Volume XV Number 3 2003 - Chiriotti Editori archivio... · 2020. 5. 15. · J.C. Cheftel...

ISSN 1120-1770

Volume XVNumber 3

2003

Sp

ediz

ion

e in

ab

. po

st. c

om

ma

26 -

art

. 2 -

leg

ge

549/

95 n

. 3/2

003

- To

rin

o

Ital. J. Food Sci. n. 3, vol. 15 - 2003 319

ITALIAN JOURNAL OF FOOD SCIENCE(RIVISTA ITALIANA DI SCIENZA DEGLI ALIMENTI)

Property of the University of PerugiaSupported in part by the Italian Research Council (CNR) - Roma - Italy

Editor-in-Chief:Paolo FantozziDipartimento di Scienze degli Alimenti, Università di Perugia, S. Costanzo, I-06126 Perugia, ItalyTel. +39 075 5857910 - Telex 662078 UNIPG - Telefax +39 075 5857939-5852067E-mail: [email protected]

Assistant Editor:S. Mary F. Traynor, F.S.E.Dipartimento di Scienze degli Alimenti, Università di Perugia, S. Costanzo, I-06126 Perugia, ItalyTel. +39 075 5857912 - Telex 662078 UNIPG - Telefax +39 075 5857939-5852067E-mail: [email protected]

Publisher:Alberto ChiriottiChiriotti Editori s.p.a., Viale Rimembranza 60, I-10064 Pinerolo, ItalyTel. +39 0121 393127 - Telefax +39 0121 794480E-mail: [email protected] - URL: www.chiriottieditori.it

Aim: The Italian Journal of Food Science is an international journal publishing original, basic andapplied papers, reviews, short communications, surveys and opinions in food science(chemistry, analysis, microbiology), food technology (engineering, processing) and related areas(nutrition, safety, toxicity, physiology, dietetics, economics, etc.). Upon request and free of charge,announcements of congresses, presentations of research institutes, books and proceedings mayalso be published in a special “News” section.

Review Policy:The Advisory Board with the Editor-in-Chief will select submitted manuscripts in relationshipto their innovative and original content. Referees will be selected from the Advisory Boardand/or from the “IJFS Official Referee List” composed of 200 qualified Italian or foreignscientists. Acceptance of a paper rests with the referees.

Frequency: Quarterly - One volume in four issues. Guide for Authors and annual indices will bepublished only in number 4 of each volume.

Impact Factor: 0.639 published in the 2002 Journal of Citation Reports, Institute for Scientific Information

Subscription Rate: 2003: Volume XV Ordinary € 120.00Supporting € 500.00

IJFS is abstracted/indexed in: Chemical Abstracts Service (USA); Foods Adlibra Publ. (USA); Gialine- Ensia (F); Institut Information Sci. Acad. Sciences (Russia); Institute for Scientific Information;CurrentContents®/AB&ES; SciSearch® (USA-GB); Int. Food Information Service - IFIS (D);Int. Food Information Service - IFIS (UK); EBSCO Publishing.

IJFS has a page charge of € 20.00 up to 5 pages; extra pages are € 30.00.Reprints (100) will be sent free of charge.

320 Ital. J. Food Sci. n. 3, vol. 15 - 2003

ITALIAN JOURNAL OF FOOD SCIENCE

ADVISORY BOARD

G. AndrichDip. di Chimicae Biotecnologie AgrarieUniversità di PisaPisa, ItalyF. AngerosaIst. Sperim. per la ElaiotecnicaCittà S. Angelo, Pescara, ItalyC.H. BellCentral Science LaboratorySand HuttonYork, UKA. BertrandInstitut d’OenologieUniversité de BordeauxTalence Cedex, FranceL.B. BullermanDept. of Food Science and TechnologyUniversity of Nebraska-LincolnLincoln, NE, USAC. CannellaIst. Scienza dell’AlimentazioneUniversità di Roma (La Sapienza)Roma, ItalyE. CarnovaleIst. Nazionale di Ricercaper gli Alimenti e la NutrizioneUnità Chimica degli AlimentiRoma, ItalyJ.C. CheftelLaboratoire de Biochimie etTechnologie AlimentairesUniversité des Sciences et TechniquesMontpellier, FranceS. CondonDepartment of Food MicrobiologyUniversity CollegeCork, IrelandR. CubaddaDip. di Scienza e TecnologiaAgro-Alimentare e MicrobiologiaUniversità del MoliseCampobasso, ItalyA. CurioniDip. di Biotecnologie AgrarieUniversità di PadovaLegnaro PD), ItalyM. Dalla RosaDip. di Protezione eValorizzazione Agro-AlimentareUniversità di BolognaBologna, ItalyG. Dall’AglioStaz. Sperim. per l’Industriadelle Conserve AlimentariParma, ItalyM. Di MatteoDip. di Ingegneria Chimicaed AlimentareUniversità di SalernoFisciano (SA), Italy

J.M. FaubionDept. of Grain Science and IndustryKansas State UniversityManhattan, Kansas, USA

P.F. FoxDepartment of Food ChemistryUniversity CollegeCork, Ireland

D. GallantLaboratoire de TechnologieAppliquée à la NutritionINRA Centre de Recherches de NantesNantes Cedex, France

S. GarattiniIst. di Ricerche Farmacologiche“Mario Negri”Milano, Italy

A.M. GattusoDip. Economia, Ingegneria eTecnologie Agro-ForestaliUniversità di PalermoPalermo, Italy

R. GiangiacomoIstituto Sperim. Lattiero-CasearioLodi, Italy

M. GobbettiDip. di Protezione delle Piante eMicrobiologia ApplicataUniversità di BariBari, Italy

T. GomesDip. di Progettazione e Gestionedei Sistemi Agro-Zootecnici e ForestaliUniversità di BariBari, Italy

M. KarelDept. of Food ScienceRutgers UniversityNew Brunswick, NJ, USA

J.W. KingLos Alamos National Lab.Chemistry DivisionLos Alamos, NM, USA

T.P. LabuzaDept. of Food and Nutritional SciencesUniversity of MinnesotaSt. Paul, MN, USA

M. LucisanoDip. di Scienze e TecnologieAlimentari e MicrobiologicheSez. Tecnologie AlimentariUniversità di MilanoMilano, Italy

R. MassiniDip. di Ingegneria IndustrialeUniversità di ParmaParma, Italy

L. MoioDip. di Scienze degli AlimentiUniversità di FoggiaFoggia, ItalyM. MoresiIst. di Tecnologie Agro AlimentariUniversità della TusciaViterbo, ItalyJ. O'BrienSchool of Biological SciencesUniversity of SurreyGuilford, Surrey, UKM. OhshimaSchool of AgricultureNagoya UniversityNagoya, JapanC. PeriDip. di Scienze e TecnologieAlimentari e MicrobiologicheSez. Tecnologie AlimentariUniversità di MilanoMilano, ItalyS. PorrettaAssociazione Italiana diTecnologie Alimentari (AITA)Milano, ItalyE. SenesiInstituto Sperim. per la ValorizzazioneTecnologica dei Prodotti Agricoli(I.V.T.P.A.)Milano, ItalyA. SensidoniDip. di Scienze degli AlimentiUniversità di UdineUdine, ItalyP. ShermanDept. of Food and Nutritional SciencesKing’s College LondonKensington, London, UKS. Spagna-MussoDip. di Scienze degli AlimentiUniversità di NapoliPortici (NA), ItalyG.W. SmithersCSIRO Div. of Food ProcessingDairy Research LaboratoryHighett, Victoria, AustraliaL. StepaniakDept. of Food ScienceAgricultural University of NorwayÅsNLH, NorwayG. VersiniDip. Lab. Analisi e RicercheIst. Agrario di S. Michele a/AdigeS. Michele all’Adige (TR), ItalyJ.R. WhitakerDept. of Food Science and TechnologyUniversity of CaliforniaDavis, CA, USA


PAPER

- Key words: electrochemical sensor, Flow Injection Analysis (FIA), ovalbumin, thiols -

ELECTROCHEMICAL METHODFOR THE EVALUATION

OF THE NUMBER OF EGGS IN EGG PASTA

DETERMINAZIONE DEL NUMERO DI UOVA NELLA PASTA ALL’UOVOMEDIANTE UN METODO ELETTROCHIMICO

M.S. COSIO*, S. BURATTI and S. MANNINODipartimento di Scienze e Tecnologie Alimentari e Microbiologiche,Università degli Studi di Milano, Via Celoria 2, 20133 Milano, Italy

* Corresponding author:Tel. +39 02 50316623; Fax +39 02 50316632; E-mail: stella.cosio@ unimi.it.

ABSTRACT

A specific and rapid method to de-termine the number of eggs in egg pas-ta based on the determination of oval-bumin thiol groups is described. Pastawater extracts were analysed by a FlowInjection Analysis (FIA) system with anelectrochemical detector operating inamperometry with a tungsten sensorworking at -0.2 V vs SCE. There was agood linear correlation between the SHcontent obtained using the proposedelectrochemical method and thenumber of eggs in egg pasta. Moreover

RIASSUNTO

Viene descritta una procedura spe-cifica basata sulla determinazione deigruppi tiolici dell’ovoalbumina per lavalutazione del numero di uova nellepaste all’uovo. Gli estratti acquosi del-la pasta vengono analizzati con un sen-sore di tungsteno operante in ampero-metria ad un potenziale fisso di -0,2 Vvs SCE in un sistema a flusso tipo FIAcon rivelazione elettrochimica. I risul-tati riportati mostrano una buona cor-relazione di tipo lineare tra il numerodi uova contenute nelle paste all’uovo


to confirm the results of the electro-chemical method, the ovalbumin con-tent of egg pasta was also evaluated byelectrophoretic SDS-Page analysis. Be-cause of the speed of analysis, the elec-trochemical method could be useful forthe rapid determination of the numberof eggs in egg pasta.

e la concentrazione di SH valutata conil metodo proposto. Inoltre, al fine diconfermare i risultati del metodo, l’ovo-albumina contenuta nelle paste all’uo-vo è stata valutata anche con l’analisielettroforetica SDS-Page. Data la velo-cità dell’analisi il metodo elettrochimi-co è utile per la rapida determinazionedel numero di uova contenute nellepaste all’uovo.

INTRODUCTION

The term “egg pasta” means a pastaproduct containing a minimum of 4whole eggs per kilogram, correspondingto 200 g of egg per kilogram of durumwheat semolina. Italian legislation (DPR,2001) has established that egg pastamust contain a minimum 0.145 g ofsterols per 100 g of dried matter. Thislimit was set considering that sterols arepresent in egg yolk at a relatively con-stant concentration (1.45-1.50 g per 100g of yolk) (POWRIE and NAKAI, 1986).However, modern hen feeding practiceshave caused a decrease in the sterolcontent in eggs to values lower than1.45 g per 100 g of egg yolk. For thisreason, egg pasta makers are now forcedto use a larger number of eggs or addegg yolk in order to meet legal require-ments.

The official method of analysis for thedetermination of the number of eggs inpasta is based on the evaluation of thesterol content (AOAC, 1990). Anothermethod reported in the literature isbased on the HPLC determination of cho-lesterol as an indicator of egg solids(HURST et al., 1985).

Considering that the informationabout the egg content in egg pasta isimportant from both the production andthe regulatory points of view, a newmethod based on the determination ofovalbumin thiol groups has been devel-

oped. Ovalbumin (M.W. 45000 D) repre-sents more than 50% (w/w) of the albu-men proteins and it is the only egg whiteprotein containing free SH groups(FOTHERGILL and FOTHERGILL, 1970;NISBET et al., 1981).

In a previous paper, COSIO et al. (2000)set up an electrochemical method to de-termine the thiol content in whey pro-teins to evaluate milk heat treatment.Likewise, in the present work a proce-dure to evaluate the SH content of eggpasta is described. It involves operatingwith a tungsten electrode at -0.2V (SCE)in a flow injection system with a carriersolution containing mercury ions.

MATERIALS AND METHODS

Apparatus and reagents

The flow-injection system consisted ofa Jasco 880 PU HPLC pump, a Rheod-yne model 7010 sample injection valveand EG&G PAR Model 400 electrochem-ical detector equipped with a tungstenwire working electrode (0.3 mm diame-ter and 20 mm length), 99.95% purity(Good Fellow, Cambridge), a referencesaturated Calomel electrode (SCE) anda platinum auxiliary electrode. Currentoutputs were displayed on a strip chartrecorder (REA 112, Radiometer, Den-mark).

All solutions were prepared from rea-


gent grade chemicals. L-cysteine, mer-cury chloride, sodium chloride and ace-tic acid were purchased from Merck(Darmstadt Germany). Ovalbumin wasfrom Sigma Chemical Co (St. Louis, MO,USA).

Samples

Ten commercial samples of egg pastacontaining 4 or 6 eggs (16.7 or 23.1%egg/kg product) were analyzed. Sevensamples of pasta containing from 0 to 6eggs, produced by the Institute of Cere-alicoltura in Rome, were also tested. Thepasta samples were dried at low temper-ature (LT) (50°C x 18 h). Moreover tenpasta samples, without egg, produced bythe same institute using different culti-vars of durum wheat flour were also an-alysed. All samples were dried at lowtemperature (as egg pasta) and at hightemperature (73°C x 11 h).

Sample preparation

All samples were ground using a lab-oratory mill and passed through a 0.188mm sieve. One gram of sample wasweighed in a centrifuge tube, dispersedin 10 mL of distilled water, mechanical-ly stirred for 30 min at room tempera-ture and centrifuged for 20 min (10,000rpm; 15°C). The supernatant was filteredthrough a Whatman n°4 filter, diluted1:1 with the carrier solution and theninjected in the flow system.

Electrochemical procedure

Amperometric flow injection measure-ments of thiol compounds were per-formed using a potential of -0.2V versusSCE and a carrier solution composed of0.1 mol L-1 acetic acid containing 0.04mol L-1 NaCl and 37 µmol L-1 HgCl2. Aflow rate of 1.0 mL min-1 and an injec-tion volume of 20 µL were used. All aque-ous standard solutions were preparedfresh each day.

Determination of the “water soluble”proteins

The “water soluble” proteins (albuminsplus globulins) were determined by theBiuret method (GORNALL et al., 1949).This method is based on a colorimetricreaction. The extent of the coloration of asolution is a function of peptide linkagesand subsequently of the protein concen-tration. Water extracts (0.5 mL) of sam-ples were added to 2 mL of the Biuret re-agent and then incubated at room tem-perature for 30 min. Absorbance at 540nm was immediately recorded using a DU650 Beckmann spectrophotometer (In-struments Inc., Fullerton, CA, USA).

Electrophoresis SDS-PAGE

Sodium dodecyl sulfate-polyacrylamidegel was prepared using 15 mL of poly-acrylamide (29.1% polyacrylamide, 0.9%N,N-methylenebisacrylamide from Bio-rad-Hercules, CA, USA), 3.3 mL of am-monium persulphate 1.5% from Fluka,30 mL of phosphate buffer (pH 7.1, 0.2M containing 0.2% sodium dodecyl sul-fate from Sigma), 70 µL of TEMED(NNN’N’-tetramethylene diamine) fromMerck and 11.63 mL of distilled water.After pre-running the gel at 200 V (cur-rent 150 mA) for 30 min, 10 µL of 0.6 mgmL-1 protein solution containing bovineserum albumin, carbonic anhydrase, lys-ozyme and ovalbumin (from Sigma), alltogether with water extracts of the pastasamples, were loaded and subjected toelectrophoresis at about 20 mA for 10 minand then 200 mA for 3 h. The gel wasstained with 1% Comassie Brilliant blueR-250 from Merck in methanol/TCA/sul-fosalicylic/water (30%/10%/5%/55%)and destained with ethanol/acetic acid/water (30%/10%/60%) (WRINGLEY et al.,1982). Electrophoretic gel images wereobtained by digitalization using a planescanner (HP Scanjet 6100C) connectedto a Pentium PC platform equipped withAdobe Photoshop software version 5.5.


The image was processed with CREAMsoftware (Cross Radio-Immuno Electro-phoresis Analysis Manager).

RESULTS AND DISCUSSION

Optimisationof the experimental conditions

The electrochemical method is basedon the measurement of the excessamount of mercury (II) ions in the carri-er solution after reactions with organicthiols (RSH) to form (RS)2Hg complexes.Complexation diminishes the Hg2+ activ-ity in the solution and thus reduces thecathodic current measured by the elec-trochemical detector. In order to optimizethe working conditions for the samplesstudied, injections of standard solutionsof HgCl2 in 0.1 mol L-1 acetic acid at apotential range from -0.1V to -0.25V(SCE) were performed. Optimal respons-es were obtained by operating with 37µmol L-1 of HgCl2 at a potential of -0.2Vand these conditions were used in allsubsequent experiments. The effect ofNaCl concentration on the HgCl2 re-sponse was evaluated by adding differ-ent concentrations of NaCl (10-80 mmolL-1) to the carrier solution. Chloride ionshave the function of increasing mercurysolubility and also of improving carriersolution conductivity. The best signal/background ratio was obtained with 40mmol L-1 of NaCl. This concentration wastherefore added to the carrier solutionin all subsequent experiments.

Using these optimized conditions, am-perometric responses of increasing con-centrations of cysteine and ovalbuminsolutions were recorded. Fig. 1 shows thatthe detector responded rapidly to the dy-namic changes of cysteine and ovalbu-min concentration, allowing about 40determinations per hour. In particularcysteine was used as reference standardin the determination of the SH content inthe pasta samples. The calibration plot

Fig. 1 - Flow injection responses for increasing con-centrations of cysteine (a) and ovalbumin (b).Cysteine concentrations: A: 0.5 µg mL-1; B: 1 µgmL-1; C: 1.5 µg mL-1; D: 2 µg mL-1; E: 2.5 µg mL-1;F: 3 µg mL-1; G: 4 µg mL-1; H: 5 µg mL-1; I: 6 µg mL-

1; L: 8 µg mL-1.Ovalbumin concentrations: A: 62.5 µg mL-1; B: 125µg mL-1; C: 250 µg mL-1; D: 500 µg mL-1.Operating conditions: Working potential -0.2V(SCE); flow rate 1 mL min-1; carrier solution: ace-tic acid 0.1 mol L-1, 37 µmol L-1HgCl2 and 40 mmolL-1NaCl; injection volume 20 µL.

(a)

(b)

was highly linear over the 0.1-8 µg mL-1

concentration range (r2=0.9998), the slopeand the intercept were 0.497 µA mL/µgand 0.025 µA, respectively. The relativestandard deviation (RSD) at the concen-tration level of 2.5 µg mL-1 was 1.5%(n=20) and the estimated detection limitcalculated using the linear regressiontechnique from MILLER and MILLER(1993) was 0.12 µg mL-1.

Optimisationof the extraction procedure

Since the methodology developed isbased on the determination of ovalbuminthiol groups, it was necessary to find an


Fig. 2 - Flow injection responses of pasta with durum wheat semoli-na (A), egg pasta with 2 eggs (B), egg pasta with 4 eggs (C), egg pastawith 6 eggs (D). Operating conditions as in Fig. 1.

extraction procedure thatcould minimize the ex-traction of thiols presentin wheat proteins.

The best results wereobtained by using wateras extractant since onlyalbumins of durumwheat (7-10% of totaldurum wheat proteins)and ovalbumin are watersoluble. Typical ampero-metric responses of pas-ta and egg pasta waterextracts are shown in Fig.2. The amperometric signal of pasta withdurum wheat semolina was about twotimes lower than that of pasta with 2 eggsand about four times lower than that ofpasta with 6 eggs. The height of the peakof pasta with durum wheat semolina isrelated to the presence of thiol groups in

durum wheat extracted with water anddepends greatly on the drying tempera-tures. It is well know that the use of ahigh or a low temperature drying profilein pasta can have a great influence onthe extraction of albumins; this is clear-ly show in Table 1, where results ob-

Table 1 - Content of “water soluble” proteins (Nsol) and SH groups in the pasta samples produced fromdifferent cultivars of durum wheat flour without egg.

Pasta samples Drying* Nsol g/100 g dried matter a SH µM cysteine eq/100 g of dried matter a

LT 1.23±0.12 40.3±3.21 HT 1.18±0.11 31.5±2.5

LT 1.25±0.14 42.6±3.32 HT 1.11±0.10 28.6±1.9

LT 1.25±0.14 39.6±3.03 HT 1.15±0.13 26.3±2.0

LT 1.26±0.13 38.0±2.94 HT 1.05±0.11 22.0±1.4

LT 1.31±0.15 46.2±3.35 HT 1.12±0.10 29.9±2.8

LT 1.21±0.13 40.0±2.96 HT 1.11±0.11 30.2±2.2

LT 1.32±0.14 38.0±2.87 HT 1.26±0.12 22.0±2.1

LT 1.28±0.13 42.5±2.98 HT 1.15±0.10 22.0±1.8

LT 1.49±0.15 43.8±3.49 HT 1.18±0.09 22.8±1.4

LT 1.55±0.15 46.8±2.910 HT 1.01±0.09 25.2±1.6

a (mean S.D., n=3); * LT= Low temperature; HT= High temperature


tained from the analysis of pasta sam-ples of known origin are reported. Sam-ples of pasta dried at low temperature(LT) had SH values in the range of 38-46.8 while those obtained at high tem-perature (HT) were in the range of 22-31.5. This is in agreement with somestudies reported in the literature (ALARYand KOBREHEL, 1987; PAGANI et al.,1986; DEXTER et al., 1981) stating thatpasta dried at high temperature has ahigher degree of protein insolubilisationwith respect to low temperature driedpasta.

Application to egg pasta samples

To verify if the electrochemical meth-od was suitable to determine the numberof eggs in egg pasta, pasta samples con-taining from 0 to 6 eggs, produced fromthe same durum wheat semolina anddried at low temperature (50°C), wereanalyzed. The results are reported inTable 2. The SH content is expressed asµmol of cysteine equivalent per 100 g ofdried matter. As expected the SH con-tent increased in relation to the numberof eggs and the increase per egg wasabout 30-40 µ mol ofcysteine equivalent per 100g of dried matter. The pas-ta sample made with 4 eggshad a SH content of about160 µmol of cysteine equiv-alent per 100 g of driedmatter. As shown in Fig. 3there was a good linear cor-relation between the SHcontent obtained by theproposed electrochemicalmethod and the number ofeggs (r=0.994).

The SH content deter-mined in commercial pas-ta samples produced with4 or 6 eggs is reported inTable 3. There was no greatvariability among samplesproduced with the same

Table 2 - Content of SH groups in the egg pastasamples produced with an increasing number ofeggs.

Experimental egg SH concentrationpasta samples (µmol cysteine eq/100 g

(number of eggs) of dried matter) a

0 44.9±3.21 85.5±2.82 109.0±3.83 133.3±4.24 162.7±6.26 203.7±5.7

a (mean S.D., n=3).

Fig. 3 - Linear correlation plot between the number of eggs in theegg pasta samples and the SH concentration determined using theproposed method.

number of eggs. This is related to thelow drying temperature used in order toavoid denaturation of proteins or enzy-matic browning that could cause the eggpasta to darken (NAMIKI, 1988; WALSH,1970).

To confirm the results of the electro-chemical method, the ovalbumin contentof egg pasta water extracts was evaluat-ed by electrophoretic SDS-PAGE analy-sis. A calibration curve of ovalbuminstandard solutions was developed


Table 3 - Content of SH groups in commercial eggpasta samples with 4 eggs (samples 1 to 6) or 6eggs (samples 7 to 10).

Commercial SH concentrationegg pasta (mol cysteine eq/100 g of driedsamples matter) a

1 140.4±5.52 132.9±6.53 141.2±5.24 130.4±4.55 134.8±6.06 136.9±3.97 160.0±5.88 193.0±7.29 188.2±8.2

10 204.2±9.9

a (mean S.D., n=3).

Fig. 4 - Linear correlation plot between the ovalbumin concentra-tion of the experimental and commercial egg pasta samples andSH concentration determined using the proposed method.

( y = 1 2 0 5 7 x - 1 2 3 . 5 7 ,r2=0.999) and the correla-tion between the SH con-tent and the ovalbuminconcentration of experi-mental and commercial eggpasta samples was evalu-ated (Fig 4). The good line-ar correlation (r=0.981)confirms that the pro-posed electrochemicalmethod based on themeasurement of the oval-bumin SH groups is use-ful for determining thenumber of eggs in egg pas-ta. Free cysteine thatcould come from egg yolkdid not seem to affect theaccuracy of the method.

CONCLUSION

The proposed electrochemical meth-od has been successfully employed forthe direct, rapid and reliable monitoringof the number of eggs in egg pasta. Thisnew method offers an alternative to thedifficult and time-consuming AOAC

method based on the determination ofthe sterol content. Furthermore the pro-posed method based on the determina-tion of the SH groups of ovalbumin, givesresults that are not related to hen feed-ing and, because of the speed of the anal-ysis, it is useful for on-line or routinedetermination of the number of eggs inegg pasta.

REFERENCES

Alary R. and Kobrehel K. 1987. The sulphydrylplus disulphide content in the proteins of du-rum wheat and its relationship with the cook-ing quality of pasta. J. Sci. Food Agric. 39:123.

AOAC. 1990. “Official Methods of Analysis”. 16th

ed, Association of Official Analytical Chemists.Arlington, VA.

Cosio M.S., Mannino S. and Buratti S. 2000. Elec-trochemical sensor detecting free sulfhydrylgroups: evaluation of milk heat treatment. J.Dairy Sci. 83: 1933.

DPR. 2001. Decreto del Presidente della Repubbli-ca No. 187/ 9 febbraio 2001.

Dexter J.E., Matsuo R.R. and Morgan B.C. 1981.High temperature drying: effect on spaghettiproperties. J. Food Sci. 46: 1741.

Fothergill L.A. and Fothergill J.E. 1970. Thiol anddisulphide content of hen ovalbumin. Biochem.J. 116: 555.

Gornall A.G., Bardawill C.D. and David M.M. 1949.


Determination of serum proteins by means ofthe biuret reaction. J. Biol. Chem. 77:751.

Hurst W.J., Aleo M.D. and Martin R.A.Jr. 1985.HPLC determination of the colesterol contentof egg noodles as an indicator of egg solids. J.Agric. Food Chem. 33: 820.

Miller J.C. and Miller J.N. 1993. “Statistics forAnalytical Chemists”. 3rd ed., p. 115. Ellis Hor-wood. Chichester U.K.

Namiki M. 1988. Chemistry of Maillard reaction:studies on the browning reaction mechanismand development of antioxidants and mutagens.Adv. Food Res. 32: 115.

Nisbet A.D., Saundry R.H., Moir Arthur J.G.,Fothergill L.A. and Fothergill J.E. 1981. The

complete amino-acid sequence of hen ovalbu-min. European J. Biochem: 115: 335.

Pagani M.A., Resmini P., Fiorino A. and Dalbon G.1986. L’essiccazione ad alta temperatura nellaproduzione di paste all’uovo: effetti su alcunecaratteristiche organolettiche e nutrizionali.Tecnica Molitoria 3: 177.

Powrie W.D. and Nakai S. 1986. The chemistry ofeggs and egg products. Egg Sci. and Techn. 6: 97.

Walsh A.E. 1970. Measurement of spaghetti color.Macaroni J. 52: 20.

Wringley C.W., Austran J.C. and Bushuk W. 1982.Identification of cereal varieties by gel electro-phoresis of the grain proteins. Adv. Cereal Sci.Technol. 211.

