991; Guinot et al.,

2003a, b; Suppakul et al., 2003). Almond paste isobtained by mixing and grinding together sweet

almonds and sugar. Percentages of the two ingredients

can vary and more substances (bitter almonds, whole

or white eggs, natural flavors, glucose and water) can

be added as a function of the desired final product(Baiano and Del Nobile, 2005). Some phases of the pro-

duction process of these cookies, such as mixing and

grinding almonds and sugar, are often done by hand

as in handicraft small confectionery. Quality decay of

almond paste is due to a weight loss strictly relatedto the loss of moisture. Also, almond paste cookies

suffer water migration, decrease of moisture content

and the consequent hardening. Moreover, almond

seeds are particularly rich in fats which are characterized

by a high degree of unsaturation that makes this productvery sensitive to oxidation (Severini et al., 2003).

Oxidation and hardening are induced by oxygen and

moisture that permeates through the packaging into

headspace. So, oxygen can be excluded from package

headspace by using MAP. Nitrogen is commonly usedin fat food preservation in addition to positioning the

oxygen scavengers within packets (Baiano and Del

Nobile, 2005).The aim of this work was the study of shelf-life of

almond pastries stored in different types of packagesand at different levels of temperatures near room condi-

tions. The parameters observed are related to color and

texture changes.


Samples

Almond pastry cookies were produced by a handi-

craft confectioner (Fiore Nuovo, Reggio Calabria,

Italy). The amounts of each ingredient were the follow-

ing: 1000 g almonds, 1300 g sugar, 200 g icing sugar,

100 g honey and 12 white eggs. Unroasted, peeledalmonds were washed by immersion in lukewarm

water for 30min, and then dried by oven. So, they

were ground alone and then mixed with sugar, icing

sugar and honey, and gently ground again to obtain a

fine meal to which the white eggs were added. Theobtained almond paste was worked in the form of 25 g

weight pastries that were cooked at 180 �C for 15min in

a professional oven and cooled at room temperature.

The cookies were not round in shape, but had

6±0.6 cm length, 3±0.3 cm width and 1.5±0.3 cmheight.

Methods

Packaging and Storage

The pastries were divided into four lots and then

wrapped into the following materials: (1) 15 mm film

thickness of polyvinylchloride (PVC); (2) aluminumfoil (ALL); and (3) plastic vessels sealed into a 170 mm

film thickness of polyamide/polyethylene (PA/PE) under

MAP with N2. Another sample was only stored intouncovered vessels (in contact with air) representing the

control. The packed pastries were stored at two different

temperatures: 20 �C to reproduce deteriorative processesthat occur at room temperature; and 30 �C, for the hot-

test season. Three samples for each trial were withdrawn

and analyzed at 5, 10, 15, 30 and 45 days of storage.

Water Activity (aw) and Dry Matter Determinations

The water activity (aw) was measured by an Aqua lab

(3TE, Decagon devices Inc., Washington) apparatus,

which uses the chilled-mirror dew-point technique tomeasure the aw of a sample. The dry matter content

was determined by oven drying at 105 �C up to constant

weight. These analyses were affected on six homogenizedpastries.

Color Determination

The color of the pastries was measured using a reflec-tion colorimeter (Minolta CR 300, Osaka, Japan). The

CIE L*a*b* coordinates were measured using D65 illu-

minant. This analysis was assessed on five points ofevery pastry and for five pastries randomly chosen on

each trial. Chroma (C*) was calculated as (a*2þ b*2)1/2.

Texture Analysis

The hardness evolution was assessed on six pastriesfor each trial using a Texture Analyzer (mod. TA.TX2,

Stable Microsystems, Surrey, UK). The Texture expert

program version 1.22 was used for data analysis. A4mm diameter cylinder probe (mod. P/4) for the punc-

turing test was used. Parameters of each test are

reported in table below. Hardness measurement of sam-ples involved plotting force (N) versus distance (mm)

and two parameters were calculated: (1) area under thecurve (as N�mm) up to 10mm of puncturing and

(2) the maximum force (N) as an index for the crust

hardness.The settings for texture measurement of almond paste

pastries were performed to measure the force (N) in

compression up to 25mm distance, with 3mm/s pre-

test speed, 1mm/s test speed, 10mm/s post-test speedand trigger auto and force set to 20 g.

234 F.V. ROMEO ET AL.




Statistica software (version 6.0, StatSoft Inc.) wasused for data processing. Data were evaluated by athree-factor completely randomized factorial design.Storage time, type of packaging and temperature condi-tion were chosen as variables. Duncan’s multiple rangetest was used to compare means when a significantvariation was highlighted by analysis of variance.


Observing the different packaging conditions at 20 �Cstorage temperature, the qualities of PVC-packed cook-ies and the control uncovered sample decrease when awis near 0.5 value, while this decrease was less at 30 �C.The ALL and MAP samples showed an unclear trend

but it was very much different with respect to the others.At both 20 and 30 �C, at the end of storage, the MAPsamples reached the highest value. During the storagetime, the MAP-packaged samples showed the highest awvalue at both temperatures, while PVC and control camethe last and showed no statistical difference after 45 daysof storage at 20 and 30 �C (Table 1).

These results are confirmed by the analysis of drymatter content showed in Table 2. In fact, it showsthat the PVC and control cookies were the sampleswith the highest loss of water with both temperatureconditions, while MAP had the lowest one. A correla-tion between the aw and dry matter values was foundonly for the control sample at 30 �C (R2

¼ 0.898;data not shown), while for all the other samples, thedecrease in aw was not correlate to the dry mattercontent. The behavior of the different types of packag-ing used suggests that MAP conditions and ALLfilm can control the water loss from the cookies to the

Table 1. Water activity of almond paste pastries wrapped in different types of packaging.

Water activity1

T (�C) Time (days) ALL MAP PVC Control Significance2

20 0 0.679 0.679 0.679 0.679 n.s.5 0.712 0.722 0.701 0.699 n.s.

10 0.657 b 0.720 a 0.698 a 0.629 b **20 0.746 a 0.744 a 0.530 b 0.513 b **30 0.700 b 0.728 a 0.554 c 0.532 d **45 0.652 b 0.750 a 0.521 c 0.530 c **

30 0 0.679 0.679 0.679 0.679 n.s.5 0.770 a 0.756 a 0.727 b 0.755 a **

10 0.756 a 0.734 ab 0.713 b 0.711 b *20 0.696 a 0.656 b 0.656 b 0.689 a *30 0.612 c 0.759 a 0.663 b 0.750 a **45 0.700 b 0.737 a 0.567 c 0.591 c **

1Data followed by different letters are significantly different by Duncan’s multiple range test.2Significance at *p<0.05; **p<0.01; n.s., not significant.

Table 2. Dry matter of almond paste pastries wrapped in different types of packaging.

Dry matter (g/100 g)

T (�C) Time (days) ALL MAP PVC Control Significance1

20 0 91.83 91.83 91.83 91.83 n.s.5 92.68 bc 93.98 a 92.42 c 93.58 ab *

10 93.65 c 93.39 c 95.00 b 95.58 a **20 92.87 b 92.38 b 96.68 a 96.93 a **30 94.18 b 93.27 c 97.49 a 97.32 a **45 94.81 b 93.15 c 97.36 a 97.35 a **

30 0 91.83 91.83 91.83 91.83 n.s.5 92.49 b 93.93 a 93.37 a 94.16 a *

10 93.16 b 92.97 b 94.16 ab 94.53 a *20 92.65 c 93.17 b 94.01 a 94.25 a **30 93.02 c 92.55 c 94.80 b 95.84 a **45 93.77 b 92.86 c 96.63 a 96.29 a **

1Data followed by different letters are significantly different by Duncan’s multiple range test.Significance at *p<0.05; **p<0.01; n.s., not significant.

Shelf-life of Almond Pastry Cookies 235



surrounding environment, according to the imperme-able characteristics of ALL and PA/PE films usedfor MAP.

Color parameters measured at 20 and 30 �C are,respectively, reported in Tables 3 and 4. The L* valuewas the highest for PVC samples at the end of

Table 3. CIE L*a*b* parameters and Chroma (C*) for the different types of packaging of almond paste pastriesduring storage at 20� C.

Time of storage1 (days)

Color parameter Sample 0 5 10 20 30 45

L* Control 54.61 59.28 a 58.05 58.93 a 56.76 a 57.38 bALL 56.44 60.15 a 56.01 54.27 c 53.76 b 57.00 bMAP 53.40 60.51 a 56.01 57.53 ab 53.83 b 54.60 cPVC 57.86 56.02 b 58.16 56.08 bc 57.07 a 63.25 aSignificance n.s. ** n. s. ** * **

a* Control 16.42 13.17 13.37 ab 12.83 b 13.95 ab 13.76 aALL 17.69 11.50 14.01 a 14.65 a 14.92 a 13.15 abMAP 16.71 12.20 13.80 a 12.46 b 13.88 ab 14.43 aPVC 16.40 12.64 12.05 b 14.70 a 13.24 b 12.56 bSignificance n.s. n.s. * ** * **

b* Control 29.23 34.35 ab 35.13 a 34.45 35.51 36.46 aALL 27.76 33.61 bc 34.64 a 34.95 35.37 35.91 abMAP 28.53 34.61 a 33.38 b 34.42 34.78 35.19 bPVC 29.76 32.80 c 34.67 a 34.73 35.06 35.34 bSignificance n.s. ** ** n.s. n.s. **

C* Control 33.59 36.81 a 37.63 a 36.84 b 38.17 ab 39.02 aALL 33.00 35.59 b 37.40 a 37.91 a 38.40 a 38.42 abMAP 33.07 36.73 a 36.17 b 36.62 b 37.49 b 38.05 bPVC 34.04 35.21 b 36.79 ab 37.73 a 37.52 b 37.56 bSignificance n.s. ** * ** * **

1Data followed by different letters are significantly different by Duncan’s multiple range test.Significance at *p< 0.05; **p< 0.01; n.s., not significant.

Table 4. CIE L*a*b* parameters and Chroma (C*) for the different types of packaging of almond paste pastriesduring storage at 30� C.

Color parameter Sample

Time of storage1 (days)

0 5 10 20 30 45

Control 54.61 58.60 a 57.00 a 54.72 52.49 b 57.52 aL* ALL 56.44 56.46 b 55.72 ab 54.67 53.94 ab 55.74 ab

MAP 53.40 55.24 b 54.17 b 53.04 53.71 ab 52.12 cPVC 57.86 59.10 a 57.62 a 54.65 55.39 a 55.46 bSignificance n.s. ** * n.s. * **Control 16.95 11.46 13.65 13.57 15.00 a 13.49 a

a* ALL 15.24 12.23 12.97 13.96 13.75 b 13.10 abMAP 16.18 13.32 13.79 13.62 13.07 b 13.93 aPVC 16.00 13.10 12.76 13.13 12.68 b 12.31 bSignificance n.s. n.s. n.s. n.s. ** *Control 30.43 33.35 b 34.87 ab 35.50 a 35.56 37.27 a

b* ALL 30.32 33.32 b 35.33 a 35.35 a 35.89 37.10 aMAP 29.74 33.03 b 34.08 b 33.71 b 35.29 35.24 bPVC 28.65 34.47 a 35.62 a 35.27 a 36.09 35.87 bSignificance n.s. ** * ** n.s. **Control 34.85 35.34 b 37.49 38.10 a 38.62 39.66 a

C* ALL 34.08 35.56 b 37.66 38.04 a 38.40 39.36 aMAP 33.93 35.67 b 36.79 36.37 b 37.69 37.92 bPVC 32.88 36.95 a 37.87 37.71 a 38.28 37.98 bSignificance n.s. ** n.s. ** n.s. **

1Data followed by different letters are significantly different by Duncan’s multiple range test.Significance at *p< 0.05; **p< 0.01; n.s., not significant.




measurements at 20 �C, but the different types of pack-aging have not showed a specific trend respect to thestorage time with both temperatures. The samplesshowed on average values lower at 30 �C than at20 �C. As far as a* and b* parameters are concerned,very little differences in the different types of packagingresulted. The data showed only an increase of b* valueat the two different temperatures and an unclear trendfor the a* parameter. The values of the latter were thelowest in PVC samples after 30 days. But more informa-tion about the sample browning is given by chromavalues (Tables 3 and 4). The ALL and the control sam-ples reached the highest value after 30 days and had asimilar trend at both temperatures. All samples showedincreased values during the observation time, showing asignificance with respect to the storage time besides thatshowed in the Tables 3 and 4. The permeability of thefilm used for packaging and the applied temperaturecertainly had an influence on the sample browning;above all at the lowest temperature the samples had aconstant increase, while the PVC sample reached theplateau after 20 days. At 20 �C, the MAP samplesshowed a positive correlation between aw and chroma(R2¼ 0.88; data not reported); this could be explained

by the role of water activity in the browning process.With MAP packaging, the permeability of the film islow and so also the oxygen contact. Moreover, the phys-ical state of cookies constituents, lipids in particular,probably does not change at 20 �C. The permeabilityof the packaging material, and the phase transitiondue to the 30 �C, influenced the process. The best per-formance of MAP samples at 30 �C is confirmed by the

other results discussed about water activity and drymatter analyses and, above all, by the texture resultsmentioned below. A difference between the samplesstorage at 20 and 30 �C was the kinetic of the brownnessreactions; in fact, in some cases, the samples reachedslightly higher values of chroma at 30 �C than after thesame time at 20 �C. It could be due to an acceleration ofthe process depending on the rise in temperature.

An example of penetration curve is showed inFigure 1. Texture analysis is reported in Figures 2 and3 as means of the area under the curve up to 10mm ofpuncturing, and in Figures 4 and 5, as means of themaximum peak force. The results showed substantialdifferences among the samples; in fact, as shown inFigure 1, the area increased from 17.32 (N�mm) infreshly baked cookies to 522.34 (N�mm) in controlcookies at 20 �C after 45 days, while only 51.49(N�mm) were reached by the MAP samples at thesame time and temperature. For all samples, the areavalue increased during storage and so the cookies under-went a progressive hardening. Two groups of treatmentappeared statistically significant. The hardest sampleswere the PVC and control; the others were grouped bythe less hard character described by the area of forceversus distance of puncturing. The differences due tothe temperature effect appear in texture figures alreadyafter 10 days of storage; in fact the 20 �C storage seemsto affect the cookies’ hardness greatly with respect to30 �C as shown in texture graphs. Similar behavior wasreported for maximum force values, in which the controland PVC cookies were always the hardest samples andso they reached the end of shelf-life faster than the other

Figure 1. Hardness measurement of cookies samples. Penetration curve plot, force (N) vs distance (mm).




two packages. The control and PVC cookies showed anincreasing trend, particularly at 20 �C (Figure 4), from4.84N in freshly baked cookies to 61.58N in 45-day-oldPVC ones, while the ALL and MAP cookies, on thecontrary, had a decreasing trend from 10 days to theend of storage. Considering firmness lower than 25Nas the acceptability limit (Baiano and Del Nobile,2005), the control and PVC cookies were acceptableonly up to 10 days at 20 �C, while at 30 �C remainedwithin the limit (Figure 5). Analyzing the shape of thecurves obtained by the puncturing test there was aninteresting observation. Starting from the 10th day, thecurve relative to the ALL and MAP cookies showed adouble hump corresponding more or less to the top andbottom part of the cookies. This shape of the curve

became present in all ALL and MAP samples at 20thday. This phenomenon was probably due to watermigration from the outer to the inner part. This kindof curve proves that the storage conditions cause a dif-ferent hardness and moisture between the crust and theinner part of the cookies, so that the applied force isnot the same throughout the whole thickness of thesample. After 30 days of storage the control and PVCsamples showed penetration curves more irregular.These appeared with two main higher peaks andothers between them. According to the texture evolu-tion, the MAP conditions are the best way to storethis kind of cookies; moreover the influence of temper-ature is a determining factor affecting the shelf-life. Infact, all the packages showed results at 30 �C better than

0

100

200

300

400

500

600

0

Time (days)

Are

a (N

x m

m)

Control ALL MAP PVC

5 10 20 30 45

Figure 2. Changes in the area under the penetration curve during the storage of cookies at 20� C.

0

20

40

60

80

100

120

140

Time (days)

Are

a (N

x m

m)

Control ALL MAP PVC

0 5 10 20 30 45

Figure 3. Changes in the area under the penetration curve during the storage of cookies at 30� C.




those obtained at 20 �C of storage, according to the pre-vious aw results. This fact could depend on the fat-phasetransition which could occur at 30 �C storage. Certainly,the physical changes of the cookies components play animportant role in the firmness of the samples studied. Inaddition to lipid state sugars also affect the texture. Thesucrose recrystallyzation is a temperature-dependentprocess; it involves in moisture migration that influencesgreatly the texture characters of the cookies (Belcourtand Labuza, 2007).Hardening is the ‘limiting factor’ for almond paste

shelf-life and results are strictly correlated to the lossof moisture. This problem particularly affects the

cookies production that is strictly related to the hotSouth of Italy, as in this case. The traditional packagingof handicraft almond paste consists of the confectioner’spaper sheets being wrapped around pastries and then thecookies being arranged on a cardboard plate. Our studyshows that of all the ALL and PA/PE barriers, the lastone coupled with MAP conditions can preserve the soft-ness and moisture content of the pastries better than thePVC film. The MAP condition, particularly consistingof removing the air and adding N2 into the bag, showsthe best results. The use of packets hermetically sealedand produced with film having excellent barrier proper-ties against moisture and oxygen could help to protect

0

5

10

15

20

25

Time (days)

For

ce (

N)

Control ALL MAP PVC

0 5 10 20 30 45

Figure 5. Changes in penetration force during the storage of cookies at 30� C.

0

10

20

30

40

50

60

70

80

Time (days)

For

ce (

N)

Control ALL MAP PVC

0 5 10 20 30 45

Figure 4. Changes in penetration force during the storage of cookies at 20� C.




the almond paste from hardness and oxidation. So, fur-ther studies need to be carried out to investigate thephysical characteristics of pastries during experiments,and the effect on texture and water redistribution deriv-ing by changing recipes, for example adding substancesas humectants or emulsifiers.

REFERENCES

Baiano A. and Del Nobile M.A. (2005). Shelf life extension of almondpaste pastries. Journal of Food Engineering 66: 487�495.

Baixauli R., Salvador A. and Fiszman S.M. (2008). Textural andcolour changes during storage and sensory shelf life of muffinscontaining resistant starch. European Food Research andTechnology 226: 523�530.

Belcourt L. and Labuza T. (2007). Effect of raffinose on sucroserecrystallization and textural changes in soft cookies. Journal ofFood Science 72: 65�71.

Farber J.M. (1991). Microbiological aspects of modified atmospherepackaging: a review. Journal of Food Protection 54(1): 58�70.

Farris S., Limbo S. and Piergiovanni L. (2008). Effect of two differenthumectant ingredients on quality of ‘‘Amaretti’’ cookies. ItalianJournal of Food Science 20: 75�90.

Gill C.O. and Tan K.H. (1980). Effect of carbon dioxide on growth ofmeat spoilage bacteria. Applied and Environmental Microbiology39(2): 317�319.

Guinot M.E., Mar0n S., Sanchis V. and Ramos A.J. (2003a). Modifiedatmosphere packaging for prevention of mold spoilage of bakery

products with different pH and water activity levels. Journal ofFood Protection 66(10): 1864�1872.

Guinot M.E., Sanchis V., Ramos A.J. and Mar0n S. (2003b). Mold-free shelf-life extension of bakery products by active packaging.Journal of Food Science 68(8): 2547�2552.

Kotsianis I.S., Giannou V. and Tzia C. (2002). Production and pack-aging of bakery products using MAP technology. Trends in FoodScience and Technology 13: 319�324.

La Rosa R., Russo A., Verdone A. and Aloschi S. (2000).Investigation on hygienic-sanitary quality of almond paste pro-duced in Sicily. Industrie Alimentari 39: 137�141.

Piga A., Catzeddu P., Farris S., Roggio T., Sanguinetti A. and Scano E.(2005). Texture evolution of ‘‘Amaretti’’ cookies during storage.European Food Research and Technology 221(3�4): 387�391.

Seiler D.A.L. (1989). Modified atmosphere packaging of bakery prod-ucts. In: Brody A.L. (ed.), Controlled/Modified Atmosphere/Vacuum Packaging of Foods, Trumbull, CT Inc.: Food andNutrition Press, pp. 119�134.

Severini C., Gomes T., De Pilli T. and Baiano A. (2003). Autoxidationof packed roasted almonds as affected by two different packagingfilms. Journal of Food Processing and Preservation 27(4):321�335.

Suppakul P., Miltz J., Sonneveld K. and Bigger S.W. (2003). Activepackaging technologies with an emphasis on antimicrobial pack-aging and its applications. Journal of Food Science 68(2):408�420.

Sutherland J.P., Patterson J.T., Gibbs P.A. and Murray J.G. (1977).The effect of several gaseous environments on the multiplicationof organisms isolated from vacuum-packaged beef. Journal ofFood Technology 12: 249�255.







DOI: 10.1177/1082013210366750

2010 16: 241 originally published online 12 August 2010Food Science and Technology InternationalI. Flores, V. Cabra, M.C. Quirasco, A. Farres and A. Galvez

Emulsifying Properties of Chemically Deamidated Corn (Zea Mays) Gluten Meal

Published by:


On behalf of:



















Emulsifying Properties of Chemically Deamidated Corn

(Zea Mays) Gluten Meal

I. Flores, V. Cabra, M.C. Quirasco, A. Farres and A. Galvez*

Departamento de Alimentos y Biotecnologıa. Facultad de Quımica, Universidad Nacional Autonomade Mexico. Circuito de la Investigacion Cientıfica s/n. Mexico. D.F. 04510 Mexico

Corn gluten meal is a by-product of starch production that is readily available. Corn protein isolates havelimited applications due to their hydrophobic nature, low solubility and limited functionality as emulsifiers.In this study, a mild acidic treatment of corn gluten meal was performed in order to achieve deamidation ofasparagine and glutamine residues and modify the interfacial behavior of this byproduct. A 0.1N HCl

treatment for 6 h at 70 �C rendered a deamidation degree of 20.4%, which increased the emulsificationactivity index of corn gluten meal from 6.8 to 16.8m2/g protein, with a remarkable increase in emulsionstability from 0 to 90.6% oil retention. Proteins participating in the emulsion were separated by SDS-

PAGE and the main polypeptides were identified as alpha and beta-zeins. After deamidation, proteindissociation and unfolding due to the obtained negative charges resulted in enhanced functionality.

Key Words: corn gluten meal, acidic deamidation, emulsifying properties, Zea mays

INTRODUCTION

In the past 5 years, the demand for food-grade pro-

teins has begun to compete progressively with other

applications, such as the production of biofuels

(von Braun et al., 2008). This situation has increased

the cost of traditional vegetable and animal protein

sources. Consequently, there is a growing need for an

integral use of less expensive protein sources. The main

challenge is still to modify nontraditional proteins to

achieve the required functional properties and render

them appropriately for their use in food formulations.

Corn gluten meal (CGM), a co-product of the corn wet-

milling industry, is an example of a high protein content

source (60% w/w) with potential food uses (May, 1987).

Dry-milling processes have been developed for corn-

starch conversion into bioethanol or high fructose

syrups, which also generate protein co-products at

approximately 28% of the total weight. These bypro-

ducts are typically known as dry distilled grains

(DDGs) with or without solubles as well as corn con-

densed distiller’s solubles that are normally used as

animal feed and are mainly composed of corn gluten

(Xu et al., 2008). The increasing demand for starch in

bio-fuel production will account for a growing availabil-ity of these protein co-products, with limited use due tothe disadvantage of having a poor functionality in thefood area. Considering CGM content in corn as of5�6% (May, 1987), the current availability of corngluten protein from the American corn industry wouldbe approximately of 19 million tons in 2008 (USDA,2009). Corn products have other potential added valueapplications, such as functional ingredients in food, rawmaterial for the production of biodegradable plasticsand films and as a source of the nutraceuticals luteinand zeaxanthin (Lu et al., 2005; Miyoshi et al., 2005).

Among the proteins available in the market, milk andsoy are widely used as food ingredients because of theirfunctionality. As for vegetable proteins, most of themrequire structural modifications in order to expand theiruse in food formulations. Several chemical and enzy-matic modification methods have been described forthe improvement of their solubility, emulsification andfoaming properties (Riha et al., 1996; Yong et al., 2006).CGM has low solubility in aqueous systems at the pHand ionic strength of most food products; consequently,attempts to improve its functional properties haveincluded modifications related to pH adjustment, parti-cle size reduction and freeze-drying (Wu, 2001; Singhet al., 2005) as well as controlled enzymatic hydrolysis(Mannheim and Cheryan, 1992). However, none of thementioned experiments has yielded a potential industrialprocess. Zeins constitute 68% of the total protein con-tent in CGM, while 27% and 1.2% correspond to glu-telins and globulins, respectively (Wu, 2001). Zeins arerich in hydrophobic amino acids, especially aliphaticamino acids, such as leucine, isoleucine and alanine

*To whom correspondence should be sent(e-mail: [email protected]).Received 29 January 2009; revised 22 June 2009.




(Wilson, 1987). This feature is responsible for the highlyaggregated state of CGM that results in poor solubility.Like other cereal proteins, such as wheat and oatproteins, corn proteins have high levels of the amide-containing amino acids, glutamine and asparagine(Riha et al., 1996). The amide bond of these sidechains is susceptible to hydrolysis, which leads to therelease of ammonia and transformation into acidicgroups. The deamidation treatment can be performedby enzymatic or chemical methods. Enzymes such astransglutaminase, protease, peptidoglutaminase andrecently protein-glutaminase (Hamada, 1992; Yonget al., 2006; Cabra et al., 2007), have been reported inthe literature for food protein deamidation. However,nonenzymatic deamidation has captured the interest offood researchers because of the convenience and feasi-bility of such modifications (Matsudomi et al., 1985;Casella and Whitaker, 1990; Cabra et al., 2007), whichtake place by addition of acids or alkalis. Regardingmaize gluten proteins, Cabra et al. (2007) reportedthat for a better deamidation of the gummy and highlyinsoluble Z19 a-zein, an alkaline treatment was required(0.5N NaOH in 70% ethanol at 70 �C during 12 h).Degree of hydrolysis (DH) values as low as 5% wereobtained under these conditions. A number of factorsmay influence deamidation reaction rate and mecha-nism, such as pH, temperature, water activity, aminoacid sequence, available ions and nonionic catalysis(Riha et al., 1996). The present investigation shows anapproach to improve the functional properties of CGMby chemical deamidation. The effect of the deamidationdegree (DD) obtained by a mildly acidic treatment usingHCl on the emulsifying activity and emulsion stability ofCGM was studied.