Paper received April 1, 2003 Accepted June 9, 2003


PAPER

- Key words: heavy metals, multivariate statistical analysis, PDO -

CHEMOMETRIC DETERMINATIONOF THE GEOGRAPHICAL ORIGIN

OF COW MILK USING ICP-OES DATAAND ISOTOPIC RATIOS:A PRELIMINARY STUDY

DETERMINAZIONE CHEMOMETRICA DELL’ORIGINE GEOGRAFICADEL LATTE VACCINO ATTRAVERSO ICP-OES E DATI ISOTOPICI:

STUDIO PRELIMINARE

M.A. BRESCIA, V. CALDAROLA1, G. BUCCOLIERI2, A. DELL’ATTI2

and A. SACCO3

Università degli Studi di Bari, Dipartimento di Chimica, Campus Universitario,Via Orabona 4, 70126 Bari, Italy

1 STMicroelectronics Srl, Modulo M6/Contrada Torrazze, Zona Industriale,Stradale Primosole 50, 95121 Catania, Italy

2 Università degli Studi di Lecce, Dipartimento di Scienza dei Materiali,via Arnesano, 73100 Lecce, Italy

3 Università degli Studi di Foggia, Facoltà di Agraria, Istituto di Produzionie Preparazioni Alimentari, Via Napoli 25, 71100 Foggia, ItalyFax +39 080 5442129; E-mail:[email protected]

ABSTRACT

The geographical origin of twenty cowmilk samples from two different geo-graphical areas of the Apulia region insouthern Italy was determined usingInductively Coupled Plasma EmissionSpectroscopy (ICP-OES) for the metals

RIASSUNTO

La differenziazione dell’origine geo-grafica di venti campioni di latte vacci-no provenienti da due differenti aree diproduzione della Puglia, nel sud Italia,è stata investigata attraverso spettro-scopia atomica ad emissione di plasma


INTRODUCTION

The growing interest of consumers ingenuine and traditional products hasincreased the need for methods to con-trol the quality of the foods producedwithin the European Community as wellas those imported from other countries.

The EU regulation 2081/92 (E.U.,1992) protects consumers through a sys-tem of effective and impartial controlsby which Protected Designation of Ori-gin (PDO) products can be distinguishedfrom non PDO products. At the sametime, this regulation meets producers’ re-quirements. That is, it is in the interestof the producers that the controls to beas objective and clear as possible so asto protect the authenticity of the prod-ucts that are to be promoted on the mar-ket.

A greater appreciation of qualified na-tional products, together with a guar-anteed reference for consumers, hasbeen developed in the field of dairy

(Ba, Mn, Zn, Al, Fe and Cu), and iso-tope ratio mass spectrometry (IRMS).The phosphorous concentration wasdetermined by colorimetry. The datawere processed using the followingmultivariate statistical methods: Prin-cipal Components Analysis (PCA), Hi-erarchical Clustering (HCA) and LinearDiscriminant Analysis (LDA). The sam-ples were divided into two groups ac-cording to their origin and were classi-fied by discriminant analysis whichgave 100% success. This method couldbe extended to classify dairy productslike mozzarella and cheese in order toverify and safeguard Protected Desig-nation of Origin (PDO) products.

induttivamente accoppiato (ICP-OES),per la determinazione dei metalli e spet-trometria di massa dei rapporti isoto-pici (IRMS). Il contenuto di fosforo to-tale è stato anche misurato per via co-lorimetrica. I dati ottenuti sono statiprocessati attraverso l’analisi statisti-ca multivariata: Analisi delle Compo-nenti Principali (PCA), Analisi di Clu-stering Gerarchico (HCA) ed Analisi Di-scriminante (DA). I campioni sono statidivisi in due gruppi in base all’origine esono stati classificati mediante l’anali-si discriminante ottenendo il 100% dicapacità di classificazione. Uno svilup-po di questo studio potrebbe consiste-re nell’estensione del metodo a prodot-ti lattiero-caseari come formaggio emozzarella contribuendo alla salva-guardia dei prodotti recanti il marchiodi Denominazione di Origine Protetta.

products. The PDO trademark has beenassigned to numerous Italian cheeses(like Parmigiano Reggiano, Mozzarelladi Bufala Campana, etc.) and manycheese producers in Italy are activelyseeking PDO classification. In order toreceive this designation, the raw mate-rials must be produced and processedin the specified region from which theproduct gets its specific name. For thisreason, authentication of the geograph-ical origin of a product is becoming amajor challenge. MANCA et al. (2001)recently identified some compositionalparameters that allow Italian cheesesamples to be distinguished accordingto their region of origin. In order to de-termine the authenticity of typical dairyproducts, the geographical origin of themilk must be determined. Studies onthe geographical origin of milk sampleshave been conducted by means of Iso-tope Ratio Mass Spectrometry (IRMS)(KORNEXL et al., 1997; ROSSMANN et al.,1998).


The question of the geographicalidentification of food products often in-volves small areas of production. Wines,for instance, are often produced in re-stricted areas defined as DOC (Denom-ination of Controlled Origin) and thereare often several DOC areas within agiven region that are not far from oneanother.

Metal composition has been widelyused to assign the geographical origin ofvarious matrices such as wine, honeyand other food products (BRESCIA et al.,2002a; REBOLO et al., 2000; PARAMASet al., 2000). Isotopic and trace elementsmeasurements have provided very inter-esting results in the case of wine (MAR-TIN et al., 1999).

The objective of this study was to de-termine whether the concentrations ofselected metals (Ba, Mn, Zn, Al, Fe andCu) and phosphorous, and the isotopicratios of carbon and nitrogen (13C/12Cand 15N/14N) could be used to identifythe specific area of origin of a milk sam-ple within the Apulia region of Italy. Theisotopic and trace element measure-ments were combined and statisticallyevaluated.


Milk samples

Twenty samples of cow milk weresupplied by some farms in Apulia, aregion in Southern Italy. Ten samplescame from five farms [Troia (T), Cas-telluccio (C), Volturara Appula (V), Fog-gia (F) and Lucera (L)] in the provinceof Foggia, in Northern Apulia and tensamples came from four farms [Al-berobello (A), Noci (N) and MartinaFranca (M1 and M2)] situated in a char-acteristic area in the centre of Apulia,called “Murgia”. For this preliminarystudy, cows in the 0-3 month lactationperiod were selected in order to elimi-nate the variability in milk composition

associated with the lactation period. Allsamples were collected in polypropyl-ene bottles and stored in a freezer at -80°C until analysis.

Phosphorus determination

Phosphorous content was determinedby adding 0.3 g of ammonium persul-phate and 2 mL of 30% sulphuric acidto 0.2 g of milk. This mixture was min-eralised for 2 h at 100°C, then neutral-ised with 30% NaOH, filtered and dilut-ed to 100 mL with deionised water. Af-terwards, 0.05 g of ascorbic acid and 10mL of an acid mixture consisting of 0.13g/L of potassium antimony tartrate, 5.6g/L of ammonium molybdate and 70.0mL/L of concentrated sulphuric acidwere added to 20 mL of the sample. Af-ter 30 min, the absorbance at λ=800 nmwas measured with a LP1W photometer(DrLange, Düsseldorf, Germany).

Metal determinations

The concentrations of Ba, Fe, Mn, Al,Cu and Zn were determined using a Lib-erty 110 (Varian Inc., Palo Alto, USA)inductively coupled plasma optical emis-sion spectrophotometer (ICP-OES),equipped with an ultra-sound nebulizerand a VGA 77P hydride generator. Quan-titative determinations were carried outusing the external standard method.Other heavy metals (Co, Ni, Pb and Cd)were determined. However, in all casesthe levels were less than the detectionlimits (1.5, 0.4, 0.08 and 0.4 mg/kg, re-spectively).

The samples were subjected to a min-eralising process: 0.5 g of freeze-driedmilk was dissolved in 6 mL of HNO3 and1 mL of H2O2 and put into a 1200 MEGAmicrowave digestion unit (Milestone,Bergamo, Italy) with a well defined tem-perature program. The operating condi-tions for measurements are reported inTable 1.

Hydrochloric and nitric acids were ul-


tra-pure grade (J.T. Baker, Deventer, TheNetherlands); deionized water was ob-tained using a Milli-Q system (Millipore,Billerica, USA).

Isotopic ratio determinations

The isotopic ratios, 13C/12C and 15N/14N, were determined using a Carlo Erba1110 elemental analyser (Carlo Erba,Milano, Italy) connected on line to aFinnigan MAT delta S mass spectrome-ter (Thermo Finnigan, San Jose, USA)according to previously described meth-odologies (BREAS et al., 1994; BRESCIAet al., 2002b). Isotopic contents (13C/12Cand 15N/14N) are expressed as isotopicdeviations, δ, defined as:

Rs - RRefδ(‰) = _______________ · 1,000

RRef

Where Rs is the isotopic ratio meas-ured for the sample and RRef is the iso-topic ratio of the following internationalstandards: Pee Dee Belemnite (PDB) forcarbon and atmospheric N2 for nitrogen.

Multivariate analysis

Multivariate statistical techniqueswere used to evaluate the possibility ofdifferentiating the milk samples accord-ing to their geographical origin. Multi-variate analyses were performed usingStatistica software (StatSoft Inc., Tulsa,

Table 1 - Operating conditions for ICP-OES measurements.

Ba Fe Mn Al Zn Cu

λ (nm) 455.4 259.9 257.6 396.1 213.9 324.7Viewing Height (mm) 10 14 12 15 10 6Search Window (nm) 0.080 0.040 0.040 0.080 0.027 0.040Amplification Signal (V) 800 800 800 650 800 600RF Power (kW) 1.2 1.2 1.2 1.0 1.2 1.2Plasma (L/min) 18.0 18.0 18.0 15.0 18.0 18.0Auxiliary (L/min) 1.50 1.50 1.50 1.50 1.50 1.50Pump Speed (rpm) 15 15 15 15 15 15

OK, USA). The chemometric approachesused were: Principal Component Analy-sis (PCA), Hierarchical Clustering Anal-ysis (HCA) and Linear Discriminant Anal-ysis (LDA).

PCA calculates orthogonal linear com-binations of the auto-scaled variables byusing a correlation matrix based on themaximum variance criterion. Such line-ar combinations are called principalcomponent scores and the coefficientsof the linear combinations are calledloadings (MASSART et al., 1988). The nu-merical loading value for each variableof a given principal component showshow much the variable has in commonwith that component.

HCA is used to identify groupingsamong the samples on the basis of mul-tivariate measurements. In the hierar-chical methods, objects are grouped onthe basis of inter-object distances inhigh-dimensional space and defined byan agglomerative algorithm. In thiswork, the Manhattan distance was com-puted and a Ward linkage procedurewas applied (FRANK and TODESCHINI,1994). The results obtained were con-firmed by hierarchical cluster analysisin which a complete linkage procedurewas applied on the Euclidean distanc-es. PCA and HCA are both unsupervisedstatistical methods that give comple-mentary information about the similar-ities and groupings of the samples con-sidered. If a trend exists, it is of inter-est to evaluate the possibility of classi-


fying the samples. LDA obtains discri-minant functions calculated to maxim-ise the distances between groups de-fined a priori. The purpose is to calcu-late class models and boundaries, giv-ing a rule of classification based on aset of known objects (training set). Thisrule can be applied to define the classi-fication of unknown objects (test set).

RESULTS AND DISCUSSIONS

The values of the parameters deter-mined in the cow milk samples are sum-marised in Table 2. The marked varianceof the values for some metals and phos-phorous in both groups reflects the widevariation in the composition of the milksamples.

PCA was applied on a data set con-sisting of twenty samples. Five PrincipalComponents (PC) were extracted cover-ing 84% of the variance. Since a five-di-mensional display is difficult to present,the first two PCs were used for the scoreplot shown in Fig. 1.

A separation between milk samplesaccording to their origin was found onPC1. All samples from Foggia except onehad positive scores on the PC1. Milksamples from Murgia werecharacterised by negativescores on the same princi-pal component. Both groupsof samples showed a sepa-ration on PC2. This behav-iour may be partly due to thedifferent farms from whichthe samples came. For thisreason samples were codedon the score plot accordingto the farm. Except for thosesamples that came from thesame farm in Martina Fran-ca (M1), that are groupedtogether, this effect is notclear and a trend cannot bedetermined.

From the loadings of the

Table 2 - Mean values and standard deviations ofthe parameters determined for the milk samplesfrom the two groups. All metals are in ppm, iso-topic ratios are in δ‰, and P is in mg/100 g.

Variables Murgia Foggia

mean SD mean SD

Ba 0.49 0.15 0.69 0.12Mn 0.11 0.04 0.09 0.04Fe 1.18 0.45 1.35 0.53Zn 16.19 7.14 17.48 5.99Al 10.89 8.81 6.51 1.69Cu 0.32 0.23 0.31 0.10δ15N 3.15 0.30 4.15 0.59δ13C -24.22 1.28 -21.09 0.78P 85.57 25.94 76.81 8.89

Fig. 1 - Scatter plot of the scores on the first two principal compo-nents PC1 and PC2 obtained from metals (Ba, Mn, Zn, Al, Fe andCu), phosphorous and isotopic ratios (13C/12C and 15N/14N). Sam-ple labels indicate the farm of origin.

variables on the first and second prin-cipal components (Fig. 2), barium, δ15Nand δ13C were the dominant variablesin the first principal component, repre-senting 24.78% of the total variance;while zinc dominated in the second prin-cipal component, representing 19.25%of the total variability. HCA was carriedout on the first five principal compo-nents. The dendrogram in Fig. 3 showsthe clustering of milk samples on thebasis of their geographical origin, con-firming the results of PCA. The compo-


Fig. 2 - Loadings of the variables on the first two principal com-ponents obtained from metals (Ba, Mn, Zn, Al, Fe and Cu), phos-phorous and isotopic ratios (13C/12C and 15N/14N).

Fig. 3 - Dendrogram of milk samples obtained using the first fivePrincipal Components. Murgia (A, N, M1, M2); Foggia (T, C, V, F, L).

sition of the clusters indicates that theparameters used gave information use-ful for classifying the samples. At a sim-ilarity level of 12, there were four clus-ters, which can be identified as follows:the first two clusters contained 10 sam-ples from Foggia and 2 from Murgia; thethird and fourth clusters included milksonly from Murgia.

Discriminant analysis was then ap-plied in order to classify the milk sam-

ples into groups. In order touse this classification meth-od appropriately, the ratiobetween the number of var-iables and the number ofsamples should be ≤1:3 (DE-FERNEZ and KEMSLEY,1997). For this reason LDAwas performed on the firstfive PCs. Samples were ran-domly split into a trainingset, formed by 66% of thesamples (7 milk samplesfrom Murgia and 7 from Fog-gia), in order to develop adiscriminant model and avalidation set, formed by theremaining 33% of the sam-ples (3 from Murgia and 3from Foggia), on which themodel could be tested. Thisdivision provided a suffi-cient number of samples inthe training set as well as arepresentative number ofmilk samples in the test set.In order to obtain a good es-timate of the stability of themodel, this validation testwas repeated three timesand each time a differentsubset was selected for thevalidation set. In all casesthe prediction ability of themodel for the training setwas 100% and the samemodel applied to the valida-tion set gave 100% success.The training and test milk

samples on the canonical function ob-tained with LDA are shown in Fig. 4.

CONCLUSIONS

This preliminary study has demon-strated that multivariate analyses ap-plied to a chemical data set are capableof extracting information that is usefulfor classifying cow milk samples accord-


ing to their geographical or-igin. The most discriminat-ing variables were the iso-topic ratios (15N/14N and 13C/12C) and Ba concentration.This confirms the fact thatisotopic data carry impor-tant information about thegeographical origin of foodproducts and its discrimi-nating potential can be en-hanced by including metalconcentrations.

The results obtained canbe considered a good start-ing point for extending thisstudy to include a broaderbase of milk samples fromcows at different stages of lac-tation and from various pro-duction areas as well as samples taken atdifferent times of the year. Other types ofmilk (buffalo, goat and sheep milk) willalso be considered. Further investigationsshould also be conducted on processeddairy products like cheeses. This could bea very important application of the proce-dure since the differentiation of geograph-ical origin is essential for the PDO classi-fication of these products. Finally, thistype of investigation could be used to de-termine misrepresentations in cheesesand other milk products which should dis-courage potential fraud and increase con-sumer confidence in PDO products.

ACKNOWLEDGEMENTS

Many thanks are due to Dr. P. D’Amico and Dr. C.Carrino for supplying the samples.

REFERENCES

Bréas O., Reniero F. and Serrini G. 1994. Isotoperatio mass spectrometry: analysis of wines fromdifferent European countries. Rapid Commun.Mass Spectrom. 8: 967.

Brescia M.A., Caldarola V., De Giglio A., BenedettiD., Fanizzi F.P. and Sacco A. 2002a. Charac-

Fig. 4 - Plot of milk samples on the canonical function obtainedusing the first five PCs to differentiate samples according to geo-graphical origin. The assignment of test set samples is also rep-resented.

terization of the geographical origin of Italianred wines based on classical and NMR determi-nations. Anal. Chim. Acta 458: 177.

Brescia M. A., Di Martino G., Guillou C., RenieroF., Sacco A. and Serra F. 2002b. Differentiationof geographical origin of durum wheat semoli-na samples using isotopic composition. RapidCommun. Mass Spectrom. 16: 1.

Defernez M. and Kemsley E. K. 1997. The use andmisuse of chemometrics for treating classifica-tion problems. Trends Anal. Chem. 16: 216.

E.U. (European Communities). 1992. Regulation2081/92. Off. J. Eur. Communities L208.

Frank I.E. and Todeschini R. 1994. “The Data Anal-ysis Handbook”. p. 339 Elsevier Science B.V.,Amsterdam, The Netherlands.

Kornexl B. E., Werner T., Rossmann A. andSchmidt H. L. 1997. Measurements of stableisotope abundances in milk and milk ingredi-ents – a possible tool for origin assignment andquality control. Z. Lebensm. Unters. Forsch. A.205: 19.

Manca G., Camin F., Coloru G.C., Del Caro A., De-pentori D., Franco M.A. and Versini G. 2001.Characterization of the geographical origin ofPecorino sardo cheese by casein stable isotope(C-13/C-12 and N-15/N-14) ratios and free ami-no acid ratios. J. Agric. Food Chem. 49: 1404.

Martin G.J., Mazure M., Jouitteau C., Martin Y.-L., Aguile L. and Allain P. 1999. Characteriza-tion of Bordeaux wines by a combined use ofisotopic and trace element measurements. Am.J. Enol. Vitic. 50: 409.

Massart D.L., Vandeginste B.G.M., Deming S.N.,Michotte Y. and Kaufman L. 1988. “Chemomet-rics: A Textbook.” p. 154. Elsevier Science B.V.,Amsterdam, The Netherlands.


Paramas A.M.G., Barez J.A.G., Garcia-VillanovaR.J., Pala T.R., Albajar R.A., and Sanchez J.S.2000. Geographical discrimination of honeysby using mineral composition and commonchemical quality parameters. J. Sci. Food Ag-ric. 80: 157.

Rebolo S., Peña R.M., Latorre M.J., Garcia S., Bo-tana A.M. and Herrero C. 2000. Characterisa-

Paper received November 26, 2002 Accepted February 24, 2002

tion of Galician (NW Spain) Ribeira Sacra winesusing pattern recognition analysis. Anal. Chim.Acta. 417: 211.

Rossmann A., Kornexl B. E., Versini G., Pichlmay-er F. and Lamprecht G. 1998. Origin assign-ment of milk from alpine regions by multiele-ment stable isotope ratio analysis (Sira). J. FoodSci. Nutr. 27: 9.


PAPER

- Key words: tartaric and citric acids; storage temperatures and atmospheric conditions -

A METHOD FOR ANTHOCYANINEXTRACTION FROM FRESH GRAPE SKIN

UN METODO PER L’ESTRAZIONE DELLE ANTOCIANINEDA BUCCE FRESCHE D’UVA

G. SPAGNA*, R.N. BARBAGALLO*, A. TODARO, M.J. DURANTE1

and P.G. PIFFERI2

Dipartimento di Orto-Floro-Arboricoltura e Tecnologie Agroalimentari -Sez. Tecnologie Agroalimentari, Università di Catania, Via S. Sofia 98,

95123 Catania, Italy1 Qualitas Consultoria S/C R. Vereador Rocha e Silva, 11035-140 -

Santos-SP, Brazil2 I.D.Tech s.rl., via degli Ortolani 3, 40139 Bologna, Italy

* Corresponding authors: Tel. +39 095 7142255; Fax + 39 095 7141960;E-mail: [email protected]; [email protected]

ABSTRACT

Tartaric and citric acid solutions wereused to extract anthocyanins from freshgrape skins, in a non continuous proc-ess. The type of solvent and concentra-tion were significant factors in the ex-traction operation, and tartaric acid wasmore efficient than citric acid in theextraction yield. Comparative trialswere conducted using sulphur dioxideand acidified ethanol as solvents. Theextraction yields using the optimal tar-taric acid solution differed slightly fromthose obtained with acidified ethanol

RIASSUNTO

Sono state impiegate soluzioni di aci-do tartarico e citrico per l’estrazione, indiscontinuo, di antocianine da buccefresche d’uva. Natura e concentrazionedel solvente costituivano i fattori signi-ficativi per la riuscita del sistema, conuna maggiore efficienza dell’acido tar-tarico rispetto al citrico quanto a resed’estrazione. Prove comparate impiega-vano come solventi anidride solforosaed etanolo acidificato. Con le soluzionidi acido tartarico i valori delle resed’estrazione differivano leggermente ri-


but were higher than those obtainedwith sulphur dioxide. Total amounts ofpolyphenols, proanthocyanidins andflavans were determined in the extractsobtained from the tartaric acid solu-tions. Solvent concentration was not asignificant factor for flavan extractionyield but was highly significant for theextraction of the other compounds. Thestability of the anthocyanins was eval-uated at four different storage temper-atures and atmospheric conditions. Thecombined effects of low pH of the ex-tract (2.4), low temperature (2.0°C) andmodified atmosphere (N2) provided along shelf-life. Based on the results, a0.75% tartaric acid solution is recom-mended for anthocyanin extractionfrom fresh grape skin, and could sub-stitute the widely used sulphur diox-ide.

spetto a quelli ottenuti impiegando eta-nolo acidificato, in ogni caso più eleva-ti di quelli ottenuti con anidride solfo-rosa. Negli estratti ottenuti dalle solu-zioni di acido tartarico sono stati ana-lizzati, polifenoli totali, proantocianidi-ne e flavani. La concentrazione del sol-vente non costituiva un fattore signifi-cativo per l’estrazione dei flavani, men-tre risultava altamente significativo aifini dell’estrazione di altri composti. Lastabilità delle antocianine è stata valu-tata secondo quattro diverse condizio-ni di temperatura e d’atmosfera. Gli ef-fetti combinati di basso pH dell’estrat-to (2,4), bassa temperatura (2,0°C) edatmosfera modificata (N2) davano luo-go ad un prodotto con un’elevata shelf-life. In base ai risultati, è raccomanda-ta per l’estrazione di antocianine dabucce fresche d’uva una soluzione diacido tartarico allo 0,75% che potreb-be sostituire l’anidride solforosa larga-mente utilizzata.

INTRODUCTION

Judging of food quality beyond itsmicrobiological and nutritional charac-teristics is greatly influenced by senso-rial properties such as colour, taste, fla-vour and texture. Depending on the food,these factors play different roles in thevaluation of overall quality. Colour, how-ever, is one of the most important prop-erties of foods and beverages and is abasis for their identification and accept-ability (CLIFFORD, 1996; HARDY, 2000).

Normally, food colour is due to natu-rally occurring pigments, but syntheticor artificial colorants are often added inorder to confer the desired colour to thefinal product. Because there is increas-ing consumer demand for natural foods,industrially practical and economic pro-duction of natural food colorants is adesirable goal for food ingredients indus-

tries. However, availability of natural pig-ment sources, extraction process re-quirements and stability of colorants andfitness of use must be taken into accountin the production of such ingredients.

Flavonoids are receiving increasing in-terest on the part of consumers and foodmanufactures for several reasons(KÜHNAU, 1976). Epidemiological studieshave suggested associations between theconsumption of polyphenol-rich foods orbeverages and prevention of diseases(STEINMETZ and POTTER, 1996; YANG etal., 2001). Anthocyanins, one of the mainclasses of flavonoids, appear to contrib-ute significantly to their powerful antiox-idative properties (VINSON, 1998; LEPIDOTet al., 1999). They are an importantsource of the naturally occurring color-ants of red fruits such as cherries, plums,strawberries, raspberries, blackberries,grapes, red currants and black currants


(JACKSON et al., 1978; FRANKEL et al.,1995; LEPIDOT et al., 1999). Restrictionsin the use of synthetic colorants in foodshave led to the interest in the potentialuse of anthocyanins as a food colorant inbeverages, syrups, fruit juices, jellies,jams, ice-creams, pastries candies andyoghurts, as well as in tooth paste, phar-maceuticals, cosmetics and similar prod-ucts (CLYDESDALE and FRANCIS, 1976;DURANTE, 1995).

Pigments present in the skins of redgrapes are industrially extracted usingextraction processes together with sul-phur dioxide. The production of this pig-ment is a way not only to recover a win-ery by-product but also to obtain a highlymarketable, valued ingredient. The com-mercial method of producing this complexof pigments may vary slightly from onemanufacturer to another, depending onthe characteristics of the pomace availa-ble and the market demand. Carpenè andComboni started to produce it in 1879using ethanol as solvent (DIECI, 1967;GAROGLIO, 1980). It was soon realized,however, that its recovery could be im-proved considerably with the use of a mildsolution of sulphur dioxide in water.

Attempts have been made to replacethe sulphur dioxide extraction processdue to health risks that this compoundcan cause, even in low concentrations.The effects of methanol, ethanol andwater, containing or not containing ace-tic and chloric acids, on the degree andrate of extraction were investigated byMetiver, as reported by HANG, (1988).CALVI and FRANCIS (1978) described aprocess of recovering anthocyanins fromgrape juice pomace with 0.01% citric acidin ethanol. The same method was usedby MAIN et al. (1978), but to produce aspray-dried colorant.

The present study was conducted todevelop a method to extract anthocy-anins from fresh grape skins using tar-taric and citric acid solutions as solvents,to evaluate if this could be a valid alter-native for the use of sulphur dioxide. The

optimal solvent concentration and theyield and rate of extraction were investi-gated, and compared with sulphur di-oxide and acidified ethanol. Stability ofthe extracts in four different storage con-ditions was evaluated.


Grape harvesting and skin removal

Vitis vinifera grapes var. Ancellotta werepurchased directly from producers in eightlocations around Reggio Emilia (Italy).Grapes were harvested at the industrialmaturation state. All 8 vineyards were lo-cated in the plains (20 m a.s.l.) on homo-geneous land, with similar climatic condi-tions (temperature and relative humidityranged from 15° to 30°C and from 40 to60%, respectively, under normal levels ofrainfall). The fresh grapes were stored at2.0°C in nitrogen atmosphere for 24 h.Grapes were sampled from 5 randomlyselected clusters of each vineyard. Berrieswere snipped from the top, middle andbottom of each cluster, and from the in-ner and outer surface. To facilitate skinand pulp tissue separation the whole ber-ries were put into an ethylene-glycol bathkept at a constant temperature of -35°Cfor 4 min; they were then immediately putin a water bath at 18°C for 30 sec. Afterthe berries were removed from the waterbath, they were thoroughly wiped withabsorbent paper. This procedure allowedonly the pulp to remain frozen, and in-creased the ease of skin removal whichwas done by squeezing the grapes betweenthe thumb and the forefinger.