Materials

Commercial (Zea mays) CGM was kindly donated byArancia Corn Products, S. A. de C. V. (Tlalnepantla,Mexico). It was ground in a disk mill (Weber BROS.and White, Metal Works. Inc., USA) to generate parti-cles smaller than 425 mm. The CGM was then keptat 5 �C in nontranslucent, tightly closed containers inorder to avoid oxidation. The other food grade materialused was corn oil from Productos de Maız S.A. de C.V.,Mexico City, Mexico.

Methods

Unless otherwise stated, reagents were purchasedfrom Sigma Chemical Co. (St. Louis, MO). In order toasses the overall quality of raw materials, proximateanalyses of CGM were performed according to AOAC

methods (AOAC International, 2006) and microbiolog-ical analytical procedures were carried out according tothe Mexican Official Standard (1996). The pH of CGMwas determined according to AOAC method 943.02(2006), in which 10 g of CGM were homogenized with50mL of distilled water at pH 7 for 2min at 8000 rpm inan Ultraturrax T 25 (Janke and Kunkel GmbH and Co.KG � IKA-Labortechnik, Staufen, Germany) and werebrought up to a final volume of 100mL. This suspensionwas magnetically stirred for 30min at room temperatureand the pH value was registered using a pH meter(Beckman Instruments, 34 pHmeter, Irvine, CA).Amino acid content was determined by HPLC after6N HCl hydrolysis (105 �C, 24 h), at Silliker, Inc.Laboratories (TX, US).

Chemical Deamidation of Corn Gluten

Deamidation of CGM was achieved according to pre-vious reports on wheat gluten (Popineau et al., 1988)and corn gluten (Flores, 1997). Treatments with 0.1Nor 0.25N HCl were applied to 5% (w/v) protein suspen-sions. The reaction was performed at 70 �C for 0, 1, 3and 6 h. Protein suspensions were homogenized with90mL of HCl (0.1 or 0.25N), which had been heatedat 60 �C, for 1min at 8000 rpm in an Ultraturrax T 25.The homogenized suspension was brought up to a totalvolume of 100mL with the corresponding concentrationof HCl, preheated at 60 �C. Suspensions were subse-quently stirred at 150 rpm at 70 �C in a shaker (NewBrunswick Scientific Model R76, Edison, NJ) for eachtime period previously stated. The reaction was stoppedby neutralization (pH 7.0) using 0.1N NaOH(Mallinckrodt Baker, Phillipsburg, NJ) on an ice bath.

Degree of Hydrolysis

Five mL of the diluted samples were added to 5mL of5% (w/v) trichloroacetic acid (TCA) (Merck Chemicals,Darmstadt, Germany). The amount of solubilized pro-tein in the supernatant, after high molecular weight pro-tein precipitation with TCA, was quantified with themodified Lowry method (Peterson, 1977). Total hydro-lysis was performed by an exhaustive 6N HCl treatmentat 100 �C (24 h). The DH was calculated as a ratiobetween the protein solubilized under each deamidationcondition and the protein quantified for the 100%hydrolyzed gluten.

Deamidation Degree

The amide groups that remained on the protein afterthe mild acidic treatments were quantified according toArntfield and Murray (1981). The ammonia releasedafter deamidation was quantified with an ammoniumelectrode (Orion Research, model 95-10, Beverly, MA).The maximum deamidation value (100%) was

242 I. FLORES ET AL.



determined by exhaustively treating CGM with 2N HClat 110 �C for 3 h. Deamidated samples were dialyzed inorder to avoid interference with the ammonium elec-trode measurements. The degree of deamidation isexpressed as the ratio of ammonia released from theacid-modified CGM suspensions treated at differenttimes and HCl concentration, to that of the completelydeamidated CGM protein suspension (Matsudomi et al.,1985). Non-treated gluten is here referred to as nativeCGM.

Functional Properties Determination: Solubility

Suspensions of 1 g protein/100mL were prepared byadding treated and native CGM to water adjusted with1N HCl (Merck Chemicals, Darmstadt, Germany) or1N NaOH (Mallinckrodt Baker, Phillipsburg, NJ) tovarious pH values in the range of 2�12, followingPopineau’s method (Popineau et al., 1988). Protein sus-pensions were incubated at 30 �C, 175 rpm for 120min,with pH adjustments to the desired pH value performedevery 30min by potentiometric means. Afterwards, sus-pensions were centrifuged at 4 �C for 30min at10 000� g in a Beckman J2-MC centrifuge (BeckmanCoulter, Inc, Fullerton, CA) and soluble proteins werequantified in the supernatant as described by Peterson(1977), using bovine serum albumin as a standard.

Emulsifying Properties Evaluation

The emulsifying activity index (EAI) was determinedby the method of Pearce and Kinsella (1978). Corn oil(10mL) was dispersed into 30mL protein suspensions(1% (w/v)), pH 7.0, with an Ultraturrax T 25 homoge-nizer at 20 000 rpm for 2min. The emulsions obtainedwere immediately diluted 250-fold with a solution ofsodium dodecyl sulfate (0.1% w/v, pH 7) and 0.1Msodium chloride. The turbidity T (T¼ 2.303A/L, A,absorptivity of the emulsion; l cm, path length of thecuvette) of the dilutions was immediately measured at500 nm in a spectrophotometer (Perkin Elmer Corp.,Norwalk, CT). EAI was defined as:

EAI ¼ 2T=�C ð1Þ

where � was the volume fraction of the oil phase (here�¼ 0.25) and C was the protein concentration in theaqueous phase. The EAI, expressed in m2/g, was relatedto the interfacial area stabilized per unit weight ofprotein.

Stability Towards Coalescence

Freshly prepared emulsions, in 12mL volumes, wereimmediately centrifuged for 10min at 3000� g (CentraCL2, International Equipment Co., Needham Heights,MA). The volume (Vs) of the separated oil phase was

measured. The ratio of 100 Vs/Vi (where Vi¼ 3mL, totaloil volume in the emulsion) was taken as a measure ofcoalescence (Dagorn-Scaviner et al., 1987).

Electrophoretic Pattern of Proteins Participatingin Emulsions

Proteins were extracted from emulsions made withnative as well as with deamidated gluten, using a mod-ification of the method reported by Saito et al. (1993).The cream portion of the emulsion was washed threetimes with 0.1M phosphate buffer, pH 7.0, gentlymixed and recovered by centrifugation at 4 �C for45min at 10 000� g. Separated water and oil were dis-carded. The emulsion was destabilized by the addition ofan equal volume of 6% (w/v) SDS and 10% (v/v) glyc-erol and boiled for 5min. This mixture was then centri-fuged (15 000� g) at 4 �C for 30min. The aqueousportion was oil extracted with an equal volume ofether. After centrifugation, the remaining aqueousphase was lyophilized and further analyzed bySDS-PAGE. The participating proteins in the emulsionwere loaded with 500 mg of protein per lane and run indenaturing 10% acrylamide gels (LaemmLi, 1970) at250V and 60mA for 6 h in a Hoefer SE600 Series SElectrophoresis Unit, with a Bio-Rad molecular weightmarker.

Statistical Data Treatment

The data from all runs represented an average valueof at least three replicates and in most cases, the coeffi-cient of variation (c.v.) was lower than 10%. An analysisof variance was performed (�¼ 0.05) with the StatGraphics Plus 5.1 Program.


Proximate Composition and Microbial Analyses

CGM showed the following composition: protein58.32±0.25, fat 1.80±0.09, ash 1.42±0.01, crude fiber2.23±0.25, moisture 10.44±0.09 and carbohydrates(determined by difference) 25.77 g/100 g of sample.CGM protein content may vary between 60% and73%. In this particular case, lower protein content wasfound; carbohydrate content contributed to one-fourthof the sample weight and all other parameters werewithin the intervals reported by other research groups(May, 1987; Mannheim and Cheryan, 1992; Wu, 2001).The pH of CGM was 3.81±0.02, resulting from wetmilling conditioning, which uses bisulfite in order toenhance the release of starch and insoluble proteinfrom the original granule matrix (Yang et al., 2005).CGM microbial content was within the interval requiredfor cereal products, flours and meals, according to the

Emulsifying Properties of Chemically Deamidated Corn 243



Mexican Official Standard (NOM-147-SSA1-1996), atlevels of less than 100 000, 100 and 1000UFC/g formesophilic bacteria, coliforms and fungi, respectively.These values confirm that the proteinic raw material isappropriate for human consumption, resulting fromgood manufacturing practices.

Acidic Treatment Deamidation and Protein Hydrolysis

In plant storage proteins, as it is expected in CGM,almost all b- and g- carboxyl residues of Asp and Gluacids are amidated (Casella and Whitaker, 1990; Wonget al., 2009). During amino acid analyses, Asn and Glnare converted into Asp and Glu, rendering impossiblethe individual quantification of the original contents ofthe amidated forms. Amino acid contents of 20.82 and5.81 g/100 g protein for Glu and Asp, respectively, werefound (Table 1). Both the high amidated amino acidcontent and low solubility limit the usage of CGM infood formulations.

Considering the previous reports of wheat and corngluten deamidation and the need to reevaluate the usesof the by-products from the new corn industrializationsuch as DDGs, the amidation value determined in this

work reinforces the feasibility of an acidic deamidationtreatment. The expected result is the improvement offunctional properties by increasing the negative chargecontent and hydration (Wong et al. 2009). After anexhaustive deamidation, which implies the total conver-sion of amide groups from Asn and Gln residues intocarboxyl groups, the ammonium released led to the cal-culation of an experimental amidation value of24.98±0.04. Total elimination of amide groups was con-sidered as a 100% of DD.

Based on previously established conditions (Flores,1997), acidic treatments were performed on CGM pro-tein. Statistically significant high experimental DDvalues were obtained with longer incubation times andhigher HCl concentrations. The highest DD valuereached was 53.4% at 6 h, with 0.25N HCl (Table 2).

Compared to other deamidation treatments reportedon different protein sources, DD values were of the sameorder of magnitude: Mimuni et al. (1994) reported asimilar deamidation behavior versus time of acidictreatment for wheat gluten, in a protein suspension of2.5% w/w, using 0.1N HCl at 70 �C. Okara, a soy milkby-product, showed 41% DD after a 48 h treatmentusing 0.1N HCl at 65 �C (Chan and Ma, 1999). Otheracids have also been used on wheat gluten, such as8.75M acetic acid, as reported by Berti et al. (2007),which was utilized to obtain 13.2% ammonia releasefrom 5% (w/v) wheat gluten at 90 �C after a 3 h treat-ment. There are some reports on the chemical deamida-tion of corn proteins (Casella and Whitaker, 1990;Cabra et al., 2007). Casella and Whitaker (1990) treatednative zein with 0.05N HCl at 95 �C for 30min, reach-ing a DD of 3.4%. DD was improved to 9.5% whenSDS was used as a catalyst. In contrast, the resultspresented herein were performed at a lower temperature(70 �C), with HCl concentrations increased to 0.1 and0.25N, in order to obtain higher DD values. Thismethod has the advantages of a less stringent thermaltreatment, as higher temperatures might cause hydroly-sis of peptide bonds as well as the avoidance of alkalinedamage to Trp and the generation of undesirable com-pounds resulting from high pH treatments.

Another approach for deamidation is an enzymatictreatment with peptidoglutaminases and transglutami-nases, which improved ammonia release after a certain

Table 2. Emulsifying properties of deamidated corn gluten meal treated with acid.

Deamination degree (%) Emulsifying actividy indexa (m2/g) Emulsified oila (%)

Deamidation time (h) 0.1 N HCl 0.25 N HCl 0.1 N HCl 0.25 N HCl 0.1 N HCl 0.25 N HCl

0 0 0 6.8±1.1 6.8±1.1 0 01 7.3±0.1 18.0±1.6 6.9±0.9 6.0±0.4 0 9.4±1.23 11.4±0.5 47.6±0.0 9.5±1.0 11.1±2.8 34.8±3.1 58.7±2.86 20.4±0.0 53.4±0.7 16.8±5.0 25.1±4.8 90.6±3.4 82.4±2.8

aThe protein samples were dispersed in 0.01 M sodium phosphate buffer, pH 7 at a concentration of 1% (w/v).

Table 1. Amino acid composition ofcorn gluten meala.

Amino acid g/100 g protein

Asp 5.81Glu 20.82Cys 0.29Ser 4.77Hys 1.79Gly 3.36Thr 3.43Arg 3.00Ala 8.36Tyr 9.44Met 2.06Val 3.29Phe 6.62Ile 2.65Leu 17.68Trp 1.62

aDetermined by HPLC; Lys and Pro: not determined.




extent of hydrolysis (Hamada, 1992; Yong et al., 2006;

Cabra et al., 2007). The chemical deamidation of corn

gluten tested in this work rendered results similar to the

ones obtained with such enzymatic treatments.

However, the enzymatic process has not been scaled-

up (Hamada, 1991), due to the fact that these enzymes

are not commercially available at an industrial scale.A certain DH is expected after the acidic treatment.

Excessive peptide bond cleavage during hydrolysis could

render a more soluble product. However, it could result

in undesirable properties, such as bitter taste and

reduced functionality (Agboola and Dalgleish, 1996;

Riha et al., 1996). DH was determined in deamidated

and native CGM samples, as previously described.

Results are shown in Figure 1. An initial DH value of

2.7% was found in the raw material. An initial hydro-

lysis was expected in native CGM due to the acidic con-

ditions during the wet milling process, which involves

sulfurous and lactic acids. Sulfurous acid is produced

by the reaction of SO2 with water; while lactic acid

can be produced in vivo by fermentation of sugars by

Lactobacillus bacteria present in corn, or it can be added

to the process (Yang et al., 2005). In all samples treated

with mild acidic conditions, resulting DH values were

limited. HCl 0.1N caused a significant change, from

2.7% to 3% DH, after a 1-h treatment; longer reaction

times produced no significant changes. Although hydro-

lysis was proportional to the stringency of the treatment,

a low DH value (4.4%) was also obtained in the more

severe conditions tested (6 h and 0.25N HCl). The

hydrolytic effect was more important during the first

three hours of reaction and no significant changes

were observed after a 6-h treatment.

Mimuni et al. (1994) and Popineau et al. (1988) inves-tigated wheat gluten deamidation, with the conclusionthat protein solubilization was mainly due to deamida-tion and that resulting hydrolysis was limited in acidictreatments at 70 �C and 0.1N HCl. Other reports havedemonstrated that mild acidic treatments, such as thoseemployed, render a negligible degree of hydrolysis fordifferent vegetable proteins, such as soy and oat(Matsudomi et al., 1985; Ma and Khanzada, 1987).Exclusion chromatography has allowed demonstratingthat deamidation conditions can dissociate oligomericproteins, due to the introduction of negative charges inside chains that cause repulsion among peptides. Thedissociation of proteins is more important on solubiliza-tion than the effect caused by hydrolysis (Matsudomiet al., 1985; Ma and Khanzada, 1987). In addition, alimited DH (less than 10%) may even enhance func-tional properties of proteins, mainly emulsifying andfoaming capacities (Foegeding et al., 2002).

Functional Properties

Protein Solubility

According to the results obtained from the solubilitycurves shown in Figure 2, native and deamidated CGMare both in a low solubility condition throughout theinvestigated pH range from 2 to 9, although a significantincreased solubility was found at pH values higher than9. This behavior could be explained by the electrostaticinteractions produced by the side chains with ionizingproperties as well as hydrogen bond formation with thesolvent. Hys is not charged; Arg and Lys are only par-tially protonated, while Asp and Glu are charged at

0.00

2.50

5.00

7.50

10.00

0 1 3 6

Treatment time (h)

Hyd

roly

sis

de

gre

e (

%)

Figure 1. Hydrolysis degree obtained with twodifferent HCl concentrations as a function of deamida-tion time: («) 0.1 N and (#) 0.25 N. Protein concentra-tion used: 5% (w/v). Acidic treatments performedat 70 �C.

0

10

20

30

40

50

60

2 3 4 5 6 7 8 9 10 11 12

pH

Sol

ubili

ty (

%)

Figure 2. Corn gluten meal (CGM) solubility as afunction of pH. CGM was deamidated for 6 h with:(�) 0.1 N HCl, (m) 0.25 N HCl, and (#) native. Proteinconcentration used 1% (w/v).




alkaline pH values. Such combinations of charged andnoncharged side chains provide native maize gluten witha maximum solubility value of 17.30%±0.75 (g of sol-uble protein per 100 g of sample) at pH 12. The curveprofile of native CGM, shown in Figure 2, is similar tothe one reported for zeins, with a low flat portion at pHbetween 4 and 7 and an increased solubility at pH valueshigher than 10 (Casella and Whitaker, 1990). Thisbehavior is expected, given the fact that CGM proteinfraction is composed by a pool of zeins and glutelins,with isoelectric points between 6.5�8.5 and 4.0�8.5,respectively (Wu, 2001). An isoelectric point value ofcommercial zein has been reported as 6.2 (Fu et al.,1999) and specifically for the 19 kD a-zein as 6.8(Cabra et al., 2006).

A marginal increase in solubility from the nativeCGM value of 5.3%±0.3 was obtained at pH 9, afterthe 0.1N and 0.25NHCl treatments, with values of9.9%±0.5 and 12.3%±0.5, respectively. At pHvalues� 10, solubility was 8-fold the value of nativegluten after deamidation with 0.25N HCl, while theless stringent treatment (0.1N HCl) promoted a slightincrease in solubility. This behavior could be explainedby the generation of negatively charged groups to someextent, because the degree of deamidation with thistreatment was only 20% (Table 2).

Emulsifying Properties: Emulsion Activity Index andEmulsion Stability

The role of proteins in emulsions involves molecularinteractions, which are constrained by their aggregationstate and their interface spanning capability. To evaluatetheir emulsifying properties, energy should be provided,through a high speed homogenization of the two immis-cible phases, with a resulting decrease in droplet size aswell as protein unfolding at the interface. In this process,protein chains quickly migrate to the oil droplet surfacewith consequent protection against coalescence. Theproteins that cannot be strongly adsorbed in an oil/water interface, whether because their side chains arestrongly hydrophilic or because they possess rigid struc-tures, will not be good emulsifiers. In order to assess theemulsifying capacity of a protein, EAI and emulsionstability (ES) have to be experimentally evaluated(Dalgleish, 2001).

EAI values, estimated in this study at pH 7 with a 0.25oil volumetric fraction, are shown in Table 2 as a func-tion of deamidation conditions. Native gluten has nota-bly limited emulsifying properties and a 0.1N HCltreatment for 6 h allowed a 2.6-fold increment in theEAI and an emulsion stability improvement from 0 to90.6% of oil retention capacity, as explained further inthe discussion. The EAI represents the interfacial surfaceproduced in an emulsion, which is expressed as theamount of square meters generated per gram of emulsi-fier (m2/g). A protein with good interfacial activity

causes a decrease in the surface tension between the oiland aqueous phase and allows for the production ofsmaller particle size emulsions, which is reflected in ahigher EAI value (Mangino, 1984; Tornberg et al.,1997), as was observed in the acidic treatment on CGMin this work. Although, the EAI exhibited the highestvalue under more stringent conditions (25.1m2/g),it was not in accordance with the highest emulsion sta-bility value, found for the 6-h treatment in the less acidicconditions. A possible explanation for this behavior isthat under stringent conditions, a certain increase inthe degree of protein hydrolysis was produced (from3.19 to 4.43DH, see Figure 1). Limited hydrolysiswould help the emulsifying capacity, because peptidesobtained would cover small interfacial surfaces wherehigh molecular weight proteins would not necessarilydo, resulting in fuller coverage of the oil droplet surface(Table 2). However, a more extensive hydrolysis wouldgenerate small peptides unable to cover large areas(Casella and Whitaker, 1990; Agboola and Dalgleish,1996; Dalgleish, 2001).

An emulsifying protein, adsorbed in the interface,reduces the phase coalescence tendency that is producedto diminish the free interfacial energy caused by the sur-face increment during emulsification. The ES valueassesses the coalescence resistance and it is expressedas the percentage of the oil fraction that remains emul-sified after a centrifugal force is applied to the system.Figure 3 shows the stability of emulsions made withdeamidated gluten under different times of 0.1N HCltreatment, after centrifugation. The greater ES valuewas obtained after a 6-h treatment. A separated waterphase and a cream containing emulsified oil by the pro-tein fraction are shown; low stability emulsions areclearly visible in the cases where oil is separated (tubesA, B and C in Figure 3). In the photograph, the pellet attime zero shows the native CGM insoluble portion thatis not involved in the emulsion. With a progressive dea-midation (1, 3 and 6 h), a smaller pellet is observed and a

Oil phase

Creamedemulsion

Waterphase

Oil phaseCreamedemulsion

Waterphase

Insolublefraction

Figure 3. Emulsion stability after centrifugation.Emulsions were made with native (A) and deamidatedCGM: deamidation treatments were for 1 h (B), 3 h (C)and 6 h (D), at pH 7.0 and 1% (w/v) protein concentra-tions with 0.1 N HCl.




large emulsified oil portion was obtained. The decreasein pellet volume may not only be due to a greater solu-bilization, which in turn was limited (3�9.5%), but alsodue to incorporation of the CGM into the emulsionas a result of a new dissociation and unfolded stateof resulting proteins. These results are in accordancewith other reports demonstrating that emulsifying prop-erties do not correlate directly to solubility (Figure 2)(Aoki et al., 1980). Previous studies (results not shown)demonstrated that the effect on ES of residual starchin CGM is negligible and that the increase on coales-cence resistance is mainly caused by deamidation(Cabra, 2002).The deamidation process of maize gluten produced

surface charges on the original protein through the gen-eration of deprotonated carboxyl groups that show neg-ative charges at neutral pH, given the pKaR values ofAsp and Glu. The repulsion between the new negativelycharged portions of the molecules, could be a factor forCGM proteins unfolding, exposing the hydrophilic aswell as hydrophobic groups, once buried in the aggre-gated native proteins (Bos and van Vliet, 2001). Besides,in an emulsion, the main driving forces are hydrophobicand electrostatic interactions in a process that gainsentropy, due to the conformational changes of the pro-tein during the adsorption (Bergenstahl and Claesson,1997). Therefore, the newly extended conformationallows the covering of a larger particle surface becausedeamidated gluten shows a better capacity for interfacespanning. Dalgleish (2001) reported an additional mech-anism for the stabilization of colloidal systems byproteins: the DLVO (Derjaguin�Landau�Verwey�Overbeek) theory, which encompasses the balance ofVan der Waals attractive forces and the electrostaticrepulsion of charges of identical signs, which are nega-tive in the case of deamidated gluten.In addition, the protein conformation in the interface

and protein surface activity could differ, depending onthe nature of the nonpolar phase that could be highlyheterogeneous in a food system (Rampon et al., 2004).

Electrophoresis Analysis of the Emulsifying Proteins

Considering the fact that CGM is composed of differ-ent protein fractions, mainly prolamines, glutelins andglobulins, these polypeptides might interact in differentways due to their structural characteristics. Figure 4shows SDS-PAGE analysis of proteins from nativeand deamidated CGM (lanes a and c), as well as theproteins extracted from emulsions made with bothCGM samples (lanes b and d), respectively. Given thefact that the amount of protein loaded on each gel lanewas standardized, the proteins that participate in theemulsion were determined, but not their concentration.The band pattern does not greatly vary from one treat-ment to another, indicating that the same protein frac-tions, from deamidated and non-deamidated gluten, do

participate in the emulsion. However, the difference intheir capacity to emulsify and more strikingly, to stabi-lize emulsions, was due to the newly generated chargesand consequently to the new states of association and/orfolding patterns.

The slightly differing apparent molecular masses ofthe constituent proteins of CGM determined bySDS-PAGE, as reported by several authors, could beexplained by the intrinsic variability in protein compo-sition among the corn varieties analyzed, as well asextraction and electrophoretic conditions used by eachresearch group (Wilson, 1991). An important issue toconsider is that the starting raw material for this workis an industrial gluten meal, obtained from an unknownmix of corn varieties, which share a common family ofdivergent genes that are expressed as structurally similarproteins with slightly different molecular masses. Forthis reason, the unique and unequivocal identity ofeach of the bands observed in this study was difficultto ascertain. The SDS-PAGE analysis indicates that themain polypeptides found are b-zein of 18 kD and a-zeinsof 19, 21 and 24 kD. The 21 kD peptide has also beenidentified as g-zein. The bands of 27 and 31 kD, whichdemonstrated lower density, could be identified as g-zeinand glutelin, respectively. A lighter band located at10 kD could be d-zein (Wilson, 1991; Wang et al.,2003). Finally, some very faint bands could be observedaround 45 and 66 kD - these proteins could be oligomersof the 21 kD a-zein, as reported by Cabra et al. (2006).

Structurally, zeins are composed of nine adjacent,topologically anti-parallel helices that are clustered

97.4MWM (kDa) a b c d

66.2 66

50

44

31

28

24

22

1918

16

10

45

31

21.5

14.4

Figure 4. SDS-PAGE analysis of proteins from: (a)native corn gluten meal (CGM), (b) proteins extractedfrom the emulsion made with native CGM, (c) deami-dated CGM and (d) proteins extracted from the emul-sion made with the modified CGM.




within a distorted cylinder. The domains of the amphi-patic helices have been previously described as favoringgood emulsifying properties that contribute to the sur-face activity of the proteins (Krebs and Phillips, 1984).Being zeins the most abundant CGM proteins partici-pating in the emulsions analyzed, their hydropathyindex was calculated in an attempt to explain theiradsorption behavior at the interface. The computer pro-gram ProtScale tool, on the exPASY Server, systemati-cally evaluates the hydrophilic and hydrophobictendencies of a polypeptide chain according to the char-acteristics of the amino acid side chains. The programuses the Kyte and Doolittle hydropathy scale (Kyte andDoolittle, 1982) in which each amino acid has beenassigned a value reflecting its relative hydrophilicity(negative values) and hydrophobicity along the aminoacid sequence, reflected in a score (Y-axis) assigned toeach residue. Accordingly, protein hydropathicity valueswould reflect its capacity to span in the oil/waterinterface.

The analysis of b-zein (panel A) and Z22 a-zein (panelB) are displayed in Figure 5. As for Z19 a-zein (Cabraet al., 2006), three wide hydrophobic regions thatinclude most of the structure were separated by threesmall hydrophilic zones. b-zein showed two very widehydrophilic regions that were interspersed by twosmall hydrophobic ones and Z22 a-zein includes sixhydrophobic regions separated by four hydrophilicones. The alternating hydropathic behavior couldexplain the presence of the zeins at the interface, butthe difference in emulsion stability between native anddeamidated CGM, would be influenced mostly by pro-tein dissociation and unfolding, given that the Kyte andDoolittle scale considers Gln and Asn as hydrophilic asGlu and Asp.

CONCLUSIONS

Chemical treatments have proven to be still an optionfor functionality improvement with industrial potential,considering the increasing amounts of CGM and DDGsthat are obtained in the bio-fuel industry. A low costmild acidic deamidation was able to induce protein dis-sociation as well as unfolding, improving the interfacialfunctionality allowing the production of an added valuefood additive. The striking high emulsion stabilityachieved by the deamidated CGM was mainly basedon the repulsive negative forces generated. Among theprotein fractions involved in emulsification, zeins werethe most abundant at the oil-water interface. Theyencompass a good theoretical alternative pattern ofhydropathy, which may explain their presence in theemulsion, though surface hydrophobicity in native anddeamidated CGM should be evaluated experimentally.

ACKNOWLEDGMENT

Authors would like to acknowledge Pamela SuarezBrito for her technical support.

REFERENCES

Agboola S.O. and Dalgleish D.G. (1996). Enzymatic hydrolysis ofmilk proteins used for emulsion formation. 1. Kinetics of proteinbreak down and storage stability of the emulsions. Journal ofAgricultural and Food Chemistry 44: 3631�3636.

Aoki E., Taneyama O. and Inami M.J. (1980). Emulsifying propertiesof soy protein: characteristics of 7S and 11S proteins. Journal ofFood Science 45: 534�538.

AOAC International (2006). Methods: 920.39, 920.86, 923.03, 943.02,945.15 and 979.09. In: Horwitz W. (ed.), Official Methods ofAnalysis of AOAC International, 18th edn. Gaithersburg, MD:AOAC International.

Arntfield A. and Murray A. (1981). Determination of amide nitrogenin plant proteins using an ammonia electrode. Canadian Instituteof Food Science and Technology Journal 14: 227�229.

1.5(a)

(b)

1

0.5

0

–0.5Sco

re

–1

–1.5

–220 40 60 80

Hphob. / kyte & Doolittle

Hphob. / kyte & Doolittle

100 120 140 160

1.2

1

0.6

0.8

0.2

0.4

0Sco

re

–0.4

–0.2

–0.6

–0.8

–150

Position100 150 200

Figure 5. Hydropathy index according to the Kyteand Doolittle scale. (a) b-zein (P08031, Swiss Protdatabase). (b) PZ22.3 a-zein (P04698, Swiss Protdatabase). Window size used: 19.




Bergenstahl B.A. and Claesson P.M. (1997). Surface forces in emul-sions. In: Friberg S.E. and Larsson K. (eds), Food Emulsions, 3rdedn. New York: Marcel Dekker, Inc, pp. 58, 67.

Berti C., Roncoroni L., Falini M.L., Caramanico R., Dolfini E.,Bardella M.T., Elli L., Terrani C. and Forlani F. (2007).Celiac-related properties of chemically and enzymatically modi-fied gluten proteins. Journal of Agricultural and Food Chemistry55: 2482�2488.

Bos M.A. and van Vliet T. (2001). Interfacial rheological properties ofadsorbed protein layers and surfactants: a review. Advances inColloid and Interface Science 91: 437�471.

Cabra V. (2002). Identificacion y estudio de las proteınas del gluten demaız involucradas en la estabilizacion de emulsiones de aceite enagua. MSc Thesis, Faculty of Chemistry, UNAM. Mexico City.

Cabra V., Arreguin R., Vazquez-Duhalt R. and Farres A. (2006).Effect of temperature and pH on the secondary structure andprocesses of oligomerization of 19 kDa alpha-zein. Biochimicaet Biophysica Acta 1764: 1110�1118.

Cabra V., Arreguin R., Vazquez-Duhalt R. and Farres A. (2007).Effect of alkaline deamidation on the structure, surface hydro-phobicity and emulsifying properties of the Z19 a-zein. Journal ofAgricultural and Food Chemistry 55: 439�445.

Casella M.L.A. and Whitaker J.R. (1990). Enzymatically and chemi-cally modified zein for improvement of functional properties.Journal of Food Biochemistry 14: 453�475.

Chan W.M. and Ma C.Y. (1999). Acid modification of proteins fromsoymilk residue (okara). Food Research International 32:119�127.

Dagorn-Scaviner C., Gueguen J. and Lefebvre J. (1987). Emulsifyingproteins of pea globulins as related to their adsorption behaviors.Journal of Food Science 52: 335�341.

Dalgleish D.G. (2001). Food Emulsions. In: Johan Sjoblom. (ed.),Encyclopedic Handbook of Emulsion Technology. New York:Marcel Dekker, Inc, pp. 209�212.

Flores I. (1997). Modificacion enzimatica y quımica del gluten de maız:posibilidades de utilizacion en la industria alimentaria. MSc Thesis,Faculty of Chemistry, UNAM. Mexico City.

Foegeding E.A., Davis J.P., Doucet D. and McGuffey M.K. (2002).Advances in modifying and understanding whey protein func-tionality. Trends in Food Science and Technology 13: 151�159.

Fu D., Weller C.L. and Wehling L.R. (1999). Zein: properties, prep-arations and applications. Food Science and Biotechnology 8:1�10.

Hamada J.S. (1991). Peptidoglutaminase deamidation of proteins andprotein hydrolysates for improved food use. Journal of the AOCS68: 459�462.

Hamada J.S. (1992). Effects of heat and proteolysis on deamidation offood proteins using peptidoglutaminase. Journal of Agriculturaland Food Chemistry 40: 719�723.

Krebs K.E. and Phillips M.C. (1984). The contribution of a-helices tothe surface activities of proteins. FEBS Letters 175: 263�266.

Kyte J. and Doolittle R.F. (1982). A simple method for displaying thehydropathic character of a protein. Journal of Molecular Biology157: 105�132.

LaemmLi U.K. (1970). Cleavage of structural proteins during theassembly of the head of bacteriophage T4. Nature 227: 680�685.

Lu Y., Yao H., Gong Y. and Sheng Q. (2005). Effects of enzymatictreatment of corn gluten meal on lutein and zeaxanthin extrac-tion. International Journal of Food Properties 8: 161�169.

Ma C.Y. and Khanzada G. (1987). Functional properties of deami-dated oat protein isolates. Journal of Food Science 52: 1583�1587.

Mangino M.E. (1984). Physico-chemical basis of whey protein func-tionality. Journal of Dairy Science 67: 2711�2722.

Mannheim A. and Cheryan M. (1992). Enzyme-modified proteinsfrom corn gluten meal. Preparation and functional properties.Journal of the AOCS 69: 1163�1169.

Matsudomi N., Sasaki T., Kato A. and Kobayashi K. (1985).Conformational changes and functional properties of acid-mod-ified soy protein. Agricultural and Biological Chemistry 49:1251�1256.

May J.B. (1987). Wet milling: process and products. In: Watson S.A.and Ramstad P.E. (eds), Corn: Chemistry and Technology, St.Paul: American Association of Cereal Chemists Inc.,pp. 377�397.

Mexican Official Standard (1996). Bienes y servicios. Cereales y susproductos. Harinas de cereales, semolas o semolinas. Alimentos abase de cereales, de semillas comestibles, harinas, semolas o sem-olinas o sus mezclas. Productos de panificacion. Disposiciones yespecificaciones sanitarias y nutrimentales. NOM-147-SSA1-1996.

Mimuni B., Raymond J., Merle-Desnoyers A.M., Azanza J.L. andDucastaing A. (1994). Combined acid deamidation and enzymichydrolysis for improvement of the functional properties of wheatgluten. Journal of Cereal Science 21: 153�165.

Miyoshi T., Toyohara K. and Minematsu H. (2005). Preparation ofultrafine fibrous zein membranes via electrospinning. PolymerInternational 54: 1187�1190.

Pearce K.W. and Kinsella J.E. (1978). Emulsifying properties of pro-teins: evaluation of a turbidimetric technique. Journal ofAgricultural and Food Chemistry 26: 716�723.

Peterson G.L. (1977). A simplification of the protein assay method ofLowry et al. which is more generally applicable. AnalyticalBiochemistry 83: 346�350.

Popineau Y., Bollecker S. and Thebaudin J.Y. (1988). Caracterisationbiochimique et fonctionnelle des proteines de gluten desamideespartiellement en conditions menagees. Sciences des Aliments 8:411�430.

Rampon V., Brossard Ch., Mouhous-Riou N., Bousseau B., LlamasG. and Genot C. (2004). The nature of the apolar phase influ-ences the structure of the protein emulsifier in oil-in-water emul-sions stabilized by bovine serum albumin. A front-surfacefluorescence study. Advances in Colloid and Interface Science108�109: 87�94.

Riha W.E., Izzo H.V., Zhang J. and Ho C.T. (1996). Nonenzymaticdeamidation of food proteins. Critical Reviews in Food Scienceand Nutrition 36: 225�255.

Saito M., Chikuni K., Momna M. and Shimizu M. (1993). Emulsifyingand oil binding properties of bovine serum albumin and its enzy-matic hydrolyzate. Biosciences, Biotechnology and Biochemistry57: 952�956.

Singh N., Kaur M. and Singh K. (2005). Physicochemical and func-tional properties of freeze-dried and oven dried corn glutenmeals. Drying Technology 2: 975�988.

Tornberg E., Olsson A. and Person K. (1997). The structural andinterfacial properties of food proteins in relation to their functionin emulsions. In: Friberg S.E. and Larsson K. (eds), FoodEmulsions, 3rd edn. New York: Marcel Dekker, Inc.,pp. 323�325.

United States Department of Agriculture. Economic Research Service(USDA) (2009). Feed Grains Database: Yearbook Tables.Available at: http://www.ers.usda.gov/Data/FeedGrains/FeedYearbook.aspx (accessed May 20, 2010).

von Braun J., Ahmed A., Asenso-Okyere K., Fan S., Gulati A.,Hoddinott J., Pandya-Lorch R., Rosegrant M.W., Ruel M.,Torero M., van Rheenen T. and von Grebmer K. (2008). Highfood prices: the what, who and how of proposed policy actions.International Food Policy Research Institute. Available at:http://www.ifpri.org/PUBS/ib/FoodPricesPolicyAction.pdf(accessed July 1, 2009).

Wang J.-F., Geil P.H., Kolling D.R.J. and Padua G.W. (2003).Analysis of zein by matrix-assisted laser desorption/ionizationmass spectrometry. Journal of Agricultural and Food Chemistry51: 5849�5854.

Wilson C.M. (1987). Proteins of the kernel. In: Watson S.A. andRamstad P.E. (eds), Corn: Chemistry and Technology. St. Paul:American Association of Cereal Chemists Inc., pp. 273�310.

Wilson C.M. (1991). Multiple zeins from maize endosperms charac-terized by reversed-phase high performance liquid chromatogra-phy. Plant Physiology 95: 777�786.

Wong H.-W., Choi S.-M., Phillips D.L. and Ma Ch.-Y. (2009). Ramanspectroscopic study of deamidated food proteins. Food Chemistry113: 363�370.




Wu V.Y. (2001). Emulsifying activity and emulsion stability of corngluten meal. Journal of the Science of Food Agriculture 81:1223�1227.

Xu Y., Hanna M.A. and Weber R. (2008). Compaction of corn distil-lers dried grains. Cereal Chemistry 85: 158�164.

Yang P., Haken A.E., Niu Y., Chaney S.R., Hicks K.B., Eckhoff S.R.,Tumbleson M.E. and Singh V. (2005). Effect of steeping with

sulfite salts and adjunct acids on corn wet-milling yields andstarch properties. Cereal Chemistry 82: 420�424.

Yong Y.H., Yamaguchi S. and Matsumura Y. (2006). Effects of enzy-matic deamidation by protein-glutaminase on structure and func-tional properties of wheat gluten. Journal of Agricultural andFood Chemistry 54: 6034�6040.







DOI: 10.1177/1082013210366770

2010 16: 251 originally published online 12 August 2010Food Science and Technology InternationalRegy Johnson, S.N. Moorthy and G. Padmaja

FloursProduction of High Fructose Syrup from Cassava and Sweet Potato Flours and their Blends with Cereal

Published by:


On behalf of:



















Production of High Fructose Syrup from Cassava and Sweet

Potato Flours and their Blends with Cereal Flours

Regy Johnson, S.N. Moorthy and G. Padmaja*

Division of Crop Utilization, Central Tuber Crops Research Institute, Sreekariyam,Thiruvananthapuram - 695 017, Kerala, India

Despite being a rich source of starch, root crops such as cassava and sweet potato have not been widelyexploited for the production of high fructose syrup (HFS), which is a highly valued sweetener for the foodand beverage industries. The major factors contributing to the cost of production of HFS are the cost andlabor-intensive steps in the production of starch, different processing temperatures and pH for the enzyme

reactions, poor extractability of starch, etc. With the objective of overcoming the cost associated with thepreparation of starch, the feasibility of using native cassava/sweet potato flours and their blends with riceflour and wheat flour, as the raw material for HFS production was investigated. The saccharified slurry

from cassava�rice flour blends contained 70�72 g reducing sugars/100 g, which was higher than thatreleased from native cassava flour (�69%). Blends of sweet potato with rice or wheat yielded saccharifiedmash with lower content of reducing sugars (60�66%). Although the percentage conversion to fructose

after isomerization was similar for cassava/sweet potato or their blends with cereal flours (42�43%),fructose yield was higher in native cassava flour and cassava�rice blends (28�29 g/100 g) than the otherflour blends.

Key Words: high fructose syrup, cassava, sweet potato, root crop�cereal blends

INTRODUCTION

Cassava (Manihot esculenta Crantz) and sweet potato(Ipomoea batatas Lam.) are tropical root crops rich instarch and serve primarily to meet the energy require-ment of around 500 million people of the tropical coun-tries (FAO, 2006). Starch is also the raw material for anumber of industrial products like bioethanol, glucosesyrup, high fructose syrup (HFS), maltodextrins, modi-fied starches, gums, adhesives, etc. (Cock, 1985;Piyachomkwan et al., 2005; Shetty et al., 2007). Withthe increasing trends in global sugar consumption cou-pled with the fluctuating sugar prices (Khalid, 2005);there is a need to look out for alternative sugar substi-tutes like glucose and high fructose syrup (HFS) as wellas raw materials like root crops. HFS is a highly valuedsweetener for the food and beverage industries, owingto its positive attributes like high sweetness, noncrystal-line nature, ability to enhance the fruit and spice flavorand keeping products fresh by maintaining consis-tent moisture (White, 1992; Hanover and White, 1993).

The reactions involved in the conversion of starch to

glucose or HFS mediated by enzymes such as a-amylase,

glucoamylase and glucoisomerase were studied by sev-

eral workers (Aschengreen et al., 1979; Dziedzic, 1981;

Barker and Petch, 1985; Labout, 1985; Antrim et al.,

1991). The potential of cassava starch for glucose or

HFS production and the process conditions associated

with the enzyme reactions were reported (Lages and

Tannebaum, 1978; Franco and Ciacco, 1987; Ghildyal

et al., 1989; Gorinstein and Cheng-yi Lii, 1992;

Waliszweski et al., 2002; Morales et al., 2008). The

major factors contributing to cost of production of

HFS were the widely different processing temperatures

and pHs for the three enzyme reactions like liquefaction

by a-amylase (pH 6.0 and 100 �C), saccharification by

glucoamylase (pH 5.0 and 60 �C) and isomerization (pH

8.0 and 60 �C) (Taniguchi, 2004). Production of starch

from corn, cassava or sweet potato is an energy-inten-

sive process and hence this also contributes towards the

high cost of production of HFS.In comparison to cassava, extractability of starch

from sweet potato is very poor (�70% of the totalstarch) due to the latex in the roots and the pectin-hemi-cellulose complex preventing the free release of starch.Direct use of the fresh roots or dry flour could elimi-nate this problem, leading to a better utilization of thetotal starch for HFS production. Studies were made byseveral workers to economize the production of HFSthrough simultaneous liquefaction�saccharification,

*To whom correspondence should be sent(e-mail: [email protected]).Received 23 February 2009; revised 21 June 2009.




use of corn flour or grits (Arasaratnam andBalasubramaniam, 1993), using improved enzymes likeStargenTM 001, which can hydrolyze granular flour orwet root slurry, etc. (Johnson et al., 2009). The objectiveof the present study was to explore the possibility ofusing the flour from cassava or sweet potato rootsdirectly for HFS production, which could eliminate theneed for converting the roots to starch. Further, thepossibility of blending cereal flours like rice or wheatto raise the starch content of the root crop flours, result-ing in a higher fructose yield was also investigated.


Cassava (variety :M4) and sweet potato(variety :Kanhangad) grown under the Institute farm,following the recommended package of practices wereused for the study. Roots were collected at harveststage (10 months for cassava and 105 days for sweetpotato), washed free of dirt, peeled and sliced to roundchips (1 cm thickness). The chips were sun-dried for 36 h,powdered in a blender, sieved (mesh size : 32) and used.Cereal flours such as rice flour (raw white rice) andwheat flour were purchased from the local market.These were mixed with cassava or sweet potato flour atlevels of 10%, 20%, 30% and 40% and sieved throughthe same mesh size sieve, to ensure uniform blending.

Methods

The starch content in the individual flours such ascassava, sweet potato, rice and wheat as well as theirblends in various proportions was estimated by themethod of Moorthy and Padmaja (2002). The sugarsin the flours or their blends were extracted using 80%ethanol and the filtrate after hydrolyzing with concen-trated hydrochloric acid (1 : 20 w/v) for 30 min was usedto estimate the total sugars by titration against potas-sium ferricyanide, with methylene blue as indicator. Thestarch in the residue was hydrolyzed using 2.0N HCl at100 �C for 30 min and the released sugars were estimatedtitrimetrically. Starch content (% dry basis; d.b) wascomputed from the sugar values by multiplying withthe Morris factor, 0.9 (Moorthy and Padmaja, 2002).

Liquefaction of Cassava/Sweet Potato

Liquezyme X (thermostable a- amylase; EC 3.2.1.1)with an activity of 200 kilo novo units (KNU) andDextrozyme GA (glucoamylase; EC 3.2.1.3) with anactivity of 270 amyloglucosidase units per gram werepurchased from M/s Novozymes A/s, Denmark.Sweetzyme T (immobilized glucose isomerase; EC5.3.1.5) with an activity of 508 glucose isomerase novou-nits (GINU) per gram was purchased from M/s NovoNordisk Biochem, USA.

A 25% (w/v) suspension of the native cassava or sweetpotato as well as their blends was prepared in distilledwater and the pH was adjusted to 6.5. After equilibrat-ing in a thermostatic water bath (JulaboSW21) at90 �C for 10 min, Liquezyme X (30.0mg) was addedand incubated at 90 �C for 1 h. The reducing groupsformed were estimated in aliquots (duplicate) by theNelson�Somogyi method (Nelson, 1944). An aliquotof the liquefied slurry was treated with alkaline copperreagent and heated in a boiling water bath for 10 min.The cuprous oxide formed was treated with arsenomo-lybdate reagent to obtain the blue colored complex,whose absorbance was measured at 620 nm and thereducing sugars were calculated using a D-glucosestandard.

Saccharification of Liquefied Slurry

The liquefied slurry was cooled to 60 �C and the pHwas adjusted to 4.0. Dextrozyme GA (0.05mL with anactivity of 270 amyloglucosidase units/g) was added andincubation continued for 48 h at 60 �C. The reducinggroups formed were estimated by the Nelson�Somogyimethod as described earlier. The glucose content of thealiquots was determined by the glucose oxidase(GOD)�peroxidase (POD) method using the glucoseoxidase (GO) assay kit (Sigma, Missouri, USA). Thekit contained glucose oxidase�peroxidase reagent (prod-uct code G 3660), o-dianisidine reagent (product codeD 2679) and glucose standard (product code G 3285)(Bergmeyer and Bernt, 1974). The glucose oxidase/per-oxidase reagent was dissolved in 39.2mL deionizedwater and stored in an amber bottle at 4 �C till use.The o-dianisidine reagent was reconstituted with1.0 mL deionized water and stored under dark at 4 �C.The assay reagent was constituted by mixing 0.8mLo-dianisidine reagent with 39.2mL of GOD/PODreagent. Two milliliters of assay reagent was added tothe test (aliquot of the saccharified slurry; 1.0mL) andstandard samples (0.05mL) and after incubating for30 min at 37 �C, the reaction was stopped by adding2.0mL sulfuric acid (12.0 N) to each tube. The absor-bance was read against a reagent blank at 540 nm.

The percentage conversion to reducing sugars orglucose was computed using the formula:

% conversion (on 100 g starch basis)

¼[Reducing sugar or glucose yieldð% d:bÞ� � 100

Starchð% d:bÞ:

ð1Þ

Isomerization of Glucose Syrup

After the saccharification, the glucose syrup was fil-tered and concentrated to 40% solids in an oven at60 �C. The pH of the concentrated syrup was adjusted

252 R. JOHNSON ET AL.



to 7.5 and kept in a thermostatic water bath at 60 �C.After equilibration, Sweetzyme T (250mg with an activ-ity of 508 glucose isomerase novo units/g) was added.The incubation was continued at 60 �C for 48 h. Thefructose formed was estimated by the cysteine carbazolemethod (Chen and Anderson, 1980). An aliquot(1.0mL) of the isomerized syrup (diluted 400 times)was treated with 0.2mL cysteine hydrochloride (1.5%),6.0mL sulfuric acid (70%) and 0.2mL absolute alcohol(0.12% (w/v)). After mixing for 10 s, carbazole reagentwas added and incubated for 1.0 h at 37 �C. The absor-bance of the colored product was measured against areagent blank at 560 nm and the fructose yield was cal-culated using D-fructose standard.


Initial starch and total sugar content in the nativeflours and their blends with cereals were determined inthree replicates, with duplicate analysis per replicate.Four replicates were maintained for the liquefaction,saccharification and isomerization studies. Statisticalanalysis was done for all parameters using one way anal-ysis of variance (ANOVA) for comparison of meanvalues among different treatments. The analyses werecarried out using the statistical package SAS Version8.1 (SAS Institute Inc., Cary, NC, USA).


Liquefaction

The native cassava (C) flour used in the study had astarch and total sugar content of 74.59% and 5.77%,respectively, of which 1.67% was constituted by reduc-ing sugars. Blending with rice (R) flour at levels of 10%,20%, 30% and 40% slightly elevated the starch content(75.21�76.71%; Table 1). Nevertheless, the totalsugar content decreased significantly from 5.77% inthe native flour to 4.88 in the 60 : 40 blend. This decreasewas due to the much lower total sugar content in thenative rice flour (�2.60%) as compared to the cassavaflour (5.77%). The reducing sugars also significantlydecreased from 1.67% in the native flour to 1.16% inthe 60 : 40 blend. On the contrary, the cassava-wheat flour blends had only lower starch content(71.25�73.98%) than cassava�rice flour blends(Table 1). The total sugar content in the native wheatflour was similar to that of cassava flour (5.10% vs5.77%), which could explain the almost identical sugarvalues in the cassava�wheat flour blends (5.54�5.57).The native sweet potato flour had a starch and total

sugar content of 70.31% and 11.43%, respectively, andthe lower starch content by around 5.0 units in comparisonto cassava, was compensated by the elevated sugar content

in sweet potato flour. Cassava and sweet potato roots havebeen reported to contain starch in the range of 25�38%and 13�30%, respectively, and total sugars ranging from1�3% in cassava and 5�15% in sweet potato (Ceredaet al., 1982; Bradbury and Holloway, 1988; Abrahamet al., 2006). Deobald et al. (1969) reported that the pre-dominant sugars in sweet potato were sucrose, maltoseand glucose. Blending with rice flour, which has a muchhigher starch content (�76%) was found to be the bestway to raise the starch content of the initial raw material.Earlier workers have also reported that the final yield ofglucose and HFS depends on the initial starch content ofthe source material (Lages and Tannebaum, 1978;Ghildyal et al., 1989; Baskar et al., 2008). Increasing thelevel of incorporation of rice flour from 20% to 30% or40% significantly (p< 0.001) elevated the starch content insweet potato blends (70.49�72.39%; Table 2). Increasedlevels of rice flour in the sweet potato� rice flour blends ledto slight decrease in the total sugar content (10.00% vs.11.43% in native sweet potato flour) while the reducingsugars decreased significantly (p< 0.001) from 3.23% inthe native flour to 2.10% in the 60 : 40 blend. Wheat flourhad almost similar starch content (70%) as sweet potatoand hence blending at the various levels did not remark-ably alter the starch content of the blends. However, thetotal sugar content decreased from 11.13 at 10% blendingto 10.22 at 40% blending (Table 2). The reducing sugarsdecreased from 3.04% to 2.46% in the sweet pota-to�wheat flour blends.