Anthocyanin extraction

A non-continuous extraction processwas employed throughout. Ten-gramsamples of fresh skin (wet weight) weresoaked in 25 mL of an aqueous tartaricor citric acid solution at the desired con-centration (between 0.25 and 2.00%).


They were soaked with mild stirring at25°C for almost 20 min, which is the timeneeded to reach mass transfer equilibri-um, as described by DURANTE (1995).Immediately the extract was transferredto a dark glass bottle, in which nitrogenwas being pumped simultaneously (a ni-trogen flush was performed throughoutthe entire bottle, before and during fill-ing), and the bottles were then stored at2.0°C. Subsequently, a new 25 mL aliq-uot of solvent was added to the skins(moisture content 22.26%; peel thickness185 µm), and the extraction process de-scribed above was repeated. This proce-dure was repeated until the extractformed was clear and a constant anthocy-anin concentration was achieved. Thesame procedure was used for an aque-ous 0.20% sulphur dioxide solution asextracting solvent. A comparative extrac-tion was carried out using a hydroalco-holic-chloride solution (70:30) as extract-ing solvent in order to determine total ex-tractable anthocyanins (GLORIES, 1978).The values of pH in all the extractionsvaried with the concentration of citric andtartaric acid in comparison to the pH ofthe control (pH 1.60 at 25°C), the hy-droalcoholic-chloride solution. Samples(5.0 g) of fresh skins (wet weight) weresoaked in 50 mL of 70% ethanol contain-ing 1% HCl (v/v) and then slightly stirredat 25°C for 30 min. After each extrac-tion, the extract was put in a dark glassbottle, in which nitrogen was beingpumped simultaneously, and then storedat 2.0°C. After that, a new 50 mL aliquotof solvent was added to the skins andthe extraction operation was carried outas above. This procedure was repeated25 times until the solvent was complete-ly clear, confirmed by an absorbance<0.05. Difference tests were performedat different liquid/solid ratios, but theratio of 5.0 g of fresh skins soaked in 1.25L of solvent was the most advantageous.

All extractions were made in triplicateand the results are reported as the arith-metic mean. The t-Test was used to esti-

mate 95% confidence intervals of meansof all determinations.

Evaluation of extract stability

Four extraction runs were carried outsimultaneously using a 0.75% tartaricacid solution as the extracting solvent(highest extraction yield) following themethod described above. The first extractsof each run were put in dark glass bot-tles and stored in four different condi-tions: in air at 2.0° and 25°C and in ni-trogen atmosphere at 2.0° and 25°C. Theheadspaces of the bottles were the sameas the external storage atmosphere. Thethree extracts from each run were addedto the bottles and after all the extractswere mixed and homogenized, a “zerotime” sample was taken for each run.Sampling was done at 30-day intervalsfor 240 days and the samples were im-mediately analysed. The stability wasevaluated as colour intensity, which is re-lated to colour quantity, according toPOMPEI et al. (1983). The analyses wereperformed without any preparation of thesamples by reading their absorbances at420 and 520 nm on a Cary 1E spectro-photometer (Varian, Mulgrave-Victoria,Australia). Colour intensity was estimat-ed as the sum of both absorbance values(E420 + E520). The first sum was taken as100% of colour intensity and extract sta-bility was evaluated as decrease in col-our intensity as a function of time.

The extracts from tartaric acid solu-tions were analysed for total anthocy-anins, total polyphenols, flavans andproanthocyanidins. With slight modifi-cation, total anthocyanin analysis wasperformed using the spectrophotomet-ric technique described by DI STEFANOet al. (1989). According to RIGO et al.(2000), the sample was diluted 20 to 40times with the hydroalcoholic-chloridesolution (70:30) cited above and absorb-ances were read at the wavelength ofmaximum absorption in the visiblerange (536-540 nm) on a Cary 1E spec-


trophotometer (Varian, Mulgrave-Victo-ria, Australia). The total anthocyaninswere directly quantified on the basis ofthe maximum absorbance in the visiblerange (536-540 nm) against a blank (hy-droalcoholic-chloride solution) Theamount of the extracted pigment wascalculated in milligrams/liter by assum-ing an average absorbance of the mix-ture of anthocyanins (average MW=500)extracted from Cabernet Sauvignongrapes to equal 18,800 M-1 cm-1 (GLO-RIES 1984; DI STEFANO et al., 1989).

Total polyphenol analysis was per-formed using the Folin-Ciocalteau rea-gent and the phenolic compounds redoxreaction method, in alkaline medium, asdescribed by SINGLETON and ROSSI(1965); results are reported as gallic acid.Flavan determinations were made usingthe strong acidic vanillin reaction meth-od described by MARGHERI and FALCIERI(1972); results are reported as catechins.Proanthocyanidins were estimated by

Fig. 1 - Anthocyanins extracted using tartaric and citric acids as solvents. (�) Tartaric acid, 20 extrac-tions for 0.75, 1.0 and 1.5% concentrations, 24 extractions for all other concentrations; (�) citric acid,25 extractions for all concentrations. Data are the means of five replicates; P<0.05.

their transformation into anthocyaninsin hot acidic media, according to themethod described by MARGHERI and FAL-CIERI (1972); results are reported as cy-anidin chloride. The solvents and rea-gents not expressly specified had a highdegree of purity (RPE) and were suppliedby Carlo Erba (Rodano, Italy). All analy-ses were done in triplicate and the re-sults are reported as the arithmeticmean. The t-Test was used to estimate95% confidence intervals of the meansof all determinations.


Anthocyanin extraction

Tartaric and citric acids were effectivein extracting anthocyanins from freshgrape skins. The pigment extracted as afunction of solvent concentration isshown in Fig. 1. Both curves had simi-


lar trends with a higher extraction yieldand rate at the lower concentrations oftartaric acid. The nature of the acids af-fected the extraction yield and rate sig-nificantly (p≥0.05). This difference couldbe due to the larger molecular size of cit-ric with respect to tartaric acid, makingit more difficult for citric acid to diffuseinto intra-lamellar regions of the skintissue, where the pigments are present.The extraction processes are based onthe stability of anthocyanins in polarsolvents. Anthocyanins are found in na-ture as cations bound to organic acidanions.

A decrease in pH from 2.90 to 2.30(pH interval of the tartaric acid solutionsused) increased extraction of the an-thocyanins, while an increase in pH de-creased the dissociated form of tartaricacid; equilibrium was reached at 0.75%concentration. The optimal concentra-tion of tartaric acid could be explainedby a hypotonic condition formed in lowconcentration solutions, which causes

water absorption by skin cells. As theacid concentration increases, the cellsswell continuously up to a maximumstate and then break, liberating the pig-ments which were in their vacuoles. In-stead, at high concentrations, a hyper-tonic situation develops, in which thecells lose water and harden, diminish-ing membrane permeability and protect-ing them from rupturing. At this point,anthocyanin molecules remain in thevacuoles and mass transfer becomes dif-ficult.

Extraction using different solvents isshown in Fig. 2. Sulphur dioxide is wide-ly used today in anthocyanin productionand the concentration used in this studyis the most commonly used in industri-al production (MAZZA et al., 1999; MAZ-ZA and MINIATI, 1993). Hydroalcoholic-chloride solutions are commonly used foranalytical purposes, in the isolation ofanthocyanins from fruits and vegetables,as they are capable of extracting all pig-ments contained in the plant materials

Fig. 2 - Anthocyanin extraction yield using different solvents. 0.20% sulphur dioxide, 15 extractions;0.75% tartaric acid, 20 extractions; 2.0% citric acid, 25 extractions. Data are the means of five repli-cates; error bars are ± S.D.; P<0.05.


(a)

(b)

(c)

(d)

Fig. 3 - Pigment extracted with tartaric acid. (a) anthocyanins; (b) total polyphenols reactive to theFolin-Ciocalteau; (c) proanthocyanidins; (d) flavans; (�) 0.25%; (x) 0.50%; (�) 0.75%; (+) 1.0%, (�) 1.50%;(�) 2.0%, cumulative values. Data are the means of five replicates; error bars are ± S.D.; P<0.05.

(DI STEFANO et al., 1989; GLORIES, 1978;LAMIRANKA, 1989; SAN JOSÉ and DI STE-FANO, 1990; GLORIES, 1984; DI STEFANOet al., 1989), but it is more expensive thattartaric or citric acids. As the best ex-traction yield (21.91 g anthocyanins/kgfresh skin) was obtained with this sol-vent, it was taken as the reference (100%of extractable anthocyanins) for the oth-er trials. Tartaric acid was more efficientthan sulphur dioxide (p≥0.05) but therewas no significant difference betweentartaric acid and the hydroalcoholic-chloride solution (p≤0.05). Sulphur di-oxide augments membrane permeabili-ty (CAPPELLI and VANNUCCHI, 2000) andbinds itself to anthocyanins, but highconcentrations are necessary when freshskins instead of spent pomace are used.

Besides anthocyanins, the extractionof other phenolic compounds with tar-taric acid were investigated and the re-sults are presented in Fig. 3. With re-spect to anthocyanins, there were threedifferent levels of extraction rate andyield (Fig. 3a). Low and high concentra-tions (0.25, 0.50 and 2.0%) gave thesame profile (p<0.05) with respect to thetotal amount extracted, but the extrac-tion rate was higher with the 2.0% so-lution. After 9 extractions, 97% of thetotal pigment was extracted, whilst withother two solutions (0.25 and 0.50%),16 extractions were necessary to reach97%. Instead, with the concentrations of0.75 and 1.0%, it was possible to extractmore pigment more rapidly. Three ex-tractions yielded more pigment than the


total obtained with all other solutions,and ten extractions were necessary toachieve 95% of the final anthocyaninconcentration. The 1.5% solution gavean intermediate profile, for both extrac-tion yield and rate. The differences in theamount extracted and the rate of extrac-tion can be explained by the theory ofhypotonic and hypertonic conditions dis-cussed above. The extraction profile oftotal polyphenols (Fig. 3b) and proan-thocyanidins (Fig. 3c) were quite similarto that of anthocyanins. The finalamounts obtained for each analyses wereproportional to those of anthocyaninsattained in the corresponding analyses.In addition, the rates of extraction werethe same as for anthocyanins. In the to-tal polyphenol extraction, the curve ofthe 1.5% solution also gave an interme-diate profile but was closer to the firstgroup of curves (0.25, 0.5 and 2.0%). Thehigh amounts of polyphenols obtainedcould be due to the low pH of the solu-tions (from 2.0 to 3.0), which enhances

polyphenol extraction (POMPEI, 1983),but the hypertonic state formed by highconcentrations of tartaric acid could havehindered extraction, suggesting that pHplays a secondary role in the efficacy ofthe process. A combination of pH andsolvent concentration indicate the 0.75%solution to be the best solvent. Flavanextractions are shown in Fig. 3d. Basedon the results, the extraction perform-ance gave a somewhat different profilewith respect to the other compounds.Solvent concentration did not affect ex-traction yield and rate significantly(p<0.05). As total amounts of polyphe-nols attained were high, the results in-dicate that anthocyanins and proan-thocyanidin are the dominant extractedconstituents and therefore extractionwith tartaric acid is preferential to thesecompounds but not to flavans. All ex-tractions were carried out at a constantlow temperature (25°C) in order to min-imize pigment degradation by heat.

The colour intensity decreased in the

Fig. 4 - Anthocyanin stability in different storage conditions (2° and 25°C) and atmosphere (air andnitrogen) at constant pH (2.4); (�) N2 at 2°C; (�) air at 2°C; (*) N2 at 25°C; (�) air at 25°C. Data are themeans of five replicates; error bars are ± S.D.; P<0.05.


different storage conditions as shown inFig. 4. Therefore, the combined effectsof low pH of the extract (2.4), low tem-perature, protection against light, highconcentration of phenols and polyphe-nols and modified atmosphere resultedin a long and stable product.

CONCLUSIONS

The use of tartaric acid as solvent inanthocyanin extraction has been shownto be a simple and inexpensive method-ology, which provides a high extractionyield and a stable extract using mild stor-age conditions. Besides, tartaric acidsolutions extract anthocyanins andproanthocyanidins more easily than fla-vans. Because, tartaric acid can be eas-ily removed from the extract, it could besuitable to substitute the commonlyused sulphur dioxide used industrially.

ACKNOWLEDGEMENTS

This research was supported by Tecnoalimentis.p.a. (Milano, Italy) and Cantine Riunite (Reg-gio Emilia, Italy), grant n° IMI 57416 (PatentIT.01268863).

REFERENCES

Calvi J.P. and Francis F.J. 1978. Stability of con-cord grape (V. labrusca) anthocyanins modelsystems. J. Food Sci. 43: 1448.

Cappelli P. and Vannucchi V. 2000. Bevande Al-coliche e Aceto. In “Chimica degli Alimenti”. p.585. Bologna, Italy.

Clifford M.N. 1996. Anthocyanins in Food. Sym-posium on Polyphenols and Anthocyanins asFood Colourants and Antioxidants. Brussels,Belgium.

Clydesdale F.M. and Francis F.J. 1976. Pigments.In “Principles of Food Chemistry, Part I, FoodChemistry”, 1st ed. Marcel Dekker, New York.

Di Stefano R., Cravero M.C. and Gentilini N. 1989.Metodi per lo studio dei polifenoli dei vini.L’Enotecnico 5: 83.

Dieci E. 1967. Sull’enocianina tecnica. Riv. Vitic.Enol.12: 567.

Durante M.J., Pifferi P.G., Spagna G. and GilioliE. 1995. Partial characterization of Vitis vinife-ra grape var. Ancellotta, Lebensm. Wiss. u. Te-chnol. 28: 635.

Frankel E.N., Waterhouse A.L. and Teissedre P.L.1995. Principal phenolic phytochemicals in se-lected California wines and their antioxidantactivity in inhibiting oxidation of human low-density lipoproteins. J. Agric. Food Chem. 43:890.

Garoglio P.G. 1980. L’enocianina. In “Enciclope-dia Vitivinicola Mondiale”. p. 131, 427. Brescia,Italy.

Glories Y. 1978. Recherches sur la matière color-ante des vins rouges. Thèse a l’Universitè deBourdeaux II. p. 126, Bourdeaux, France.

Glories Y. 1984. La couleur des vins rouges. Con-nais. Vigne Vin. 18: 195.

Hang Y.D. 1988. Recovery of food ingredients fromgrape pomace. Process Biochem. 23: 2.

Hardy G. 2000. Nutraceutical and functional foods:introduction and meaning. Nutrition 16: 688.

Jackson M.G., Timberlake C.F., Bridle P. and Val-lis L. 1978. Red wine quality: correlation be-tween color, aroma and flavor and pigment andother parameters of young Beaujolais. J. Sci.Food Agric. 29: 715.

Kühnau J. 1976. The flavonoids: a class of semi-essential food components; their role in humannutrition. World Rev. Nutr. Diet. 24: 117.

Lamiranka O. 1989. Anthocyanins of Vitis rotun-difolia hibrid grapes. Food Chem. 33: 225.

Lepidot T., Harel S., Akiri B., Granit R. and Kan-ner J. 1999. pH-dependent forms of red wineanthocyanins as antioxidants. J. Agric. FoodChem. 47: 67.

Main J.H., Clydesdale F.M. and Francis F.J. 1978.Spray-drying anthocyanin concentrates for useas food colorants. J. Food Sci. 43: 1693.

Margheri G. and Falcieri E. 1972. Polyphenolicsubstances in quality red wines during stor-age. Vini d’Italia 81: 561.

Mazza G. and Miniati E. Grapes. 1993. Introduc-tion. In “Anthocyanins in Fruits, Vegetables andGrains”. p. 362. CRC Press, Boca Raton FL.

Mazza G., Fukomoto L., Delaquis, P., Girard B. andEwert B. 1999. Anthocyanins, phenolics, andcolor of Cabernet Franc, Merlot, and Pinot Noirwines from British Columbia. J. Agric. FoodChem. 47: 4009.

Pompei C., Casiraghi E.M. and Lucisano M. 1983.Influenza di alcuni parametri sull’estrazione ela conservazione dell’enocianina, I. Andamentodel processo di estrazione. Riv. Vit. Enol. diConegliano 36: 71.

San José M.L.G. and Di Stefano R. 1990. Factorsaffecting anthocyanins stability in solution. Riv.Vit. Enol. 3: 63.

Singleton V.L. and Rossi J. 1965. Colorimetry of


total phenolics with phosphomolybidic-phos-photungstic acid reagents. Am. J. Enol. Vitic.16: 144.

Steinmetz K.A. and Potter J.D. 1996. Vegetables,fruit, and cancer prevention: a review. J. A. Diet.Assoc. 96: 1027.

Vinson J.A. 1998. Flavonoids in foods as in vitro andin vivo antioxidants. Adv. Exp. Med. Biol. 439: 151.

Yang C.S., Landau J.M., Huang M.-T. and New-mark H.L. 2001. Inhibition of carcinogenesis bydietary polyphenolic compounds. Ann. Rev.Nutr. 21: 381.

Revised paper received February 24, 2003 Accepted May 29, 2003


PAPER

- Key words: anthocyanins, flavonol glycosides, red-berry varieties,tartaric esters of hydroxycinnamic acids (HCTA) -

PHENOLIC COMPOUNDSIN RED-BERRY SKINS OF UVA DI TROIA

AND BOMBINO NERO GRAPES(VITIS VINIFERA L.)

COMPOSTI FENOLICI IN BUCCE DI UVE A BACCA NERADELLE VARIETÀ UVA DI TROIA E BOMBINO NERO (VITIS VINIFERA L.)

P. TAMBORRA, M. ESTI1*, M. MINAFRA and F. SINESIO2

Istituto Sperimentale per l’Enologia, Via Vittorio Veneto 26, 70051 Barletta, Italy1 Dipartimento di Scienze e Tecnologie Agroalimentari,

Università degli Studi della Tuscia, Via S. Camillo De Lellis, 01100 Viterbo, Italy2 Istituto Nazionale di Ricerca per gli Alimenti e la Nutrizione (INRAN),

Via Ardeatina 546, 00178 Roma, Italy* Corresponding author: Tel. +(39) 0761-357497; Fax +(39) 0761-357498;

E-mail: [email protected]

ABSTRACT

The amounts of anthocyanins, flavo-nol glycosides and hydroxycinnamoyltartaric acids in the skins of two redgrape varieties (Vitis vinifera L.) fromsouthern Italy, Bombino nero and Uvadi Troia, were determined. Arbor andtrellis training systems in the differentgrowing areas were also compared. Skinextracts were analyzed by a directHPLC-DAD method for the simultane-ous detection of the selected flavonoidsand phenolic acids. Although the phe-nolic profiles were similar in the two

RIASSUNTO

Il contenuto e la composizione inantociani, flavonoli glicosilati e acidiidrossicinnamil tartarici, sono stati de-terminati sulle bucce di Uva di Troia eBombino nero (Vitis vinifera L.), varie-tà a bacca rossa tipiche del sud Italia.Le uve, oggetto di sperimentazione,sono state campionate in differentiaree di produzione ed in vigneti consistemi di allevamento diversi. Un me-todo diretto HPLC-DAD è stato impie-gato per l’analisi simultanea dei diffe-renti costituenti fenolici presenti negli


varieties, the absolute content of acylat-ed forms of malvidin and peonidin-3-(6-O-p-coumaroyl-glucoside) was high-er in Uva di Troia. This particularacylated anthocyanin content could beuseful to promote the varietal charac-ter of the respective rosé and young redwines. Hydroxycinnamoyl tartaric ac-ids were significantly affected by thearea of production and training system.

estratti delle bucce. Il profilo fenolicodelle bucce delle due varietà è risulta-to simile, tuttavia il contenuto delleforme acilate della malvidina e dellapeonidina-3-p-cumarilglucoside nel-l’Uva di Troia era significativamentepiù elevato rispetto al Bombino nero.Questa peculiare dotazione in antocia-nine acilate potrebbe costituire un ca-rattere varietale distintivo anche per irelativi vini rossi e rosati. L’area di pro-duzione e il sistema di allevamentohanno, invece, significativamente in-fluenzato la dotazione di acidi idrossi-cinnamil tartarici in entrambe le va-rietà.

INTRODUCTION

The phenolic composition of a winedepends mainly on the phenolic con-tent of the grapes from which the wineis made, juice extraction and wine-making techniques, and on the nu-merous reactions that occur duringwine aging. The initial phenolic con-tent of the berries is of great taxonomicimportance in viticulture for classify-ing grape cultivars (MACHEIX et al.,1990; BOSS et al., 1996). Moreover, thesynthesis and concentration of phe-nols in red grapes depend on a numberof environmental factors, particularlyfruit ripeness, climatic conditions, soilfeatures and crop load (JACKSON andLOMBARD, 1993; REYNOLDS et al.,1996; ROUBELAKIS-ANGELAKIS andKLIEVER, 1986).

In making rosè and young or agedred wines, the phenolic compounds inthe grape skin, particularly anthocy-anins, have a relevant role in the choiceof berry crushing and pressing tech-niques and vatting time (GAO et al.,1997; GOMEZ-PLAZA et al., 2000). More-over, in recent studies specific health-related properties, especially with re-

spect to heart disease, have been as-cribed to particular flavonoids fromgrape skin: anthocyanins (WANG et al.,1997) and flavonols (HERTOG et al.,1997).

A number of studies have demonstrat-ed the key role of the phenolic com-pounds in grapes in the characterizationand classification of widespread inter-national grape varieties (MAZZA et al.,1999); however, few data (ROMANI et al.,1996) are available about typical Italiangrapes used for wines destined to a nicheconnoisseur market.

The aim of this study was to investi-gate two autochthonous red grape vari-eties (Vitis vinifera L.) from Apulia (TAM-BORRA et al., 2000), a key wine-produc-ing region in southern Italy, in order toimprove the winemaking techniques,and to promote the varietal characterof the respective young red wines. Toachieve this purpose the qualitative andquantitative anthocyanin, flavonol gly-coside and hydroxycinnamoyl tartaricacid (HCTA) profiles of berry skins ofBombino nero and Uva di Troia varie-ties, used for the production of Control-led Denomination of Origin Castel delMonte wines, were investigated in rela-


tion to various production areas andtraining systems.


Plant materialGrape samples of Bombino nero cv.

and Uva di Troia cv. were harvested attechnological ripeness in 5 and 8 vine-yards, respectively. The vineyards werelocated in the areas delimited for theregion of Controlled Denomination ofOrigin Castel del Monte, and arbor andtrellis training systems were used. Ineach vineyard, 300 berries were collect-ed from 30 bunches produced by vineswith similar vegetative and productivecharacteristics (age, amount of buds).Berries were picked, cutting and leav-ing part of the peduncle, from differ-ent zones of the bunch. A random se-lection of 3 replicates of 10 berry subsamples were weighed before phenolicextraction.

Extraction of phenolic compoundsPhenolic extraction was carried out

according to DI STEFANO and CRAVE-RO (1991). The berry skins were man-ually removed from the pulp, driedwith filter paper, and then immersedin an HCl/water/ethanol solution (1/30/ 70) for 20 hours (in the dark andat room temperature) to keep the ex-tracts.

Chemicals and reference compoundsThe reagents (Carlo Erba, Milan, Ita-

ly) were analytical or HPLC grade, as re-quired. Malvidin 3-glucoside chlorideand rutin were purchased from Extra-synthèse (Genay, France); caffeic acid,and quercitin dihydrate were from Flu-ka (Buchs, Switzerland).

HPLC-DAD analysisFor the phenolic fractionation, in or-

der to save solvents and to reduce timeand the number of analyses, the ex-

tract was diluted 5 times with water,filtered (0.2 µm) and analyzed by di-rect HPLC-DAD separation. Therefore,the compound identification andquantification were carried out withminimal pre-treatment of the sample(LAMUELA-RAVENTOS and WATER-HOUSE, 1994; TAMBORRA and MINA-FRA, 1998; RITCHEY and WATER-HOUSE, 1999).

Detection and quantification of phe-nolic compounds were carried out byHPLC using an HP 1090 L apparatus(MILFORD, MA, USA) with a photodiodearray detector and an injection valve (20µL loop). The stationary phase was aSymmetry C18 column, 4.6x100 mm, 3.5µm particle size from Waters (MILFORD,MA, USA). The flow rate was 0.5 mL min-1

and the solvents used for the separationwere: Solvent A = 50 mM dihydrogenammonium phosphate adjusted to pH2.6 with orthophosphoric acid; SolventB = 20% A with 80% acetonitrile; andSolvent C = 0.2 M orthophosphoric acidadjusted with ammonia to pH 1.5. Table1 gives the solvent gradient conditions.

The eluates were detected at 25°C, at280 nm for all phenols, at 313 nm forhydroxycinnamic derivatives, at 365 nmfor flavonols and at 520 nm for anthocy-anins. In addition, UV-visible absorp-

Table 1 - Solvent gradient conditions with lineargradient.

Time Solvent A Solvent B Solvent C(min) (%) (%) (%)

0 100 0 05 100 0 0

15 96 4 025 92 8 025.01 0 8 9245 0 20 8050 0 30 7055 0 40 6060 0 80 2065 100 0 0


tion spectra (260-600 nm) of each peakwere obtained. The identification wasperformed by chromatographic separa-tion by HPLC and by comparing spec-tra with authentic standards, whenavailable, and with data from the liter-ature.

Each single anthocyanin, hydroxycin-namoyl tartaric acid (HCTA) and flavo-nol was quantitatively determined (mg/kg of berry) from the areas of the chro-matographic peaks, using calibrationcurves prepared from standards of mal-vidin 3-monoglucoside, caffeic acid andquercitin, respectively. The phenol con-centration in each sample is the meanof three replicates.

StatisticsA three-way Analysis of Variance (va-

riety * production area * training sys-tem), fixed effects and interactions, wasperformed on each chemical parameterto get an indication of the variation inthe material and level of error of thedata.

In addition, Principal ComponentAnalysis (PCA) of the correlation ma-trices was carried out on all variablessimultaneously in order to summarisethe information on differences betweensamples. Each variable was standard-ised at unit variance prior to analy-sis.

The analyses were performed usingSPSS (1991), for univariate analyses,and UNSCRUMBLER Software version7.01 (CAMO ASA, Trondheim, Norway)for PCA.


The HPLC-chromatograms of Bombinonero berry skin extracts are reported inFig. 1.

Results of the three-way Analyses ofVariance (Table 2) showed significantdifferences between the varieties fortheir content in cinnamoyl tartaric ac-

Fig. 1 - HPLC-chromatograms of Bombino nero ber-ry skin extract, recorded at 313 nm (A) peaks:1=cis-caftaric acid, 2=trans-caftaric acid, 3=cis-coutaric acid, 4=trans-coutaric acid; at 365 nm (B)peaks: 1=myricetin-3-O-glucuronide, 2=myricetin-3-O-glucoside, 3=rutin, 4=quercitin-3-O-glucuro-nide, 5=quercitin-3-O-glucoside, 6=kaempferol-3-O-glucuronide, 7=kaempferol-3-O-glucoside; andat 520 nm (C) peaks: 1=delphinidin-3-O-glucoside,2=cyanidin-3-O-glucoside, 3=petunidin-3-O-gluco-side, 4=peonidin-3-O-glucoside, 5=malvidin-3-O-glucoside, 6=peonidin-3-O-acetylglucoside, 7=mal-vidin-3-O-acetylglucoside; 8=malvidin-3-O-caffe-oylglucoside, 9=petunidin-3-(6-O-p-coumaroyl-glu-coside), 10=peonidin-3-(6-O-p-coumaroyl-gluco-side), 11=malvidin-3-(6-O-p-coumaroyl-glucoside)

A

B

C

313 nm

365 nm

520 nm


factors (DI STEFANO and MAGGIOROT-TO, 1995), the skin content of hydrox-ycinnamate is significantly affected bythe area of production and training sys-tem.