Blending with rice or wheat flour did not signifi-cantly influence the formation of reducing sugarsduring liquefaction by Liquezyme X. The liquefiedslurry from native cassava flour contained 12.46%reducing sugars, of which 10.79% was formed as aresult of liquefaction, the rest being accounted by thereducing sugar present in the native flour. It wasfound from the study that irrespective of the level of

Table 1. Starch and sugar content of cassava flourand its blends.

Treatments

Starch (% d.b.)Total sugars

(% d.b.)Reducing

sugars (% d.b.)

Mean±SD, n¼3

Native cassava flour 74.59±0.71 5.77±0.16 1.67±0.08Cassava�rice flour

blend90 : 10 (T1) 75.21±0.36 5.60±0.07 1.54±0.0780 : 20 (T2) 75.42±0.36 5.65±0.06 1.42±0.0470 : 30 (T3) 76.71±0.76 5.10±0.11 1.29±0.0560 : 40 (T4) 76.27±0.65 4.88±0.03 1.16±0.04

Cassava�wheat flourblend90 : 10 (T5) 73.98±0.70 5.57±0.02 1.63±0.0880 : 20 (T6) 72.97±0.34 5.54±0.01 1.60±0.0570 : 30 (T7) 72.98±0.49 5.56±0.02 1.56±0.0460 : 40 (T8) 71.25±0.86 5.57±0.02 1.52±0.05LSD (5%) 1.081 0.1220 0.0869

High Fructose Syrup from Cassava and Sweet Potato Flour 253



incorporation of rice flour or wheat flour, the liquefiedslurry from cassava�cereal blends contained around10.12�11.23% reducing sugars, formed as a result ofthe Liquezyme X action (Figures 1 and 2). The liquefiedslurry from native sweet potato flour had a total reduc-ing sugar content of 17.83% of which 14.60% wasformed during liquefaction using Liquezyme X. Theblends of sweet potato flour with rice or wheat flouryielded liquefied slurry with similar levels of total reduc-ing sugars (16.49�18.05%) and reducing sugars formedas a result of Liquezyme X action (14.39�15.10%;Figures 3 and 4). Increasing the levels of rice or wheatin sweet potato flour was found to decrease the reducingsugar yield in the liquefied slurry.

Saccharification

The saccharified slurry from native cassava flour con-tained 68.76% reducing sugars (Table 3) of which

67.09% was contributed by the liquefaction and sacchar-ification reactions. The total reducing sugar content inthe saccharified slurry from cassava�rice flour blendswas significantly (p< 0.001) higher (70.56�72.00%)than that from the native flour. It was found that

Table 2. Starch and sugar content of sweet potatoflour and its blends.

Treatments

Starch (% d.b.)Total sugars

(% d.b.)Reducing

sugars (% d.b.)

Mean±SD, n¼ 3

Native sweetpotato flour

70.31±0.55 11.43±1.04 3.23±0.09

Sweet potato�riceflour blend90 : 10 (T1) 70.49±0.32 11.16±0.10 2.95±0.0380 : 20 (T2) 70.86±0.56 10.15±0.46 2.66±0.0570 : 30 (T3) 72.00±0.58 10.27±0.10 2.38±0.0460 : 40 (T4) 72.39±0.89 10.11±0.09 2.10±0.05

Sweet potato�wheat flour blend90 : 10 (T5) 70.50±0.32 11.13±0.12 3.04±0.0980 : 20 (T6) 70.31±0.00 10.96±0.10 2.84±0.0470 : 30 (T7) 70.50±0.64 10.47±0.87 2.65±0.0460 : 40 (T8) 69.58±0.31 10.22±0.10 2.46±0.05LSD (5%) 0.8950 0.8320 0.0911

0

2

4

6

8

10

12

14

16

18

20

T5 T6 T7 T8

Treatments

Red

ucin

g su

gar

cont

ent

(g/1

00g)

Figure 4. Reducing sugar content in the liquefiedslurry (sweet potato�wheat flour blend): (#) Totalreducing sugar content. («) Reducing sugarsformed by the action of Liquezyme X.

0

2

4

6

8

10

12

14

T1 T2 T3 T4

Treatments

Red

ucin

g su

gar

cont

ent

(g/1

00g)

Figure 1. Reducing sugar content in the liquefiedslurry (cassava�rice flour blend): (#) Total reducingsugar content. («) Reducing sugars formed by theaction of Liquezyme X.

0

2

4

6

8

10

12

14

T5 T6 T7 T8

Treatments

Red

ucin

g su

gar

cont

ent

(g/1

00g)

Figure 2. Reducing sugar content in the liquefiedslurry (cassava�wheat flour blend): (#) Total reducingsugar content. («) Reducing sugars formed by theaction of Liquezyme X.

0

5

10

15

20

T1 T2 T3 T4

Treatments

Red

ucin

g su

gar

cont

ent

(g/1

00g)

Figure 3. Reducing sugar content in the liquefiedslurry (sweet potato�rice flour blend): (#) Total reduc-ing sugar content. («) Reducing sugars formed by theaction of Liquezyme X.




reducing sugars were formed to the extent of

69.02�70.84% during the liquefaction�saccharificationsteps. The percentage conversion to reducing sugars

(after nullifying the initial sugars) was computed as per

the formula given in Equation (1) and it ranged from

90.77% to 92.88% for the various cassava�rice flour

blends. This was significantly higher than the percentage

conversion obtained with native cassava flour (89.94%;

Table 3). The reducing sugar content in the saccharified

slurry from cassava�wheat flour blends was significantlylower than that formed during the saccharification of

cassava�rice flour blends. The initial starch content of

the cassava�wheat blends was also less than the cassava-

rice blends, accounting for the lower content of reducing

sugars in the saccharifiedmash (Tables 1 and 3). The total

reducing sugars and that formed by the action of

Liquezyme X and Dextrozyme GA in the saccharified

slurry from cassava�wheat flour blends (90 : 10 and

80 : 20) were similar to that from the native cassava

flour (�69.00% and 67.00%, respectively), while higher

levels of addition of wheat flour such as 30% and 40%

significantly decreased the formation of reducing sugars.

The percentage conversion to reducing sugars was also

less in the cassava�wheat blends (90 : 10 and 80 : 20),

while it was less compared to the cassava�rice blends athigher levels of incorporation of wheat flour (Table 3).

Earlier studies in our laboratory showed that the sacchar-

ified mash from native cassava starch had 72.96% reduc-

ing sugars, with a percentage conversion of 96.25%

(Johnson et al., 2004). Although the use of cassava

flour decreased the percentage conversion to 90% (6%

decrease as compared to starch), the difficulties and cost

involved in the preparation of starch from cassava tubers

suggest that cassava flour in the native form or its blends

with rice flour will be a better option.Sweet potato�rice/wheat flour blends yielded sacchari-

fied mash with significant less content of total reducing

sugars (65.58�68.58% in the sweet potato�rice blends

and 62.70�65.22% in the sweet potato�wheat blends)

than cassava cereal blends. The reducing sugars formed

after saccharification was 62.63�66.48% in sweet pota-

to�rice blends and 60.24�62.18% in sweet potato�wheatblends (Table 4). Accordingly, the percentage conversion

was also significantly lower in the case of native sweet

potato (87.83%) as well as sweet potato� rice

(88.85�91.84%) and sweet potato�wheat (86.58�88.20%)

blends, when compared with cassava (Table 3). The low

starch content in the initial raw material from sweet

potato led to the low conversion to reducing sugars,

during liquefaction and saccharification as compared to

cassava. Although the sweet potato�rice blends had a

higher percentage conversion to reducing sugars than

native, sweet potato flour, indicating its scope for glucose

and HFS production, the values were significantly less for

the sweet potato�wheat blends. Native sweet potato starch

was reported to yield saccharifiedmashwith 72.24% reduc-

ing sugars, with a percentage conversion of 95.30%

(Johnson et al., 2005). Although, the percentage conver-

sion, when computed on starch basis was less for sweet

potato flour or its blends with cereal flours, the reducingsugars present in the initial raw materials before enzyme

action, could also account towards the final reducing

sugars/glucose. Further, the level of glucose in the sacchar-

ified mash is critical for high fructose yields.The true glucose content in the saccharified mash,

assayed by the GOD�POD method (including that

Table 3. Reducing sugar content in the saccharifiedslurry from cassava flour and its blends.

Treatments

Reducing sugars (g/100)

Percentageconversion to

reducing sugarsaTotal

Formed byliquefaction andsaccharification

Nativecassava flour

68.76±0.63 67.09±0.63 89.94±0.85

Cassava�riceflour blendsT1 70.56±0.52 69.02±0.52 91.77±0.69T2 71.34±0.23 69.92±0.23 92.71±0.31T3 70.92±0.31 69.63±0.31 90.77±0.41T4 72.00±0.44 70.84±0.44 92.88±0.58

Cassava�wheatflour blendsT5 69.00±0.31 67.37±0.31 91.07±0.42T6 68.94±0.30 67.34±0.30 92.28±0.42T7 67.26±0.49 65.70±0.49 90.02±0.68T8 65.88±0.46 64.36±0.46 90.33±0.65LSD (5 %) 0.6224 0.6224 0.8401

aBased on sugars formed by liquefaction and saccharification. Mean±SDfrom four replicates.

Table 4. Reducing sugar content in the saccharifiedslurry from sweet potato flour and its blends.

Treatments

Reducing sugars (g/100)

Percentageconversion to

reducing sugarsaTotal

Formed byliquefaction andsaccharification

Native sweetpotato flour

64.98±0.53 61.75±0.53 87.83±0.75

Sweet potato�riceflour blendsT1 65.58±0.36 62.63±0.36 88.85±0.51T2 66.18±0.49 63.52±0.49 89.64±0.70T3 67.14±0.41 64.76±0.41 89.94±0.57T4 68.58±0.41 66.48±0.41 91.84±0.56

Sweet potato�wheatflour blendsT5 65.22±0.30 62.18±0.30 88.20±0.43T6 64.38±0.23 61.54±0.23 87.53±0.33T7 63.72±0.31 61.07±0.31 86.62±0.44T8 62.70±0.41 60.24±0.41 86.58±0.59LSD (5%) 0.5726 0.5726 0.8064

aBased on sugars formed by liquefaction and saccharification. Mean±SDfrom four replicates.




originally present in the initial source material and thatformed after the liquefaction�saccharification reactions)showed that the native cassava or sweet potato flouryielded mash having 67.41% and 60.87% glucose,respectively. It was found that the saccharified mashfrom cassava�rice blends had 66.45�68.14% glucose,while that from cassava�wheat blends had 66.33% glu-cose at 10% level of addition of wheat to cassava(Table 5). Higher levels of addition of wheat led to sac-charified mash with only 60.90�63.35% glucose.Significant increase in glucose yield over native cassavaflour was obtained with cassava�wheat blend (60 : 40).The sweet potato�rice/wheat blend yielded mash having58.84�60.96% glucose (Table 5). Sweet potato rootshave been reported to contain a higher percentage ofthe nonreducing sugar, sucrose than maltose and glu-cose (Truong et al., 1997). It is likely that during theinitial liquefaction reaction by Liquezyme X conductedat 90 �C for 1 h, the native invertase may be gettinginactivated. Hence, the contribution of the glucose pre-sent in the original raw material towards the final glu-cose content in the saccharified mash may beinsignificant. Native sweet potato starch could yield68.12 g glucose/100 g after saccharification with astarch conversion efficiency of 89.87% (Johnson et al.,2005). Nevertheless, the starch in sweet potato is extract-able only to the extent of 65�70%, due to the granulesbeing trapped in the pectin�hemicellulose matrix(Balagopalan, 2000). Use of flour or flour blends is abetter option, as the tedium associated with starchextraction, poor starch yield, cost factor, etc. could beeliminated. The yield of glucose in the saccharified mashfrom sweet potato was less than that from cassava, dueto the low starch content in the former roots. Thesaccharification of sweet potato flour for ethanol pro-duction was investigated by Bindumole andBalagopalan (2001), who reported a yield of 78.79% of

reducing sugars using Termamyl and amyloglucosidasetreatment at 45 �C for 72 h, without accounting for thereducing sugars in the initial raw material. The reactiontime was reduced to 49 h in our study (1.0 h for lique-faction and 48 h for saccharification), with a relativeyield of 64.98 g reducing sugars per 100 g sweet potatoflour.

Isomerization

The fructose yield in the isomerized syrup from nativecassava flour was 29.10%, with a percentage conversionof 43.20% to fructose (Table 6). It was found that theyield of fructose from cassava�rice blends was also sim-ilar, falling in the range of 28.68�29.62%, with a per-centage conversion of 42.98�43.86%. The fructose yieldwas significantly less in the isomerized syrup from cas-sava�wheat flour blends, compared to native cassavaflour and proportionate decrease occurred with increasein the level of addition of wheat flour from 10% to40 %. The fructose content in the cassava�wheat(60 : 40) mash was only 26.32%, which resulted fromthe low glucose content in the saccharified mash usedfor isomerization (Table 5). Nevertheless, the conversionefficiency remained the same (42.61�43.81%), indicatingthat the isomerization reaction was adequately opti-mized. The yield of fructose was found to be dependenton the glucose content of the saccharified mash, whichin turn was proportional to the starch content in the rawmaterial. Cassava�rice flour blends were found to begood source material for HFS production.

The yield of fructose in the isomerized mash fromsweet potato was in the range of 25.18�26.54% for thesweet potato�rice flour blends and 25.10�25.84% forthe sweet potato�wheat blends, in comparison to26.26% in the mash from sweet potato flour (Table 7).Increasing the level of incorporation of rice flour

Table 6. Fructose yield and percentage conversionto fructosea after isomerization from cassava�cereal

flour blends.

TreatmentsFructose yield(g/100 g flour)

Percentageconversiona

Native cassava flour 29.10±0.33 43.20±0.49Cassava�rice flour blends

T1 28.68±0.14 42.98±0.21T2 29.16±0.14 43.86±0.21T3 29.02±0.19 43.24±0.28T4 29.62±0.21 43.23±0.30

Cassava�wheat flour blendsT5 28.46±0.16 42.96±0.25T6 27.74±0.22 43.81±0.35T7 26.64±0.11 42.61±0.18T8 26.32±0.24 43.23±0.39LSD (5%) 0.2944 0.4500

aBased on the glucose content. Each value is Mean±SD from four replicates.

Table 5. Glucose content in the saccharified mashfrom cassava/ sweet potato�cereal flour blends.

TreatmentsGlucose contenta (g/100 g flour)

Cassava Sweet potato

Rice flour blendsT1 66.73±0.20 59.83±0.33T2 66.45±0.11 60.14±0.19T3 67.12±0.29 60.48±0.11T4 68.14±0.11 60.96±0.13

Wheat flour blendsT5 66.33±0.14 59.91±0.19T6 63.35±0.20 59.69±0.11T7 62.53±0.24 59.29±0.17T8 60.90±0.19 58.84±0.20LSD (5%) 0.2824 0.2658

aTotal of the glucose originally present in the flour blends and that formedafter liquefaction and saccharification. Each value is Mean±SD from fourreplicates.




increased the fructose yield as well and significantlyhigher yields were obtained for sweet potato�riceblends (70 : 30 and 60 : 40). The percentage conversionof glucose to fructose was not influenced by the starchcontent in the roots or glucose content in the saccharifiedmash. Dziedzic (1981) also reported that the percentageisomerization was not directly proportional to theDextrose Equivalent (D.E) of the glucose syrup.Irrespective of the composition of the blends, the per-centage conversion ranged from 42.02% to 43.52%.Khalid (2005) reported a composition of 60.7% glucose,36.9% fructose and 2.4% maltose in the high fructosesyrup obtained from cassava starch. However, a higheryield of 42.0�43.0% was obtained from cassava or sweetpotato flour or their blends, in our study. Besides thecost associated with the conversion of roots to starch,the low yield of starch due to processing loss or poorextractability (sweet potato) could be eliminated by thedirect use of the root flour or its blends with cereal flours.Berghofer and Sarhaddar (1988) also reported that directconversion of cassava roots/chips for HFS productionwas a feasible proposition. On the contrary, Ghildyalet al. (1989) found that the low raw material cost ofcassava flour, as compared to starch, was totally upsetby the higher capital investment on plant and machinery,higher quantity of activated carbon, etc. Despite this, amajor advantage of our study seems to be the reductionin the reaction time and ability to use flour, so that theproblems related to the poor post harvest life of cassavaroots and poor extractability of starch from sweet potatoroots could also be appropriately eliminated.

CONCLUSIONS

In order to cope with the increasing demand for HFS,there is a need to look out for alternative substrates.

Though cassava and sweet potato are concentratedsources of starch, their potential for HFS productionhas not been fully tapped. A major fraction of the costtowards HFS production is contributed by the rawmaterial, that is starch. If the conversion to starchcould be bypassed by the direct use of the flour, thecost could be considerably reduced. The scope ofdirect use of native cassava or sweet potato flour ortheir blends with cereal flours for HFS production wasinvestigated in this study. In comparison to pure cassavastarch, the native cassava flour or its blends with riceflour yielded only 2�5 units less of reducing sugars andthe percentage conversion to reducing sugars for theflour and its blends was 90�92% (96% for cassavastarch). Sweet potato flour and its blends were less effec-tive than cassava, as the yield of reducing sugars wasless, resulting from the low initial starch content insweet potato roots. Accordingly, the glucose and fruc-tose yields were also higher from cassava than sweetpotato. Irrespective of the nature of the source material(flour or its blends), the percentage conversion of glu-cose to fructose during isomerization was 42�43%. Thestudy showed that the flour from cassava or sweetpotato could be used in native form or as blends withrice/wheat flour, with a better yield from cassava thansweet potato.

ACKNOWLEDGMENTS

The first author (Regy Johnson) acknowledges theResearch Fellowship by the Council of Scientific andIndustrial Research (CSIR), Govt. of India. The authorsare grateful to the Director, CTCRI for the facilitiesprovided for the study and Dr J. Sreekumar, SeniorScientist (Agricultural Statistics), CTCRI for the statis-tical analyses.

REFERENCES

Abraham K., Edison S., Unnikrishnan M., Sheela M.N., Vimala B.,Sreekumari M.T. and Naskar S.K. (2006). Technical BulletinSeries 24. In: Tuber Crop Varieties Released by the CentralTuber Crops Research Institute, Thiruvananthapuram, Kerala:Central Tuber Crops Research Institute.

Antrim R.L., Solheim B.A., Solheim L., Auterinen A.L., Cunefare J.and Karppelin S. (1991). A new Bacillus licheniformis Alpha-Amylase capable of low pH liquefaction. Starch/Starke 43(9):355�360.

Arasaratnam V. and Balasubramaniam K. (1993). Synergistic actionof a-amylase and glucoamylase on raw corn. Starch/Starke 45(6):231�233.

Aschengreen N.H., Helwiig Nielsen B., Rosendal P. and Ostergaard J.(1979). Liquefaction, saccharification and isomerisation of starchfrom sources other than maize. Starch/Starke 31(9): 64�66.

Balagopalan C. (2000). Integrated Technologies for Value Additionand Post Harvest Management in Tropical Tuber Crops.Thiruvananthapuram, Kerala: Central Tuber Crops ResearchInstitute, p. 147.

Table 7. Fructose yield and percentage conversionto fructose after isomerization from sweet

potato�cereal flour blends.

TreatmentsFructose yield(g/100 g flour)

Percentageconversiona

Native sweet potato flour 26.26±0.31 43.16±0.52Sweet potato�rice flour blends

T1 25.18±0.16 42.02±0.27T2 26.18±0.23 43.52±0.38T3 26.30±0.17 43.45±0.29T4 26.54±0.18 43.54±0.29

Sweet potato�wheat flour blendsT5 25.84±0.11 43.12±0.19T6 25.38±0.08 42.53±0.13T7 25.44±0.27 42.89±0.46T8 25.10±0.26 42.69±0.45LSD (5%) 0.3066 0.5115

aBased on the glucose content. Each value is Mean±SD from four replicates.




Barker S.A. and Petch G.S. (1985). Enzymatic processes for high fruc-tose corn syrup. In: Laskin A.I. (ed.), Enzymes and ImmobilizedCells in Biotechnology, California: The Benjamin/CummingsPublishing Company Inc, pp. 93�107.

Baskar G., Muthukumaran C. and Renganathan S. (2008).Optimization of enzymatic hydrolysis of Manihot esculenta rootstarch by immobilized a- amylase using response surface meth-odology. International Journal of Chemical and BiomolecularEngineering 1(3): 156�160.

Berghofer E. and Sarhaddar S. (1988). Production of glucose and highfructose syrup by enzymatic direct hydrolysis of cassava roots.Process Biochemistry 32: 910�915.

Bergmeyer H.U. and Bernt E. (1974). D- glucose: Determinationwith GOD and POD. In: Begmeyer H.U. (ed.), Methods ofEnzymatic Analysis, Vol. 3, New York: Academic Press,pp. 1205�1212.

Bindumole V.R. and Balagopalan C. (2001). Saccharification of sweetpotato flour for ethanol production. Journal of Root Crops 27(1):89�93.

Bradbury J.H. and Holloway W.D. (1988). Chemistry of tropical rootcrops: significance for nutrition and agriculture in the Pacific.ACIAR Monograph Ser. No. 6. Canberra.

Cereda M.P., Conceicao F.A.D., Cagliari A.M., Heezen A.M. andFioretto R.B. (1982). Comparative study of sweet potato(Ipomoea batatas) varieties to estimate their utilization in thefood industry. (Portuguese). Turrialba 32: 365�370.

Chen W.P. and Anderson A.W. (1980). Extraction of hemicelluloseryegrass straw for the production of glucose isomerase and theuse of the resulting straw residue for animal feed. Biotechnologyand Bioengineering 22(3): 519�531.

Cock J.H. (1985). Cassava -New Potential for a Neglected Crop.Boulder: Westview Press, p. 191.

Deobald H.J., Hasling V.C., Catalano E.A. and McLemore T.A.(1969). Relationship of sugar formation and sweet potatoalpha-amylase activity during processing for flake production.Food Technology 23: 826�828.

Dziedzic S.Z. (1981). Production and physico-chemical properties ofisomerized glucose syrups. Starch/Starke 33(11): 369�372.

FAO (2006). Production Year Book. Rome, Italy: FAO.

Franco C.M.L. and Ciacco C.F. (1987). Studies on the susceptibility ofgranular cassava and corn starches to enzymatic attack. Part 1:Study of the conditions of hydrolysis. Starch/Starke 39(12):432�435.

Ghildyal N.P., Ramakrishna M. and Lonsane B.K. (1989).Comparative economics of the production of high fructosesyrup from cassava chips and cassava starch. Starch/Starke41(2): 64�68.

Gorinstein S. and Cheng-yi Lii. (1992). The effects of enzyme hydro-lysis on the properties of potato, cassava and amaranth starches.Starch/Starke 44(2): 461�466.

Hanover L.M. and White J.S. (1993). Manufacturing, compositionand applications of fructose. American Journal of ClinicalNutrition 58(5): 724S�732S.

Johnson R., Moorthy S.N. and Padmaja G. (2004). Enzyme kinetics inthe liquefaction and saccharification of cassava starch for HighFructose Syrup (HFS) production. In: 1st National Seminar onRoot and Tuber Crops NSRTC 1, Bhubaneswar, Orissa, 29�31October.

Johnson R., Moorthy S.N. and Padmaja G. (2005). Optimized param-eters for the enzyme catalysed liquefaction and saccharificationof sweet potato starch. Journal of Root Crops 31(1): 7�13.

Johnson R., Moorthy S.N. and Padmaja G. (2009). Comparative pro-duction of glucose and high fructose syrup from cassava andsweet potato roots by direct conversion techniques. InnovativeFood Science and Emerging Technologies 10: 616�620.

Khalid N.M. (2005). Cassava starch for high fructose syrup produc-tion. In: Proceedings of the 2nd International Symposium Sweetpotato and Cassava, Kuala Lumpur, Malaysia, 14�20 June.

Labout J.M. (1985). Conversion of liquefied starch into glucose usinga novel glucoamylase system. Starch/Starke 37(5): 157�161.

Lages A.C.A. and Tannenbaum S.R. (1978). Production of glucosefrom tapioca (cassava starch) and farinha de mandioca (cassavameal). Journal of Food Science 43: 1012�1014.

Moorthy S.N. and Padmaja G. (2002). A rapid titrimetric method forthe determination of starch content in cassava tubers. Journal ofRoot Crops 28(1): 30�37.

Morales S., Alvarez H. and Sanchez C. (2008). Dynamic models forthe production of glucose syrups from cassava starch. Food andBioproducts Processing 86(1): 25�30.

Nelson N. (1944). A photometric adaptation of the somogyi methodfor determination of glucose. Journal of Biological Chemistry 153:375�380.

Piyachomkwan K., Walapatit S., Vetthaisong T., Keawsompong S.and Sriroth K. (2005). Advanced technology in ethanol produc-tion from cassava chips. In: Proceedings of the 2nd InternationalSymposium Sweet potato and Cassava, Kuala Lumpur, Malaysia,14�15 June.

Shetty J., Chotani G., Gang D. and Bates D. (2007). Cassava as analternative feedstock in the production of renewable transporta-tion fuel. International Sugar Journal 109(1307): 3�11.

Taniguchi H. (2004). Carbohydrate research and industry in Japanand the Japanese Society of Applied Glycoscience. Starch/Starke 56(1): 1�5.

Truong V.D., Walter Jr W.M. and Hamann D.D. (1997). Relationshipbetween instrumental and sensory parameters of cooked sweetpotato texture. Journal of Textural Studies 28: 163�185.

Waliszweski K.N., Alvarado M.G. and De la Cruz Medina J. (2002).Kinetics of enzyme hydrolysis of cassava flourstarch-optimiza-tion and modelling. International Journal of Food Science andTechnology 27(4): 465�472.