Phenol profiles of the two varieties forthe different areas of production andtraining systems are listed in Tables 3and 4.

Among the phenolic acids in the ber-ry skin, the most abundant non-flavo-noid phenol was trans-caftaric acid(62% of total HCTA). BOURSIQUOT(1987) found the percentage of this acidto vary from 36% in the skins of Carig-

Table 2 - Analysis of Variance. F-values for effects (variety, production area, training system and inter-actions) on phenolic compounds in red-berry skins of Uva di Troia and Bombino nero grapes (Vitisvinifera L.).

Variables F-Variety F-Production F-Training F-Variety F-Variety F-AreaArea System x Area x System x System

cis-caftaric acid 82.8 *** 2.5 0.0 2.8 0.1 2.8trans-caftaric acid 33.9 ** 7.1 * 11.0 * 5.0 6.3 1.1cis-coutaric acid 93.6 *** 19.0 * 33.7 ** 24.6 ** 15.6 * 1.2trans-coutaric acid 37.3 ** 12.7 * 9.6 * 13.7 * 4.4 0.9delphinidin-3-O-glucoside 0.3 3.7 0.0 0.5 0.5 0.1cyanidin-3-O-glucoside 0.1 2.9 0.0 0.4 1.6 0.3petunidin-3-O-glucoside 0.3 4.6 0.0 0.6 0.0 0.1peonidin-3-O-glucoside 2.1 4.3 2.4 1.5 0.9 3.0malvidin-3-O-glucoside 0.1 2.8 0.1 0.0 0.4 0.8delphinidin-3-O-acetylglucoside 1.9 0.2 0.9 0.5 2.4 0.6peonidin-3-O-acetylglucoside 1.1 0.0 1.0 0.5 1.5 0.2malvidin-3-O-acetylglucoside 13.2 * 0.0 1.8 0.4 6.4 0.7malvidin-3-O-caffeoylglucoside 15.9 * 1.3 0.9 0.6 8.5 * 0.8delphinidin-3-(6-O-p-coumaroyl-glucoside) 4.3 0.4 2.0 0.0 1.7 0.4petunidin-3-(6-O-p-coumaroyl-glucoside) 2.1 0.1 0.7 0.4 3.4 0.0peonidin-3-(6-O-p-coumaroyl-glucoside) 15.2 * 0.1 6.1 0.0 13.6 * 2.7malvidin-3-(6-O-p-coumaroyl-glucoside) 11.1 * 0.0 4.5 0.2 9.2 * 1.2myricetin-3-O-glucuronide 0.1 0.1 0.2 0.9 0.7 1.4myricetin-3-O-glucoside 1.6 1.2 0.7 0.3 1.1 1.3rutin 0.5 0.9 0.0 1.4 2.3 1.6quercitin-3-O- glucuronide 0.9 1.1 0.1 1.6 3.7 2.6quercitin-3-O- glucoside 0.9 5.6 1.8 4.7 1.0 2.9kaempferol-3-O-glucuronide 0.0 0.0 0.1 1.7 0.7 1.2kaempferol-3-O-glucoside 0.0 0.2 0.1 0.4 0.0 1.2

* P<0.05; ** P<0.01; *** P<0.001.

ids, acylated forms of malvidin and pe-onidin-3-(6-O-p-coumaroyl-glucoside),thus confirming the marked geneticbasis of the synthesis of these com-pounds present in the grapes (CALÒ etal, 1994) and wines (MATTIVI et al.,1995). There were also significant dif-ferences at the 95% confidence level forarea of production (trans-caftaric, cis-coutaric and trans-coutaric acids), andtraining system (trans-caftaric andtrans-coutaric acids, at 95% and cis-coutaric acid at 99% confidence, re-spectively). These results indicate that,besides being determined by genetic


Table

3 -

Ph

enol

ic c

once

ntr

ati

on (m

g/kg)

of

ber

ry s

kin

s fr

om B

ombin

o n

ero

grapes

.

Trai

ning

sys

tem

Arbo

rTr

ellis

Prod

uctio

n Ar

eaR

uvo

di P

uglia

Andr

iaR

uvo

diAn

dria

Cor

ato

Aver

age

SDC

alen

dano

S. D

omen

ico

Pugl

iaS.

Dom

enic

oBo

sco

(n=5

)(n

=5)

Cal

enda

no

Phen

olic

Com

poun

dsci

s -ca

ftaric

aci

d0.

75±0

.10.

75±0

.10.

99±0

.10.

42±0

.00.

85±0

.10.

80.

2tra

ns-c

afta

ric a

cid

25.3

±2.1

20.1

±1.9

31.9

±2.5

14.2

±1.5

20.4

±2.8

22.4

6.6

cis -

cout

aric

aci

d2.

8±0.

31.

9±0.

23.

3±0.

31.

6±0.

22.

8±0.

52.

50.

7tra

ns-c

outa

ric a

cid

12.8

±1.5

8±1.

115

.6±1

.15.

9±1.

48.

5±1.

310

.23.

9de

lphi

nidi

n-3-

O-g

luco

side

125.

5±15

.266

.4±4

.290

.7±8

.210

7.3±

10.2

127.

9±18

.610

3.6

25.7

cyan

idin

-3-O

-glu

cosi

de63

.3±4

.91.

7±0.

29.

4±1.

215

.5±1

.89.

8±1.

119

.924

.7pe

tuni

din-

3-O

-glu

cosi

de10

7.4±

9.8

62.2

±4.2

79.4

±5.1

88.5

±5.4

116.

5±11

.290

.821

.7pe

onid

in-3

-O-g

luco

side

172.

7±21

.332

.6±2

.842

.3±3

.254

.5±5

.346

.6±3

.969

.758

.1m

alvi

din-

3-O

-glu

cosi

de62

0.9±

36.7

419.

8±28

.956

3.9±

31.8

544.

6±28

.466

3.5±

37.5

562.

592

.6To

tal n

on-a

cyla

ted

anth

ocya

nins

1,08

9.8

582.

778

5.7

810.

496

4.3

846.

619

2.1

delp

hini

din-

3-O

-ace

tylg

luco

side

8.5±

1.3

6.9±

0.9

7.8±

1.1

14.5

±1.3

11.3

±1.2

9.8

3.1

peon

idin

-3-O

-ace

tylg

luco

side

8.9±

1.1

9.6±

2.1

7.8±

1.2

15.1

±1.4

7.7±

1.0

9.8

3.1

mal

vidi

n-3-

O-a

cety

lglu

cosi

de80

.8±5

.777

.3±5

.488

.5±6

.215

1.1±

19.5

101.

5±10

.199

.830

.1m

alvi

din-

3-O

-caf

feoy

lglu

cosi

de25

.2±2

.917

.3±1

.827

.9±2

.537

.7±3

.133

.3±2

.928

.37.

8de

lphi

nidi

n-3-

(6-O

- p-c

oum

aroy

l-glu

cosi

de)

0.5±

0.1

4.4±

0.7

2.9±

0.4

2.9±

0.4

0.5±

0.0

2.2

1.7

petu

nidi

n-3-

(6-O

-p-c

oum

aroy

l-glu

cosi

de)

4.7±

0.8

14.2

±1.5

21.3

±1.8

20.3

±2.4

18.6

±1.9

15.8

6.8

peon

idin

-3-(6

-O-p

-cou

mar

oyl-g

luco

side

)28

.1±2

.324

.3±2

.131

.1±2

.636

.9±2

.831

.8±3

.130

.44.

7m

alvi

din-

3-(6

-O-p

-cou

mar

oyl-g

luco

side

)28

5.5±

21.6

259.

6±18

.432

4.5±

26.1

338.

5±30

.127

1.5±

19.8

295.

934

.1To

tal a

cyla

ted

anth

ocya

nins

441.

841

3.6

511.

861

7.0

476.

249

2.1

79.0

myr

icet

in-3

-O-g

lucu

roni

de1.

95±0

.20.

36±0

.00.

6±0.

11.

31±0

.71.

84±0

.31.

20.

7m

yric

etin

-3-O

-glu

cosi

de11

.06±

1.0

3.91

±0.3

10.3

2±1.

212

.34±

1.5

14.2

9±1.

610

.43.

9ru

tin3.

14±0

.40.

36±0

.00.

61±0

.21.

25±0

.81.

68±0

.51.

41.

1qu

erci

tin-3

-O- g

lucu

roni

de2.

25±0

.30.

73±0

.12.

01±0

.32.

41±0

.42.

78±0

.42.

00.

8qu

erci

tin-3

-O- g

luco

side

26.0

6±2.

15.

03±0

.618

.11±

1.6

13.8

5±1.

719

.54±

1.1

16.5

7.8

kaem

pfer

ol-3

-O-g

lucu

roni

de3.

43±0

.60.

36±0

.10.

61±0

.21.

57±0

.62.

43±0

.81.

71.

3ka

empf

erol

-3-O

-glu

cosi

de4.

13±0

.81.

38±0

.21.

82±0

.42.

62±0

.73.

01±1

.02.

61.

1

Each

val

ue is

the

mea

n ±

SD o

f thr

ee re

plic

ates

of 1

0 be

rry

sub

sam

ples

.


Table

4 -

Ph

enol

ic c

once

ntr

ati

on (m

g/kg)

of ber

ry s

kin

s fr

om U

va d

i T

roia

gra

pes

.

Traini

ng sy

stem

Arbo

rTre

llis

Prod

uctio

n ar

eaM

inerv

inoRu

vo d

i Pug

liaRu

vo d

i Pug

liaRu

vo d

i Pug

liaAn

dria

Andr

iaRu

vo d

iAn

dria

Aver

age

SDM

urge

II Pi

ano

Calen

dano

Capo

casa

leRi

vera

S. D

omen

icoPu

glia

S. D

omen

ico(n

=8)

(n=8

)

Phen

olic C

ompo

unds

cis-c

afta

ric a

cid0.

1±0.

00.

1±0.

00.

1±0.

00.

1±0.

00.

1±0.

00.

1±0.

00.

1±0.

00.

1±0.

00.

10

trans

-caf

taric

acid

6.1±

0.8

6.9±

0.9

6.9±

0.8

10.3

±1.6

5.1±

0.9

6.7±

0.8

17.9

±1.6

17.5

±1.5

9.7

5.2

cis-c

outa

ric a

cid0.

9±0.

10.

9±0.

20.

76±0

.10.

91±0

.20.

74±0

.10.

79±0

.11.

87±0

.21.

9±0.

21.

10.

5tra

ns-c

outa

ric a

cid3.

6±0.

53.

5±0.

53.

4±0.

45.

28±0

.82.

89±0

.33.

81±0

.67.

81±0

.97.

8±0.

84.

82.

0de

lphini

din-3

-O-g

lucos

ide12

0.6±

12.6

86.2

±6.4

187.

7±29

.114

3.3±

22.4

69.6

±7.2

93.7

±10.

214

9.1±

25.6

69.2

±7.9

114.

942

.6cy

anidi

n-3-

O-glu

cosid

e28

.1±2

.816

.6±2

.029

.4±2

.317

.4±1

.99.

3±1.

516

.6±2

.142

.8±3

.59.

8±1.

821

.311

.4pe

tunid

in-3-

O-glu

cosid

e97

.4±9

.381

.5±8

.915

1±19

.711

7.3±

11.5

64.3

±7.1

81.8

±8.4

127.

6±15

.461

.4±7

.097

.831

.8pe

onidi

n-3-

O-glu

cosid

e85

.7±8

.250

.8±5

.194

.3±9

.875

.8±8

.232

.2±2

.852

.7±5

.951

.3±5

.632

.3±2

.659

.423

.4m

alvidi

n-3-

O-glu

cosid

e55

7.2±

34.6

480.

9±30

.179

8.2±

59.8

766.

9±63

.443

4.1±

29.7

564.

8±39

.755

0.6±

38.1

491.

1±29

.858

0.5

132.

6To

tal n

on-a

cylat

ed a

ntho

cyan

ins88

9.0

716.

012

60.6

1120

.760

9.5

809.

692

1.4

663.

887

3.8

225.

5de

lphini

din-3

-O-a

cetyl

gluco

side

18.8

±2.9

15.3

±2.3

26.4

±4.5

15.7

±2.1

10.4

±1.4

19±3

.59.

8±1.

28.

7±1.

415

.55.

9pe

onidi

n-3-

O-ac

etylg

lucos

ide19

.9±3

.515

.8±2

.929

.9±3

.114

.3±2

.211

.8±1

.120

.4±2

.510

.1±1

.68.

7±1.

516

.46.

9m

alvidi

n-3-

O-ac

etylg

lucos

ide21

6.8±

38.2

201.

3±30

.125

7.5±

41.3

201.

3±29

.416

7.9±

27.6

243.

7±38

.513

0.6±

17.3

134.

8±19

.219

4.2

46.8

malv

idin-

3-O-

caffe

oylgl

ucos

ide56

.4±6

.950

.7±7

.362

.6±7

.547

.5±7

.042

.6±5

.949

.5±6

.238

.2±4

.831

.4±4

.247

.49.

9de

lphini

din-3

-(6-O

- p-c

oum

aroy

l-gluc

oside

)18

.3±2

.515

.1±2

.29.

6±2.

07.

8±1.

57.

3±1.

312

.8±2

.00.

5±0.

07.

8±1.

19.

95.

5pe

tunid

in-3-

(6-O

- p-c

oum

aroy

l-gluc

oside

)26

.3±4

.018

.7±2

.318

.6±2

.827

.6±3

.814

.9±2

.622

.8±5

.215

.3±2

.920

.4±4

.820

.64.

7pe

onidi

n-3-

(6-O

- p-c

oum

aroy

l-gluc

oside

)54

.1±5

.952

.1±5

.059

.3±8

.356

.3±7

.242

.8±6

.258

.1±7

.426

.9±4

.836

.6±5

.148

.311

.7m

alvidi

n-3-

(6-O

-p-c

oum

aroy

l-gluc

oside

)46

8.4±

35.8

470.

9±39

.252

5.4±

46.3

527.

3±49

.540

9.4±

38.9

560.

6±55

.227

2.6±

44.2

365.

9±50

.145

0.1

96.3

Tota

l acy

lated

ant

hocy

anins

879.

083

9.9

989.

389

7.8

707.

198

6.9

504.

061

4.3

802.

317

6.8

myric

etin-

3-O-

glucu

ronid

e1.

75±0

.20.

48±0

.01.

21±0

.10.

62±0

.00.

66±0

.11.

09±0

.31.

24±0

.31.

2±0.

21.

00.

4my

ricet

in-3-

O-glu

cosid

e15

.99±

2.0

9.31

±1.1

14.0

1±1.

913

.22±

2.3

8.82

±1.5

12.6

5±2.

311

.81±

2.0

10.7

7±1.

912

.12.

4ru

tin1.

17±0

.20.

48±0

.30.

78±0

.21.

21±0

.20.

48±0

.31.

08±0

.21.

64±0

.41.

32±0

.81.

00.

4qu

ercit

in-3-

O-glu

curo

nide

2.56

±0.2

2.01

±0.2

2.28

±0.1

2.47

±0.1

1.88

±0.1

2.43

±0.2

1.81

±0.1

1.86

±0.1

2.16

0.3

quer

citin-

3-O-

gluco

side

13.2

4±2.

311

.11±

2.6

10.1

1±2.

517

.04±

2.1

5.49

±1.3

12.3

5±2.

916

.41±

3.1

16.9

2±3.

712

.84.

0ka

empfe

rol-3

-O-g

lucur

onide

1.94

±0.2

1.25

±0.1

1.45

±0.1

1.21

±0.1

0.98

±0.0

2.64

±0.2

1.75

±0.1

1.86

±0.2

1.6

0.5

kaem

pfero

l-3-O

-gluc

oside

2.75

±0.3

1.58

±0.2

5.05

±1.1

1.81

±0.4

3.49

±0.8

1.09

±0.1

1.92

±0.4

2.82

±0.5

2.6

1.3

Each

value

is th

e m

ean

± SD

of t

hree

repli

cate

s of 1

0 be

rry su

b sa

mple

s.


nane berries to 83% in the skins of Al-fohonse Lavallée. Although the trans-caftaric:trans-coutaric ratio was about2:1 in both varieties, the average con-tent of HCTA in Bombino nero (Table 3)was double that of Uva di Troia (Table 4).DI BENEDETTO and TAMBORRA (1992)and TAMBORRA et al. (1994) also founda ratio higher than 1 in Primitivo ber-ries, whereas the ratio was lower than1 in Aleatico, Malvasia di Lecce, Malva-sia di Brindisi, Negroamaro and Gagli-oppo di Cirò. It should be noted that,as these compounds are involved in thebrowning reaction of must and wine,they can affect their sensorial proper-ties (MACHEIX et al., 1991; VRHOVSEK,1998).

Delphinidin, cyanidin, petunidin,peonidin, and malvidin 3-monogluco-sides are the only anthocyanins foundin Pinot noir (RIBÉREAU-GAYON, 1982),while these anthocyanins, along withvarious forms acylated by acetic, cou-maric and caffeic acids, are found inCabernet Sauvignon (NAGEL and WULF,1979; TAMBORRA and DI BENEDETTO,1991), Merlot and the Syrah varieties(MAZZA, 1995). Acylated derivativeswere also found in Bombino nero andUva di Troia berries, and a total of 13anthocyanins were quantified by HPLCanalysis. The anthocyanin compositionof both varieties, which was similar toCabernet Sauvignon, mainly consist-ed of malvidin 3-monoglucoside (on av-erage, 66.4% of the non-acylatedforms) and a rather high relativeamount of acetate derivatives. Withregard to the acylated forms, the mal-vidin derivatives, 3-acetylglucoside(about 20%), 3-p-coumarylglucoside(about 60%) and 3-caffeoylglucoside(about 6%), were the principle onesfound in both varieties. The acylatedforms (the greatest quantity in Uva diTroia berries) contributed the most todistinguishing the two varieties. Infact, the average concentration of theacylated anthocyanins in Uva di Troia

(802 mg/kg) was about twice as muchas that in Bombino nero (492 mg/kg),with non-acylated/acylated anthocy-anin ratios of about 1:1 and 2:1, re-spectively.

The flavonol varietal profile of grapes(PIGELLA et al., 1998; TAMBORRA and MI-NAFRA, 1998) and wines (MCDONALD etal., 1998; VUORINEN et al., 2000) hasbeen demonstrated to be useful for cul-tivar characterization and classificationof wines on the basis of their geographi-cal origin.

In Uva di Troia and Bombino nero,quercitin-3-O-glucoside and myrice-tin-3-O-glucoside were the most rep-resented flavonols with non-significantdifferences due to variety. Moreover,the average quercitin (3-glucuronideplus 3-glucoside forms): myricetin (3-glucuronide plus 3-glucoside forms)ratio was slightly higher than 1 in bothvarieties.

Principal Component analysis wasperformed on all variables simultane-ously to classify the samples (Figs. 2and 3). A sample classification map wasobtained in relation to acylated an-thocyanin (negative loadings) and cin-namoyl tartaric acid (positive loadings)contents on the first dimension (Fig. 2)that describes most of the differences(41% explained variance). From thesample distribution plot (Fig. 3), Uvadi Troia samples scored in the oppo-site position to Bombino nero (all pos-itively scored on the abscissa). Thesevarietal differences could be attribut-ed to different biosynthetic pathwaysof anthocyanin and cinnamate esters(MACHEIX et al., 1990).

Non-acylated anthocyanins and glyco-sidic flavonols (positive loading on thesecond dimension) did not contribute toproduct discrimination by variety, asconfirmed by results from ANOVA.

There were few differences betweensamples of the two varieties trained us-ing the trellis system (located in a cen-tral position on the plot). Note that the


Fig. 2 - PCA - Loadings of variables for factors 1 and 2.

Variety identification: Uva di Troia (in bold); Bombino nero (block capitals) Training system iden-tification: trellis (underlined); arbor (italic).

amount of HCTA in Uva di Troia washigher in the grapes grown using thetrellis system.

Area of production gave no particulargroupings, probably due to a limited var-iability of soil and climatic conditions inthe growing regions.

CONCLUSIONS

The study has demonstrated thatred-berry skins of the two varieties, Uvadi Troia and Bombino nero, had an

analogous composition in anthocy-anins, flavonol glycosides and hydrox-ycinnamoyl tartaric acids. However, theabsolute content of cinnamoyl tartaricacids, acylated forms of malvidin andpeonidin-3-(6-O-p-coumaroyl-gluco-side), significantly different in the twovarieties, highlight a genetic basis forthe synthesis of these compounds inthe grapes.

The particular content of acylated an-thocyanins of the two varieties could beuseful to promote the varietal charac-ter of the respective young red wines.

-0.4

-0.4

-0.4 -0.4


Furthermore, HCTA were significantlyaffected by the area of production andtraining systems in both varieties.

REFERENCES

Boss P.K., Davies C. and Robinson S.P. 1996. Anal-ysis of the expression of anthocyanin pathwaygenes in developing Vitis vinifera L. cv Shirazgrape berries and the implications of pathwayregulation. Plant Physiol. 111:1059.

Boursiquot J.M. 1987. Contribution à l’étude desesters hydroxycinnamoyltartriques chez le genre

Fig. 3 - PCA - Scores of samples for factors 1 and 2.

Variety identification: Uva di Troia (in bold); Bombino nero (block capitals) Training system iden-tification: trellis (underlined); arbor (italic).

Vitis. Recherche d’application taxonomique.Thése Doct. Ing., ENSA, Montpellier, France.

Calò A., Tomasi D., Cravero M.C. and Di StefanoR. 1994. Varietal analysis and classification ofthe species Vitis by determination of anthocy-ans and of hydroxycinnamoyl tartaric acids inthe skin of red-berry cultivars. Riv. Vitic. Enol.3:13.

Di Benedetto G. and Tamborra P. 1992. Phenolicpattern of some red grape varieties from Apulianregion. L’Enotecnico 12:85.

Di Stefano R. and Cravero M.C. 1991. The grapephenolics determination. Riv. Vitic. Enol. 2:37.

Di Stefano R. and Maggiorotto G. 1995. Anthocy-anins, hydroxycinnamic acids and flavonols in

8

-8

-8 8


berries, leaves, stems and shoots of vine. Riv.Vitic. Enol. 2:51.

Gao L., Girard B., Mazza G. and Reynolds A.G.1997. Changes in anthocyanins and color char-acteristics of Pinot Noir wines during differentvinification processes. J. Agric. Food Chem.45:2003.

Gomez-Plaza E., Gil-Munoz R., Lopez-Roca J.M.and Martinez A. 2000. Color and phenolic com-pounds of a young red wine. Influence of wine-making techniques, storage temperature, andlength of storage time. J. Agric. Food Chem.48:736.

Hertog M.G., Sweetnam P.M., Fehily A.M., ElwoodP.C. and Kromhout D. 1997. Antioxidant fla-vonols and ischemic heart disease in a Welshpopulation of men: Caerphilly study. Am. J.Clin. Nutr. 65:1489.

Jackson D.I. and Lombard P.B. 1993. Environmen-tal and management practices affecting grapecomposition and wine quality. A review. Am. J.Enol. Vitic. 44: 409.

Lamuela-Raventos R.M. and Waterhouse A.L. 1994.A direct HPLC separation of wine phenolics. Am.J. Enol. Vitic. 45:1.

Macheix J.J., Fleuriet A. and Billot J. 1990. Bio-synthesis of phenolic compounds in fruits. In“Fruit Phenolics”, p. 170. CRC Press, Boca Ra-ton, FL.

Macheix J.J., Sapis J.C. and Fleuriet A. 1991.Phenolic compounds and polyphenoloxidase inrelation to browning in grapes and wines. Crit.Rev. Food Sci. Nutr. 30: 441.

Mattivi F., Monetti A. and Nicolini G. 1995. Phe-nolic composition and characterization of redwines. L’Enotecnico. 6: 69.

Mazza G. 1995. Anthocyanins in grapes and grapeproducts. Crit. Rev. Food Sci. Nutr. 35: 341.

Mazza G., Fukumoto L., Delaquis P., Girard B. andEwert B. 1999. Anthocyanins, phenolics, andcolor of Cabernet Franc, Merlot, and Pinot Noirwines from British Columbia. J. Agric. FoodChem. 47: 4009.

McDonald M.S., Hughes M., Burns J., Lean M.E.J.,Matthews D. and Crozier A. 1998. Survey of thefree and conjugated myricetin and quercitincontent of red wines of different geographicalorigins. J. Agr. Food Chem. 46:368.

Nagel C.W. and Wulf L.W. 1979. Changes in theanthocyanins, flavonoids and hydroxycinnam-ic acid esters during fermentation and aging of

Merlot and Cabernet Sauvignon. Am. J. Enol.Vitic. 30:111.

Pigella R., Bosso A., Di Stefano R., Corino L. andMalerba G. 1998. Pinot noir varietal character-ization by polyphenols and aroma precursorsstudy. Riv. Vitic. Enol. 1:45.

Reynolds A.G., Stephane Y., Watson B., Price S.F.and Wardle D.A. 1996. Fruit environment andcrop level effects on Pinot noir. III. Compositionand descriptive analysis of Oregon and BritishColumbia wines. Am. J. Enol. Vitic. 47:329.

Ribéreau-Gayon P. 1982. The anthocyanins ofgrapes and wines. In “Anthocyanins as FoodColors”, p. 209. Markakis P. (Ed.), AcademicPress, New York.

Ritchey J.G. and Waterhouse A.L. 1999. A stand-ard red wine: monomeric phenolic analysis ofcommercial Cabernet Sauvignon wines. Am. J.Enol. Vitic. 50:91.

Romani A., Mancini P., Tatti S. and Vincieri F.F.1996. Polyphenols and polysaccharides in Tus-can grapes and wines. Ital. J. Food Sci. 1:13.

Roubelakis-Angelakis K.A. and Kliever W.M. 1986.Effects of exogenous factors on phenylalanineammonia-lyase activity and accumulation ofanthocyanins and total phenolics in grape ber-ries. Am. J. Enol. Vitic. 37:275.

Tamborra P. and Di Benedetto G. 1991. Phenolicpattern of some red grape varieties from Apulianregion. L’Enotecnico 10:89.

Tamborra P., Di Benedetto G., Damis P.D. andFuoco C. 1994. Research into phenolic compo-sition of some red grape varieties from Calabriaregion. L’Enotecnico 11:93.

Tamborra P. and Minafra M. 1998. HPLC mono-meric phenolic separation of red grapes andwines from south of Italy. L’Enotecnico 7-8:93.

Tamborra P., Dimitri P., Gianfrate C. and Trotta L.2000. The characterization of Apulian viticul-tural productions, first results of zoning. Vigne-vini. 3:100.

Vrhovsek U. 1998. Extraction of Hydroxycin-namoyltartaric acids from berries of differentgrape varieties. J. Agr. Food Chem. 46:4203.

Vuorinen H., Maatta K. and Torronen, R. 2000.Content of the flavonols myricetin, quercitin,and kaempferol in Finnish berry wines. J. Agr.Food Chem. 48:2675.

Wang H., Cao G.H. and Prior R.L. 1997. Oxygenradical absorbing capacity of anthocyanins. J.Agric. Food Chem. 45:304.

Paper received June 27, 2002 Accepted February 24, 2003


PAPER

- Key words: Mediterranean mussel, Mytilus galloprovincialis, potassium sorbate, sodium lactate -

EFFECT OF SODIUM LACTATEAND POTASSIUM SORBATE ON QUALITY

CHARACTERISTICS AND SHELF-LIFEOF MEDITERRANEAN MUSSEL

(MYTILUS GALLOPROVINCIALIS) MEATDURING CHILLED STORAGE

IN POUCHES WITH WATER

EFFETTO DEL LATTATO DI SODIO E SORBATO DEL POTASSIOSULLE CARATTERISTICHE QUALITATIVE E LA VITA DI SCAFFALE

DELLA CARNE DI VONGOLE MEDITERRANEE(MYTILUS GALLOPROVINCIALIS)

REFRIGERATA CON ACQUA IN SACCHE DI PLASTICA

A. VASAKOU, K. VARELTZIS1 and J. G. BLOUKAS*Department of Food Science and Technology, School of Agriculture,Aristotle University of Thessaloniki, GR 541 24 Thessaloniki, Greece

1 Department of Food Hygiene and Food Technology of Animal Origin,School of Veterinary Medicine, Aristotle University of Thessaloniki,

GR 541 24 Thessaloniki, Greece*Corresponding author: Tel. +30 2310 998773, Fax +30 2310 998791;

E-mail: [email protected]

ABSTRACT

The quality characteristics and shelf-life of Mediterranean mussel (Mytilusgalloprovincialis) meat (shucked mus-sels), stored at 5±2°C in plastic pouch-

RIASSUNTO

Sono state studiate le caratteristichequalitative e la vita di scaffale della car-ne di vongole mediterranee (Mytilusgalloprovincialis) conservata in acqua a


es with water, were studied. The con-trol was 150 g of mussel meat with 150mL of water. Sodium lactate (1% and2%) and potassium sorbate (0.35%)were added to 150 g of mussel meat in150 mL of water. Potassium sorbate sig-nificantly (p<0.05) retarded the growthof mesophilic aerobic bacteria and psy-chrotrophic bacteria, and especially thegrowth of lactic acid bacteria. In mus-sel meat with potassium sorbate, thepH was slightly reduced and the totalvolatile basic nitrogen (TVBN) and thio-barbituric acid (TBA) values were sig-nificantly (p<0.05) lower than all othertreatments. Based on sensory at-tributes the shelf-life of the product withpotassium sorbate was increased from6 to 8 days. The potassium sorbate con-tent of the mussel meat was initially lessthan 0.185%, and after the 5th day, lessthan 0.15%. The shelf-life of the con-trol was 6 days. Sodium lactate at bothlevels had no effect (p>0.05) on micro-bial counts, TVBN and TBA and shelf-life, but the pH was higher (p<0.05)than the control at the end of the stor-age period and the mussels were salti-er.

5±2°C in sacche di plastica. Al control-lo (150 g di carne di vongole in 150 mLdi H2O) sono stati aggiunti, rispettiva-mente 1 e 2% di lattato di sodio e sor-bato di potassio (0,35%). Il sorbato dipotassio ha ritardato in modo signifi-cativo (p<0.05) la crescita di batteri ae-robi mesofilici e psicrotropici ed in par-ticolare la crescita di batteri lattici. Neicampioni con sorbato, il pH si è debol-mente ridotto e i valori di TVBN e TBAsono stati significativamente (p<0,05)più bassi di quelli degli altri trattamenti.Sulla base degli attributi sensoriali, lavita di scaffale delle vongole aggiuntedi sorbato di potassio è aumentata da6 ad 8 gg. Il contenuto in sorbato dipotassio è stato inizialmente inferioreallo 0,185% e, dopo il quinto giorno, allo0,15%. La vita di scaffale del controlloè stata di 6 giorni. Il lattato di sodio, adambedue i livelli di concentrazione, nonha mostrato effetti (p>0,05) sulla contamicrobica totale, e i valori di TVBN eTBA e sulla vita di scaffale. Il valore delpH, invece, è stato significativamente(p<0,05) più elevato del controllo allafine del periodo di conservazione le von-gole sono apparse più salate.

INTRODUCTION

The Mediterranean mussel (Mytilusgalloprovincialis) contains about 79-83%moisture, 11-13% protein of high biolog-ical value, 2.0-2.3% carbohydrates, 1.5-1.8% fat, 2.5% ash and 0.35% non-pro-tein nitrogen compounds (METAXATOU,1998), and a high iron content, about110 ppm (KARAKOLTSIDIS et al., 1995).It is considered a low-fat and low-cho-lesterol food (KING et al., 1990; HOLLANDet al., 1994) and gives about 87 kcal/100 g (HOLLAND et al., 1994). More than40.8% of its fatty acids are polyunsatu-rated (MILETIC et al., 1991), with eicos-apentaenoic (C20:5, n-3) and docosahex-

aenoic (C22:6, n-3) acids being the prev-alent ones (KARAKOLTSIDIS et al., 1995;ORBAN et al., 2002). Mussel meat has ahigh n-3/n-6 fatty acid ratio with posi-tive effects on human health. Raising thelevel of n-3 fatty acids in the blood con-tributes to low incidence of cardiovas-cular diseases and mortality from athero-sclerosis and heart disease, reduced in-flammation associated with rheumaticarthritis and psoriasis and low bloodpressure and viscosity (KINSELLA, 1986;ACKMAN, 1995).

In Greece, more than 25000 ton ofmussels are produced annually, thegreatest part of which is exported to othercountries. In Greece they are retail mar-


keted in plastic pouches which containwater and either live mussels in theshells or shucked mussels. The productwhich is sealed in pouches, placed inboxes and covered with crushed ice forchilling and transport, has a shelf-life of4 to 6 days.

Sodium lactate is currently consideredas a GRAS (Generally Recognized AsSafe) substance and is used as an emul-sifier, flavour enhancer, humectant andpH control agent at levels not exceedingcurrent good manufacturing practice(DUXBURY, 1988; BREWER et al., 1991).It has also been shown to retard bacte-rial growth and to increase the shelf lifeof beef and pork meat and meat prod-ucts (BREWER et al., 1991; MACA et al.,1997; BLOUKAS et al., 1997; LIN and LIN,2002), turkey (MAAS et al., 1989) andsalmon (PERLOY et al., 1994). Sodiumlactate has been reported to delay growthand toxin production by Clostridium bot-ulinum in cooked fish products and tocontrol growth of Listeria monocytogenesin meat and poultry frankfurters (SOFOS,1995) and fish products (PERLOY et al.,1994; NYKANEN et al., 2000) and growthof Escherichia coli O157:H7 in commer-cial beef patty formulations (BYRNE et al.,2002). WILLIAMS et al. (1995) used 2%sodium lactate to preserve fresh catfishfillets and found that it affects its shelf-life and consumer acceptance under sim-ulated retail conditions.

Sorbic acid and its salts, known assorbates, are regarded as GRAS sub-stances and were approved for use infoods in 1950. Traditionally, they serveas broad spectrum inhibitors of yeastsand moulds and are useful to inhibitaerobic, catalase positive bacteria andpathogenic microorganisms, such assalmonellae, Staphylococcus aureus andClostridium botulinum (SOFOS, 1989).They have been used throughout the foodindustry to extend the shelf-life of manyfoods, including bakery items, cereals,butter, cheese and meat products (RO-BACH and SOFOS, 1982; LIEWEN and

MARTH, 1985). Potassium sorbate hasbeen used for the preservation of manyproducts (FEY and REGENSTEIN, 1982;FLETCHER et al., 1988; SAWAYA et al.,1993; KARABULUT et al., 2001; ANDERSet al., 2002), to control the growth of Lis-teria monocytogenes on cooked Grawfish(Procambarus clarkii) tail meat (DORSAet al., 1993), and as an additive in edi-ble films and wrapping materials (BALD-WIN et al., 1996; OZDEMIR and FLOROS,2001).

KASTANIDOU-MANOUSOU et al. (1982)studied the keeping quality of Mediter-ranean mussel meat stored at refrigera-tion temperatures (3°-4°C), but theshucked mussels were not kept inpouches with water. There have been noreports in the literature on the effects ofsodium lactate and potassium sorbateon the quality characteristics and shelf-life of Mediterranean mussel meat.

The objective of this study was to in-vestigate the effect of the levels of sodi-um lactate (1% and 2%) and potassiumsorbate (0.35%) on the quality charac-teristics of Mediterranean mussel(Mytilus galloprovincialis) meat and theireffect on shelf-life under chilled storage(5°±2°C) in plastic pouches with water.


Experimental design

Fresh (live) mussels were harvestedfrom approved shellfishing waters in anarea near Thessaloniki. Within 1 hourthe mussels were taken to the laborato-ry in plastic boxes covered with crushedice, and were immediately shucked byforcing a knife between the two valves ofthe mollusk shell, cutting appropriatelythe adductor muscle from each valve andthus freeing the meat. The shuckedmussels were carefully washed with po-table tap water and separated into sam-ples of 150 g, containing about 14-16pieces of shucked mussels, each.


Four treatments were carried out on150 g samples of shucked mussels,putting them into plastic pouches as fol-lows:

1. C: Control treatment; the musselswere mixed with 150 mL of water.

2. SL 1%: The mussels were mixedwith a 150 mL solution containing 1%sodium lactate.

3. SL 2%: The mussels were mixedwith a 150 mL solution containing 2%sodium lactate.

4. PS 0.35%: The mussels were mixedwith a 150 mL solution containing 0.35%potassium sorbate.

A 60% pure commercial sodium lac-tate solution (PURAC Bioquimica S.A.,Montmelo, Spain) was used for treat-ments 2 and 3, and pure solid potassi-um sorbate E202 (Merck) was used fortreatment 4.

Twelve pouches with shucked musselswere prepared for each treatment. Allpouches were carefully closed with a heat(conductance) sealer and kept in chilledstorage at 5°±2°C. Two pouches pertreatment were taken on the 0, 3, 5, 7and 10th day of storage for analyses. Theexperiment was repeated three times inthree different months from April to Sep-tember.

Physicochemical analysis

pH was measured in a homogenateprepared by blending 20 g of musselmeat with 80 mL of distilled water for 30s. Readings were taken with a WTW,model pH 521 digital pH meter and aWTW, type E 56 combination electrode(WTW-Wissensehaftlich-TechnischeWerkstaetten GmbH, Weilheim, Germa-ny). Total volatile basic nitrogen was de-termined according to the method de-scribed by BOTTA et al. (1984) on 10 g ofmussel meat applying direct distillationwith MgO, using the micro-Kjeldahl dis-tillation apparatus and titrating with0.1N H2SO4. Results are expressed as mgN/100g. The thiobarbituric acid (TBA)

value was determined according to themethod described by BOTSOGLOU et al.(1994) on 2 g of mussel meat using 5%aqueous TCA (8 mL) and 0.8% BHT inhexane (5 mL). The sample was Ultra-Turraxed for 30 s and centrifuged at3000 g for 3 min. The bottom layer wasmade to 10 mL with 5% TCA and a 2.5mL aliquot was taken and mixed with1.5 mL 0.8% aqueous TBA. Followingincubation for 30 min at 70°C, the sam-ple was cooled and submitted to third-derivative spectrophotometry against ablank reaction mixture. The results areexpressed as mg of malonaldehyde/kg.The potassium sorbate content was de-termined according to the AOAC (1990)procedure. Duplicate samples from eachtreatment were analyzed for all physico-chemical parameters.

Microbiological analysis

A 20 g sample of mussel meat washomogenized in a stomacher Lab-blender400 (Seward Medical, London, UK) with180 mL sterile 0.1% peptone water for 2min. Appropriate dilutions of sampleswere prepared in sterile 0.1% peptonewater blank and plated in duplicate ondifferent growth media. The followingmedia and incubation conditions wereused: (1) Plate Count Agar (Oxoid) at30°C for 72 h, for total aerobic mesophiliccount; (2) Plate Count Agar (Oxoid) at7°C for 10 days, for psychrotrophic bac-terial count; (3) deMan Rogosa Sarpe(MRS) Agar (Oxoid) at 25°C for 5 daysand anaerobic incubation (BBL “Gas-Pack” System with H2/CO2 cartridges),for lactic acid bacterial count. Resultsare expressed as log10 numbers of colo-ny forming units/gram (cfu/g).

Sensory shelf-life

An experienced ten-member panel wasused to evaluate mussel meat duringstorage. At each sampling time storedmussels were compared with fresh ones


just after being put in a pouch and mixedwith water (reference sample). They wereevaluated for appearance and odourwhen the pouches were opened, andodour and taste after cooking. Musselmeat with acceptable odour was washedwith tap water, put in boiling water for 5min, cooled under running tap water and

Table 1 - Sensory evaluation scales used to evalu-ate the mussel meat.

Odour during opening of pouches5 = odour of fresh shucked mussels4 = slightly off-odour, but still pleasant3 = moderately off-odour, but acceptable2 = slightly sour, not acceptable1 = strongly sour or putrid, highly objectionable

Odour and taste of cooked mussel meat

5 = excellent4 = good3 = acceptable2 = fair1 = unacceptable

served to the panelists in a random man-ner together with cooked fresh musselmeat for comparison and evaluation ofodour and taste. Tap water was also pro-vided to the panelists to cleanse the pal-ate between samples. The sensory eval-uation scales used for evaluating theodour during opening of pouches and theodour and taste of cooked mussel meatare given in Table 1. The time in days forthe mussels to be considered at the lim-it of acceptability was defined as theshelf-life of shucked mussel meat underthe specified storage conditions. In termsof panel scores given, this correspondedto the score of 3, both for odour duringopening of the pouches and odour andtaste of the cooked mussel meat.

Statistical analysis

Data were analyzed by a split-plot de-sign in a completely randomized system,with treatment as a whole plot and stor-age time and treatment by storage time

Table 2 - Analysis of variance on the effect of sodium lactate (1 and 2%) and potassium sorbate (0.35%)on the microbiological, physicochemical and sensory attributes of Mediterranean mussel (Mytilus gallo-provincialis) meat in pouches with water during chilled storage (5±2°C) (F-value for independent varia-bles and interactions).

Source of variance

Parameter Treatmenta Storage time Interaction(A) (B) (AxB)

MicrobiologicalTotal aerobic mesophilic count 493* 385.35*** 3.06**Psychrotrophic bacteria 2.97NS 283.48*** 3.48**Lactic acid bacteria 40.90*** 278.59*** 6.09***

PhysicochemicalpH value 25.30*** 511.40*** 36.24***Total volatile basic nitrogen (TVBN) 4.89* 245.76*** 5.21***Thiobarbituric acid value (TBA) 2.67NS 122.18*** 6.71***

Sensory attributesOff-odour during opening of pouches 2.24NS 550.82*** 5.29***Odour and taste of cooked mussels 19.94*** 20.69*** 8.63***

*Significant at p<0.05; **significant at p<0.01; ***significant at p<0.001; NS, not significant;a C: Control; Sodium lactate 1%; Sodium lactate 2%; Potassium sorbate 0.35%.


as the sub-plot. Means were comparedby using the LSD0.05 test. Data analyseswere performed using the MSTAT (1985)program (Version 3.0, Michigan StateUniversity, East Lansing, Ml, USA). Sim-ple correlations were determined be-tween selected response variables.


Sodium lactate and potassium sorb-ate slightly affected (p<0.05) the totalaerobic mesophilic count, had no effecton the psychrotrophic bacterial count,but significantly affected (p<0.001) thelactic acid bacterial count (Table 2). Stor-age time and the interaction of treatmentby storage time significantly affected(p<0.001) all microbial counts.

Total aerobic mesophilic count, psy-chrotrophic bacterial count and lacticacid bacterial counts of treatments withsodium lactate did not differ (p>0.05)from the control (Fig. 1). WILLIAMS et al.(1995) also reported that 1% and 2%sodium lactate had no effect on phychro-trophic bacteria during chilled storageof fish fillets. However, they found that2% sodium lactate affected the total aer-obic mesophilic count. LEMAY et al.(2002) found that sodium lactate had noeffect on aerobic mesophilic counts,while it delayed the growth of Brochothrixthermosphacta and lactic acid bacteriain a cooked and acidified chicken mod-el. They suggest that the spectrum ofactivity of sodium lactate is probablynarrow or specific to a certain group oforganisms, and that with time, othermicroorganisms take over. Two mainmechanisms have been proposed for thebacteriostatic mode of action of lactate:a) passage across the cell membrane ofthe undissociated acid, which is disso-ciated within the cell to reduce the in-tracellular pH, and b) acting as a hu-mectant by lowering the water activity(SHELEF, 1994; LEMAY et al., 2002). Thetreatment with potassium sorbate had

lower (p<0.05) total aerobic mesophiliccounts, phychrotrophic bacterial countsand lactic acid bacterial counts than allother treatments throughout storage.Potassium sorbate significantly (p<0.05)inhibited the growth of lactic acid bacte-ria even at day “0”. At the end of the stor-age period (10th day) the treatment withpotassium sorbate had about a 1-2 log10cfu/g lower phychrotrophic bacterialcount and a 3.0-3.5 log10 cfu/g lower lac-tic acid bacterial count than the othertreatments.

Sodium lactate, potassium sorbateand storage time, as well as the interac-tion of treatment by storage time, signif-icantly affected (p<0.001) the pH of themussel meat (Table 2). The initial pH ofthe mussel meat was about 6.5 (Fig. 2a).KASTANIDOU-MANOUSOU et al. (1982)reported similar pH values. The treat-ment with potassium sorbate caused theleast reduction in pH during the storageperiod, while the control treatmentcaused the largest. At the 10th day of stor-age the pH of the control dropped to 5.0,that with sodium lactate to about 5.6 andthat with potassium sorbate to 6.0. Ahigh negative correlation (r = -0.95,p<0.05) was found between the pH andall microbial counts. No difference in pHwas found between the treatments with1% and 2% sodium lactate throughoutthe storage period. Similar results werereported by WILLIAMS et al. (1995) dur-ing storage of fish fillets.

Total volatile basic nitrogen (TVBN)was slightly (p<0.05) affected by treat-ment, while the storage time and the in-teraction of treatment by storage timehad a significant (p<0.001) effect on it(Table 2). After the 3rd day of storage themussel meat with potassium sorbate hadsignificantly (p<0.05) lower TVBN thanother treatments (Fig 2c). A high posi-tive correlation was found between TVBNand total aerobic mesophilic count (r = -0.95, p<0.05) and TVBN and psychro-trophic bacterial count (r = -0.96,p<0.05).


Fig. 1 - Effect of sodium lactate and potassium sorbate on (a) Total aerobic mesophilic count, (b)Psychrotrophic bacterial count, and (c) Lactic acid bacterial count of Mediterranean mussel meat dur-ing chilled storage in pouches with water. �: Control; �: Sodium lactate 1%; ∆: Sodium lactate 2%; X:Potassium sorbate 0.35%.

Log

10 c

fu/g

Log

10 c

fu/g

Log

10 c

fu/g


Fig. 2 - Effect of sodium lactate and potassium sorbate on (a) pH, (b) Thiobarbituric Acid (TBA), and (c)Total Volatile Basic Nitrogen (TVBN) of Mediterranean mussel meat during chilled storage in poucheswith water. �: Control; �: Sodium lactate 1%; ∆: Sodium lactate 2%; X: Potassium sorbate 0.35%.

TV

BN

(mg

N/1

00 g

)M

alo

nald

ehyd

e m

g/k

gp

H


Sodium lactate and potassium sorb-ate had no effect (p>0.05) on the thio-barbituric acid (TBA) value, while stor-age time and the interaction of treat-ment by storage time significantly(p<0.05) affected it (Table 2). The TBAvalue continuously increased in the con-trol and 1% sodium lactate treatment(Fig. 2b). It remained constant for 5 daysin the treatment with 2% sodium lac-tate and for 7 days in the treatment withpotassium sorbate and after that timeit increased rapidly. The treatment with1% sodium lactate had a higher (p<0.05)TBA value than the control and 2% so-dium lactate. Similar results and simi-lar TBA values were reported by WIL-LIAMS et al. (1995) for fish filets. Theantioxidant activity of sodium lactate isattributed in part to the Fe3+-lactatecomplexation or lactyl radical reductionof Fe3+ to Fe2+ and subsequent chela-tion of Fe2+ by the carboxylate species(NNANNA et al. 1994). The treatmentwith potassium sorbate had the lowest(p<0.05) TBA value on the 7th and 10th

days of storage.In the treatment with 0.35% potassi-

um sorbate, the potassium sorbatewhich had been absorbed by the musselmeat was determined. It was found thatup to the 3rd day of storage the potassi-um sorbate content of the mussel meatwas 0.16-0.18% and decreased to 0.14-0.15% between the 5th and 10th days ofstorage. MUNZNER et al. (1990) investi-gated the possible genotoxic action ofpotassium sorbate. In all the in vitro andin vivo tests they found no signs of gen-otoxicity. However, BINSTOK et al. (1998)showed that sorbates can react with ni-trites to form mutagenic reaction prod-ucts and they suggested that the com-bined use of nitrites and sorbates shouldbe avoided.

Sodium lactate and potassium sorb-ate had no effect (p>0.05) on odoursduring opening of the pouches, whilestorage time and the interaction oftreatment by storage time significant-

ly (p<0.05) affected it (Table 2). Theodour scores during opening of thepouches continuously decreased withstorage time for all treatments. Closeto the 6th day of storage the control andthe treatments with sodium lactate hadodour scores at the limit of acceptabil-ity. On the contrary, the treatment withpotassium sorbate had an acceptableodour score up to the 8th day of storage(Fig. 3a). Sodium lactate, potassiumsorbate, storage time and the interac-tion of treatment by storage time sig-nificantly (p<0.05) affected the odourand taste of cooked mussel meat (Ta-ble 2). The cooked mussel meat thatwas treated with potassium sorbatehad significantly higher (p<0.05) scoresfor odour and taste throughout stor-age than all other treatments. Themussel meat of this treatment had anacceptable odour and taste score up tothe 8th day of storage. All other treat-ments gave cooked meat with an un-acceptable odour and taste after the 6th

day of storage (Fig. 3b).Based on these sensory attributes,

the shelf-life of Mediterranean mussel(Mytilus galloprovincialis ) meat(shucked mussels), stored at 5±2°C inplastic pouches with water, was deter-mined to be 6 days for the control treat-ment and the treatments with sodiumlactate. However, the shelf-life in thetreatment with potassium sorbate wasincreased to 8 days. Sodium lactate (1%and 2%) had no effect (p>0.05) on theshelf-life of the product compared tothe control. The panelists noticed thatmussel meat with sodium lactate wassaltier, even on the first day of storage,and became more pronounced duringstorage. This is probably due to theaddition of sodium ions (Na+) from so-dium lactate. WILLIAMS et al. (1995)also reported that fish filets with sodi-um lactate had a more pronouncedsalty taste compared to the control andthat this taste was confirmed by 60%of the panelists.


Fig 3 - Effect of sodium lactate and potassium sorbate on (a) Odour during opening of pouches, and (b)Odour and taste of cooked Mediterranean mussel meat during chilled storage in pouches with water. :Control; : Sodium lactate 1%; ∆: Sodium lactate 2%; X: Potassium sorbate 0.35%. Horizontal dottedlines represent the limit of acceptability of mussel meat for each sensory attribute and vertical lines theshelf-life of mussel meat for each treatment.


CONCLUSION

Potassium sorbate (0.35%) can beused to increase the shelf-life of Medi-terranean mussel (Mytilus galloprovincia-lis) meat (shucked mussels), stored at5±2°C in plastic pouches with water, upto 8 days. Such an improvement in shelflife can have an important economicimpact by reducing losses attributed tospoilage and by allowing the product toreach distant and new markets. On thecontrary, sodium lactate at 1 or 2% hadno effect on the shelf-life of the productand increased its saltiness.

ACKNOWLEDGEMENTS

Thanks are expressed to the Director, Professor S.Georgakis, and the staff of the Laboratory ofFood Technology of Animal Origin, School ofVeterinary Medicine, Aristotle University ofThessaloniki, for help they offered to completethis work.

REFERENCES

Ackman R.G. 1995. Composition and nutritivevalue of fish and shellfish lipids. In “Fish andFishery Products, Composition, Nutritive Prop-erties and Stability”. A. Ruiter (Ed.). p. 117. CabInternational, Wallingford, UK.

Andres S.C., Giannuzz L. and Zarifzky N.E. 2002.Quality parameters of packaged refrigeratedapple cubes in orange juice. Lebensm.-Wiss. u.Technol. 35:670.

AOAC. 1990. Official Methods of Analysis, 15th ed.Association of Official Analytical Chemists. Ar-lington, VA.

Baldwin E.A., Nisperos M.O., Chen X. and Hogen-maier R.D. 1996. Improving storage life of cutapple and potato with edible coating. Postharv.Biol. Technol. 9:151.

Binstok G., Campos C., Varela O. and Gerschen-son L.N. 1998. Sorbate nitrite reactions in meatproducts. Food Res. Int. 31: 581.

Bloukas J.G., Paneras E.D. and Fournitzis G.C.1997. Sodium lactate and protective cultureeffects on quality characteristics and shelf-lifeof low-fat frankfurters produced with olive oil.Meat Sci. 45:223.

Botsoglou NA., Fletouris D.J., Papageorgiou C. E.,Vassilopoulos V.N., Mantis A.J. and TrakatellisAG. 1994. A rapid, sensitive and specific thio-

barbituric acid method for measuring lipid per-oxidation in animal tissues, food and feedstuffsamples. J. Agric. Food Chem. 42:1931.

Botta J.R., Lauder J.T. and Jewer M.A. 1984. Ef-fect of methodology on total volatile basic nitro-gen (TVB-N) determination as an index of qual-ity of fresh Atlantic cod (Gadus morhua). J. FoodSci. 49:734.

Brewer S.M., McKeith F., Martin S.E., DallmierA.W. and Meyer J. 1991. Sodium lactate effectson shelf-life, sensory and physical characteris-tics of fresh pork sausage. J. Food Sci. 56:1176.

Byrne C.M., Bolton D.J., Sheridan J.J., Blair I.S.and McDowell D.A. 2002. Determination of theeffect of sodium lactate on the survival of heatresistance of Escherichia coli O157:H7 in twocommercial beef patty formulations. Food Micro-biol. 19:211.

Dorsa W.J., Marshall D.L. and Semien M. 1993.Effect of potassium sorbate and citric acidsprays on growth of Listeria monocytogenes oncooked Crawfish (Procambarus clarkii) tail meatat 4°C. Lebensm. -Wiss. u Technol. 26: 480.

Duxbury D.D. 1988. Natural sodium lactate ex-tends shelf-life of whole ground meats. FoodProcess. 49: 91.

Fey M.S. and Regenstein J.M. 1982. Extendingshelf-life of fresh wet red hake and salmon us-ing carbon dioxide and oxygen modified atmos-phere and potassium sorbate ice at 1°C. J. FoodSci. 47:1048.

Fletcher G.C., Murrel W.G., Statham J.A., StewartB.J. and Bremner HA. 1988. Packaging of scal-lops with sorbate: An assessment of the hazardfrom Clostridium botulinum. J. Food Sci. 53:349.

Holland B., Welch A.A., Unwin I.D., Buss D.H., PaulA.A. and Southgate DAT. 1994. “The Composi-tion of Foods”. 5th ed. McCance and Widdow-sons (Ed.). p. 214. The Royal Society of Chem-istry and Ministry of Agriculture, Fisheries andFood, Cambridge.

Karabulut O.A., Lune S. and Droby S. 2001. Eval-uation of the sodium bicarbonate, potassiumsorbate and yeast antagonists for decreasingpost-harvest decay of sweet cherries. Postharv.Biol. Technol. 23:233.

Karakoltsidis P.A., Zotos A. and Constantinidis SM.1995. Composition of the commercially impor-tant Mediterranean finfish, crustaceans andmolluscs. J. Food Compos. Anal. 8:258.