Wardrip E.K. (1971). High fructose corn syrup. Food Technology 25:501�504.

White J.S. (1992). Fructose syrup: production, properties and applica-tions. In: Schenck F.W. and Hebeda R. (eds), Starch HydrolysisProducts- Worldwide Technology, Production and Applications,New York: VCH Publishers, Inc, pp. 177�200.







DOI: 10.1177/1082013210366779

2010 16: 259 originally published online 9 July 2010Food Science and Technology InternationalI. Yilmaz and M. Demirci

Quality Characteristics of MeatballEffect of Different Packaging Methods and Storage Temperature on Microbiological and Physicochemical

Published by:


On behalf of:



















Effect of Different Packaging Methods and Storage

Temperature on Microbiological and Physicochemical Quality

Characteristics of Meatball

I. Yilmaz* and M. Demirci

Faculty of Agriculture, Department of Food Engineering, Namik Kemal University59030 Tekirdag, Turkey

The objective of this research was to determine physicochemical changes and microbiological quality of the

different packaged meatball samples. Meatball samples in polystyrene tray were closed with polyethylenefilm (PS packs), vacuumed and modified atmosphere packaged, (MAP) (65% N2, 35% CO2), and heldunder refrigerated display (4 �C) for 8, 16 and 16 days for PS packs, vacuum and MAP, respectively.Microbial load, free fatty acids and thiobarbituric acid values of the samples tended to increase with

storage time. Bacteria counts of the raw meatball samples increased 2 log cycles at the end of storagecompared with initial values. Meatball samples can be stored without any microbiological problem for7 days at 4 �C. Results from this study suggested that shelf-life assigned to modified-MAP and vacuum-

packed meatballs may be appropriate. Meatball samples underwent physical deformation when they werepacked before vacuum process. With these negative factors considered, MAP is superior to other two packsmethods.

Key Words: meatball, modified atmosphere packaging, vacuum, shelf-life, microbiological quality

INTRODUCTION

Extending the shelf-life of fresh meat has been one of

the important aims in the meat industry in recent years

(Ahmad and Marchello, 1989). Microbial spoilage of

fresh meat is one of the major problems of the meat

industry. The average storage and retail shelf-life of

fresh pork is approximately 10�14 days. The limiting

factor is bacterial spoilage (Huffman, 1974).Modified atmosphere packaging (MAP) is well-

known as a method to extend the shelf-life of a variety

of foods including fresh red meat (Luno et al., 2000;

Yilmaz, 2001; Iacurto et al., 2005). In MAP, different

gases and gas combinations are used depending on the

type and properties of the meat product. Atmospheres

combine oxygen (O2), carbon dioxide (CO2) and nitro-

gen (N2) to maintain the quality of fresh red meat. CO2

is known for its inhibitory effect on microbial growth;

nevertheless atmospheres with high levels of CO2 (low

O2) can cause meat discoloration (Iacurto et al., 2005).

The main purpose of MAP of chilled meat is to extendthe microbiological shelf-life and the color stability ofthe packaged meat. In processed meat products, CO2

and N2 combinations are the most common gasesused. The gas composition normally used in MAP is70% O2 � 30% CO2 or (60�70%) CO2 � (30�40%)N2. The 70% O2 � 30% CO2 gives the product anextended shelf-life compared to air (Gill, 1996).Although the elevated concentration of oxygen prolongscolor stability of the meat, it is also expected to increasethe rate of lipid oxidation (Zhao et al., 1994), whichcauses a rancid off-flavor in meat. The compositionbased on (60�70%) CO2 � (30�40%) N2 furtherextend the shelf-life of packaged meat; however, thelack of oxygen and the elevated CO2 concentration,makes the meat become dark pink or pale brown afterpackaging (Dai and Nan, 2005).

The objective of the present study was to examine theeffects of different packaging methods (vacuum andmodified atmosphere) on microbiological spoilage,physicochemical and chemical changes of meatballsstored at 4 �C for up to 8�16 days.


Veal meat is ground with different seasonings (groundblack pepper 0.1%, red pepper 2% and cumin 0.4%) andsome other additives (onion 3%, garlic 0.5%, salt 2%

*To whom correspondence should be sent(e-mail: [email protected]).Received 2 April 2009; revised 29 June 2009.




and toasted bread crumbs 8%) and kneaded for

30min. The mix is reground and stored in a cold room

(þ4 �C) for 1 day and then is shaped into 2 cm diameter

meatballs with a weight of 18�20 g before cooking.

Meatball samples were packed in vacuum, polystyrene

tray and modified atmosphere containing 65% N2 �35% CO2. Meatball samples under refrigeration were

transported to laboratory and handled immediately.

Meatball samples were packaged under vacuum and

MAP using low O2 permeable (14.38mL/m2/24 h at

23 �C, 0% RH; moisture vapor transmission rate of

4.1 g/m2/24 h at 38 �C, 90% RH) polyvinylchloride

tray (02 transmission rate 250 cc/cm2/24 h at 23 �C, 0%

RH; CO2 transmission rate 2.100 cc/cm2/24 h at 23 �C,

90% RH) and lidding material composed of polyethyl-

ene (oxygen permeability 7.06mL/m2/24 h at 23 �C, 0%

RH; moisture vapor transmission rate of 2.1 g/m2/24 h

at 38 �C, 90% RH (Cryovac). The packaging machine

type, TIROPAC 1000 S (Turkey) was used to vacuum

and MAP samples. The combination of gases used to

MAP meatball was 65% N2 � 35% CO2. As a third

packaging method, samples were put in polystyrene

tray and wrapped with polyethylene film (PS packs).

A tray contained 10 meatballs. All packageded samples

were stored at 4 �C.Totalbacteria, coliforms,Escherichia coli,Staphylococcus

aureus, psychrophilic bacteria, Salmonella, Clostridium

perfringens counts of the meatball samples were determined

according to the FDA (1998) procedure and reported

as cfu/g.Total aerobic counts were determined on plate count

agar (PCA) and incubated at 35 �C for 48 h. The num-

bers of coliforms were determined with violet red bile

agar (VRBA) after the incubation at 35 �C for 24 h. The

presence of E. coli was examined by transferring 1mL of

each sample dilution to sterile Petri dishes followed by

pouring 10mL of violet red bile agar (with 4-methylum

belliferryl-ß-D-glucuronide (VRB-MUG)) tempered

48 �C into plates. The plates were swirled, allowed to

solidify, overlaid with 3�5 of VRB-MUG and then incu-

bated at 37 �C for 24�48 h. The plates were the exam-

ined for typical coliform colonies which were counted to

obtain a presumptive coliform count. Isolates that were

Gram-negative and which produced acid and gas in lac-

tose broth were recorded as confirmed coliforms. These

plates were also examined under long wave ultraviolet

(UV) light for the presence of fluorescent colonies, indi-

cating possible presence of E. coli. Those with positive

fermentation and gas production in lactose broth

were further characterized as E. coli using indole,

methyl red, Voges-Proskauer and citrate (IMVIC) iden-

tification tests.Total Staphylococcus counts were determined on

Staphylococcus Medium 110 and incubated at 37 �C

for 48 h (Speck, 1976). The presence of S. aureus was

tested by surface plating on prepoured and dried

Baird-parker agar with egg-yolk tellurite enrichment.

The plates were incubated at 35 �C for 48 h after

which were examined for typical S. aureus colonies.

Suspicious colonies were transferred slants for

S. aureus confirmation by Gram stain, catalase reaction

and coagulase test.For the Salmonella, 25 g samples were enriched in

Selenite Cystine Broth for 24 h at 35 �C and then the

cultures were streaked onto bismuth sulfite agar and

incubated at 35 �C for 24 h. The typical Salmonella

colonies were subjected to subsequent biochemical tests

by using Triple Sugar Iron and Lysine Iron Agar slants.

However, due to the some inconveniences, the presump-

tive Salmonella cultures from the agar slants could

not be subjected to serological tests for the final

confirmation.Psychrophilic bacteria were determined on PCA after

incubation at 7 �C for 10 days. The proteolitic bacteria

were determined on Skim Milk Agar after incubation at

37 �C for 24 h (Speck, 1976). The yeasts and molds were

counted on potato dextrose agar (PDA) after 5 days of

incubation at 25 �C. The presence of C. perfringens was

tested by Trypticase Sulfite Neomycin Agar and the

plates were anaerobically incubated at 37 �C for 48 h.Moisture and pH measurements were done according

to the methods described by AOAC (1990).For determination of free fatty acids (FFA), a sample

(10 g) was mixed with anhydrous Na2SO4 (100 g), fat was

extracted in CHC13, (100mL) and filtered. In order to

remove nonfatty acids, that may have come from for-

mulation ingredients, an aliquot of the CHCI3, extract

was washed three times with 4�5 volumes of distilled

water in a separatory funnel. Nonfatty acids were parti-

tioned into the aqueous layer and fatty acids into the

CHC13, layer. The FFAs, as a percent of oleic acid,

were estimated (AOAC, 1990).The color of the meatballs were measured using a

DP-900 D25 Aoptical Sensor Reston (Virginia, USA),

which was used to determine Commission International

de l’Eclairage (CIE) L (lightness), a (redness) and b (yel-

lowness) values.The distillation method of Tarladgis et al. (1960) as

Modified by Ke et al. (1977) was used to determine the

degree of lipid oxidation in meat samples. Results were

expressed as thiobarbituric acid reactive substances

(TBARS; milligrams of malonaldehyde per kilogram

of meat).The data obtained from three replications were ana-

lyzed by ANOVA using the SPSS statistical package

program and differences among means were compared

using the Duncan’s multiple range test. The model

included storage temperature (4 �C) and storage period

(0, 1, 2 or 16 days) as main effects (storage time

and packaging systems) and all their interactions

260 I. YILMAZ AND M. DEMIRCI



(Soysal, 1992). Treatment means were separated by leastsquare means procedures if a difference was detected atp< 0.05.


The initial physicochemical and chemical characteris-tics of meatball samples were as follows (average of ninesamples): pH 5.74, moisture 57.69%, protein 18.52%,ash 2.82%, fat 17.63%, FFA 0.350%, TBA 0.234mgmalonaldehyde/kg, L-value 42.37, a-value 7.57 andb-value 15.35. The effects of the storage temperature(4 �C) on the chemical quality of meatball samples arein Table 1. In general, there were significant (p< 0.05)

alterations in some quality parameters of the meatballsamples during storage.

pH

Availability of nutrients affect the selection andgrowth of bacteria; pH of meatball samples decreasedduring storage period (Table 1). While in the beginningof storage period pH of samples were 5.74, 5.72 and5.72, at the ending of storage period pH of meatballsamples decreased to 4.87 (8 days), 4.52 (16 days) and4.43 (16 days) in PS packs, MAP and vacuum packaged,respectively. Storage time� packaging systems interac-tion was significant for pH (p< 0.05). pH valuedecreased by the action of microorganisms, whichcaused acidity development. Similarly, Rubio et al.

Table 1. Chemical and physicochemical characteristics of different packaged meatball samples.

Sample Days Moisture FFA TBA L a b pH

PS packs 1 57.69ab 0.350n 0.234pr 42.37cd 7.57m 15.35a 5.74a2 57.17abcdef 0.395mn 0.263p 42.17cdef 8.91l 14.88bc 5.68c3 56.85cdef 0.437klmn 0.410o 40.82ijk 11.24ij 14.85bc 5.57f4 55.79g 0.521jklmn 0.456no 41.43fghi 11.74hi 14.84bc 5.48i5 54.57h 0.590ijkl 0.803h 41.98defg 13.38def 14.52cde 5.31n6 53.25i 0.705ghij 0.816h 41.31ghij 13.26def 14.74bcd 5.19o7 52.08j 0.752fghi 0.852gh 41.85defg 12.64fg 14.83bc 5.17p8 51.41j 0.815fgh 1.241e 43.75a 9.15l 13.65gh 4.87u

MAP 1 57.07abcdef 0.357n 0.424 o 41.83defg 8.86l 15.05ab 5.72b2 56.94abcdef 0.425klmn 0.522lm 40.61jk 11.33ij 14.87bc 5.63e3 56.92abcdef 0.442klmn 0.571kl 40.31kl 13.63bcd 14.89bc 5.54g4 56.88bcdef 0.578ijklm 0.616jk 41.31ghij 14.41ab 14.95abc 5.49h5 56.77def 0.618ijk 0.604jk 43.16ab 12.96defg 13.88fg 5.43k6 56.58defg 0.657hij 0.616jk 40.90hijk 14.26abc 14.77bc 5.40m7 56.48efg 0.890fg 0.691i 42.22cde 13.66bcd 14.99abg 5.06r8 56.43fg 1.615e 1.210e 43.62a 10.67jk 14.27fg 4.92s9 56.41fg 1.443cd 1.315d 42.74bc 11.24ij 14.32fg 4.77v

10 56.08g 1.635c 1.334d 42.56bc 11.39ij 14.56cde 4.70 v11 55.91g 1.903bc 1.422d 42.13cd 10.80jk 14.71bcd 4.61v12 55.47gh 2.168b 1.655c 41.89defg 10.24jk 13.89fg 4.60y13 55.12gh 2.250b 1.834bc 41.82defg 9.75l 14.07de 4.59y14 54.84gh 2.440ab 1.912b 41.55defg 9.82l 14.09de 4.57y15 54.41h 2.665a 2.111ab 42.07cdef 8.93l 14.81bc 4.55y16 53.45h 2.353ab 2.437a 42.63cd 9.07l 14.76bcd 4.52y

Vacuum 1 57.73a 0.355m 0.180s 40.48kl 10.03k 14.76bcd 5.72b2 57.63abc 0.413lmn 0.186rs 39.79l 12.74efg 14.32def 5.64d3 57.40abcd 0.440klmn 0.350p 41.00hijk 13.59cd 14.96abc 5.46j4 57.34abcd 0.542jklmn 0.476mn 40.97hijk 15.03a 14.77bcd 5.43k5 57.28abcde 0.607ijkl 0.637j 42.82bc 12.31gh 13.36h 5.42l6 56.98abcdef 0.641hij 0.741i 41.63efgh 13.47cde 14.65bcde 5.16q7 56.65defh 0.656hij 0.901g 41.41ghi 14.27abc 14.66bcde 4.88t8 56.63def 0.918f 1.024f 43.50ab 10.75jk 12.80i 4.74v9 56.50fg 1.064d 1.395d 42.25cd 10.43jk 12.75i 4.62y

10 56.27fg 1.335cd 1.419d 42.71bc 11.26ij 13.27h 4.57y11 55.97g 1.573cd 1.509cd 42.83bc 11.52ij 14.54cde 4.58y12 55.52g 1.656c 1.572cd 41.94defg 11.75ij 13.87fg 4.58y13 54.81gh 2.026bc 1.613c 41.90defg 10.28jk 13.71fg 4.49z14 54.22gh 2.100b 1.690c 41.86defg 10.32jk 14.70bcd 4.47z15 53.55h 2.692a 1.783bc 42.41cd 10.75jk 12.92hi 4.45z16 53.65h 2.398ab 1.990b 41.80efg 11.07ij 14.09de 4.43z

Means within the same column with different letters are significantly different (p< 0.05).

Microbiological and Physicochemical Quality Characteristics of Meatball 261



(2007) attributed the decrease in pH of Salchichon (dry

cured Spanish sausage) with increasing storage time to

the activity of lactic acid bacteria. This decrease in pH

was linear during the 16 days of storage. These results

were in agreement with the results of Houben and Van-

Dijk (2001) and Pexara et al. (2002) and disagreement

with the results of Fernandez-Lopez et al. (2006) for

ostrich burgers and Colak et al. (2008) for meatball

added nisin and lactoferrin.

Thiobarbituric Acid

Oxidative rancidity of lipids is a serious problem

during storage of meat and meat products and TBA

value is the most commonly used parameter to measure

it. The development of rancidity as a result of lipid oxi-

dation has been found to limit storage life of beef held in

frozen state (Gokalp et al., 1978). The meatballs samples

contained 17.63% fat. Changes in the rancidity of meat-

ball samples were measured using the TBA analysis.

TBA values for all samples increased up to 8 and 16

days but at different rates. Storage time� packaging sys-

tems interaction was significant for TBA (p< 0.05).

TBA values during storage for 16 days at 4 �C summa-

rized in Table 1. The TBA value also increased signifi-

cantly (p< 0.05) to final value of 1.241, by day 8.

According to Ockerman (1976) meat products with

TBA> 1mg malonaldehyde/kg could be considered as

rancid. The determination of thiobarbituric acid reactive

substances (TBA-RS) is one of the most widely used

methods for measuring lipid oxidation in muscle

foods. Relationships between TBA-RS numbers and

warmed-over flavor have been reported in several stud-

ies (Ang, 1992). Time had a significant effect on

TBA values of all packaged samples. Galvin et al.

(1998) suggested increased suspectibility of thigh

muscle to oxidation was due to several factors includ-

ing higher total lipid and phospholipids (and there-

fore, higher unsaturated fatty acids) and higher iron

contents.The 2-thiobarbituric acid reactive substances values

(TBARS) represent the content of secondary lipid oxi-

dation products, mainly aldehydes (or carbonyls), which

contribute to off-flavors in oxidized meat and meat

products. The effect of the packaging systems on

TBARS value of meatball samples over 8-16 days of

refrigerated storage (4 �C) is shown in Figure 1.In a previous study, TBA values for all samples

increased at different rates during the storage for 180

days at �18 �C. The TBA value also increased signifi-

cantly (p< 0.05) to final value of 1.018, 0.812

and 0.759mg malonaldehyde/kg PS packs, MAP and

vacuum packed samples, respectively (Yilmaz and

Demirci, 2009). Similar results were reported by Aksu

et al. (2005).

Hunter L, a, b

Color of meat and meat products is very importantfrom the consumers point of view. Color parameters ofmeatball are displayed in Table 1. Lightness increasedduring storage in vacuum packed samples slowing sig-nificantly higher values by the 8th day storage. Therewas a significant (p< 0.05) storage time� packaging sys-tems interaction for L-, a- and b-value. Initial a-value forthe meatball samples ranged from 7.57 to 10.03. Zakryset al. (2008) reported that instrumental a-value displayeda negative correlation with days.

CO2 and N2 atmospheres combinations of these gaseskeep myoglobin in the reduced state (Sorheim et al.,1996). Generally, L, a and b color values of samplesincreased slightly throughout storage, indicating thatmeat became more pale and that the meat color changedfrom red to yellow (Lambert et al., 1992). Brooks (1993)demonstrated that concentrations of CO2 over 30%caused increased discoloration, but systems containingup to 20% had little effect on meat color.

Zhao et al. (1994) reported that the gaseous environ-ment within a MAP is not static and product respira-tion, microbial metabolism and gas mobilizationcontinually act to change the composition of the atmo-sphere within the pack. Therefore, L, a and b value arenot stable for storage time in this research for all packedmeatball samples. In addition to O2 permeability andmoisture vapor transmission of packaging material,there was decrease of pH value and increase of microbialload which caused instable color value.

Moisture Content

The moisture contents were lower in the PS packsthan the other samples. The lowest moisture contentwas found in the PS packs samples (51.41%) for8 days. There were appreciable differences in moisturecontents of sample (p< 0.05). There was a significant(p< 0.05) storage time�packaging systems interactionfor moisture.

0

0.5

1

1.5

2

2.5

3

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Days

Lipi

d ox

idat

ion

(mg

mal

onal

dehy

de/k

g)

Figure 1. Lipid oxidation (TBARS) of different pack-aged meatball samples. (¨) PS Packs, (#) MAP, (m)vacuum.




The moisture losses of meatball samples were affected(p< 0.05) by packaging type and storage time. Themoisture loss of meatballs packaged in MAP andvacuum remained fairly stable, between 3% and 4%during the packaging period, whereas moisture lossduring storage in PS packs was 6% or higher andincreased during display in air-permeable packaging.Similar results were reported by Yilmaz and Demirci(2009).

FFA

The increase in FFA was linear during the storagetime. The FFA levels were low at meatball samples inPS packs, then increased slowly during storage in allpackaged meatball samples stored at 4 �C. From theanalysis of variance, a highly significant (p< 0.05) pack-aging method storage time interaction was observed.Similar results were reported by Sawaya et al. (1995)and Kesava Rao et al. (1996).FFA, microbial load and TBA values of the samples

tended to increase with storage time while % moisturedecreased (Table 1). For example, initial FFA values ofthe packaged meatball samples were 0.350 but theygradually increased to 0.815, 1.615 and 0.918 at the8th day of the storage in PS packs, MAP and vacuum,respectively (Table 1). FFA values of MAP and vacuumpackaged meatball samples were significantly (p< 0.05)higher than that of PS packed meatball samples for allstorage times. Storage time� packaging systems interac-tion was significant for FFA (p< 0.05).

Microbiological Quality

The initial microbiological characteristics of meatballsamples are as follows; 3.9� 105 cfu/g total aerobicmesophile bacteria, 8.8� 103 cfu/g coliform group bac-teria, 1.5� 102 cfu/g E. coli, 3.9� 103 cfu/g total staphy-lococci, 5.3� 102 cfu/g S. aureus, 6.2� 105 cfu/gpsychrophilic bacteria, 1.6� 103 cfu/g yeast and moldand 1� 103 cfu/g proteolitic bacteria. Salmonella spp.and C. perfringens were not found in meatball samples.The shelf-life of the meatball samples stored at 4 �C

is given in Table 2. There was a significant (p< 0.05)storage time�packaging systems interaction for aero-bic plate count, coliform bacteria, E.coli, S.aureus, psy-chrophilic bacteria, yeast and mold. Storage time�packaging systems interaction was not significantfor total staphylococci and proteolitic bacteria(p> 0.05). Ground beef is a product of relatively poorcolor stability and microbiological condition (Gill andJones, 1994).As can be seen from Table 2, in PS packed meatball

samples, the total aerobic bacteria counts reached at theend of the 6th day were higher than 107 cfu/g, which isconsidered a spoilage level for this type of product.Similar results were reported by Fernandez-Lopez

et al. (2006). Mesophile aerob bacteria count of theraw meatball samples increased by 102 cfu/g at the endof storage compared to beginning. These results were inagreement with the results of Yilmaz and Demirci(1995). Putrefaction of raw meatball samples weredetected when the mesophile aerobic bacteria countreached to 107 cfu/g in the 10th day and 8th day forMAP and vacuum, respectively. The success of MAPtechnology application on meat depends mainly on itsinitial microbiological contamination and of tempera-ture control during storage and distribution; howeverthe growth of microorganisms is also influencedby intrinsic meat factors (Lucke, 1995; Borch et al.,1996; Dainty et al., 1996). It is not surprising rathersupportive that MAP preserves microbiological qualitybetter than aerobic/vacuum packaging methods (Davies,1995).

By packaging the beef in modified atmosphere andstorage at low temperature, the shelf-life can be pro-longed considerably (Nissen et al., 1996). Both vacuumand MAP are used successfully with meatballs for stor-age time.

The vacuum packaging offers benefits through theelimination of oxygen. Bacteria grow much moreslowly in the absence of oxygen. Therefore, bacterio-logical spoilage is greatly reduced. Unrefrigerationconditions the shelf-life of meatball, which is limited to3�4 days by growth of aerobic, psychotropic spoilagemicroorganisms. In refrigerated conditions, the shelf-life of meatball is limited from 7 to 8 days. A totalnumber of 107 cfu/g of meat is considered to be upperlimit of microbiological acceptability (Lambert et al.,1992). Because of the long shelf-life of these packagingproducts, the potential for psychrophilic pathogens togrow to high numbers increases. None of the meatballsamples contained C. perfringens, in agreement withresults of Yilmaz et al. (2002) and Yilmaz and Demirci(2009), but all of meatball samples containedSalmonella.

Turkish meat ball governmental guideline (TSE.,1992) states that total bacteria, psychrophilic bacteriaand yeast and mold counts are limited as 106 cfu/g,105 cfu/g and 102 cfu/g respectively and there shouldnot be any E. coli, S. aureus, Salmonella and sulphidereducing anaerobes. The levels of yeasts and molds were2� 105 cfu/g (for 8 days), 4.2� 105 cfu/g (for 16 days)and 4.2� 105 cfu/g (for 16 days) PS packs, MAP andvacuum, respectively. The growth of yeast is greatlyaffected by the modified atmosphere (CO2þN2), withO2 being restricted in an anaerobic atmosphere.Psychrophilic bacteria counts of samples stored at4 �C increased to 1.3� 107 cfu/g by day 8 (PS packs).In this study, no clear differences were found amongthe three packaging systems for proteolitic bacteriacounts in storage conditions studied. Our results are inagreement with the results of Yilmaz and Demirci(2009).




CONCLUSION

Vacuum packaging and MAP have been used in themeat industry for quite a long time and have beenaccepted by the consumers. Both packaging techniques

have the potential to increase the shelf-life of manychilled foods without adversely affecting the quality.

Storage under modified atmospheres is also used forcured meat products. Comparisons of the shelf-life ofcured meat products in vacuum packs and modified

atmosphere packs have resulted in different findings.As a conclusion, meatball samples can be stored with-

out any microbiological problem for 7 days at 4 �C. PSpacks can be treated and deformed easily and

particularly in the markets no attention is paid duringhandling. Findings from the present study showed thatthe quality properties of meatball samples during stor-age may be improved by MAP.

REFERENCES

Ahmad H.A. and Marchello J.A. (1989). Effects of gas atmospherepackaging on psychrotrophic growth and succession on steaksurface. Journal of Food Science 54: 274�276, 310.

Aksu M.I., Kaya M. and Ockerman H.W. (2005). Effect of modifiedatmosphere packaging and temperature on the shelf life of slicedpastırma produced from frozen/thawed meat. Journal of Musclefoods 16: 192�206.

Table 2. Microbiological quality of different packaged meatball samples.