Kastanidou-Manousou Chr., Karaioannoglou Pr.and Koidis P. 1982. Keeping quality of shuckedmussels stored at refrigerating temperatures.Bull. Hell. Vet. Med. Soc. 25:200.

King I., Childs M.T., Dorsett C., Ostrander J.G.and Monsen E.R. 1990. Shellfish. Proximatecomposition, minerals, fatty acid and sterols.J. Amer. Diet. Assoc. 90:677.

Kinsella E.J. 1986. Food components with poten-tial therapeutic benefits: The n-3 polyunsatu-rated fatty acids offish oils. Food Technol.40:146.


Lemay M-J., Choquette J., Delaquis P.J., GariepyC., Rodrique N. and Saucier L. 2002. Antimi-crobial effect on natural preservatives in acooked and acidified chicken meat model. Int.J. Food Microbiol. 78: 217.

Liewen M.B. and Marth E.H. 1985. Growth andinhibition of microorganisms in the presence ofsorbic acid: A review. J. Food Prot. 48:364.

Lin K-W. and Lin S-N. 2002. Effects of sodium lac-tate and trisodium phosphate on the physico-chemical properties and shelf life of low-fatChinese-style sausage. Meat Sci. 60:147.

Maas M.R., Glass K.A. and Doyle M.P. 1989. Sodi-um lactate delays toxin production by Clostrid-ium botulinum in cook-in-bag turkey products.AppI. Environ. Microbiol. 55: 2226.

Maca J.V., Miller, R.K., Maca J.D. and Acuff G.R.1997. Microbiological, sensory and chemicalcharacteristics on vacuum-packaged cookedbeef top rounds treated with sodium lactate. J.Food Sci. 62:591.

Metaxatou A. 1998. The mussels. In “Proceedingsof 1st Balkan Congress on Aqui-cultures”. p. 87.Thessaloniki, Greece. 12-15 September.

Miletic I. Miric M., Lalic Z. and Sobajic S. 1991.Comparison of lipids and proteins of severalspecies of molluscs, marine and terrestrial fromthe Adriatic sea and Serbia. Food Chem. 41:303.

MSTAT. 1985. “Design and Analysis of AgronomicResearch” Michigan State University. East Lans-ing, Ml.

Munzner R., Guigas C. and Renner H.W. 1990.Re-examination of potassium sorbate and so-dium sorbate for possible genotoxic potential.Food Chem. Toxicol. 28:397.

Nnanna I.A., Ukutu D.O., McVann K.B. and She-let L.A. 1994. Antioxidant activity of sodium lac-tate in meat and model systems. Lebensm.-Wiss. u. Technol. 27:78.

Nykanen A., Weckman K. and Lapvetelainen A.2000. Synergistic inhibition of Listeria monocy-togenes on cold-smoked trout by nisin and so-dium lactate. Int. J. Food Microbiol. 61:63.

Orban E., Di Lena G., Nevigato T., Casini I., MarzettiA. and Caproni R. 2002. Seasonal changes inmeat content, condition index and chemicalcomposition of mussels (Mytilus galloprovincia-lis) cultured in two different Italian sites. FoodChem. 77: 57.

Ozdemir M. and Floros I.D. 2001. Analysis of mod-eling of potassiumn sorbate diffusion throughedible whey protein films. J. Food Engin. 47:149.

Perloy GA., Peterson M.E., Holland P.J. and Ek-lund M.W. 1994. Inhibition of Listeria monocy-togenes in cold-process (smoked) salmon bysodium lactate. J. Food Prot. 57:108.

Robach M.C. and Sofos J.N. 1982. Use of sorbatesin meat products, fresh poultry and poultryproducts: A review. J. Food Prot. 45:374.

Sawaya W-N., Abu-Ruwaida A.S., Baroon Z.H.,Khalafawi M.S. and Murad M. 1993. Shelf-lifeof eviscerated broiler carcasses as affected byvacuum packaging and potassium sorbate. Leb-ensm.-Wiss. u.Technol. 26:517.

Shelef L.A. 1994. Antimicrobial effects of lactates:a review. J. Food Prot. 57:445.

Sofos J.N. 1989. Potential use of sorbates in meatproducts. In “Sorbate Food Preservatives”. p.167. CRC. Press, Boca Raton, FL.

Sofos J.N. 1995. Effects of chemical additives. In“Proceedings of European Consortium for Con-tinuing Education in Advanced Meat Scienceand Technology Course”. Athens, Greece. 25-28 June.

Williams S.K., Rodrick G.E. and West R.L. 1995.Sodium lactate affects shelf life and consumeracceptance of fresh catfish (Ictalurus nebulosus,marmoratus) fillets under simulated retail con-ditions. J. Food Sci. 60:636.

Paper received August 7, 2002 Accepted February 11, 2003


PAPER

- Key words: Mediterranean mussel, mussel meat, Mytilus galloprovincialis, smoking -

EFFECT OF SMOKING ON QUALITYCHARACTERISTICS AND SHELF-LIFE

OF MEDITERRANEAN MUSSEL(MYTILUS GALLOPROVINCIALIS)

MEAT UNDER VACUUM IN CHILLED STORAGE

EFFETTO DELL’AFFUMICATURA SULLE CARATTERISTICHE QUALITATIVEE SULLA VITA DI SCAFFALE DELLA CARNE DI VONGOLE MEDITERRANEE

(MYTILUS GALLOPROVINCIALIS) CONSERVATE SOTTOVUOTO E REFRIGERATE

A. KYRIAZI-PAPADOPOULOU *, K. VARELTZIS1,J.G. BLOUKAS2 and S. GEORGAKIS1

Veterinary Medicine Administration of Thessaloniki, Department of Veterinary -Public Health, Prefecture of Thessaloniki, Egeou 14 and Solomou 1,

GR 55134 Thessaloniki, Greece1 Department of Food Hygiene and Food Technology of Animal Origin,

School of Veterinary Medicine, Aristotle University of Thessaloniki, GR 54124Thessaloniki, Greece

2 Department of Food Science and Technology, School of Agriculture, AristotleUniversity of Thessaloniki, GR 54124 Thessaloniki, Greece

* Corresponding author: Tel. +30 2310 478 325; Fax +30 2310 478 334

ABSTRACT

A smoking procedure was developedfor Mediterranean mussel meat (Mytilusgalloprovincialis). Fresh mussels werewashed in running water and steamedat 80°C (air temperature) for 10 min,the shells were removed and the meat

RIASSUNTO

È stata sviluppata la seguente tec-nica di affumicatura per la carne divongole del mediterraneo: le vongolefresche erano lavate in acqua correntee bollite a 80°C per 10 min, Dopo averrimosso le conchiglie, la carne era po-


INTRODUCTION

Mediterranean mussel meat (Mytilusgalloprovincialis) contains proteins witha high biological value and lipids with ahigh content of polyunsaturated fattyacids with more than 20 C atoms (MI-LETIC et al., 1991). Eicosapentaenoic(C20:5, n-3) and docosahexaenoic(C22:6, n-3) are the prevalent fatty ac-ids among the polyunsaturated fatty ac-ids (KARAKOLTSIDIS et al., 1995; FERN-ANDES-REIRIZ et al., 1996; ORBAN et al.,

was put in a brine with 4% salt for 15min, drained for 5 min, dried in asmokehouse at 60°-65°C for 13 min,smoked at air temperature 65°-80°C for17 min to obtain a F70°C value of 40.2min, vacuum packed in pouches andcold stored at 2°-3°C. The season(spring and autumn) of harvesting hadno effect (p>0.05) on the physicochem-ical and microbiological parameters ofthe smoked mussels. Steaming andsmoking affected (p<0.05) the chemi-cal composition of the mussel meat, re-duced the TBA value, the total platecount and lactic acid bacteria count andincreased the TVBN, but had no effect(p>0.05) on the pH value. During stor-age of mussel meat for 90 days at 2°-3°C the total plate count increased by0.8 log10 cfu/g and the lactic acid bac-teria count by 0.4 log10 cfu/g, the pHdecreased by 0.2 units, the TBA valueincreased by 0.2 mg of malonaldehyde/kg and the TVBN increased from 18 to29 mg N/100 g. Based on sensory at-tributes the shelf-life of Mediterraneanmussel meat under the specified smok-ing procedure and storage conditionscan be extended to 70 days.

sta in salamoia salina al 4% per 15min, scolata per 5 min ed essiccata a60°-65°C per 13 min ed affumicata atemperatura ambiente per 17 min, ot-tenendo un valore di F70°C pari a 40,2min. Finalmente la carne era confezio-nata sottovuoto e conservata a 2°-3°C.La stagione di raccolta delle vongole(autunno e primavera) non ha mostratoalcun effetto (p>0,05) sulle caratteri-stiche fisico-chimiche e microbiologi-che della carne. La cottura e l’affumi-camento ne hanno invece influenzato(p<0,05) la composizione chimica, ri-ducendo, da un lato, il valore TBA, laconta microbica totale e la conta deibatteri lattici ed aumentando, dall’al-tro, il valore TVBN. Nessun effetto(p>0,05) si è osservato sul pH. Dopoconservazione della carne di vongoleper 90 gg a 2°-3°C, la conta microbicatotale è incrementata di 0,8 log10 cfu/g, la conta di batteri lattici di 0,4 log10cfu/g, mentre il pH è diminuito di 0,2unità. Il valore di TBA è aumentato di0,2 mg di aldeide malonica/kg. Il va-lore di TVBN è aumentato da 18 a 29mg N/100 g. Sulla base delle valuta-zioni sensoriali la vita di scaffale dellacarne di vongole del mediterraneo, con-servata nelle condizioni descritte, puòessere estesa a 70 gg.

2002; FREITES et al., 2002). They alsohave a low level of n-6 fatty acids, a highratio of n-3/n-6 fatty acids (ORBAN et al.,2002) and a high iron content, about 110ppm, (KARAKOLTSIDIS et al., 1995), andare also considered to be a low-fat andlow-cholesterol food (HOLLAND et al.,1994; ORBAN et al., 2002). The meat ofMytilus galloprovincialis is also appreci-ated by consumers for its organolepticproperties which remain even afterprocessing (ORBAN et al., 2002).

More than 25,000 tons of Mediterra-


nean mussel are produced annually inGreece, the largest part of which is ex-ported to other countries, usually at alow price. On the Greek retail marketmussels are mainly sold in plastic pouch-es with water, which contains either livemussels or shucked mussels. Sealedpouches containing mussels, placed inboxes and covered with crushed ice forchilling and transport, have a shelf-lifeof about 4 to 6 days (VASAKOU et al.,2003). Frozen mussels are also producedin a limited quantity.

Smoking, as well as salting and dry-ing, are used as preservation techniquesof fish and other sea foods that have re-mained virtually unaltered from prehis-tory to the present day. Today the pres-ervation of fish and other seafood prod-ucts is still the prime objective of smok-ing in most parts of the world. Smokedseafood products are regarded as deli-catessen food items, to be consumed onspecial occasions, or presented as tempt-ing alternatives in our diet (HORNER,1992).

KASTANIDOU-MANOUSSOU et al.(1982) studied the keeping quality ofMediterranean mussel meat stored atrefrigerator temperatures of 3°-4°C andfound that its shelf-life was limited to 4days. VASAKOU et al. (2003) studied theeffect of sodium lactate and potassiumsorbate on the quality characteristics ofMediterranean mussel meat. They foundthat the shelf-life of shucked musselsunder chilled storage (5°±2°C) in plas-tic pouches with water could be extend-ed to 8 days by the addition of 0.35%potassium sorbate. However, no re-search has been conducted on the ef-fect of smoking on the quality charac-teristics and shelf-life of Mediterraneanmussel meat.

The objective of this study was a) tostandardize a smoking procedure forMediterranean mussel meat and b) toinvestigate the quality characteristicsand the shelf-life of the smoked productunder chilled storage (2°-3°C).


Experimental design

Fresh (live) mussels were harvestedfrom approved shellfishing waters in anarea near Thessaloniki. The length of themussels was about 5-6 cm and their to-tal weight was about 35 g. The musselswere transported within 1 hour to thelaboratory in isothermal plastic boxesand immediately separated into samplesof about 300 g. These samples were usedin a series of preliminary experiments todetermine the optimum conditions forprocessing and smoking of mussels asdescribed below.

The fresh (live) mussels, about 12 kg,were washed in running tap water andthe byssus thread removed. In order toopen the shells and free the mussel meat,cleaned mussels were steamed in anoven with live steam at an air tempera-ture of about 80°C for 10 min. The mus-sel meat was placed in a 4% sodium chlo-ride brine at room temperature for 15min in order to improve its taste andconsistency. The ratio of mussel meat tobrine was 1:2. The brined mussel meatwas placed in a canister, drained for 5min in a refrigerator and after that placedon racks in a special smokehouse, whichwas heated by electricity. The musselmeat was initially dried at an air tem-perature of 60°-65°C for 13 min. Afterdrying and when the air temperature was65°C, the mussel meat was hot smokedfor 17 min with smoke which was pro-duced in a smoke generator by burningwood shavings under forced air flow. Thesmokehouse was switched-off when theair temperature reached 80°C. A ther-mocouple (Ellab cmc 821, Denmark) wasused to monitor the core temperature ofthe mussel meat. The F70°C value, basedon the destruction of D-steptotococcus(REICHERT and THUEMEL, 1986) wascalculated from the core temperaturedata during hot smoking of the musselmeat. The smoked mussel meat was left


in ambient conditions to cool at roomtemperature and weighed in order tocalculate the processing yield. It wasthen separated into samples of about 75g, which were vacuum packed (-1 barvacuum level) in film pouches (Inplast,Oberaudorf, Germany) with a Weromat-ic vacuum packaging machine (WernerBonk, Bochum, Germany) and stored inthe dark at 2°-3°C for 90 days. The filmpouches were made of polyethylene /polyamide (80/80 µm thickness) with anoxygen transmission rate of 77 cm3/m2/24h/1atm (23°C, 0%RH) (KOTZEKIDOUand BLOUKAS, 1996).

Samples were taken for analysis fromfresh (live) mussels and mussels aftersteaming and smoking, as well as fromsmoked mussel meat after 5, 15, 30, 45,60 and 90 days in chilled storage.

Physicochemical analyses

Moisture, protein, fat (ether-extracta-ble) and sodium chloride content weredetermined according to standard AOAC(1990) procedures. pH was measured ina homogenate prepared by blending 20g of mussel meat with 80 mL of distilledwater for 30 s. Readings were taken witha WTW, model pH 521, digital pH meterand a WTW, type E 56, combination elec-trode (WTW-Wissensehaftlich-Techin-iche Werkstaetten GmbH, Weilheim, Ger-many). Total volatile basic nitrogen(TVBN) was determined according to themethod described by BOTTA (1995) andthe results are expressed as mg N/100g.Thiobarbituric acid (TBA) value was de-termined according to the method de-scribed by RAHARJO and SOFOS (1993)and the results are expressed as mg ofmalonaldehyde/kg.


A 20 g sample of mussel meat washomogenized in a stomacher Lab-blender400 (Seward Medical, London, UK) with180 mL sterile 0.1% peptone water for 2

min. Appropriate dilutions of sampleswere prepared in sterile 0.1% peptonewater blank and plated in duplicate ondifferent growth media. The followingmedia and incubation conditions wereused: (1) Plate Count Agar (Oxoid) at30°C for 72h, for total plate count (totalaerobic mesophilic count); (2) deManRogosa Sarpe (MRS) agar (Oxoid) at 25°Cfor 5 days under anaerobic conditions(BBL “Gas-Pack” System with H2/CO2cartridges), for lactic acid bacteria count.Results are expressed as log10 numbersof colony forming units/gram (cfu/g).

Sensory shelf-life

An experienced five-member panel wasused to evaluate smoked mussel meatduring processing and storage. A prepar-atory session was held prior to testingso that each panel member could thor-oughly discuss and clarify each attributeto be evaluated. Testing was initiatedafter the panelists had agreed on thespecifications.

The panelists were offered a packageof smoked mussel meat but they did notknow its storage time. They were askedto evaluate the appearance and colourof the smoked mussel meat in the pack-age according to the following scale: 5 =excellent appearance, light colour simi-lar to that of fresh mussel meat; 4 = verygood appearance, slightly dark colour; 3= good appearance, fairly dark but ac-ceptable colour; 2 = not good appearance,dark colour rather unacceptable; 1 =completely unacceptable appearance,very dark colour completely unaccepta-ble. The package was subsequentlyopened and the mussel meat was servedto the panelists. Panelists were asked toevaluate the following attributes usingthe respective scale: consistency: 5 = soft,4 = slightly dry, 3 = dry but acceptable,2 = rather dry, 1 = too dry and brittle;odour and taste: 5 = excellent, 4 = good,3 = acceptable, 2 = fair, 1 = unaccepta-ble, smoke intensity: 7 = very intensive-


ly smoked and unacceptable, 6 = inten-sively smoked and unacceptable 5 =slightly smoked, 4 = acceptable smokeintensity, 3 = slightly unsmoked, 2 =rather unsmoked and unacceptable, 1 =completely unsmoked and unacceptable.

The time in days determined by thepanelists to consider the quality to be atthe limit of acceptability was defined asthe shelf-life (BLOUKAS et al., 1997) ofmussel meat under the specified smok-ing and storage conditions. In terms ofscores given, this was when scores of 3and 5 were given by the panelists forsmoke intensity, and a score of 3 forappearance and colour, consistency andodour and taste, or when the score ex-ceeded the above limits, as measured byinterpolation, between the scores givenat two successive storage times.


Four replications were carried out forchemical composition and sodium chlo-ride content of the fresh (live) musselsharvested in April and May 1997. Thesmoking of mussel meat and its preser-vation under vacuum in chilled storage(2°-3°C) was repeated 10 times. Six rep-lications were made for samples harvest-ed from March to May and four for sam-ples harvested from September to Octo-ber 1997. Data collected for physico-chemical and microbiological parametersof fresh (live) mussels were analyzed byone-way analysis of variance in order totest for significant differences betweenthe two harvesting seasons (spring andautumn) and processing of the mussels.Data collected from fresh mussels andmussel meat after steaming and smok-ing were also analyzed by one-way anal-ysis of variance in order to test for sig-nificant processing fluctuations for eachparameter considered. In both casesmeans were compared using the LSD0.05test. Data collected during the chilledstorage of smoked mussel meat wereanalyzed by a two factor factorial analy-

sis in a completely randomized design.The two factors were: (1) the season(spring and autumn) of processing themussels and (2) the storage time. Meanswere compared using the Duncan test.Data analysis was performed with thesoftware SPSS program (Version 11 forWindows).


The chemical composition of fresh(live) mussels as well as their physico-chemical and microbiological parametersare given in Table 1.

Fresh mussels, harvested in April andMay, had a moisture content of77.38%±2.83, with a very low coefficientof variation (CV%). MILETIC et al. (1991)reported the same moisture content(77.92%) for Mytilus galloprovincialis.However, KARAKOLTSIDIS et al. (1995)found that mussels harvested during thesame season had a moisture content of87%±0.1. The protein content of the freshmussel meat was 9.27%±1.16 or about41%, expressed as % of dry matter. KAR-AKOLTSIDIS et al. (1995) reported a pro-tein content of 8%±0.2 for Mediterrane-an mussel and MILETIC et al. (1991) re-ported 42.06-44.27% on dry matter. Thefat content of fresh mussels was1.51%±0.65 and was close to that report-ed by KARAKOLTSIDIS et al. (1995). Ex-pressed on the dry matter basis, the fatcontent was 6.67%. MILETIC et al. (1991)found that Mytilus galloprovincialis had9.61% fat content on dry matter. Therewas a high (43%) coefficient of variationfor fat content, indicating that the fatcontent of fresh mussels is quite varia-ble. The sodium chloride content of freshmussels was 1.42%±0.26, with a coeffi-cient of variation (CV%) 18.30%, and thewater phase salt content (brine) was1.80%±0.17. A significant seasonal fluc-tuation has been reported for the chem-ical composition of mussels (KARAKOLT-SIDIS et al., 1995; FERNANDES-REIRIZ et


Table 1 - Effect of steaming and smoking on the physicochemical and microbiological parameters ofMediterranean mussel meat (Mytilus galloprovincialis).

Fresh mussels Steamed mussels Smoked mussels

Parameters Mean SD CV% Mean SD CV% Mean SD CV%

Processing yield (%) 9.5 2.10 4.6Moisture (%) 77.38a 2.83 3.65 71.10b 1.55 2.18 62.60c 3.19 5.11Protein (%) 9.27c 1.16 12.51 11.96b 0.77 6.49 15.59a 1.63 10.46Fat (%) 1.51c 0.65 43.00 2.01bc 0.71 35.35 3.29a 1.31 39.82Sodium chloride % 1.42 0.26 18.30 1.26 0.22 17.82 1.62 0.31 19.14Brine (%)* 1.80 0.17 13.8 1.74 0.12 11.02 2.52 0.23 15.6pH value 6.47 0.10 1.54 6.22 0.16 2.57TVBN (mg N/100g) 12.84b 1.06 8.25 18.35a 0.78 4.25TBA value** 1.18a 0.61 51.16 0.32b 0.14 43.75Total plate count*** 3.14a 0.57 18.16 1.94b 0.55 35.03Lactic acid bacteria 2.58a 0.26 10.0 1.90b 0.27 14.67

* Brine or water phase salt content (%) was calculated using the formula: [(NaCl%)/( NaCl% + H2O%)] x 100,**Expressed as mg of malonaldehyde/kg, ***Expressed as log10 cfu/g,SD=standard deriation, CV%= coefficient of variation,a-c Means within the same row with different superscript letters are significantly different (p<0.05)

al., 1996; ORBAN et al., 2002), reflectingthe stage of gonadal development as wellas the availability of food.

The effect of steaming and smoking onprocessing yield, composition and phys-icochemical and microbiological param-eters of mussel meat is given in Table 1.The average processing yield of smokedmussel meat during the spring was9.5%±1.12. This low processing yield isattributed to the small meat portion offresh (live) mussels, ranging from 25 to35 g/100 g of fresh weight, and to theirhigh moisture content, which accountsfor high water losses during processing.The meat content of mussels shows sea-sonal variations (OKUMUS and STIRLING,1998; ORBAN et al., 2002), which affectsthe processing yield of smoked musselmeat.

Steaming and smoking of musselsunder the specified conditions reduced(p<0.05) the moisture content and in-creased (p<0.05) the protein and fat con-tent, as a result of water removal. Themoisture content of smoked mussel meatwas 62.60%±3.19, the protein content

15.59%±1.63 and the fat content3.29%±1.31. The sodium chloride con-tent decreased to 1.26%±0.22 duringsteaming and increased to 1.62%±0.31after brining and smoking, but thesechanges were not significant (p>0.05).However, brining and smoking signifi-cantly increased (p<0.05) the waterphase salt content (brine %) from1.74%±1.12 to 2.52%±0.23.

The season of harvesting had no ef-fect (p>0.05) on the microbiological andphysicochemical parameters of fresh(live) mussels (Table 2).

The total plate count and lactic acidbacteria count of fresh (live) mussels waslow, 3.14±0.57 and 2.58±0.26 log10 cfu/g respectively (Table 1). VASAKOU et al.(2003) reported a total plate count of 3.7log10 cfu/g and a lactic acid bacteriacount of 3.22 log10 cfu/g for fresh mus-sels harvested in different shellfishingareas.

The growth retarding and lethal effectof smoking on spoilage and pathogenicmicroflora depends on the content of saltin the water phase of the product, the


Table 2 - Analysis of variance on the effect of season and storage time on the physicochemical andmicrobiological parameters of fresh (live) mussels and smoked musel meat (F-values of indepententvariables and interaction).

Source of variance

Parameter Fresh mussels Smoked mussel meat

Aa A Ba AxB

Total plate count 0.35ns 0.29ns 4.37** 0.67nsLactic acid bacteria 1.02ns 0.06ns 1.50ns 0.18nspH 4.50ns 0.83ns 0.511ns 0.97nsTVBN 0.01ns 3.64ns 5.98*** 0.31nsTBA 0.30ns 3.54ns 1.00ns 0.22ns

*p<0.05; **p<0.01; ***p<0.001; ns=not significant.a A=season; B=storage time.

temperature, humidity and density of thesmoke, the duration of smoking, the con-centration of active components in thesmoke and the time and temperature ofheating. A very intensive heat processwas applied during hot smoking of mus-sel meat equal to F70°C = 40.2 min (Table3), which is significantly higher than theF70°C =14 min value required for a 5Ddestruction of D-streptococcus (RE-ICHERT and THUEMEL, 1986).

As result of the intensive heat applied,smoking significantly reduced the totalplate count and the lactic acid bacteriacount (p<0.05) of the mussel meat (Ta-ble 1). During subsequent storage of thesmoked mussel meat under the speci-fied conditions, the total plate count con-tinuously increased from 1.94 to 2.77log10 cfu/g (Fig. 1a). Storage time signif-icantly (p<0.05) affected the total platecount, while the season of harvesting andprocessing, as well as the interaction ofseason and storage time had no effect(Table 2). The lactic acid bacteria countof smoked mussel meat also increasedduring chilled storage from 1.90 to 2.28log10 cfu/g (Fig. 1b), but neither the sea-son of processing nor the storage timehad a significant effect on it (Table 2).The very small increase in total plate

Table 3 - Temperature change during smoking ofmussel meat and calculated F70°C value.

Time Smokehouse Mussel meat Lethality(min) air temperature core temperature Rate*

(°C) (°C) (min)

2 23 16.88 54 30.1

16 63 54.2 0.01017 64 57.6 0.05718 65 60.7 0.12019 68 63.5 0.22620 70 65.5 0.35721 72 67.5 0.56522 74 69.1 0.81223 76 70.6 1.15124 77 71.9 1.55125 78 73.1 2.04626 79 74.0 2.51227 80 74.9 3.09728 80 75.8 3.81729 79 76.6 4.59930 78 77.2 5.27531 77 77.8 6.10632 65 78.2 6.69233 54 76.4 4.59934 40 60.0 0.100

Total 40.256

* For D. streptococcus (REICHERT and THUEMEL,1986).


Fig. 1 - Effect of smoking on (a) total plate countand (b) lactic acid bacteria count of Mediterrane-an mussel meat under vacuum in chilled storage(2°-3°C). ◆ spring; ■ autumn.

count and lactic acid bacteria count dur-ing chilled storage, 0.83 and 0.38 log10cfu/g, respectively, is due to the very lowinitial microbial population, as well asto the combined action of brining andsmoking of the mussel meat, packagingunder vacuum and storage at low tem-peratures (2°-3°C). Salting reduces thewater activity of the product, while dur-ing smoking, the surface of the productis dried, providing a physical barrier tothe passage of microorganisms and ahostile environment for aerobic micro-bial proliferation, while antimicrobial

substances, such as phenols and formal-dehyde, are transferred to the productinhibiting microbial growth (HORNER,1992 ).

Among the pathogenic bacteria, Lis-teria monocytogenes is able to survive,and in many cases to increase in num-bers, on vacuum-packed smoked sea-food products under conditions ofchilled storage (HARTEMINK andGEORGSSON, 1991; RORVIK et al., 1991;PETERSON et al., 1993; EMBAREK, 1994;THAM et al., 2000). NG and SEATH (1995)isolated L. monocytogenes from smokedmussels and a small outbreak of liste-riosis was reported by BRETT et al.(1998) to be associated with smokedmussels. BREMER and OSBORNE (1995)found that the thermal-death time of L.monocytogenes in green shell mussels(Perna canaliculus) prepared for hotsmoking was D60°C = 5.49 min and z =4.25°C. The D60°C = 5.49 min value forL. monocytogenes, which is equivalentto D65.5°C = 0.27 min, is much lower thanthe D65.5°C = 0.5-1.0 min of mesophilicnon-spore-bearing spoilage bacteria(HELDMAN and HARTEL, 1997), in whichD-streptococcus are included. There-fore, the heat process applied to themussel meat during smoking, F70°C =40.2 min (Table 3), is adequate for thesafety of the product from L. monocy-togenes. Other pathogenic bacteria as-sociated with smoked seafood productsare the non-proteolytic Clostridium bot-ulinum type E, Staphylococcus aureusand Salmonella species (KOLODZIESKAet al., 2002). All these pathogenic bac-teria are also less heat resistant thanD-streptococcus and therefore the heatprocess applied during smoking of themussel meat is adequate for the safetyof the product.

The pH of fresh mussels was6.47±0.10 with a very low coefficientof variation (1.54%). Similar pH val-ues for fresh mussels have been re-ported by VASAKOU et al. (2003) andKASTANIDOU–MANOUSOU et al. (1982).

(a)

(b)


Smoking slightly reduced the pH val-ue of mussels to 6.22±0.16, but thedifference was not significant (Table1). During chilled storage of smokedmussel meat under vacuum, the pHwas slightly reduced by about 0.2units (Fig. 2a). This small change inpH during chilled storage of smokedmussel meat is probably due to the lowinitial level of lactic acid bacteria aswell as to the combined action of allthe above-mentioned factors whichinhibit their growth. Values of pH 6.0-6.3 have also been reported for vacu-um-packed smoked salmon (CIVERA etal., 1995).

The total volatile basic nitrogen(TVBN) of fresh mussels was 12.84±1.06mg N/100g. Similar values (14.67 mgN/100g) were reported by VASAKOU etal. (2003). According to CONNEL (1990),some TVBN, about 20 mg N/100g, ispresent even in very fresh fish. Smok-ing significantly increased (p<0.05) theTVBN to 18.35±0.78 mg N/100g (Table1). Storage time had a significant effect(p<0.001) on TVBN levels (Table 2).TVBN remained almost constant thefirst 45 days of chilled storage, afterwhich it continuously increased(p<0.05) to 29.43 mg N/100g (Fig. 2b).However, this value is lower than theupper limit of 35-40 mg N/100g beyondwhich round, whole chilled fish can beconsidered too spoiled for most uses,and much lower than the upper limit of100-200 mg N/100g for a variety of salt-ed and dried fish (CONNEL, 1990). CIV-ERA et al. (1995) suggested 45 mg N/100g as the acceptable limit of TVBNfor cold smoked fish products and CAN-TONI et al. (1993) recommended 40 mgN/100g as the maximum TVBN contentfor smoked salmon. A low correlation (r= 0.47, p<0.01) was found betweenTVBN and total plate count.

The thiobarbituric acid (TBA) value offresh mussels, expressed as mg ofmalonaldehyde/kg, was 1.18±0.61 witha high (51.16%) coefficient of variation.

Fig. 2 - Effect of smoking on (a) pH value, (b) totalvolatile basic nitrogen (TVBN) and (c) TBA value ofMediterranean mussel meat under vacuum inchilled storage (2°-3°C) ◆ spring; ■ autumn.

VASAKOU et al. (2003) found a TBA val-ue for fresh mussels of 1.08 mg ofmalonaldehyde/kg. According to CON-NEL (1990) the TBA values should be low-er than 1-1.5 mg of malonaldehyde/kgfor fresh fish. Smoking significantly re-duced (p<0.05) the TBA value of musselmeat to 0.32±0.14 mg of malonaldehyde


/kg (Table 1). This reduction may be dueto the fact that volatile compounds whichare responsible for the TBA value are lostduring the steaming and smoking ofmussels. During chilled storage of thesmoked mussel meat for 90 days underthe specified conditions the TBA valueremained almost constant, since it in-creased by only about 0.2 mg of malonal-dehyde/kg (Fig. 2c). This is probably due

Fig. 3 - Effect of smoking on (a) appearance andcolour, (b) consistency, (c) odour and taste and (d)smoke intensity of Mediterranean mussel meatunder vacuum in chilled storage (2°-3°C).

to the antioxidant action of the smokewhich contains phenolic antioxidantcompounds that delay the autoxidationand rancidity of highly unsaturated fat-ty acids (HORNER, 1992).

During storage of the smoked mus-sel meat for 90 days under the speci-fied conditions, the consistency of thesmoked mussel meat remained constant(Fig. 3b). All other sensory attributesalso remained constant for 45 days.However, between the 68th and 74th daythe colour of the meat changed fromlight to dark (Fig. 3a), the odour andtaste decreased (Fig. 3c) and the inten-sity of smoke increased and exceededthe limit of acceptability (Fig. 3d). Basedon these sensory attributes the shelf-life of Mediterranean smoked musselmeat, in chilled storage (2°-3°C) undervacuum, was restricted to about 70days. These results are in agreementwith those of CIVERA et al. (1995) whofound that the effective shelf-life of vac-uum-packed smoked salmon is nearly40-50 days and it can be increased to80 days if the fish product is kept attemperatures of 2°-3°C.

CONCLUSION

Smoking can be used for the process-ing of Mediterranean mussel meat and theproduction of a delicatessen food. Theshelf-life of smoked Mediterranean mus-sel meat under vacuum in chilled storage(2°-3°C) can be extended to about 70 days.

REFERENCES

AOAC. 1990. “Official Methods of Analysis”, 15th

ed. Association of Official Analytical Chemists,Arlington, VA.

Bloukas J.G., Paneras E.D. and Fournitzis G.C.1997. Sodium lactate and protective cultureeffects on quality characteristics and shelf-lifeof low-fat frankfurters produced with olive oil.Meat Sci. 45: 223.

Botta J.R. 1995. “Evaluation of Seafood Freshness


Quality”. p. 26. VCH Publishers, Inc. New York.USA.

Bremer P.J. and Osborne C.M. 1995. Thermal-death times of Listeria monocytogenes in greenshell mussels (Perna canaliculus) prepared forhot smoking. J. Food Prot. 58: 604.

Brett M.S.Y., Short P. and McLauchlin J. 1998. Asmall outbreak of listeriosis associated withsmoked mussels. Int. J. Food Microbiol. 43: 223.

Cantoni C., Moret S. and Comi G. 1993. Indici bat-teriologi e chimici per valutare la qualità del sal-mone affumicato. Industrie Alimentari 23: 842.

Civera T., Parisi E., Amero G.P. and Giaccone V.1995. Shelf-life of vacuum-packed smokedsalmon: microbiological and chamical changesduring storage. Arch. Lebensmittelhyg. 46: 1.

Connel J.J. 1990. “Control of Fish Quality”. 3rd ed.p. 185. Whitestable Litho, London.

Embarek P.K.B. 1994. Presence, detection andgrowth of Listeria monocytogenes in seafood: areview. Int. J. Food Microbiol. 23: 17.

Fernandez-Reiriz M.J. Labarta U. and BabarroJ.M.F. 1996. Comparative allometries in growthand chemical composition of mussels (Mytilusgalloprovincialis Lkm) cultured in two zones inthe Ria sada (Galicia, NW Spain). J. ShellfishRes. 15: 349.

Freites L., Fernandez-Reiriz M.J. and Labarta U.2002. Fatty acid profiles of Mytilus galloprovin-cialis (Lmk) mussel of subtidal and rocky shoreorigin. Comparative Biochemistry and Physiol-ogy Part B: Biochemistry and Molecular Biolo-gy 132: 453.

Hartemink R. and Georgsson F. 1991. Incidenceof Listeria monocytogenes in seafood and sea-food salads. Int. J. Food Microbiol. 12: 189.

Heldman D.R. and Hartel R.W. 1997. “Principlesof Food Processing”. p. 21. Chapman & Hall,International Thomson Publishing, N. York.

Holland B., Welch A.A., Unwin I.D., Buss D.H., PaulA.A. and Southgate D.A.T. 1994. “The Compo-sition of Foods”. 5th ed. p. 214. McCance andWiddowson’s, The Royal Society of Chemistry.Ministry of Agriculture, Fisheries and Food.Cambridge.

Horner W.F.A. 1992. Preservation of fish by cur-ing (drying, salting and smoking). In “FishProcessing Technology”. G.M. Hall (ed), p. 61.Blackie Academic & Professional, New York.

Karakoltsidis P.A., Zotos A. and ConstantinidisS.M. 1995. Composition of the commerciallyimportant Mediterranean finfish, crustaceansand molluscs. J. Food Composition and Analy-sis 8: 258.

Kastanidou-Manousou Chr., Karaioannoglou Pr.and Koidis P. 1982. Keeping quality of shucked

mussels stored at refrigerating temperatures.Bulletin of the Hellenic Veterinary Medical So-ciety 25: 200.

Kolodziejska I., Niecikowska C., Januszewska E.and Sikorski Z.E. 2002. The microbial and sen-sory quality of mackerel hot smoked in mild con-ditions. Lebensm.-Wiss. u.-Technol. 35: 87.

Kotzekidou P. and Bloukas J.G. 1996. Effect ofprotective culture and packaging film permea-bility on shelf-life of sliced vacuum-packedcooked ham. Meat Sci. 42: 333.

Miletic I. Miric M., Lalic Z. and Sobajic S. 1991.Comparison of lipids and proteins of severalspecies of molluscs, marine and terrestrial fromthe Adviatic sea and Serbia. Food Chem. 41:303.

Ng D.L.K. and Seath H.L. 1995. Isolation and iden-tification of Listeria monocytogenes from a rangeof foods in Singapore. Food Control 6: 171.

Okumus I. and Stirling H.P. 1998. Seasonal varia-tions in meat weight, condition index and bio-chemical composition of mussels (Mytilus edu-lis L.) in suspended culture in two Scottish sealochs. Aquaculture 159: 249.

Orban E., Di Lena G., Nevigato T., Casini I., MarzettiA. and Caproni R. 2002. Seasonal changes inmeat content, condition index and chemicalcomposition of mussels (Mytilus galloprovincia-lis) cultured in two different Italian sites. FoodChem. 77: 57.

Peterson M.E., Pelroy G.A., Paranjpye R.N., PoyskyF.T., Almond J.S. and Eklund M.W. 1993. Pa-rameters for control of Listeria monocytogenesin smoked fishery products: sodium chlorideand packaging method. J. Food Prot. 56: 938.

Raharjo S. and Sofos J. N. 1993. Methodology formeasuring malonaldehyde as a product of lipidperoxidation in muscle tissues. A Review. MeatSci. 35: 145.

Reichert J.E. und Thuemel H. 1986. Zur Ermit-tlung des Erhitzungs Effectes von Frishware (F-wert). Fleiserei 37: 350.

Rorvik L.M., Yndestad M. and Skejerve E. 1991.Growth of Listeria monocytogenes in vacuum-packed, smoked salmon during storage at 4°C.Int. J. Food Prot. 14: 111.

Tham W., Ericsson H., Loncarevic S., UnnerstadH. and Danielsson-Tham M.-L. 2000. Lessonsfrom an outbreak of listeriosois related to vac-uum-packed gravad and cold-smoked fish. Int.J. Food Microbiol. 62: 173.

Vasakou A., Vareltzis K. and Bloukas J.G. 2003.Effect of sodium lactate and potassium sorbateon quality characteristics and shelf-life of Med-iterranean mussel meat (Mytilus galloprovincia-lis) during chilled storage in pouches with wa-ter. Ital. J. Food Sci. 15: 359.

Revised paper received December 16, 2002 Accepted March 11, 2003


PAPER

- Key words: crumb firmness, modified atmosphere packaging, oxygen scavenger, shelf life -

SHELF LIFE EXTENSIONOF DURUM WHEAT BREAD

ESTENSIONE DELLA SHELF LIFE DI PANEDI SEMOLA DI GRANO DURO

M.A. DEL NOBILE*, T. MARTORIELLO1, S. CAVELLA3, P. GIUDICI2 and P. MASI3

Istituto di Produzioni e Preparazioni Alimentari, University of Foggia,Via Napoli 25, 71100 Foggia, Italy

1 Consorzio “G. P. Ballatore”, Corso Vittorio Emanuele 121,94010 Catenanuova (En), Italy

2 University of Modena and Reggio Emilia, ItalyDepartment of Agricultural Science, Via Kennedy 17, 42100 Reggio Emilia, Italy

3 Department of Food Science, University of Naples “Federico II”,Parco Gussone, 80055 Portici (Na), Italy

* Corresponding Author: Tel. +39 0881 589233; Fax +39 0881 740211;E-mail: [email protected]

ABSTRACT

The aim of this study was to investi-gate packaging methods to prolong theshelf life of durum wheat bread pack-aged in a bag type package without re-frigeration beyond three days (the cur-rent limit for this bread). Two differentgas compositions (40%CO2/60%N2 and80%CO2/20%N2) and an oxygen scav-enger were tested in conjunction withtwo high barrier multilayer films, giv-ing six different packaging combina-tions. The shelf life tests were carriedout at 30°C, monitoring the crumb elas-

RIASSUNTO

Nel presente lavoro è presentato unstudio sull’effetto sia dell’atmosferamodificata sia di assorbitori d’ossigenosulla shelf life di pane ottenuto da se-mola di grano duro. A tale scopo sonostate utilizzate due diverse composizionidi gas (40%CO2/60%N2 e 80%CO2/20%N2) e un assorbitore d’ossigeno. Itest per valutare la shelf life sono statieseguiti a 30°C, ogni campione venivavisionato giornalmente per rilevarel’eventuale presenza di muffe. Inoltre,sono state eseguite prove di compres-


tic modulus and evaluating the sam-ples for visible signs of mold. The re-sults showed that it is possible to pro-long the shelf life of durum wheat breadfrom three days to about 18 days byusing modified atmosphere packagingwith high carbon dioxide concentrationor by means of oxygen scavengers.

sione allo scopo di determinare il gra-do di avanzamento del fenomeno di raf-fermimento della mollica. I risultati ot-tenuti mostrano che è possibile prolun-gare la shelf life del pane preso in esa-me dagli attuali tre giorni a 18 giorniutilizzando sia atmosfera protettiva conelevata concentrazione di CO2 sia glioxygen scavenger.

INTRODUCTION

Generally the shelf life of bread is lim-ited by several deterioration processesincluding microbiological spoilage, theloss of water and staling. Staling refersto a broad set of sensory and chemicalchanges that affect the crispiness of thecrust, the textural properties of thecrumb and the aroma. Enzymatic treat-ment of bread has proven to be very ef-fective in delaying crumb firming, andthe use of antimicrobial agents is effec-tive in inhibiting mold growth.

Several regions of southern Italyhave started research programs aimedat valorizing typical products, i.e.,products manufactured in a limited ge-ographical area by using traditionaltechnologies and raw materials grownin the same area. One of these prod-ucts is bread made from two durumwheat cultivars (75% Simeto and 25%Duilio) and fermented using sour-dough. This durum wheat bread hasgreat potential for export since in thelocal area consumer demand is satis-fied. It is generally characterized by asoft crust, and a crumb with low po-rosity and high water content. This typeof bread is usually packed in a bag-type package made of either micro-per-forated or continuous film. Because ofthe short shelf life, this durum wheatbread is currently sold only at localmarkets. In order for this bread to beexported, the shelf life should be ex-

tended beyond the present time of threedays.

Preliminary studies have shown that(for this bread) spoilage takes place af-ter a few days of packaging. It is causedeither by the presence of visible signs ofmold or by crumb attribute modification.The former spoilage phenomenon is themain cause of unacceptability when thebread is packaged using continuous film,while the latter phenomena, accompa-nied by a rapid loss of aroma, occurswhen it is packed using micro-perforat-ed film.

Generally, it is difficult and expensiveto delay mold growth and the crumbfirming process without modifying thebread composition and technology. Infact, crumb textural properties are af-fected by several factors such as: tem-perature, crumb water content, the de-gree of starch retrogradation and theextent of interaction between starch andlipids (PONTE et al., 1993). Microbialspoilage caused by the growth of aero-bic microorganisms such as mold, de-pends on the temperature, concentrationof oxygen in the package head space andon the extent of bread contaminationprior to packaging.

According to the data in literature, itis possible to delay both microbialgrowth and the crumb firming processby packaging under carbon dioxide inconjunction with high barrier films(SMITH et al., 1983; AVITAL et al., 1990;KNORR and TOMLINS, 1985; CENCIC et


al., 1996). It has been reported thatmodified atmosphere packaging usingCO2 and N2 has been successfully ap-plied to bakery products to prolong theirshelf life (AVITAL et al., 1990; KNORR andTOMLINS, 1985). In some cases, modi-fied atmosphere packaging fails to pro-long the shelf life of food items, espe-cially porous products, where some ox-ygen usually remains inside the pack-age. Gas flushing and analogous phys-ical methods are rarely used to main-tain the oxygen level below the thresh-old value because they are not econom-ical (BERENZON and SAGUY, 1998). Toovercome this problem, oxygen scaven-gers have been successfully used to re-duce the oxygen partial pressure in thepackage headspace (BERENZON andSAGUY, 1998).

The purpose of this research was toinvestigate ways of extending the shelflife of durum wheat bread beyond threedays without refrigeration by acting onlyon the packaging technology withoutmodifying the bread composition andtechnology used to make it.


Materials

Commercially available sliced durumwheat bread was supplied by Valle delDittaino (Cooperativa Agricola Valle delDittaino, Enna, Italy), and it was used asreceived. Two commercially available bar-rier multilayer films were used. Film 1,composed of PE(40 µm)/EVOH(5 µm)//PE(40 µm)//PET(12 µm), was suppliedby Di Mauro (Di Mauro Officine Grafiche,Cava dei Tirreni, Italy); film 2, composedof PP(75 µm)/Nylon6,6 (15 µm), was sup-plied by Cryovac (Cryovac Sealed AirCorp., Passirana di Rho, Italy). As report-ed by HERNANDEZ (1994), the water andoxygen permeability of water sensitivefilms such as polyamides and EVOH de-pends on the water activity inside and

outside the package. In these cases, thewater and oxygen permeability coefficientcannot be determined by a single meas-urement. A more accurate analysis of thepermeation process is necessary to cor-rectly determine the barrier properties ofthe film in actual working conditions. Acomplete characterization of the water andoxygen transport properties of the multi-layered films used goes beyond the aim ofthis paper and will therefore not be con-sidered here. A rough estimate was madeof the water and oxygen permeability at30°C based on the data reported in theliterature (BAKER, 1986; BRANDRUP andIMMERGURT, 1975): film1/O2 = 1·10-10

cm3(STP)·cm·cm-2·s-1·atm-1; film1/H2O =9·10-7 cm3(STP)·cm·cm-2·s-1·atm-1; film2/O2= 4·10-9 cm3(STP)·cm·cm-2·s-1·atm-1; film2/H2O = 6·10-7 cm3(STP)·cm·cm-2·s-1·atm-1.

The oxygen scavenger used was Age-less FX100 produced by Mitsubishi GasChemical Company (Tokyo, Japan). It issold in the form of a powder containedin a little sachet.

Six different packaging combinationswere investigated by combining the de-scribed packaging materials; for the sakeof simplicity they will be referred to as:A1 = Film 1 + 40% CO2/60% N2; B1 =Film 1 + 80% CO2/20% N2; C1 = Film 1+ oxygen scavenger; R1 = Film 1 + air;A2 = Film 2 + 40% CO2/60% N2; B2 =Film 2 + 80% CO2/20% N2; C2 = Film 2+ oxygen scavenger; R2 = Film 2 + air.R1 and R2 were used as controls.

Packaging procedure

Modified atmosphere packaging: 1.08cm thick bread slices were packed indi-vidually in 35x30 cm pouches using amodel MG 400 Vuotomac (Vuotomac,Alessandria, Italy). The pouches wereevacuated, gas flushed with a previous-ly prepared gas mixture (40% CO2/60%N2 and 80%CO2/20%N2) and thenheat-sealed. The time taken to evacuateand flush the pouches with the gas mix-ture was the same for all the packages.


This ensured that the amount of oxygenentrapped in the porous structure of thebread did not change from one pouch toanother.

Oxygen scavenger: 1.08 cm thickbread slices were packed in 35x30 cmpouches. After introducing the oxygenscavenger into the pouch, it was heatsealed using a model 421 Magulta seal-er (Gaudius Saldatrici, Milan, Italy).

Reference sample: 1.08 cm thick breadslices were packed in 35x30 cm pouch-es using both types of film and then heatsealed using a model 421 Magulta seal-er (Gaudius Saldatrici, Milan, Italy).

Mechanical tests

Uniaxial compression tests were car-ried out using a model 4301 UniversalTesting Machine (Instron Ltd ItalianBranch, Milan, Italy, maximum load al-lowed 100 N) at a crosshead speed of 10mm/min. Cylindrically shaped samples,with a diameter of 2.93 cm and a heightof 1.08 cm, were punched out of thebread slice immediately after removingthe slice from the pouches, and thencompressed according to the above pro-cedure.

Evaluation of the crumb water content

Immediately after removing the slicefrom the pouches, part of the crumbwas separated from the slice; it wasweighed and then placed in an oven at110°C. The weight of the crumb wasmonitored till it reached a constantvalue. The crumb water content wasevaluated by means of the followingrelationship:

Pin. - Pfin.

Pin.

where: Pin. is the initial weight of thecrumb, Pfin. is the final steady weight at-tained in the oven.

Microbiological spoilage

For each type of packaging combina-tion tested, three bread slices were pack-aged separately according to the aboveprocedures and analyzed daily for visi-ble signs of mold. The bread was con-sidered unacceptable if one of the threeslices showed visible signs of mold. Aconventional pour plate microbial countwas not carried out for two reasons: first-ly, during the homogenization phase, themycelia always broke up, giving rise to avery large number of colony formingunits (cfu) (BOURGEOIS et al., 1988); sec-ondly, since the bread slice was not uni-formly contaminated, estimation of theaverage microbial count would not berepresentative of the local microbial pop-ulation. In fact, the method used in thepresent study to determine microbialspoilage of the bread tested is very closeto the consumer’s perception of breadquality.


As reported above, bread can be re-jected by the consumer either on thebasis of a change in the textural prop-erties of the crumb or because of thepresence of visible signs of mold. Thesetwo aspects will be discussed separate-ly.

Microbial shelf life

For the purpose of this study micro-bial shelf life is defined as the timeelapsed from the time the bread ispackaged to the time the packagedslice shows visible signs of mold. Thefirst column in Table 1 reports thevalues of the microbial shelf life ob-tained for the different packaging com-binations tested. For samples C1 andC2, a value is not reported since thebread slices stored in these packagesdid not show any visible signs of mold


Table 1 - Microbial shelf life, textural shelf life and shelf life of the samples.

Sample Microbial Shelf Life Textural Shelf Life Shelf Lifedays days days

A1 12 14.6 12Film 1 + 40% CO2/60% N2 [12.0-17.7]*

B1 22 19.5 19.5Film 1 + 80% CO2/20% N2 [17.2-22.6]* [17.2-22.6]*

C1 > 38 16.6 16.6Film 1 + oxygen scavenger [14.6-18.7]* [14.6-18.7]*

R1 3 - 3Film 1 + air (control)

A2 12 18.0 12Film 2 + 40% CO2/60% N2 [15.8-21.0]*

B2 22 18.9 18.9Film 2 + 80% CO2/20% N2 [16.5-21.4]* [16.5-21.4]*

C2 > 38 19.8 19.8Film 2 + oxygen scavenger [17.5-22.3]* [17.5-22.3]*

R2 3 - 3Film 2 + air (control)

* The values reported in the brackets are the lower and upper limit of the confidence interval (95%).

during the entire time of observation(i.e., 38 days). The microbial shelf lifeof the two reference packages (threedays) was considerably shorter thanthat of the other packages. The resultsshow that the oxygen scavenger is aneffective solution for reducing thegrowth of aerobic spoilage microorgan-isms such as mold. In modified atmos-phere packaging the microbial growthrate diminished because of the bacte-riostatic effect of carbon dioxide cou-pled with a decrease in oxygen fromthe package head space. In fact, themicrobial shelf life of bags containinga high concentration of CO2 had alonger shelf life. In the case of the ox-ygen scavenger, the active compoundput into in the package reacts contin-

uously with the oxygen present insidethe package, reducing its concentra-tion, and consequently the microbialgrowth rate. These results suggest thatoxygen scavengers are more effectivein prolonging the microbial shelf lifeprobably because they eliminate theoxygen that remains trapped insidethe porous structure of the food prod-uct during packaging.

Data reported in Table 1 show that theoxygen barrier properties of the twomultilayer films used did not influencethe microbial shelf life of the bread pack-aged in modified atmosphere, suggest-ing that the oxygen trapped in the po-rous structure of the bread is quite im-portant for microbial spoilage of thepacked food product.


Textural shelf life

Textural shelf life is defined as thetime elapsed from the time the breadis packaged to the time it is rejectedby the consumer on the basis ofchanges in the textural properties ofthe crumb. As reported in the litera-ture, the textural properties of thecrumb are frequently evaluatedthrough compression tests (BAKER etal., 1986; 1987; BAKER and PONTE,1987; PIAZZA and MASI, 1995). Theresulting stress-strain curves dependon the porous structure of the crumb(or, more simply, on the relative den-sity of the crumb, defined as the ratiobetween the crumb density and thedensity of its solid matrix) as well ason the elastic modulus of its solidmatrix (HILYARD, 1982). However, dur-ing storage, the relative density of thecrumb does not change substantially,while the elastic modulus of its solidmatrix increases due to several phe-nomena including the loss of water,the retrogradation of starch and in-crease in the extent of the interactionsbetween gluten and starch (PONTE etal., 1993). In view of the above sce-nario of the firming process, decay incrumb quality during storage can, inprinciple, be followed by monitoringthe elastic modulus of its solid matrix.However, since the relative density ofthe crumb does not change duringstorage, the elastic modulus of thesolid matrix of the crumb is directlyproportional to the elastic modulus ofthe crumb (HILYARD, 1982). For theabove reason, in this study increasein crumb firmness was followed bymonitoring its elastic modulus duringstorage.

The elastic modulus of the crumbcan be evaluated from the initial slopeof the compressive stress-strain curve.However, since the surfaces on whichthe compressive load is applied are notperfectly parallel (as they should be),

estimation of the elastic modulus ofthe crumb based on such a techniqueis generally affected by an error thatis quite significant. For this reason, amathematical model capable of de-scribing the entire compressive stress-strain curve is proposed. Consideringboth the complexity of the deforma-tion mechanisms and their depend-ence on the porous structure of thecrumb, a suitable mathematical mod-el can be obtained by factoring thecompressive behavior of the crumb intwo parts: an elastic modulus andsome function of the strain (HILYARD,1982):

σ (ε) = EC·ε·ξ(ε) (1)

where: ε is the engineering strain, i.e. theabsolute deformation divided by the ini-tial height of the sample, σ is the engi-neering stress, EC is the elastic modulusof the crumb and ξ(ε) is a function thatreflects the influence of the buckling ofthe cell elements of the crumb and theirsubsequent compaction on the shape ofthe compressive stress-strain curve. Tobe representative of the above phenom-ena, the function ξ(ε) has to fulfill thefollowing requirements:

dξ(ε)ξ(0) = 0, [ ]

ε=0

= 0.dε

In the present investigation the follow-ing empirical relationship for ξ(ε) is pro-posed:

ξ(ε) = {-K1·ε1.1·exp(ε·K2)+exp[(K3·ε)

K4]} (2)

where: the values of Ki are constant andhave to be regarded as fitting parame-ters. Substituting equation (2) in equa-tion (1) the following equation is ob-tained:

σ(ε) = EC·ε·{-K1·ε1.1·exp(ε·K2)+

+exp[(K3·ε)K4]} (3)


Fig. 1 shows the compressive stress-strain curve relative to sample C1 alongwith the most suitable fit of the dataobtained by means of equation (1). It isevident that the ability of the proposedmodel to fit the experimental data isquite satisfactory.