Microbial counts (cfu/g)

Sample Days TMAB Coliform E.coli Total stap. S. aureusPsycrophilic

bacteria Yeast moldProteoliticbacteria

PS packs 1 3.9� 105n 8.8�103o 1.5�102r 3.9�103st 5.3�102g 6.2� 105 kl 1.6� 103q 1�103

2 2.6� 106ikl 3�104mn 2.1�103ij 9.3�103klmn 1�103fg 1.6� 106f 1.6� 104ghi 1.4�103

3 6.6� 106fghi 1.7�105gh 2.5�103i 1.1�104ijklm 1.6�103f 3.8� 106d 3.3� 104d 4.9�103

4 8.5� 106 fghij 3.2�105efg 3.2�103h 2.1�104f 1.9�103f 7.9� 106c 3.5� 104d 5.9�103

5 8.9� 106efgh 4.8�105de 1.2�104d 2.2�104ef 2�103f 8.3� 106c 6.2� 104c 6.3�103

6 1.1� 107defgh 5.6�105de 2�104c 2.3�104ef 2.7�103ef 1� 107b 6.9� 104c 1.5�103

7 1.8� 107defg 7.1�105d 2.8�104b 2.4�104ef 5.9�103d 1.1� 107ab 1.2� 105b 1.6�104

8 3� 107d 7.4�105d 4�104a 8.7�104ab 8.3�103cd 1.3� 107a 2� 105a 3.7�104

MAP 1 7� 105mn 9.3�103o 2.6�102q 2.8�103t 8.9�102ij 4.8� 105m 3.9� 103p 2.9�103

2 1.2� 106lmn 3.9�104lm 6.6�102n 3.7�103st 1�103fg 5.8� 105l 4.9� 103no 6.9�103

3 1.2� 106lmn 1.4�105hi 7.6�102n 3.9�103st 1�103fg 8.1� 105j 6.3� 103lm 8.3�103

4 1.3� 106klm 1.7�105gh 1.1�103l 5.1�103qrs 8.1�102g 1� 106i 8.1� 103k 8�103

5 1.4� 106klm 2�105fgh 1.5�103k 8.3�103mnop 1.3�103fg 1.2� 106hi 1.1� 104j 1.3�104

6 1.4� 106klm 3.5�105ef 2.2�103ij 8.3�103mnop 1.3�103fg 1.3� 106gh 1.4� 104hij 2.2�104

7 1.4� 106klm 3.3�105ef 7.4�103fg 1.1�104jklmn 2�103f 1.4� 106fg 1.6� 104gh 2.1�104

8 4� 106ghij 4.2�105de 8.1�103ef 1.3�104hijkl 3�103ef 1.4� 106fg 2.6� 104e 2.7�104

9 8.3� 106bcdef 4.6�105c 9.3�103e 1.3�104hijk 3.9�103e 1.5� 106f 2.2� 104e 2.8�104

10 7.4� 107cd 4.8�105c 2.3�104bc 1.4�104ghijk 5.5�103d 1.7� 106f 2.9� 104e 5.3�104

11 8.9� 107c 5.1�105c 1.7�104c 1.4�104fghij 6.3�103d 2� 106ef 5.9� 104c 3.4�104

12 1� 108bc 5.8�105c 1.9�104c 1.6�104fghi 8.9�103cd 2� 106ef 8.7� 104bc 3.9�104

13 1.1� 108bc 1�106b 2�104c 2�104f 1.1�104c 2� 106ef 1.5� 105b 5.4�104

14 1.7� 108b 1.2�106b 2�104c 5�104cd 3�104ab 2.2� 106ef 1.7� 105b 5.5�104

15 2.2� 108ab 2.1�106ab 2�104c 6�104bc 3.3�104ab 2.3� 106ef 2.2� 105ab 6.5�104

16 2.9� 108a 3.2�106ab 3.4�104ab 1.2�105a 4.2�104a 4.6� 106cd 4.2� 105a 7.6�104

Vacuum 1 1� 106lmn 1.6�104no 3.6�102p 4.2�103rs 5.5�102g 3.2� 105n 3.8� 103p 2�103

2 1� 106lmn 2.7�104mn 5�102o 6�103pqr 6.9�102g 6.5� 105kl 4.2� 103qp 2.9�103

3 1.5� 106klm 7�104kl 8.1�102mn 6.5�103opq 1.1�103fg 7.1� 105jk 5.5� 103mn 3.1�103

4 1.8� 106klm 7.6�104ik 9.7�102lm 7.6�103nop 1.4�103fg 1.4� 106fg 6.8� 103klm 4.5�103

5 1.7� 106klm 1.1�105hij 1.8�103jk 8.9�103lmno 1.9�103f 1.4� 106gh 7.8� 103kl 1.3�104

6 2.7� 106hijk 3.2�105efg 6�103g 9.6�103klmn 2�103f 1.5� 106fg 1.3� 104ij 1.5�104

7 6.9� 106fghi 4.1�105de 1�103de 1.1�104ijklm 2.8�103ef 2.9� 106e 1.9� 104fg 3.4�104

8 2.6� 107de 5.1�105de 2.3�103bc 1.4�104fghij 3�103ef 3� 106e 2.3� 104ef 4.1�104

9 4.3� 107cd 6.8�105c 2.5�103i 1.5�104fghij 4�103e 3.1� 106e 3.6� 104cd 5.1�104

10 5.7� 107cd 9.8�105c 2.5�103i 1.7�104fgh 4�103e 3.2� 106e 5.5� 104c 5.1�104

11 8.2� 107c 1.7�106b 2.5�103i 1.8�104fgh 4.9�103de 3.2� 106e 9.1� 104bc 5.4�104

12 8.8� 107c 1.8�106b 2.4�103i 1.9�104fg 5.5�103d 3.1� 106e 9.1� 104bc 6.2�104

13 1.1� 108bc 2.2�106ab 3.6�103h 2�104fg 7.9�103cd 3.6� 106d 1.6� 105b 7�104

14 1.2� 108bc 2.6�106ab 3.9�103h 3.1�104e 8.3�103cd 3.8� 106d 2.2� 105ab 8.3�104

15 1.5� 108b 2.7�106ab 4�103h 3.9�104de 1.9�104bc 5.9� 106bc 2.5� 105ab 8.7�104

16 2.1� 108ab 3.9�106a 4.9�103gh 8.4�104ab 2.8�104b 6.3� 106bc 4.2� 105a 1.4�104

Means within the same column with different letters are significantly different (p<0.05).




Ang C.Y.W. (1992). Reheating effect on thiobarbutiric acid reactivesubstance of refrigerated stored, cooked broiler meat. Journal ofFood Protection 55(11): 924�926.

AOAC (1990). Official Methods for the Analysis, 15th edn. Arlington,Washington: Association of Official Analytical Chemists,pp. 931�971.

Borch E., Kant-Muermans M.L. and Blixt Y. (1996). Bacterial spoil-age of meat and cured meat products. International Journal ofFood Microbiology 33: 103�120.

Brooks J. (1993). The effect of carbon dioxide on the color changes orbloom of lean meat. Journal of the Chemical Society 52: 17�19.

Colak H., Hampikyan H., Bingol E.B. and Aksu H. (2008). The effectof nisin and bovine lactoferrin on the microbiological quality ofTurkish-style meatball (Tekirdag kofte). Journal of Food Safety28: 355�375.

Dai R. and Nan Q. (2005). Color stability of chilled beef packaged in anatmosphere containing low level of carbon monoxide.In: Proceedings of 51st International Congress of Meat Science andTechnology, MarylandUSA: Baltimore, pp. 864�870, 7�12August.

Dainty R.H. (1996). Chemical/biochemical detection of spoilage.International Journal of Food Microbiology 33: 19�30.

Davies A.R. (1995). Fate of foodborne pathogens on modified-atmo-sphere packaged meat and fish. International Biodeterioration andBiodegradation 36: 407�410.

Fernandez-Lopez J., Jimenez S., Sayas-Barbera E., Sendra E. andPerez-Alvarez J.A. (2006). Quality characteristics of ostrich(Struthio camelus) burgers. Meat Science 73: 295�303.

FDA (1998). Bacteriological Analytical Manual, 8th edn.Gaithersburg, MD, USA: FDA.

Galvin K., Morrissey P.A. and Buckley D.J. (1998). Effect of dietarya-tocopherol supplementation and gamma irradiation on a-toco-pherol retention and lipid oxidation in cooked minced chicken.Food Chemistry 62: 185�190.

Gill C.O. (1996). Extending the storage life of raw chilled meats. MeatScience 43: 99�109.

Gill C.O. and Jones T. (1994). The display of retail packs of groundbeef after their storage in master packages under various atmo-spheres. Meat Science 37: 281�295.

Gokalp H.Y., Ockerman H.W., Plimpton R.F., Parfett N.A. andCahill V.R. (1978). Effect of different packaging methods onobjective quality characteristics of frozen and stored cow beef.Journal of Food Science 43: 297�300.

Houben J.H. and Van-Dijk A. (2001). Effect of dietary vitamin Esupplementation and packaging on the color stability of slicedpasteurized beef ham. Meat Science 58: 403�407.

Huffman D.L. (1974). Effect of gas atmospheres on microbiologicalquality of pork. Journal of Food Science 39: 723�725.

Iacurto M., Settineri D., Mormile M. and Giacomo A.D. (2005).Effects of packaging techniques, freezing temperature and stor-age time on beef shelf life. Italian Journal of Animal Science 4(2):275�277.

Ke P.J., Ackman R.J., Linke B.H. and Nash D.M. (1977). Differentiallipid oxidation in various parts of frozen mackerel. Journal ofFood Technology 12: 37.

Kesava Rao V., Kowale B.N., Babu N.P. and Bisht G.S. (1996). Effectof cooking and storage on lipid oxidation and development ofcholesterol oxidation products in water buffalo meat. MeatScience 43: 179�185.

Lambert A.D., Smith J.P. and Doods K.L. (1992). Physical, chemicaland sensory changes in irradiated fresh pork packaged in modi-fied atmosphere. Journal of Food Science 57: 1294�1299.

Lucke F.K. (1995). Microbiological changes during storage and spoil-age of meat and meat products. In: Bauer F. and Burt S. (eds),

Shelf Life of Meat and Meat Products: Quality Aspects,Chemistry, Microbiology and Technology. Utrecht:ECCEAMST, pp. 57�74.

Luno M., Roncales P., Djenane D. and Beltran J.A. (2000). Beef shelflife in low O2 and high CO2 atmospheres containing different lowCO concentrations. Meat Science 55: 413�419.

Nissen H., Sorheim O. and Dainty R. (1996). Effects of vacuum, mod-ified atmospheres and storage temperature on the microbial floraof packaged beef. Food Microbiology 13: 183�191.

Ockerman H.W. (ed.) (1976). Microbiology. In: Quality Control ofPostmortem Muscle Tissue, Vol. 4, 10th edn. Ohio: Dept. ofAnimal Science, The Ohio State University and OhioAgricultural Research and Development Center, pp. 136�410.

Pexara E.S., Metaxopoulos J. and Drosinos E.H. (2002). Evaluationof shelf life of cured, cooked, sliced turkey fillets and cookedpork sausages-piroski stored under vacuum and modified atmo-spheres at 4 �C and 10 �C. Meat Science 62: 33�43.

Rubio B., Martınez B., Garcıa-Cachan M.D., Rovira J. and Jaime I.(2007). The effects of high pressure treatment and of storageperiods on the quality of vacuum-packed ‘salchichon’ made ofraw material enriched in monounsaturated and polyunsaturatedfatty acids. Innovative Food Science and Emerging Technologies 8:180�187.

Sawaya W.N., Elnawawy A.S., Abu-Ruwaida A.S., Khalafawi S. andDashti B. (1995). Influence of modified atmosphere packaging onshelf-life of chicken carcasses under refrigerated storage condi-tions. Journal of Food Safety 15: 35�51.

Sorheim O., Kropf D.H., Hunt M.C., Karwoski M.T. and WarrenK.E. (1996). Effects of modified gas atmosphere packaging onpork loin colour, display life and drip loss. Meat Science 43:203�212.

Soysal I. (1992). Biometrinin Temel Prensibleri (Principles of BiometricAnalysis). Tekirdag: Trakya Uni. Tekirdag Zir. Fak. Yay.

Speck M.L. (1976). Conpendium of Methods for the MicrobiologicalExamination of Foods. Washington D.C: American PublicHealth Association Inc.

Tarladgis B.G., Watts B.M. and Younathan M.T. (1960). A distilla-tion method for the quantitive determination of malonaldehydein rancid foods. Journal of the American Oil Chemists Society 37:44�48.

TSE (1992). TS (10581) Turkish Uncooked Meatball Standard. Ankara:Turk Standartlari Enstitusu.

Yilmaz I. (2001). Et ve urunlerinde modifiye atmosfer paketleme(MAP). Gıda 2000 Gıda Teknolojisi ve Tarım Dergisi. Sayı 25:34�40.

Yilmaz I. and Demirci M. (1995). Tekirdag koftesinin fiziksel, kimya-sal ve mikrobiyolojik ozelliklerinin belirlenmesi uzerine bir aras-tırma. T.U.Tekirdag Ziraat Fakultesi Dergisi 4: 17�29.

Yilmaz I. and Demirci M. (2009). Effect of different packaging meth-ods and storage temperature (�18 �C) on microbiological andphysicochemical quality characteristics of meatball. In: Yilmaz I.(ed.), Quality of Meat and Meat Products. Kerala, India:Transworld Research Network, pp. 137�148.

Yilmaz I., Yetim H. and Ockerman H.W. (2002). The effect of differ-ent cooking procedures on microbiological and chemical qualitycharacteristics of Tekirdag meatball. Nahrung 46: 276�278.

Zakrys P.I., Hogan S.A., O’Sullivan M.G., Allen P. and Kerry J.P.(2008). Effects of oxygen concentration on the sensory evaluationand quality indicators of beef muscle packed under modifiedatmosphere. Meat Science 79: 648�655.

Zhao Y., Wells J.H. and McMillin K.W. (1994). Applications ofdynamic modified atmosphere packaging systems for fresh redmeats: review. Journal of Muscle Foods 5: 299�328.







DOI: 10.1177/1082013210368884

2010 16: 266 originally published online 12 August 2010Food Science and Technology InternationalG. Cano-Sancho, S. Marin, A.J. Ramos and V. Sanchis

Biomonitoring of Fusarium spp. Mycotoxins: Perspectives for an Individual Exposure Assessment Tool

Published by:


On behalf of:



















Biomonitoring of Fusarium spp. Mycotoxins: Perspectives for

an Individual Exposure Assessment Tool

G. Cano-Sancho, S. Marin, A.J. Ramos and V. Sanchis*

Food Technology Department, University of Lleida, Rovira Roure 191, 25198 Lleida. Spain

Fusarium species are probably the most prevalent toxin-producing fungi of the northern temperate regionsand are commonly found on cereals grown in the temperate regions of America, Europe and Asia. Amongthe toxins formed by Fusarium we find trichothecenes of the A-type or B-type, zearalenone, fumonisins or

nivalenol. The current exposure assessment consists of the qualitative and/or quantitative evaluation basedon the knowledge of the mycotoxin occurrence in the food and the dietary habits of the population. Thisprocess permits quantifying the mycotoxin dietary intake through deterministic or probabilistic methods.Although these methods are suitable to assess the exposure of populations to contaminants and to identify

risk groups, they are not recommended to evaluate the individual exposition, due to a low accuracy andsensitivity. On the contrary, the use of biochemical indicators has been proposed as a suitable method toassess individual exposure to contaminants. In this work, several techniques to biomonitor the exposure to

fumonisins, deoxynivalenol, zearalenone or T-2 toxin have been reviewed.

Key Words: biomarkers, Fusarium, trichothecene, zearalenone, fumonisin

INTRODUCTION

Fusarium species are probably the most prevalenttoxin-producing fungi of the northern temperate regionsand are commonly found on cereals grown in the tem-perate regions of America, Europe and Asia (Creppy,2002). Among the toxins formed are trichothecenesof the A-type, such as T-2 toxin (T-2), HT-2 toxin(HT-2), T-2 triol, T-2 tetraol, neosolaniol (NEO),di- and 15-monoacetoxyscirpenol (DAS, MAS), scirpen-triol (SCIRP); trichothecenes of the B-type such asdeoxynivalenol (DON), 3- and 15-acetylDON (3- and15-ADON), nivalenol (NIV), fusarenon-X (FUS-X) aswell as zearalenone (ZEA) and fumonisins (FB)(Gelderblom et al., 1988; De Nijs et al., 1996; Glenn2007).

Fumonisins

Fumonisins (FB) are a group of mycotoxinsmainly produced by F. verticillioides and F. proliferatum,which usually contaminate corn. Among them, themost important are fumonisin B1 (FB1) and B2 (FB2)

(Nelson et al., 1992). FB occur mainly in maize

and maize-based foods, therefore populations with

high maize consumption can be exposed to signifi-cant amounts of these mycotoxins through the inges-

tion of fumonisin contaminated maize (Marasas,

1996; Shephard et al., 1996; Visconti et al., 1996;

WHO, 2001).Toxicity of FB has been widely reviewed by Soriano

et al. (2005), Voss et al. (2007), Stockmann-Juvala and

Savolainen (2008) and Wan Norhasima et al. (2009).Human exposure to fumonisin contaminated commodi-

ties has been linked to esophageal and liver cancer in

South Africa and China (Sydenham et al., 1991;

Yoshizawa et al., 1994). Acute and chronic toxicity of

FB has been largely demonstrated in several animal spe-

cies, including carcinogenicity and cardiovascular toxic

effects (Gelderblom et al., 1988, 1991). FB1 is a cancerpromoter but a poor cancer initiator. It is not genotoxic

because FB1 does not induce unscheduled DNA synthe-

sis in primary rat hepatocytes (Norred et al., 1992).

Based on toxicological evidence, the International

Agency for Research on Cancer (IARC) has classified

FB1 as possibly carcinogenic (group 2B) to humans

(IARC, 1993). The Joint FAO/WHO ExpertCommittee on Food Additives (JECFA) evaluated FB

and allocated a provisional maximum tolerable daily

intake (PMTDI) of 2 mg/kg body weight/day for FB on

the basis of the nonobserved effect level (NOEL) of

0.2 mg/kg body weight/day and a safety factor of 100

(WHO, 2001). This PMTDI for FB has recently beenconfirmed by the Scientific Committee on Food

(ECSCF; European Commission, 2003).

*To whom correspondence should be sent(e-mail: [email protected]).Received 13 October 2009; revised 22 December 2009.




Zearalenone

Zearalenone (ZEA) is a nonsteroidal mycotoxin pro-duced by Fusarium spp. that is found commonly in anumber of cereal crops, such as maize, barley, oats,wheat, rice and sorghum, being most frequently encoun-tered on corn (Kuiper-Goodman et al., 1987; Tanakaet al., 1988; Krska et al., 2003). ZEA and some ofits metabolites have shown to competitively bind to oes-trogen receptors. Thus, the toxicity is associated withreproductive problems in specific animals and possiblyin humans (Wood, 1992). Fertility problems havebeen observed in animals such as swine and sheep(Krska et al., 2003). ZEA may be an important etiologicagent of intoxication in infants or fetuses exposed to thismycotoxin, with results in premature thelarche, pub-arche and breast enlargement (CAST, 2003). Riskassessment of ZEA by the EC SCF concluded on aPTDI of 0.2mg/kg body weight whereas the tolera-ble daily intake (TDI) established by JECFA was0.5mg/kg body weight (EFSA, 2004).

Type-A Trichothecenes: T-2 and HT-2 Toxins

T-2 toxin and HT-2 toxin are type-A trichothecenemycotoxins produced by different Fusarium species,that may contaminate a variety of cereal grains, espe-cially in cold-climate regions or during wet storage con-ditions (Bottalico, 1998; WHO, 2001; SCOOP, 2003).They have been widely studied in animals, but despitetheir toxic effects, the toxicology has never been assessedin humans. T-2 is a potent inhibitor of protein synthesisand mitochondrial function both in vivo and in vitro andshows immunosuppressive and cytotoxic effects.Moreover, it has been reported that the toxin has extre-mely toxic effects on skin and mucous membranes(Visconti et al., 1991; Visconti, 2001; Sudakin, 2003;Eriksen and Pettersson, 2004). In poultry, T-2 toxinhas been reported to inhibit DNA, RNA and proteinsynthesis in eukaryotic cells, to affect the cell cycle andto induce apoptosis both in vivo and in vitro (Sokolovicet al., 2008). It has been shown that through deacetyla-tion of T-2, HT-2 is obtained as major metabolite; how-ever, little information is available regarding toxicity ofHT-2 alone (Visconti, 2001; Sudakin, 2003). JECFA,after assessing the toxic effect of both mycotoxins, hasconcluded that the toxic effects of T-2 and HT-2 couldnot be differentiated. Thus, the PMTDI for these toxins,combined or separately, was set at 0.06 mg/kg bodyweight/day (FAO, 2001).

Type-B Trichothecenes: Deoxynivalenol (vomitoxin)

The mycotoxin deoxynivalenol (DON) is a type-Btrichothecene, produced by molds of Fusarium genus,mainly F. graminearum or F. culmorum (Greenhalghet al., 1986) when grown on various cereals crops

(wheat, maize, barley, oat and rye). Although DON isnot as toxic as other trichothecenes such as T-2 toxin,HT-2 toxin or fusarenon-X, this mycotoxin is one of themost common contaminants of cereals worldwide(Jelinek et al., 1989; Scott, 1989; IARC, 1993). Uponingestion it can cause severe toxicosis in humans andfarm animals. Acute effects of food poisoning inhumans are abdominal pain, dizziness, headache,throat irritation, nausea, vomiting, diarrhoea andblood in the stool (Rotter et al., 1996). The TDI of1 mg/kg body weight based on a reduction of bodyweight gain (Iverson et al., 1995) was established bythe EC SCF (SCF, 2002).

BIOLOGICAL MARKERS AS ANEXPOSURE ASSESSMENT TOOLFOR MYCOTOXINS

Exposure assessment is the qualitative and/or quanti-tative evaluation of the likely intake of chemical agentsvia food as well as exposure from other sources if rele-vant (WHO, 1997). Thereby, through knowledge of themycotoxin occurrence in the food and dietary habits ofthe population, we can quantify the mycotoxin dietaryintake. To assess food consumption, four different typesof data can be used: food supply data, data from house-hold consumption surveys, data from dietary surveysamong individuals and the collection of duplicate diets(Hulshof and Lowik, 1998). The current exposureassessment schemes are largely deterministic and uncer-tainty and/or variability issues are accounted for bymeans of cautionary measures which are implicitlyembedded in calculation schemes and rules (Verdoncket al., 2005). More recently, probabilistic methods asMonte Carlo simulations have been developed to quan-tify the sources of uncertainty and variability of humanexposure (Verdonck et al., 2006).

Although these methods are suitable to assess theexposure of populations to contaminants and to identifyrisk groups, they are not recommended to evaluate theindividual exposition due to a low accuracy and sensi-tivity. The use of biochemical indicators has beenproposed as a suitable method to assess individualexposure to contaminants. The WHO defined in 1993,a biomarker as ‘any parameter that can be used to mea-sure an interaction between a biological system and anenvironmental agent, which may be chemical, physicalor biological’. This method allows effective exposureconsidering variability among food contaminationlevels, cooking effect, individual consumption, varia-tions in toxicokinetics or toxicodynamics (WHO, 1993;Paustenbach and Galbraith, 2006). Among the potentialvaluable application of biomarkers in epidemiologicstudies and in clinical trials, there is the possibility ofmeasuring them earlier than the observed true endpoint

Biomonitoring of Fusarium spp. Mycotoxins 267



of interest, given their property of relating the effect of

exposures or treatments on cellular and molecular

changes to the true endpoint/outcome (Merlo et al.,

2006).Interpretation of biomarkers of effect is hampered by

lack of knowledge on the metabolism of most non-nutri-

ents and their mechanisms of action in humans in vivo.

Before a biochemical indicator can be used as a measure

of dietary intake, it must be evaluated with respect to its

sensitivity to the intake of those contaminants. If these

indicators are to be used as measures of dietary expo-

sure, however, the epidemiologist is obviously responsi-

ble for ensuring that the exposure measure is a valid

representation of long-term intake. Several strategies

are available to define the relationships between long-

term dietary intake and biological levels: (i) animal stud-

ies; (ii) geographic correlation of intake and biological

marker; (iii) correlation with individual intake; (iv) die-

tary manipulation in humans and (v) repeated measures

(Walter, 1998).Regarding biomonitoring of mycotoxin intake, suc-

cessful results have been reported about biomarkers of

ochratoxin A and aflatoxins. Higher levels of ochratoxin

A have been found in blood samples from people with

kidney or urinary disorders than in healthy people,

showing good correlation among dietary intake and

blood levels of this toxin (Scott, 2005). The use of

serum aflatoxin B1-albumin adducts as biomarkers of

aflatoxin exposure has been validated in experimental

and human sample analyses (Wild et al., 1990a, b,

1992). The use of urinary aflatoxin B1-N7-guanine

adduct validated in the laboratory with human samples,

provides a measure of acute exposure to aflatoxin B1

(AFB1) and reflects a relatively short-term (24�48 h)exposure (Groopman et al., 1992a, b, 1993).

Another problem to conduct an accurate exposure

assessment is the presence of conjugated forms of

mycotoxins, known as ‘masked’ mycotoxins. More

important ‘masked’ mycotoxins have been reported to

be produced by Fusarium species. For example, more

common mycotoxin conjugation products in mammals

are glucuronides, as found in ZEA-4-glucoside and

DON-3-glucoside. These conjugated metabolites are

usually stable under extraction conditions, maintaining

the capability to produce toxic effects (Berthiller et al.,

2009). Unfortunately they cannot be detected through

routine analysis making necessary alternative methods.

Biomonitoring Exposure to Fumonisins

Methods to biomonitor the exposure to FB have been

reviewed previously in several cases (Turner et al., 1999;

Shephard et al., 2007). Mainly, two analytical proce-

dures have been reported as fumonisin biomarker:

fumonisin B1 and sphingoid bases ratios.

Fumonisin B1 as Biomarker

Absorption, distribution and excretion of FB1 havebeen widely studied in several animal species includingrats, laying hens, vervet monkeys, swine or piglets(Prelusky et al., 1994; Shephard et al., 1994a, b; Fodoret al., 2006, 2008) and FB2 toxicokinetics has been stud-ied in rats and vervet monkeys (Shephard et al., 1995;Shephard and Snijman, 1999). These toxicokineticsstudies have shown that FB1 had low oral bioavailabil-ity, with values ranging from 3% to 6% (Prelusky et al.,1994; Fodor et al., 2008) and short half-life when dosedintraperitoneally or intravenously. Half-life in rats hasbeen reported at 18min and 40min in vervet monkeysand estimated by regression analysis in 70 kg human as128min giving an animal’s weight to fit the prediction(Shephard et al., 1992; Delongchamp and Young, 2001).

As FB1 is mainly excreted in feces, HPLC withfluorescence detection method was initially devel-oped as suitable tool to exposure assessment to FB.Determination of FB1 in faeces was applied on exposureassessment of rural and urban populations from SouthAfrica with mean fumonisin levels in maize for con-sumption of 2.2 and 0.3mg/kg, respectively. Resultsshowed significant differences among FB1 concentra-tions in rural and urban feces (p¼ 0.014). Consideringfecal samples were taken 24 h after maize consumption,FB1 could be expected to be a suitable short-term bio-marker of this toxin exposure (Chelule et al., 2000).

Moreover, FB have been detected in human hairusing LC-MS-MS analytical method. Hair sampleswere obtained from South African population highlyexposed to fumonisin with probable daily intake for70 kg individuals of 13.8mg/kg body weight per day.Results showed that mean values of FB1 ranged from33.0 to 22.2mg/kg hair, with maximum values of93.5 mg/kg hair, concluding that human hair analysiscould be an useful tool to measure the cumulative expo-sure to FB (Sewram et al., 2003).

Urinary FB1 have been reported recently as a suffi-ciently sensitive tool to assess the human exposition toFB. A liquid chromatograph-mass spectrometry methodand extraction on Oasis MAZ cartridges was performedto determine urinary FB1. Urinary FB1 was correlatedwith maize intake (p¼ 0.001) and the correlationremained significant after adjusting for age, educationand place of residence (Gong et al., 2008).

Sphingoid Base Levels and Ratios in Plasma

Due to rapid elimination and low bioavailability ofFB, an indirect indicator of human exposition to thesetoxins has been required. FB have a remarkable struc-tural similarity to sphingolipids (Merrill et al., 1996;Riley et al., 2001). This group of mycotoxins, especiallyFB1 potently inhibits the enzyme ceramide (CER)synthase, which catalyzes the acylation of sphinganine

268 G. CANO-SANCHO ET AL.



and reacylation of sphingosine. The inhibition of CERsynthase by FB increases the intracellular sphinganineconcentration, process described as the main contributorto the toxicity and carcinogenicity of FB1 (Wang et al.,1991; Merrill et al., 1993; Yoo et al., 1996; Riley et al.,2001). Based on this biological perturbation, particu-larly elevation of sphinganine (Sa) to sphingosine (So)or Sa 1-phosphate to So 1-phosphate ratios in tissues,urine and blood, have been proposed as potential bio-markers of fumonisin exposure in various animalspecies (Wang et al., 1992; Riley et al., 1993; Morganet al., 1997; Wang et al., 1999; Van der Westhuizen et al.,2001; Kim et al., 2006; Tran et al., 2006; Cai et al., 2007).This biomarker was validated initially in Wistar rats bySolfrizzo et al. (1997) and recently in F344 rats by Caiet al. (2007), obtaining more sensitive results in urinethan in serum for acute and sub-chronic exposure toFB1. Furthermore, several studies have been conductedto assess the effectiveness of this biomarker on humanpopulation without successful results to obtain an accu-rate validation due to the low sensitivity when appliedto individuals (Van der Westhuizen et al., 1999, 2008;Abnet et al., 2001; Qiu and Liu 2001; Solfrizzo et al.,2004; Missmer et al., 2006; Nikiema et al., 2008).Van der Westhuizen et al. (1999) initially conducted a

study to assess Sa : So ratio in human plasma and urinefrom three different populations from Africa (Centane,n¼ 154; Bomet, n¼ 29 and KwaZulu-Natal, n¼ 27) withmean fumonisin intake of 3.8, 0.06mg/kg body weight/day and nondetected levels, respectively. Despite thesedifferences among exposures, nonsignificant differencesin Sa : So ratios were found, showing mean levels ofserum ratios of 0.34, 0.43 and 0.28 in Centane, Bometand KwaZulu-Natal population and urinary ratios of0.41 and 0.38 in Centane and Bomet.More recently, they conducted a cross-sectional study

in two areas from the same region of South Africa(Bizana, n¼ 150 and Centane, n¼ 152), concludingthat although significant and contrasting differences inplasma and urinary sphingoid base levels in the areaswere observed, there was no significant difference inthe mean total fumonisin levels in the maize consumed,mean plasmatic ratio and urinary ratio from Bizanapopulation (Van der Westhuizen et al., 2008).Croatia is a country located in the region affected by

endemic nephropathy (EN), chronic renal disease geo-graphically restricted to several European Eastern coun-tries. Ribar et al. (2001) conducted a study to determinethe possible modifications in the concentrations of uri-nary and serum Sa, So and Sa : So ratio of healthysubjects and EN patients from EN endemic area inCroatia. Eighty-nine serum samples and 30 urine sam-ples were obtained from men and women affected(n¼ 1), suspected (n¼ 7) or at risk (n¼ 12) to EN aswell as healthy (n¼ 27) and control from nonendemicarea (n¼ 20). Sphingolipids were extracted from serumand urine according to the method of Riley et al. (1994).

Results showed nonstatistically significant difference inthe serum Sa : So ratio in either men or women from theendemic area as compared with the control group of sub-jects. While urinary Sa : So ratio was found to differ sig-nificantly in the male group of healthy, suspected andaffected people, among women it differs significantly insubjects at risk and suspected to EN. The authors did notreport conclusions but suggested that study subjectscould be presumed to have been exposed to FB andsphingolipid metabolism impairment could be postu-lated as an early indicator of EN (Ribar et al., 2001).

Mexico is one of the most important countries regard-ing maize consumption. Human consumption is approx-imately 300 g/day providing 56% of the calories.Population of 38 Mexican volunteers (categorized withdifferent maize based food consumption level: high,medium and low) participated in a trial to determineurinary Sa : So ratio. Urine samples were collected atthree stages: A) at the beginning of the experimentwith normal diet, B) after two weeks without consump-tion of any type of maize based food and C) one weekafter the re-assumption of normal maize based food con-sumption. Urine samples were analyzed according tomethodology described by Solfrizzo et al. (1997).Results showed that there were no significant differencesamong the groups in the estimated mean fumonisinintake and the Sa : So ratio. Sa : So ratio was signifi-cantly higher during exposed stage A and C, with respec-tive mean fumonisin intake of 6 and 5.1mg/kg bw/daythan Sa : So ratio obtained during nonexposed period B(Landeros et al., 2005).

Other study was performed in China with 15 femalesand 13 males exposed to FB1 in corn diets over 1 monthto analyze So and Sa in human urine and monitor theSa : So ratio. The estimated daily FB1 intake was rangedbetween 0.4 and 457 mg/kg body weight/day in femalesand between 0.5 and 740 mg/kg body weight/day inmales. Urinary Sa : So ratio did not change over themonth in females (0.2 initially and 0.18 at the end),while mean urinary ratio increased from 0.11 to 0.21in males. However, it could be ascribed to a singleparticipant with a high value, as they reported (Qiuand Liu, 2001).

Solfrizzo et al. (2004) assessed urinary sphingoid basesof population from northern Argentina (n¼ 74) andsouthern Brazil (n¼ 116) as exposed population withmean fumonisin intake of 0.56 g mg/kg body weight/day and urinary sphingoid bases of population fromsouthern Italy (n¼ 66) and central Argentina (n¼ 20)with low or no fumonisin exposure (control group).Mean Sa : So ratio in regions with exposure to FB was1.24, significantly higher than regions without exposure,where the Sa : So ratio was 0.36. However, mean Sa : Soratio from northern Argentina was 0.69, not signifi-cantly different from the control population and signif-icantly lower than the value 1.57 showed in the southernBrazil population. Therefore, the highest value obtained




in southern Brazil cannot be associated to fumonisinexposure, existing with other confounding factors.

Moreover, the ratios have been assessed in simulta-neous matrices as buccal cells, urine and serum in pop-ulation from Burkina Faso, without showing anyassociation between urinary Sa : So ratios and fumonisinintake, but suggesting a positive trend between fumoni-sin intake and Sa : So ratios in serum (Nikiema et al.,2008).

Latest study was performed to assess Sa : So ratio andfrequency of detection in urine samples from urban andrural population from Portugal. A total of 68 urine sam-ples were collected from male and female adult healthyvolunteers from urban (n¼ 38) and rural (n¼ 30) zone.Optimized extraction method, based on the proceduresdescribed by Castegnaro et al. (1996, 1998) and Qiu andLiu (2001), followed by derivatization with naphtalene-2,3-dicarboxyaldehyde (NDA) and injection to HPLC-FD system, was carried out to detect and quantifyurinary Sa and So. Sa : So ratio was between 0.11 and0.95, with a mean value of 0.43±0.22. Significant dif-ferences were not found when the results of Sa, So andSa : So ratio of males, females as well as combined(males and females together) were compared betweenrural and urban population (Silva et al., 2009).

In our latest study (unpublished data), performedwith two exposure groups from the same region(exposed and nonexposed group from Catalonia,Spain), significant differences were observed amongmean plasma Sa : So ratios. Results showed significantdifferences in sphingosine levels in groups consideringboth sexes combined or among males (p< 0.05), whileno significant differences were observed in femalesbetween groups (p> 0.05). Thus these results suggestthat the decrease of the ratios could be due to a decreaseof sphingosine level, as should be expected according tothe mechanism of action. However, wide ranges ofSa : So ratios and bad correlation coefficients wereobserved when linear regression was fitted, which sug-gests that this biomarker is low sensitive and impreciseto apply over individuals.

Further studies are required to better understand allphysiological factors that lead to Sa : So ratios variationsas reported by Abnet et al. (2001) as well as biochemicalprocesses that can modify sphingoid metabolism asextensive cell death, metabolization by other bioactivemolecules or alteration by other components of the diet(Merrill et al., 2001). Other main problem is the lack ofinformation about sphingoid bases basal levels in tis-sues, urine and blood of healthy human population.

Biomonitoring Exposure to DON

Absorption, distribution, accumulation, metabolismand elimination of DON have been reported in a widerange of animal species. Toxicodynamic studies haveshown low absorption in poultry (<1%), in sheep

ranged 6�10% and at least 29% was absorbed bydairy cows when DON toxin was administered; on theother hand, high absorption has been estimated in swine(47�82%). Plasma elimination of DON tended to beslower in pigs, taking approximately 7 times longerthan sheep, 2 times longer than cow and slightlylonger than laying hens to clear the toxin after a singleoral dose. Numerous studies have reported that swine isvery sensitive to DON in contaminated feedstuffs. Thedistribution characteristics of DON in swine are alsodifferent than in other species. Only a small proportionof the dose can be found in the blood, although the toxinis extensively absorbed (Yoshizawa et al., 1981; Preluskyet al., 1985, 1986, 1988, 1994; Friend et al., 1986).

DON and DON-glucuronide excretion in the urinerepresented 37% and 50% of the ingested DON inrats (Meky et al., 2003) and swine respectively(Goyarts and Danicke, 2006). Goyarts and Danicke(2006) have confirmed that not all animals are able todetoxify DON to the metabolite de-epoxy-DON andthat this metabolism occurs principally in the large intes-tine, where unlikely absorption proceeds. Furthermore,it was shown that quantitative urinary recovery ofDON can be considered as an indicator for its systemicabsorption, as it approximates the bioavailability as esti-mated by the kinetic study. Assuming a high compara-bility of digestion and excretion in humans and swine,they concluded that although DON is poorly detoxified,it is rapidly excreted and so is not found in remarkableconcentrations in serum after 24 h.

Regarding human population, an earlier study wasperformed to develop, to validate and to measure uri-nary concentrations of DON and its metabolites in15 females from Henan (Linxian) region, where thestaple diet was based on corn and wheat (high-riskregion of esophageal cancer, n¼ 11), or rice (low-riskregion of esophageal cancer, n¼ 4). The mean levels ofDON detected in the samples from high-risk and low-risk areas were 37 ng/mL and 12 ng/mL, respectively.Through these values and specific assumptions regard-ing excretion, urine production and recoveries, theauthors estimated a daily exposure ranged from 1.9 to13.0 and 0.6�2.5mg/kg/day for high- and low-risk pop-ulation respectively, in the line of previous studies thathave been estimated this exposure through classic meth-ods (Meky et al., 2003).

Urinary DON was widely surveyed in a large-scalestudy conducted in UK and compared with cereal-based food intake (Turner et al., 2008a). Three popula-tion groups were selected according to low, medium orhigh cereal intake, estimated previously through 7-dayweighted food diary. From each group, 100 individualswere selected and urinary samples were collected duringthe period on the basis of available data in the 7-daydiary that was provided. DON was detected in 296 of300 (98.7%) urine sample, with geometric mean of9.42 mg DON/day (nd-65.97 mg/day). Cereal intake was




significantly associated with urinary DON (p< 0.0005),showing mean levels of 6.55, 9.63 and 13.24 mg/day inlow, medium and high cereal intake groups, respectively.The food groups associated with urinary DON were pre-dominantly wheat based, particularly the three mainbread groups (white, wholemeal and other bread). Acrude estimation was made based on: (i) the amount ofurinary DON, (ii) an assumption that 50% of theingested DON was being excreted in the urine (Mekyet al., 2003; Goyarts and Danicke, 2006) and (iii) theurinary DON originated from DON intake in the previ-ous 24 h. For the 300 individuals the mean intake wasestimated as 319 ng/kg body weight/day, below TDI forDON ingestion of 1 mg/kg body weight (SCF, 2002) andslightly higher than previous estimation that showedDON daily intakes of 176 and 142 ng/kg body weight/day for males and females respectively (SCOOP, 2003).Briefly, more detailed analysis of these data will be pub-lished. In this report, food diary information (n¼ 255)for (a) the day of urine collection, (b) the previous 24-hperiod and (c) the day of urine collection plus the pre-vious 24 h combined, were further examined to assesswhether the recent intake of cereal correlated morestrongly with urinary DON, compared with (d) thelonger term assessment of usual cereal intake from7-day food diaries. Results suggest that the inter-indivi-dual variation in urinary DON was somewhat betterexplained by recent cereal intake compared with usualcereal intake assessed over 7 day (Turner et al., 2009).An intervention study was conducted to assess the

effect of wheat-restricted diet over DON urinary levels.The study was performed with 25 healthy adult volun-teers and involved 2 days of normal diet and 4 days of awheat-restricted diet. Food diaries were kept for normaldiet days and for the two latest days of intervention diet.Initial morning urinary samples were collected the fol-lowing day of each period. Samples were analyzed andadjusted using a creatinine concentration in mg/mL ofurine and subsequent data were expressed as ng DON/mg creatinine. Results showed that during interventiondiet period there was a low percentage of detected sam-ples (36%) while during normal diet all samples haddetectable levels of DON. Mean levels and ranges were1.0 (nd-8.4) and 10.8 (0.7�61.3) ng/mL for interventionand normal diet respectively � results in the line of pre-vious study conducted in UK (Turner et al., 2008b).Recently, another intervention trial conducted with 22

urine samples from UK volunteers to correlate urinaryDON level with one or more metabolite in the urine waspublished. A1H-Nuclear Magnetic Resonance-basedchemometrics approach (metabolomics) was utilized toexamine samples from individuals eating a normal diet.Urinary DON was determined using an in-house immu-noaffinity-LC-MS assay (Turner et al., 2008b). Modelwas built on 16 individuals, eight with low urinary DONand eight with high urinary DON level; and validatedwith a further six urine samples, of which there were

three in each category of DON level. Through statisticalanalysis two possible biomarkers were identified: hippu-rate and mannitol, the first one being the more interest-ing candidate (Hopton et al., 2010).

Turner et al. (2008c) emphasized urinary DON as agood tool to assess exposure to this contaminant at theindividual level. In contrast, they reported several uncer-tainties to resolve the full validating of this biomarker toapply in epidemiological studies. Their questions were:(a) what is the relationship at the individual levelbetween DON intake and the urinary biomarker?; (b)what are the pharmacokinetics of DON and DON-glu-curonide excretion and what are the consequent tempo-ral variations in this biomarker?; (c) Does the ratioof DON to DON-glucuronide in urine vary by individ-ual? Therefore, they concluded that in humans, suchstudies require validated methods of exposure assess-ment to compare exposure to toxins both individuallyand in combination with health outcomes (Turner et al.,2008c).

Biomonitoring the Exposure to ZEA

Absorption of ZEA has been reported as extensiveand quick in rats and rabbits (Kuiper-Goodman et al.,1987; Ramos et al., 1996), being estimated in the rangeof 80-85% in pigs (Biehl et al., 1993). In mammals,ZEA is mainly metabolized into a-zearalenol (a-ZEA)and b-zearalenol (b-ZEA), while the first is the mostpredominant in pigs, the second is the most predomi-nant metabolite in cows (Jodlbauer et al., 2000;Kleinova et al., 2002; Zollner et al., 2002). Earlier stud-ies of Ueno et al. (1983) showed that there are two typesof ZEA reductase differing in optimum pH. In humansas in pigs, ZEA probably can be absorbed after oraladministration and can be metabolized in intestinalcells into a-ZEA and b-ZEA and would be excreted sig-nificantly in bile and urine (Doll et al., 2003).

In a previous study conducted with one male volunteer,100mg of ZEA were administered and a-ZEA and b-ZEAconcentrations were determined in the urine at 6, 12 and24h after the administration. The respective concentra-tions of ZEA, a-ZEA and b-ZEA were 3.7, 3mg/mL andnot detected after 6h; 6.9, 6 and 2.7mg/mL after 12 h; and2.7, 4 and 2 mg/mL after 24 h (Mirocha et al., 1981).

Furthermore, ZEA and its metabolites were studied inserum from 32 girls affected by central precocious pub-erty (CPP) and in 31 healthy female. Results showedincreased serum levels of ZEA and a-ZEA in 6 girlswith CPP. ZEA levels correlated with patient heightand weight. The authors concluded that ZEA is sus-pected to be a triggering factor for CPP developmentin girls and may also represent a growth promoter inexposed patients (Massart et al., 2008).

ZEA dietary intake was estimated by JECFA toEuropean region, reporting ranges of 0.004�0.029 and0.006�0.055mg/kg body weight per day for adults and




infants respectively (CAST, 2003). Despite the high con-sumption of cereals in European countries, few studieshave been conducted to assess the exposure to thismycotoxin, neither through conventional method norbiomarkers. Thus more studies are required to accu-rately characterize the risk of this endocrine disruptorand confirm it as a dangerous problem for human health(Minervini et al., 2005).

Biomonitoring the Exposure to T-2 and HT-2 Toxins

T-2 toxin is more rapidly absorbed than DON after itsingestion by most species, its plasmatic half-life beingless than 20min. The fraction of T-2 toxin eliminatedas parent compound in the urine was showed as negli-gible. In spite of administration of a lethal oral dose inswine (2.4mg/kg) and toxic oral doses (up to 3.6mg/kg)in calves, no parent T-2 toxin was detected in plasma orurine (Beasley et al., 1986; Larsen et al., 2004). T-2 toxincan be detected in pig blood before 30min after theiringestion (Eriksen et al., 2004). The main reactions intrichothecene metabolism are hydrolysis, hydroxylationand deep oxidation. Typical metabolites of T-2 toxinin an organism are HT-2 toxin, T-2-triol, T-2-tetraol,3-pm-hydroxy-T-2 and 30-hydroxy-HT-2 toxin. Thereare significant differences in the metabolic pathwaysof T-2 toxin between ruminants and nonruminants.Ruminants are more resistant to the adverse effects ofT-2 toxin due to microbial degradation within rumenmicroorganisms (Dohnal et al., 2008). The patterns ofdistribution and excretion suggest that T-2 toxin and/orits metabolites are excreted into the intestine through thebile and that the liver is a major organ for excretion ofthe toxin (Chi et al., 1978). No studies have been con-ducted with humans until now to assess the presence ofthis toxin or its metabolites in biological fluids.

CONCLUSIONS

Conventional methods of exposure assessment arebased on the combination of food analysis data withdietary intake data. That combination of data can bedeterministic or probabilistic; however, both caseshave been proven to be limited, due to low sensitivityand accuracy. A useful method to assess effective expo-sition of human populations to contaminants could bethrough the study of the effect on biological moleculesor monitoring these toxins directly on biological fluids.These biological markers, known as biomarkers, allowthe assessment of exposure of human populations tomycotoxins considering the variability within dietaryintake, cooking effect, intestinal absorption, metabo-lism or distribution over individuals. Thus, the under-standing of mechanism of action, toxicokinetics and

toxicodynamics of the mycotoxins, is required todevelop useful biomarkers.

Successful methods have been developed to biomoni-tor exposure to ochratoxin A and aflatoxins. However,few studies and unsuccessful results have been obtainedwith biomarkers of Fusarium toxins exposure. Urinarylevels of FB have been reported as an effective methodto assess short-term intake of this toxin. Despite theratio Sa : So has been validated as biomarker of fumo-nisin exposure in animal species, unsuccessful resultshave been reported among human populations.Further studies are required to understand accuratelybasal levels of these contaminants, interactions withother contaminants or variability sources.

Regarding DON, several studies have reported dataon absorption, toxicokinetics, toxicodynamics andmetabolism in animals, but few studies have been con-ducted in human populations. Urinary level of DON hasbeen used as biomarker to assess the exposure of humanpopulations showing successful results, with positivecorrelations among estimated dietary intake of thetoxin and urinary levels. Inspite of the interest ofresearchers in ZEA contamination of food and its tox-icity in animal species, very few studies have been con-ducted to assess the real impact on human population.Finally, no studies have been conducted to assess possi-ble biomarkers to assess the exposition to T-2 toxin orHT-2 toxin.

ACKNOWLEDGMENTS

The authors would like to acknowledge ExposureAssessment of Spanish Population to FusariumToxins Project, National Plan of Spanish Government(AGL2008-05030-C02-01), Catalonian Food SafetyAgency of Generalitat de Catalunya HealthDepartment and University of Lleida for their financialsupport.

REFERENCES

Abnet C.C., Borkowf C.B., Qiao Y.-L., Albert P.S., Wang E., MerrillA.H., Mark S.D., Dong Z.-W., Taylor P.R. and Dawsey S.M.(2001). A cross-sectional study of human serum sphingolipids,diet and physiologic parameters. Journal of Nutrition 131:2748�2752.

Beasley V.R., Swanson S.P. and Corley R.A. (1986). Pharmacokineticsof the trichothecene mycotoxin, T-2 toxin, in swine and cattle.Toxicon 24: 13�23.

Berthiller F., Schuhmacher R., Adam G. and Krska R. (2009).Formation, determination and significance of masked andother conjugated mycotoxins. Analytical and BioanalyticalChemistry 395: 1243�1252.

Biehl M.L., Prelusky D.B., Koritz G.D., Hartin K.E., Buck W.B. andTrenholm H.L. (1993). Biliary excretion and enterohepaticcycling of zearalenone in immature pigs. Toxicology andApplied Pharmacology 121: 152�159.




Bottalico A. (1998). Fusarium diseases of cereals: species complex andrelated mycotoxin profiles, in Europe. Journal of Plant Pathology80: 85�103.

Cai Q., Tang L. and Wang J. (2007). Validation of fumonisin bio-markers in F344 rats. Toxicology and Applied Pharmacology225: 28�39.

CAST. (2003). Mycotoxins: risks in plant, animal and human system.Task Force Report Ames, IA No. 139, pp. 48�58.

Castegnaro M., Garren L., Galendo D., Gelderblom W.C.A., CheluleP., Dutton M.F. and Wild C.P. (1998). Analytical method for thedetermination of sphinganine and sphingosine in serum as apotential biomarker for fumonisin exposure. Journal ofChromatography Part B 720: 15�24.

Castegnaro M., Garren L., Gaucher I. and Wild C.P. (1996).Development of a new method for the analysis of sphinganineand sphingosine in urine and tissues. Natural Toxins 4: 284�290.

Chelule P.K., Gqaleni N., Chuturgoon A.A. and Dutton M.F. (2000).The determination of fumonisin B1 in human faeces: a short termmarker for assessment of exposure. Biomarkers 5: 1�8.

Chi M.S., Robison T.S. and Mirocha C.J. (1978). Excretion andtissue distribution of radioactivity from tritium-labeled T-2toxin in chicks. Toxicology and Applied Pharmacology 45:391�402.

Creppy E.E. (2002). Update of survey, regulation and toxic effects ofmycotoxins in Europe. Toxicology Letters 127: 19�28.

Delongchamp R.R. and Young J.F. (2001). Tissue sphinganine as abiomarker of fumonisin-induced apoptosis. Food Additives andContaminants 18: 255�261.

De Nijs M., Rombouts F. and Notermans S. (1996). Fusarium moldsand their mycotoxins. Journal of Food Safety 16: 15�58.

Dohnal V., Jezkova A., Jun D. and Kuca K. (2008). Metabolic path-ways of T-2 toxin. Current Drug Metabolism 9: 77�82.

Doll S., Danicke S. and Schnurrbusch U. (2003). The effect of increas-ing concentrations of Fusarium toxins in the diets of pigletson histological parameters of uterus. Mycotoxin Research 19:73�76.

Eriksen G.S. and Pettersson H. (2004). Toxicological evaluation oftrichothecenes in animal feed. Animal Feed Science andTechnology 114: 205�239.

European Commission: Scientific Committee for Food (2003).Updated opinion of the Scientific Committee on Food onFumonisin B1, B2 and B3. Available at http://ec.europa.eu/food/fs/sc/scf/out185_en.pdf (accessed May 20, 2010).

European Food Safety Authority (EFSA) (2004). Opinion of the sci-entific panel on contaminants in the food chain on a request fromthe commission related to zearalenone as undesirable substancein animal feed. The EFSA Journal 73: 1�41.

Fodor J., Balogh K., Weber M., Mezes M., Kametler L., Posa R.,Mamet R., Bauer J., Horn P., Kovacs F. and Kovacs M.(2008). Absorption, distribution and elimination of fumonisinB1 metabolites in weaned piglets. Food Additives andContaminants Part A 25: 88�96.

Fodor J., Meyer K., Riedlberger M., Bauer J., Horn P., Kovacs F. andKovacs M. (2006). Distribution and elimination of fumonisinanalogues in weaned piglets after oral administration ofFusarium verticillioides fungal culture. Food Additives andContaminants 23: 492�501.

Food and Agriculture Organization (FAO) (2001) Safety Evaluationof Certain Mycotoxins in Food. In: The 56th Meeting of the JointFAO/WHO Expert Committee on Food Additives (JECFA),Geneva, Switzerland, p. 557.

Friend D.W., Trenholm H.L., Thompson B.K., Prelusky D.B. andHartin K.E. (1986). Effect of (DON)-contaminated diet fed togrowing-finishing pigs on their performance at market weight,nitrogen retention and DON excretion. Canadian Journal ofAnimal Science 66: 1075�1085.

Gelderblom W.C.A., Jaskiewicz K., Marasas W.F.O., Thiel P.G.,Horak R.M., Vleggaar R. and Kriek N.P.J. (1988).Fumonisins - novel mycotoxins with cancer-promoting activityproduced by Fusarium moniliforme. Applied and EnvironmentalMicrobiology 54: 1806�1811.

Gelderblom W.C.A., Kriek N.P.J., Marasas W.F.O. and Thiel P.G.(1991). Toxicity and carcinogenecity of the Fusarium moniliformemetabolite, fumonisin B1 in rats. Carcinogenesis 12: 1247�1251.

Glenn A.E. (2007). Mycotoxigenic Fusarium species in animal feed.Animal Feed Science and Technology 137: 213�240.

Gong Y.Y., Torres-Sanchez L., Lopez-Carrillo L., Peng J.H., SutcliffeA.E., White K.L., Humpf H., Turner P.C. and Wild C.P. (2008).Association between tortilla consumption and human urinaryfumonisin B1 levels in a Mexican population. CancerEpidemiology, Biomarkers and Prevention 17: 688�694.

Goyarts T. and Danicke S. (2006). Bioavailability of the Fusariumtoxin deoxynivalenol (DON) from naturally contaminatedwheat for the pig. Toxicology Letters 163: 171�182.

Greenhalgh R., Levandier D., Adams W., Miller J.D., Blackwell B.A.,McAlees A.J. and Taylor A. (1986). Production and characteri-zation of deoxynivalenol and other secondary metabolites ofFusarium culmorum (CMI 14764, HLX 1503). Journal ofAgricultural and Food Chemistry 34: 98�102.

Groopman J.D., Hasler J.A., Trudel L.J., Pikul A., Donahue P.R. andWogan G.N. (1992a). Molecular dosimetry in rat urine of afla-toxin-N7-guanine and other aflatoxin metabolites by multiplemonoclonal antibody affinity chromatography and immunoaffi-nity/high performance liquid chromatography. Cancer Research52: 267�274.

Groopman J.D., Jiaqi Z., Donahue P.R., Pikul A., Lisheng Z., Jun-shiC. and Wogan G.N. (1992b). Molecular dosimetry of urinaryaflatoxin-DNA adducts in people living in GuangxiAutonomous Region, People’s Republic of China. CancerReserarch 52: 45�52.

Groopman J.D., Wild C.P., Hasler J., Junshi C., Wogan G.N. andKensler T.W. (1993). Molecular epidemiology of aflatoxin expo-sures: validation of aflatoxin-N7-guanine levels in urine as a bio-marker in experimental rat models and humans. EnvironmentalHealth Perspectives 99: 107�113.

Hopton R.P., Turner E., Burley V.J., Turner P.C. and Fisher J. (2010).Urine metabolite analysis as a function of deoxynivalenol expo-sure: an NMR-based metabolomics investigation. Food Additivesand Contaminants Part A 27: 255�261.

Hulshof K.F.A.M. and Lowik M.R.H. Nutritional surveillance inindustrialised countries In: Sader M.J., Strain J.J. andCaballero B. (eds), Encyclopedia of Human Nutrition, London:Academic Press, pp. 1413�1422.

International Agency for Research on Cancer (IARC) (1993). Somenaturally occurring substances: food items and constituents, het-erocyclic aromatic amines and mycotoxins. IARCMonographs onthe Evaluation of Carcinogenic Risks to Humans 56: 397�444.

Iverson F., Armstrong C., Nera E., Truelove J., Fernie S., Scott P.,Stapley R., Hayward S. and Gunner S. (1995). Chronic feedingstudy of deoxynivalenol in B6C3F1 male and female mice.Teratogenesis, Carcinogenesis and Mutagenesis 15: 283�306.

Jelinek C.F., Pohland A.E. and Wood G.E. (1989). Worldwide occur-rence of mycotoxins in foods and feeds-an update. JournalAssociation of Official Analytical Chemists 72: 223�230.

Jodlbauer J., Zollner P. and Lindner W. (2000). Determination ofzeranol, taleranol, zearalenone, a- and b-zearalenol in urineand tissue by high-performance liquid chromatography-tandemmass spectrometry. Chromatographia 51: 681�687.

Kim D.-H., Yoo H.-S., Lee Y.-M., Kie J.-H., Jang S. and Oh S.(2006). Elevation of sphinganine 1-phosphate as a predictive bio-marker for fumonisin exposure and toxicity in mice. Journal ofToxicology and Environmental Health Part A 69: 2071�2082.

Kleinova M., Zollner P., Kahlbacher H., Hochsteiner W. and LindnerW. (2002).Metabolic profiles of the mycotoxin zearalenone and ofthe growth promoter zeranol in urine, liver and muscle of heifers.Journal of Agricultural and Food Chemistry 50: 4769�4776.

Krska R., Pettersson H., Josephs R.D., Lemmens M., Mac Donald S.and Welzig E. (2003). Zearalenone in maize: stability testing andmatrix characterization of a certified reference material. FoodAdditives and Contaminants 20: 1141�1152.

Kuiper-Goodman T., Scott P.M. and Watanabe H. (1987). Riskassessment of the mycotoxin zearalenone. RegulatoryToxicology and Pharmacology 7: 253�306.




Landeros P., Reyes W., Torres A.M., Rojo F. and Chulze S. (2005).Sphinganine/Sphingosine (Sa/So) ratio and intake of fumonisincontaminated tortillas in a Mexican population. ActaToxicologica Argentina 13: 8�11.

Larsen J.C., Hunt J., Perrin I. and Ruckenbauer P. (2004). Workshopon trichothecenes with a focus on DON: summary report.Toxicology Letters 153: 1�22.

Marasas W.F.O. (1996). Fumonisins: history, world-wide occurrenceand impact. Advances in Experimental Medicine and Biology 392:1�17.

Massart F., Meucci V., Saggese G. and Soldani G. (2008). Highgrowth rate of girls with precocious puberty exposed to estro-genic mycotoxins. Journal of Pediatrics 152: 690�669.

Meky F.A., Turner P.C., Ashcroft A.E., Miller J.D., Qiao Y.L., RothM.J. and Wild C.P. (2003). Development of a urinary biomarkerof human exposure to deoxynivalenol. Food and ChemicalToxicology 41: 265�273.

Merlo D.F., Sormani M.P. and Bruzzi P. (2006). Molecular epidemi-ology: new rules for new tools? Mutation Research: Molecularand Fundamental Mechanisms 600: 3�11.

Merrill A.H., Sullards M.C., Wang E., Voss K.A. and Riley R.T.(2001). Sphingolipid metabolism: roles in signal transductionand disruption by fumonisins. Environmental HealthPerspectives 109: 283�289.

Merrill A.H., Wang E., Gilchrist D.G. and Riley R.T. (1993).Fumonisins and other inhibitors of de novo sphingolipid biosyn-thesis. Advances in Lipid Research 26: 215�234.

Merrill A.H., Wang E., Vales T.R., Smith E.R., Schroeder J.J.,Menaldino D.S., Alexander C., Crane H.M., Xia J., LiottaD.C., Meredith F.I. and Riley R.T. (1996). Fumonisin toxicityand sphingolipid biosynthesis. Advances in ExperimentalMedicine and Biology 392: 297�306.

Minervini F., Giannoccaro A., Cavallini A. and Visconti A. (2005).Investigations on cellular proliferation induced by zearalenoneand its derivatives in relation to the estrogenic parameters.Toxicology Letters 159: 272�283.

Mirocha C.J., Pathre S.V. and Robison T.S. (1981). Comparativemetabolism of zearalenone and transmission into bovine milk.Food and Cosmetics Toxicology 19: 25�30.

Missmer S.A., Suarez L., Felkner M., Wang E., Merrill A.H.,Rothman K.J. and Hendricks K.A. (2006). Exposure to fumoni-sins and the occurrence of neural tube defects along theTexas-Mexico border. Environmental Health Perspectives 114:237�241.

Morgan M.K., Schroeder J.J., Rottinghaus G.E., Powell D.C.,Bursian S.J. and Aulerich R.J. (1997). Dietary fumonisins disruptsphingolipid metabolism in mink and increase the free sphinga-nine to sphingosine ratio in urine but not in hair. Veterinary andHuman Toxicology 39: 334�336.

Nelson P.E., Plattner R.D., Shackelford D.D. and Desjardins A.E.(1992). Fumonisin B1 production by Fusarium species otherthan F. moniliforme in section Liseola and by some related spe-cies. Applied and Environmental Microbiology 58: 984�989.

Nikiema P.A., Worrilow L., Traore A.S., Wild C.P. and Turner P.C.(2008). Fumonisin exposure and the sphinganine/sphingosineratio in urine, serum and buccal cells in adults from BurkinaFaso, West Africa. World Mycotoxin Journal 1: 483�491.

Norred W.P., Plattner R.D., Vesonder R.F., Bacon C.W. and VossK.A. (1992). Effects of selected secondary metabolites ofFusarium moniliforme on unscheduled synthesis of DNA by ratprimary hepatocytes. Food and Chemical Toxicology 30: 233�237.

Paustenbach D. and Galbraith D. (2006). Biomonitoring and bio-markers: exposure assessment will never be the same.Environmental Health Perspectives 114: 1143�1149.

Prelusky D.B., Hamilton R.M.G., Trenholm H.L. and Miller J.D.(1986). Tissue distribution and excretion of radioactivity follow-ing administration of 14C-labeled deoxynivalenol to whiteLeghorn hens. Fundamental of Applied Toxicology 7: 635�645.

Prelusky D.B., Hartin K.E., Trenholm H.L. and Miller J.D. (1988).Pharmacokinetic fate of 14C-labeled deoxynivalenol in swine.Fundamental of Applied Toxicology 10: 276�286.

Prelusky D.B., Trenholm H.L. and Savard M.E. (1994).Pharmacokinetic fate of 14C-labelled fumonisin B1 in swine.Natural Toxins 2: 73�80.

Prelusky D.B., Veira D.M. and Trenholm H.L. (1985). Plasma phar-macokinetics of the mycotoxin deoxynivalenol following oral andintravenous administration to sheep. Journal of EnvironmentalScience and Health Part B 20: 603�624.

Qiu M.F. and Liu X.M. (2001). Determination of sphinganine, sphin-gosine and Sa/So ratio in urine of humans exposed to dietaryfumonisin B1. Food Additives and Contaminants 18: 263�269.

Ramos A.J., Hernandez E., Pla-Delfina J.M. and Merino M. (1996).Intestinal absorption of zearalenone and in vitro study of non-nutritive sorbent materials. International Journal ofPharmaceutics 128: 129�137.

Ribar S., MesariC M. and Bauman M. (2001). High-performanceliquid chromatographic determination of sphinganine and sphin-gosine in serum and urine of subjects from an endemic nephrop-athy area in Croatia. Journal of Chromatography Part B 754:511�519.

Riley R.T., An N.H., Showker J.L., Yoo H.S., Norred W.P.,Chamberlain W.J., Wang E., Merrill Jr A.H., Motelin G.,Beasley V.R. and Haschek W.M. (1993). Alteration of tissueand serum sphinganine to sphingosine ratio: an early biomarkerof exposure to fumonisin-containing feeds in pigs. Toxicology andApplied Pharmacology 118: 105�112.

Riley R.T., Enongene E., Voss K.A., Norred W.P., Meredith F.I.,Sharma R.P., Spitsbergen J., Williams D.E., Carlson D.B. andMerrill A.H. (2001). Sphingolipid perturbations as mechanismsfor fumonisin carcinogenesis. Environmental Health Perspectives109: 301�308.

Riley R.T., Wang E. and Merrill A.H. (1994). Liquid chromato-graphic determination of sphinganine and sphingosine: use ofthe free sphinganine to sphingosine ratio as biomarker for con-sumption of fumonisins. Journal of AOAC International 77:533�540.

Rotter B.A., Prelusky D.B. and Pestka J.J. (1996). Toxicology ofdeoxynivalenol (vomitoxin). Journal of Toxicology andEnvironmental Health Part A 48: 1�34.

Scientific Committee on Food (SCF) (2002). Opinion of the ScientificCommittee on Food on Fusarium toxins. Part 6: Group evalua-tion of T-2 toxin, HT-2 toxin, nivalenol and deoxynivalenol.SCF/CS/CNTM/MYC/27 Final Report.

Scientific Cooperation (SCOOP) (2003). Collection of occurrence dataof Fusarium toxins in food and assessment of dietary intake bythe population of EU member states. SCOOP Task 3.2.10 FinalReport. Available at: http://ec.europa.eu/food/fs/scoop/task3210.pdf (accessed May 20, 2010).

Scott P.M. (1989). The natural occurrence of trichothecenes.In: Beasley V.R. (ed.), Trichothecene Mycotoxicosis:Pathophysiologic Effects, Boca Raton: CRC Press FL, pp. 1�26.

Scott P.M. (2005). Biomarkers of human exposure to ochratoxin A.Food Additives and Contaminants 22: 99�107.

Sewram V., Mshicileli N., Shephard G.S. and Marasas W.F.O. (2003).Fumonisin mycotoxins in human hair. Biomarkers 8: 110�118.

Shephard G.S. and Snijman P.W. (1999). Elimination and excretion ofa single dose of the mycotoxin fumonisin B2 in a non-humanprimate. Food and Chemical Toxicology 37: 111�116.

Shephard G.S., Thiel P.G., Stockenstrom S. and Sydenham E.W.(1996). Worldwide survey of fumonisin contamination of cornand corn-based products. Journal of AOAC International 79:671�687.

Shephard G.S., Thiel P.G. and Sydenham E.W. (1992). Initial studieson the toxicokinetics of fumonisin B1. Food and ChemicalToxicology 30: 277�279.

Shephard G.S., Thiel P.G., Sydenham E.W., Alberts J.F. and CawoodM.E. (1994a). Distribution and excretion of a single dose of themycotoxin fumonisin B1 in a non-human primate. Toxicon 32:735�741.

Shephard G.S., Thiel P.G., Sydenham E.W. and Snijman P.W. (1995).Toxicokinetics of the mycotoxin fumonisin B2 in rats. Food andChemical Toxicology 33: 591�595.




Shephard G.S., Thiel P.G., Sydenham E.W., Vleggaar R. and AlbertsJ.F. (1994b). Determination of the mycotoxin fumonisin B1 andidentification of its partially hydrolysed metabolites in the faecesof non-human primates. Food and Chemical Toxicology 32: 23�29.

Shephard G.S., Van der Westhuizen L. and Sewram V. (2007).Biomarkers of exposure to fumonisin mycotoxins: a review.Food Additives and Contaminants 24: 1196�1201.

Silva L.J.G., Lino C.M. and Pena A. (2009). Sphinganine-sphingosineratio in urine from two Portuguese populations as biomarker tofumonisins exposure. Toxicon 54: 390�398.

Sokolovic M., Garaj-Vrhovac V. and Simpraga B. (2008). T-2 toxin:incidence and toxicity in poultry. Arhiv za Higijenu Rada iToksikologiju 59: 43�52.

Solfrizzo M., Avantaggiato G. and Visconti A. (1997). In vivo valida-tion of the sphinganine/sphingosine ratio as a biomarker to dis-play fumonisin ingestion. Cereal Research Communications 25:437�441.

Solfrizzo M., Chulze S.N., Mallmann C., Visconti A., De GirolamoA., Rojo F. and Torres A. (2004). Comparison of urinary sphin-golipids in human populations with high and low maize con-sumption as a possible biomarker of fumonisin dietaryexposure. Food Additives and Contaminants Part A 21:1090�1095.

Soriano J.M., Gonzalez L. and Catala A.I. (2005). Mechanism ofaction of sphingolipids and their metabolites in the toxicity offumonisin B1. Progress in Lipid Research 44: 345�356.

Stockmann-Juvala H. and Savolainen K. (2008). A review of the toxiceffects and mechanisms of action of fumonisin B1. Human andExperimental Toxicology 27: 799�809.

Sudakin D.L. (2003). Trichothecenes in the environment: relevance tohuman health. Toxicology Letters 143: 97�107.

Sydenham E.W., Shephard G.S., Thiel P.G., Marasas W.F.O. andStockenstrom S. (1991). Fumonisin contamination of commercialcorn-based human foodstuffs. Journal of Agricultural and FoodChemistry 39: 2014�2018.

Tanaka T., Hasegawa A., Yamamoto S., Lee U.S., Sugiura Y. andUeno Y. (1988). Worldwide contamination of cereals by theFusarium mycotoxins nivalenol, deoxynivalenol and zearalenone.1. Survey of 19 countries. Journal of Agricultural and FoodChemistry 36: 979�983.

Tran S.T., Tardieu D., Auvergne A., Bailly J.D., Babile R., Durand S.,Benard G. and Guerre P. (2006). Serum sphinganine and thesphinganine to sphingosine ratio as a biomarker of dietary fumo-nisins during chronic exposure in ducks. Chemico-BiologicalInteractions 160: 41�50.

Turner P.C., Burley V.J., Rothwell J.A., White K.L.M., Cade J.E. andWild C.P. (2008b). Dietary wheat reduction decreases the level ofurinary deoxynivalenol in UK adults. Journal of ExposureScience and Environmental Epidemiology 18: 392�399.

Turner P.C., Burley V.J., Rothwell J.A., White K.L.M., Cade J.E. andWild C.P. (2008c). Deoxynivalenol: rationale for developmentand application of a urinary biomarker. Food Additives andContaminants 25: 86�871.

Turner P.C., Nikiema P. and Wild C.P. (1999). Fumonisincontamination of food: progress in development of biomarkersto better assess human health risks.Mutation Research 443: 81�93.

Turner P.C., Rothwell J.A., White K.L., Gong Y., Cade J.E. and WildC.P. (2008a). Urinary deoxynivalenol is correlated with cerealintake in individuals from the United Kingdom. EnvironmentalHealth Perspectives 116: 21�25.

Turner P.C., Taylor E.F., White K.L.M., Cade J.E. and Wild C.P.(2009). A comparison of 24 h urinary deoxynivalenol with recentversus average cereal consumption for UK adults. British Journalof Nutrition 102: 1276�1279.

Ueno Y., Tashiro F. and Kobayashi T. (1983). Species differences inzearalenone-reductase activity. Food and Chemical Toxicology 21:167�173.

Van der Westhuizen L., Brown N.L., Marasas W.F.O., Swanevelder S.and Shephard G.S. (1999). Sphinganine/sphingosine ratio inplasma and urine as a possible biomarker for fumonisin exposurein humans in rural areas of Africa. Food and Chemical Toxicology37: 1153�1158.

Van der Westhuizen L., Shephard G.S., Rheeder J.P., SomdyalaN.I.M. and Marasas W.F.O. (2008). Sphingoid base levels inhumans consuming fumonisin-contaminated maize in ruralareas of the former Transkei, South Africa: a cross-sectionalstudy. Food Additives and Contaminants Part A 25: 1385�1391.

Van der Westhuizen L., Shephard G.S. and Van Schalkwyk D.J.(2001). The effect of a single gavage dose of fumonisin B2 onthe sphinganine and sphingosine concentrations in vervet mon-keys. Food and Chemical Toxicology 39: 455�459.

Verdonck F.A.M., Sioen I., Baert K., Van Thuyne N., Bilau M.,Matthys C., De Henauw S., De Meulenaer B., Devlieghere F.,Van Camp J., Vanrolleghem P.A., Van Sprang P., Verbeke W.and Willems J. (2006). Uncertainty and variability modelling ofchemical exposure through food. Archives of Public Health 64:159�173.

Verdonck F.A.M., Van Sprang P.A. and Vanrolleghem P.A. (2005).Uncertainty and precaution in European environmental riskassessment of chemicals. Water Science and Technology 52:227�234.

Visconti A. (2001). Problems associated with Fusarium mycotoxins incereals. Bulletin of the Institute for Comprehensive AgriculturalSciences, Kinki University 9: 39�55.

Visconti A., Doko M.B., Solfrizzo M., Pascale M. and Boenke A.(1996). European inter-comparison study for the determinationof Fumonisins in Maize. Mikrochimica Acta 123: 55�61.

Visconti A., Minervini F., Lucivero G. and Gambatesa V. (1991).Cytotoxic and immunotoxic effects of Fusarium mycotoxinsusing a rapid colorimetric bioassay. Mycopathologia 113:181�186.

Voss K.A., Smith G.W. and Haschek W.M. (2007). Fumonisins: tox-icokinetics, mechanism of action and toxicity. Animal FeedScience and Technology 137: 299�325.

Walter M.D. (1998). Food Frequency Methods. In: Kelsey J.L.,Marmot M.G., Stolley P.D. and Vessey M.P. (eds), NutritionalEpidemiology, Oxford: Oxford University Press, pp. 74�94.

Wan Norhasima W.M., Abdulamir A.S., Abu Bakar F., Son R. andNorhafniza A. (2009). The health and toxic adverse effects ofFusarium fungal mycotoxin, fumonisins, on human population.American Journal of Infectious Diseases 5: 280�288.

Wang E., Norred W.P., Bacon C.W., Riley R.T. and Merrill J.A.H.(1991). Inhibition of sphingolipid biosynthesis by fumonisins.Implications for diseases associated with Fusarium moniliforme.Journal of Biological Chemistry 266: 14486�14490.

Wang E., Riley R.T., Meredith F.I. and Merrill J.A.H. (1999).Fumonisin B1 consumption by rats causes reversible, dose-depen-dent increases in urinary sphinganine and sphingosine. Journal ofNutrition 129: 214�220.

Wang E., Ross P.F., Wilson T.M., Riley R.T. and Merrill J.A.H.(1992). Increases in serum sphingosine and sphinganine anddecreases in complex sphingolipids in ponies given feed contain-ing fumonisins, mycotoxins produced by Fusarium moniliforme.Journal of Nutrition 122: 1706�1716.

Wild C.P., Hudson G.J., Sabbioni G., Chapot B., Hall A.J., WoganG.N., Whittle H., Montesano R. and Groopman J.D. (1992).Dietary intake of aflatoxins and the level of albumin-bound afla-toxin in peripheral blood in the Gambia, West Africa. CancerEpidemiology, Biomarkers and Prevention 1: 229�234.

Wild C.P., Jiang Y.-Z., Allen S.J., Jansen L.A.M., Hall A.J. andMontesano R. (1990a). Aflatoxin-albumin adducts in humansera from different regions of the world. Carcinogenesis 11:2271�2274.

Wild C.P., Jiang Y.-Z., Sabbioni G., Chapot B. and Montesano R.(1990b). Evaluation of methods for quantitation of aflatoxin-albumin adducts and their application to human exposure assess-ment. Cancer Research 50: 245�251.

Wood G.E. (1992). Mycotoxins in foods and feeds in the UnitedStates. Journal of Animal Science 70: 3941�3949.

World Health Organization (WHO) (1993). International Programmeon Chemical Safety, Biomarkers and Risk Assessment: Conceptsand Principles, Environmental Health Criteria 155. Geneva,Switzerland: World Health Organization.




World Health Organization (WHO) (1997). Guidelines for predictingdietary intake of pesticide residues. 2nd revised edn, Geneve,Switzerland: World Health Organization, Document WHO/FSF/FOS/97.7.

World Health Organization (WHO) (2001). Fumonisins. SafetyEvaluation of Certain Mycotoxins in Food. WHO FoodAdditives Series 47, FAO Food and Nutrition Paper 74.Geneva, Switzerland: WHO, pp. 103�279.

Yoo H., Norred W.P., Showker J. and Riley R.T. (1996). Elevatedsphingoid bases and complex sphingolipid depletion as contrib-uting factors in fumonisin-induced cytotoxicity. Toxicology andApplied Pharmacology 138: 211�218.

Yoshizawa T., Mirocha C.J., Behrens J.C. and Swanson S.P. (1981).Metabolic fate of T-2 toxin in a lactating cow. Food andCosmetics Toxicology 19: 31�39.

Yoshizawa T., Yamashita A. and Luo Y. (1994). Fumonisin occur-rence in corn from high-risk and low-risk areas for human esoph-ageal cancer in China. Applied and Environmental Microbiology60: 1626�1629.

Zollner P., Jodlbauer J., Kleinova M., Kahlbacher H., Kuhn T.,Hochsteiner W. and Lindner W. (2002). Concentration levels ofzearalenone and its metabolites in urine, muscle tissue and liversamples of pigs fed with mycotoxin-contaminated oats. Journal ofAgricultural and Food Chemistry 50: 2494�2501.




food science and technology international vol16 issue3

Documents