Figs. 2 and 3 show the crumb elas-tic modulus of the packaged bread slic-es (–EC) plotted as a function of storagetime for all the packaging combina-tions, while Figs. 4 and 5 show thecrumb water content of the packagedslice plotted as a function of storagetime. Each of the values of –EC reportedin Figs. 2 and 3 were obtained by tak-ing the average of the values of the elas-tic modulus of four cylinders punchedout of a single slice of bread. For eachcylinder the elastic modulus was ob-tained by fitting the compressivestress-strain curve by means of equa-tion (1). The error bars appearing inFigs. 2 and 3 are the standard errorsof the mean (MANDEL, 1964) charac-terizing each value of –EC. The data rel-ative to the two reference samples are

not reported in the figures because oftheir short microbial shelf life, threedays, for which it was useless to meas-ure the –EC, and consequently the crumbwater content, for periods of time longerthan three days. As shown in these fig-ures, for all the samples studied therewas an increase in the storage time ofthe crumb elastic modulus with a de-crease in the crumb water content.

As is the case of white bread (GIO-VANELLI and PAGLIARINI, 1996), the un-acceptability threshold of the elasticmodulus of the crumb, –EC

Threshold, was setequal to three times the value of the elas-tic modulus of the fresh crumb, –EC

Fresh

(i.e., –ECThreshold = 3·–EC

Fresh). To determinethe textural shelf life of the packagedbread, it was assumed that the data re-ported in Figs. 2 and 3 can be describedby means of the following expression:

–EC = A1 + A2·t (4)

where A1 and A2 values are constantand are regarded as the fitting param-eters. It is worth noting that A1 is equal to

Fig. 1 - Compressive stress plotted as a function of strain for sample C1.


Fig. 2 - EC plotted as a function of storage time of samples packaged with film 1.

Fig. 3 - EC plotted as a function of storage time of samples packaged with film 2.


Fig. 4 - Mass fraction of water present in the crumb versus storage time of samples packaged with film 1.

Fig. 5 - Mass fraction of water present in the crumb versus storage time of samples packaged with film 2.


–ECFresh. Equation (4) can be rearranged

as follows:

2–EC = A1·(1+ ·t) (5)

tS.L.

where tS.L. is the textural shelf life ofthe packaged bread slice. Equation (5)was used to fit the data reported inFigs. 2 and 3; the fitting was done bytaking into account the error that char-acterized each data point. The resultsobtained are reported in Table 1; theconfidence intervals (95%) were evalu-ated on the basis of 500 converginginteractions. Unfortunately, due to thewidth of the confidence intervals it wasnot possible to take into considerationthe differences in the textural shelf lifeof the packaging combinations studied.The average value of the textural shelflife, which was evaluated by averagingthe values obtained for the six packag-ing combinations tested, is equal to17.9 days.

CONCLUSIONS

This investigation shows that it ispossible to prolong the shelf life of thedurum wheat bread investigated from3 to 18 days acting only on the pack-aging technology. Depending on thepackaging combination tested, unac-ceptability of the product was causedeither by the presence of visible signsof mold or by a change in the texturalproperties of the crumb. In particular,the oxygen scavenger was much moreeffective than the modified atmospherepackaging with regard to suppressionof molds, but an extension of the shelflife beyond 18 days was limited by thetextural deterioration of the bread. Onthe other hand, in the case of modifiedatmosphere packaging, shelf life waslimited by microbial growth in low car-bon dioxide concentration and the firm-

ing process in high carbon dioxide con-centration.

The results point out that using eithermodified atmosphere packaging or oxy-gen scavengers, it is possible to prolongthe shelf life to about 18 days. Furtherextension of the shelf life is limited bytextural changes. The choice betweenthese two packaging technologies can bemade only on an economical basis.

ACKNOWLEDGEMENTS

The authors wish to thank Dr. Bianchi (CryovacSealed Air Corporation) for supplying the polyole-fin/polyamide/polyolefin films, Dr. Celano (DiMauro Officine Grafiche) for supplying the poly-propylene/EVOH/polypropylene/PET films, andDr. Pecorino (Cooperativa Agricola Valle del Dit-taino) for supplying the durum wheat bread usedin this investigation. The work was funded underthe “POP Misura 10-4 della Regione Sicilia” grants.

REFERENCES

Avital Y., Mannheim C. H. and Miltz J. 1990. Ef-fect of carbon dioxide atmosphere on staling andwater relations in bread. J. Food Sci. 2:413.

Baker A.E., Diben R.A. and Ponte J.G. 1987.Comparison of bread firmness measurementsby four instruments. Cereal Foods World32:486.

Baker A.E., Doerry W. T. and Kemp K. 1986. In-stron factors involved in measuring crumb firm-ness. Cereal Foods World 31:193.

Baker A.E. and Ponte J.G. 1987. Measurementsof bread firmness with the universal testingmachine. Cereal Foods World 32:491.

Baker M. 1986. “The Wiley Encyclopedia of Pack-aging Technology”. p. 274, 478. John Wiley &Sons. New York.

Berenzon S. and Saguy I.S. 1998. Oxygen absorb-ers for extention of crackers shelf life. Lebens-mittel-Wissenschaft und Technologie 31:1.

Bourgeois C.M., Mescle J.F. and Zucca J. 1988.“Microbiologie Alimentaire 1 – Aspect Micro-biologique de la Sécurité ed de la Qualité Ali-mentaires”. p. 182-183Technique et Documen-tation. Lavoisier.

Brandrup J. and Immergurt E.H. 1975. “PolymerHandbook”. p. III-232 – III-238. John Wiley &Sons. New York.

Cencic L., Bressa F. and Dalla Rosa M. 1996. In-fluence of modified atmosphere on bread stal-ing during storage, Ital. Food & Bev. Tech. 7:20.


Giovanelli G. and Pagliarini E. 1996. Valutazionedella soglia di raffermimento della mollica dipane. Industrie Alimentari, 35:635.

Hernandez R.J. 1994. Effect of water vapor on thetransport properties of oxygen through polya-mides. J. Food Eng. 22:495.

Hilyard N.C. 1982. Stiffness and strength – Flexi-ble polymer foams. In “Mechanics of CellularPlastics”. (N. C. Hilyard Ed.). p. 83-86. AppliedScience Publisher LTD. London.

Knorr D. and Tomlins R.I. 1985. Effect of carbondioxide modified atmosphere on the compressi-bility of stored packed goods. J. Food Sci. 50:1172.

Mandel J. 1964. “The Statistical Analysis of Ex-

Paper received August 12, 2002 Accepted December 17, 2002

perimental Data”. p. 62-63. Dover Publications.New York.

Piazza L. and Masi P. 1995. Moisture redistribu-tion throughout the bread loaf during stalingand its effect on mechanical properties. CerealChem. 72:320.

Ponte J.C., Payne J.D. and Ingelin M.E. 1993. TheShelf Life of Bakery Foods. In “Shelf Life Stud-ies of Foods and Beverages”. (G.CharalambousEd.). p. 1144-1160. Elsevier Science Publish-ers B.V.. Amsterdam.

Smith J.P., Jackson E.D. and Ooraikul B. 1983.Storage study of a gas-packaged bakery prod-uct. J. Food Sci. 48:1370.


PAPER

- Key words: biogenic amines, cheese, proteolysis, SDS-PAGE, Toma Piemontese -

PROTEOLYSIS AND PRODUCTIONOF BIOGENIC AMINES

IN TOMA PIEMONTESE PDO CHEESEDURING RIPENING

PROTEOLISI E SVILUPPO DI AMMINE BIOGENE NEL CORSO DELLAMATURAZIONE DEL FORMAGGIO TOMA PIEMONTESE DOP

M. ARLORIO*, J.D. COïSSON, F. TRAVAGLIA, M. CAPASSO1,M. RINALDI and A. MARTELLI

Dipartimento di Scienze Chimiche, Alimentari, Farmaceutiche,Farmacologiche (DiSCAFF), Università del Piemonte Orientale “A. Avogadro”,

Via Bovio 6, 28100 Novara, Italy1 Dipartimento di Chimica Farmaceutica e Tossicologica,

Via D. Montesano 49, 80131 Napoli, Italy* corresponding author

ABSTRACT

Proteolysis of Toma Piemontese PDOcheese (large/long–ripened cheese andsmall/short-ripened cheese types) dur-ing ripening was studied by monitor-ing changes in nitrogenous compounds(protein content, SDS-PAGE, pH 4.6soluble-nitrogen and ripening index).Changes in the concentration of fourbiogenic amines and their precursoramino acids were also determined,showing the presence of histamine andtyramine (up to a total concentrationof 145 mg kg-1) in long-ripened cheese.

RIASSUNTO

La proteolisi del formaggio Toma Pie-montese DOP (tipologia “grande a lungamaturazione” e tipologia “piccola a bre-ve maturazione”) è stata studiata, du-rante la maturazione, analizzando la va-riazione della frazione azotata (contenutoin proteine totali, N solubile a pH 4.6,indice di maturazione) nonché della fra-zione proteica per SDS-PAGE. Sono statedeterminate anche alcune ammine bio-gene ed i loro aminoacidi precursori, evi-denziando la presenza di istamina e ti-ramina (valore massimo complessivo


The total concentration of biogenicamines in the commercial samples an-alyzed did not exceed 219 mg kg-1. Incomparison to short-ripened Toma,long-ripened Toma had a higher levelof proteolysis (ripening index up to20.8% after 64 days ripening) with thecharacteristic disappearance of the αs1-casein band in the electrophoresis pro-files. The increase in proteolysis dur-ing ripening was confirmed by PCA,showing the possibility of using thistechnique to assess the age of unknowncheese samples. The main variablescontributing to the first principal com-ponent were ripening index, % solubleN, phenylalanine and αs1-casein.

pari a 145 mg kg-1) nella tipologia a lun-ga maturazione. Il contenuto totale inammine biogene dei campioni commer-ciali analizzati non ha superato i 219mg kg-1. Lo stesso tipo di formaggio hainoltre presentato il maggior grado diproteolisi (IM 20.8% a 64 giorni di ma-turazione), con la caratteristica scom-parsa della banda relativa all’αs1-casei-na dai profili elettroforetici. L’uso dellaPCA ha confermato l’aumento della pro-teolisi durante la maturazione, indican-do la possibilità futura di datare cam-pioni di formaggio di età sconosciuta.Le principali variabili che incidono sul-la componente principale sono: indicedi maturazione, % di N solubile, fenila-lanina e αs1-caseina.

INTRODUCTION

Proteolysis in cheese is a fundamen-tal process which, together with glycoly-sis and lipolysis, contributes to the tex-ture, aroma and taste of ripened chees-es. Proteolysis may be regarded as themost important biochemical event in theripening of many cheese varieties (FOXet al., 1994).

Toma Piemontese is a typical PDO Ital-ian cheese (Italian law DPCM, 10/05/1993, CE regulation 1263/96), widelyconsumed in northern regions of Italy.The production of Toma Piemontese in2001 was about 1,233 t (official dataobtained from Associazione RegionaleProduttori Latte Piemonte).

Toma Piemontese is a semi-hardcheese (coagulation temperature: 32°-35°C), made from cows’ milk (raw or pas-teurized milk, acidity: 3.6°SH/50). Thereare two different types of this cheese: thefirst, called “Toma Piemontese PDO”(40% minimum fat as dry matter), is pro-duced from whole milk; the second,called “Toma Piemontese semigrasso”(20% minimum fat as dry matter), is pro-

duced from partially skimmed cows’milk. Each type of Toma cheese is cur-rently produced in two different sizes:small cheese (1.5-2 kg, 15 days mini-mum ripening time) and large cheese (6-8 kg, 60 days minimum ripening time).About 70% of the total production is thelarge-type, of which 30% is “Toma semi-grasso”. The long-ripened Toma is cur-rently sold and consumed within 30 daysafter the required 60 days of ripening.

Only a few scientific papers have re-ported data on the chemical composi-tion of Toma Piemontese cheese and areprevious to the PDO certification (DEL-FORNO, 1966; 1977; BATTAGLINI andFORTINA, 1992). However, since 1995 theAgricultural Department of the PiedmontRegion has coordinated interdisciplinaryresearch on this topic (SOSTER, 1998).The preliminary results showed the ex-istence of at least three different typesof Toma Piemontese cheese: “classic”,“cheese factory” and “Biellese” which dif-fer for some chemical parameters (drymatter, fat, protein, DL lactic acid). Theuse of sensory analysis (based on the useof 20 descriptor parameters) and clus-


ter analysis has partially confirmed thedifferences among these three Tomacheeses (AMBROSOLI et al., 1998).

As reported in many studies on dif-ferent cheeses (MAYER, 1996; DEWET-TINCK et al., 1997; MOONEY et al., 1998),the evolution of primary proteolysis canbe readily monitored using SDS-PAGEand/or urea-PAGE. The formation andaccumulation of biogenic amines incheese (as well as in other fermented/seasoned foods) is a well-studied sub-ject (BRINK et al., 1990; HALASZ et al.,1994; VALE and GLORIA, 1998). The for-mation of some biogenic amines (hista-mine, tyramine, tryptamine and 2-phe-nylethylamine) and their related precur-sor amino acids have been studied inother Italian cheeses, for example, Gor-gonzola PDO cheese (MARTELLI et al.,1993).

This paper describes proteolysis inToma Piemontese PDO during ripening.Variations in the total and soluble nitro-gen fractions, as well as some parame-ters related to the proximate composi-tion, were monitored. The final goal wasto study the changes in the nitrogenousfraction during ripening in order to ob-tain some ripening markers, useful forassessing the age of this cheese. Thesame parameters were also determinedon some commercial samples of un-known age, purchased directly from re-tail shops.


Cheese sampling

Samples of Toma Piemontese cheesewere manufactured from whole milk ona commercial scale in an industrial plant,according to the standard protocol re-ported in PDO regulations. These sam-ples were kindly supplied from “Consor-zio di Tutela del Toma Piemontese”. Inone trial consisting of “small” type cheese(3 different whole cheeses of about 1.5

kg) samples were taken at 4, 11 and 19days of ripening. “Large” type cheeses(from 5 different whole cheeses of about7 kg) were sampled at 4, 19, 29, 39, 49and 64 days of ripening. This trial wasdone with cheese produced in the win-ter and another trial with cheese pro-duced in spring.

A total representative sample of 100 gwas obtained by grating and homogenis-ing the pieces collected from each wholecheese; each final sample was then ana-lysed, avoiding any loss of moisture.

All commercial samples were pur-chased at retail shops and supermar-kets located throughout the PiedmontRegion.

Compositional analysis

Cheese moisture, ash, fat and totalacidity were determined according to theItalian Official Methods for cheese anal-ysis (Official Italian Journal, 1986).

Total and pH 4.6-soluble nitrogenvalues were determined by the Kjeldahlmethod (AOAC, 1990) using the Kjeltecsystem (Tecator, Sweden); protein wascalculated using the conversion factor,6.38.

Total fat was determined using theSoxhlet apparatus with dichloromethanefor 14 hours.

Biogenic amines (tyramine, histamine,tryptamine and 2-phenylethylamine) andtheir precursor amino acids were deter-mined using an ion-pair HPLC methodpreviously optimised for Gorgonzola PDOcheese (ARLORIO et al., 1998). Eachsample (10 g) was extracted with 50 mLof trichloroacetic acid (5%, w/v). Allchemical reagents (reagent grade) werepurchased from Fluka Chemie AG(Buchs, Switzerland).

SDS-PAGE electrophoresis

SDS-PAGE was performed accordingto LAEMMLI (1970) using a Mini ProteanII Dual Slab Cell (Bio-Rad Laboratories


S.r.l., Segrate, Italy) on a urea-solublefraction of cheese (obtained using a 6 Murea solution, pH 8.5). Fluor-S MultiIm-ager and Quantity One Software (Bio-Rad) were used to analyse all gels. Datacollected are expressed as relative per-cent for each lane and were used for PCAstatistical analysis.

Statistical methods

All results are reported as the meanof three (or more) independent determi-nations ± standard deviation. One-wayAnova test was conducted on long-rip-ened cheese, using Microsoft Excel Soft-ware. Principal Component Analysis(PCA), performed employing Mathemat-ica Software, was used to analyse all datacollected from the study.


The values for the principal parame-ters (fat, protein, moisture) of well-rip-ened cheeses (19 days and 64 days, re-spectively, for two different types ofToma) were in accord with the minimumvalues prescribed by the PDO rules (CERules 1263, 1/7/96) (Tables 1, 2). Aslight increase in the total nitrogen con-tent was evident for both types of TomaPiemontese (Tables 1, 2); this is normalfor cheese due the loss of moisture dur-ing ripening (MADKOR et al., 1987; HAAS-NOOT et al., 1989; FOX et al., 1996; GOB-BETTI et al., 1997). The pH 4.6 solublenitrogen fraction, as well as the ripen-ing index (R.I.; ratio of soluble to totalnitrogen), showed a constant increase.The R.I. of the short-ripened Toma in-creased from 8.1 to 14.4% in 19 days;the R.I. of long-ripened Toma increasedfrom 9.8 to 20.8% in 64 days. Althoughnot comparable, both increases in R.I.are important parameters for PCA.

Differences in the degree of ripeningobserved in the winter and spring long-ripened cheeses may be related to the

influence of environmental parameters(quality of the original milk, microbialquality and count, climatic conditions)on the cheesemaking processes. Thesewinter and spring-produced cheeses didnot exhibit significant differences usingAnova statistical tests, although thespring series showed a higher level ofsome proteolysis parameters. The datareported in Table 2 are the average ofspring and winter produced long-ripenedcheeses.

The total acidity of the two types ofToma cheese had an opposite trend (Ta-bles 1, 2). It decreased during the ripen-ing of short-ripened Toma which couldbe partially related to the increase inproteolysis and the release of basic com-pounds. In contrast, it increased duringripening in long-ripened Toma. The de-gree of proteolysis at 19 days was simi-lar for both cheese types; therefore, thereason why acidity increased in largeToma may be the higher moisture con-tent, with higher whey and lactose re-tention in the cheese. These conditionscould promote greater lactic acid produc-tion.

Tyrosine, histidine, phenylalanine andtryptophan showed no marked changesin the short-ripened Toma (data notshown). Long-ripened Toma cheese hada significant increase in these aminoacids during ripening (Table 3). This maybe a useful marker, as reported for oth-er cheeses (BÜTIKOFER and FUCHS,1997; INNOCENTE, 1997). Phenylalanine,for example, had a constant and signifi-cant increase during ripening, which hasalso been reported in other Italian chees-es analysed using the same approach(Gorgonzola PDO, Raschera PDO, BraPDO, etc., unpublished data) or in someSpanish cheeses (GARDE et al., 2002).Apparently, decarboxylation of phenyla-lanine did not occur, in contrast to tyro-sine and histidine (CHRISTENSEN et al.,1999). JOOSTEN (1987) did not identifyphenylalanine-decarboxylating bacterialstrains; the formation of 2-phenylethyl-


Table 1 - Some parameters (%) of Toma Piemontese PDO (small, whole short-ripened cheese type)during ripening (mean±S.D.).

Days Moisture Total fat Ash Protein Soluble nitrogen R.I.a Total acidityb

4 43.6±0.4 28.9±0.2 2.77±0.04 20.3±0.3 0.26±0.02 8.1±0.39 3.62±0.0111 43.1±0.2 29.2±0.1 3.13±0.08 21.1±0.1 0.31±0.02 9.3±0.45 1.73±0.1919 42.2±0.1 29.5±0.2 3.68±0.23 21.5±0.3 0.49±0.03 14.4±0.82 1.06±0.28Significance ** * *** ** *** *** ***

a R.I.: Ripening index (ratio between soluble and total nitrogen x 100). b expressed in meq/100g of cheese. Signifi-cance of time of ripening: ANOVA test, *** p<0.001, ** p<0.01, * p<0.05. n=3 samples for each day.

Table 2 - Some parameters (%) of Toma Piemontese PDO (large, whole long-ripening cheese type) duringripening (mean±S.D.).

Days Moisture Total fat Ash Protein Soluble nitrogen R.I. a Total acidityb

4 46.0±0.3 28.2±0.5 2.02±0.43 20.9±0.7 0.32±0.07 9.8±2.0 1.94±0.1519 45.0±0.4 28.2±0.7 2.95±0.03 21.1±0.6 0.52±0.04 15.6±1.1 2.39±0.1029 44.7±0.5 28.6±0.2 3.14±0.18 21.0±0.3 0.62±0.13 18.7±3.8 2.45±0.0739 44.1±0.4 29.3±0.2 3.40±0.05 21.3±0.3 0.69±0.16 20.5±4.7 2.53±0.1249 43.8±0.6 29.4±0.3 3.58±0.17 21.5±0.4 0.67±0.13 19.8±3.8 2.68±0.1364 43.0±0.9 29.4±0.3 3.71±0.21 21.9±0.2 0.71±0.05 20.8±1.5 2.74±0.31Significance *** *** *** ** *** *** ***

a R.I.: Ripening index (ratio between soluble and total nitrogen x 100). b expressed in meq/100g of cheese. Signifi-cance of time of ripening: ANOVA test, *** p<0.001, ** p<0.01, * p<0.05. Data are the average of winter and springproduced cheeses. n=6 samples for each day.

Table 3 - The concentration (mg kg-1) of biogenic amines and related precursor amino acids in TomaPiemontese PDO (large whole, long-ripening cheese type) during ripening (mean±S.D).

Days Tyrosine Histidine Phenylalanine Tryptophan Tyramine Histaminea

4 85±39 24± 1.4 98±40 8.8± 3.0 Tr Nd19 170±23 56±29 295±27 18 ± 1.6 10± 1 12±129 141±23 98±49 333±35 16 ± 3.7 26± 4 14±339 180±25 108±31 439±42 23 ± 5.5 34± 8.8 15±149 188±25 93±56 504±60 28.7± 9.0 60±14 22±964 254±23 253±94 709±36 57.3±21 108±46 37±9Significance *** *** *** *** *** **

a Present only in spring trial (n=3). Significance of time of ripening: ANOVA test, *** p<0.001, ** p<0.01, * p<0.05. Tr:traces, Nd: not detected. Data are the average of winter and spring produced cheeses. n=6 samples for each day.

amine could be related to the action ofother enzymes, for example tyrosine-de-carboxylase, that has shown a slightactivity towards phenylalanine, structur-ally-related to tyrosine.

Biogenic amines were not found inshort-ripened Toma samples, presuma-bly because of the short time of ripening(19 days). In long-ripened Toma cheese,tyramine ranged from 10 to 108 mg kg-1;


the presence of histamine was detectedonly in the long-ripened spring Toma,ranging from 12 to 37 mg kg-1. The in-crease in the concentration of eachamine during ripening was significant(respectively p<0.001 for tyramine andp<0.01 for histamine). 2-Phenylethyl-amine and tryptamine were absent in allsamples.

General hygienic conditions during in-dustrial/artisanal cheesemaking and de-carboxylation by some starter and non-starter bacterial strains are two funda-mental parameters directly related to theaccumulation of biogenic amines (VOIGTand EITENMILLER, 1977; JOOSTEN andSTADHOUDERS, 1987; HALASZ et al.,1994). For this reason, tyramine is a po-tential marker for the control of hygienicconditions during cheese production. Theformation of biogenic amines in cheese isenhanced by temperature (mainly if>18°C), pH (>5) and low salt content asdemonstrated by ANTILA et al. (1984) andJOOSTEN (1987). The ripening and stor-age temperatures are parameters whichinfluence the formation of biogenicamines in cheese (MARTELLI et al., 1993).

The maximum total concentration ofbiogenic amines in the samples of long-ripened Toma Piemontese at the end ofthe ripening period (64 days, about 145mg kg-1 as the sum of histamine andtyramine) does not present any toxico-logical risk. A total content of 500 mgkg-1 of biogenic amines is considered acritical concentration able to cause the“cheese reaction”. About 200 mg kg-1 ofbiogenic amines represents a risk for thehealth of the consumer only in the caseof simultaneous consumption of MonoAmino Oxidase-inhibitor (i-MAO) drugs(enzymes involved in the catabolism ofamines) as well as in the case of contin-uous excessive consumption of cheeseby sensitive patients (SMITH, 1980; BRINKet al., 1990).

The SDS-PAGE profile of the total pro-teins obtained for short-ripened Tomashowed a slight decrease of the αs1-ca-

sein band (about 30 kDa) (Fig. 1, lane3). The SDS-PAGE pattern of the pH 4.6-soluble nitrogen fraction did not providea useful fingerprint of proteolysis (datanot shown). The degree of proteolysis inshort-ripened Toma Piemontese was lim-ited, in agreement with the other chem-ical indexes (Table 1). The SDS-PAGEdata for long-ripened Toma Piemontesecheese showed a sharp decrease of theα-casein band (about 30 kDa); the β-ca-sein band (molecular weight 28 kDa) wasobserved in all samples (Fig. 1b).

Even in long-ripened Toma, the pH4.6-soluble nitrogen fraction analyzed bySDS-PAGE did not give any significantresults (data not shown). An analysis ofthis soluble fraction by RP-HPLC is sure-ly preferable, as previously suggested(BICAN and SPAHNI, 1993; SINGH et al.,1993; 1995).

Some commercial samples of Toma(unknown age, large type) were then an-alysed for free amino acids, biogenicamines, total nitrogen and pH 4.6-solu-ble nitrogen, R.I., SDS-PAGE. Theyshowed a different degree of proteolysis(Table 4); the R.I. ranged from 19.5 to35.1%, indicating a large variability ofproteolysis. The data for most sampleswere quite close, as also confirmed byPCA analysis (Fig. 2). Most commercialsamples had R.I. values higher than theR.I. of the 64 day-ripened samples (Ta-ble 2). On the contrary, free amino acidvalues of the commercial samplesshowed more variability (Table 5).

Commercial samples showed the pres-ence of tyramine (from 31 to 178 mg kg-1)and histamine (from traces to 67 mg kg-1).2-Phenylethylamine and tryptaminewere absent in all commercial samples,as well as in the control samples ana-lyzed.

The total concentration of biogenicamines in the commercial samples didnot exceed 219 mg kg-1 (Table 5); thisvalue, however, was higher than the val-ues obtained for the 64 day-ripened con-trol samples (Table 3).


Table 4 - Values (%) of chemical parameters related to the proteolysis in some commercial samples ofToma Piemontese PDO cheese. R.I.: ripening index.

Samples Protein Soluble (pH 4.6) nitrogen R. I.

1 21.8±0.2 0.69±0.04 20.22 22.2±0.3 1.22±0.02 35.13 25.2±1.5 0.94±0.06 23.84 24.3±0.3 0.99±0.06 26.15 23.1±0.6 0.96±0.01 26.56 24.4±0.6 1.12±0.03 29.27 23.8±0.6 0.73±0.02 19.58 24.4±0.6 1.19±0.04 31.29 28.0±0.5 1.08±0.02 24.6

Fig. 1a - SDS-PAGE of Toma Piemontese PDO cheese (small cheese, short-ripened type) during ripen-ing. Lane 1: 4 days; lane 2: 11 days; lane 3: 19 days. 1b - SDS-PAGE of Toma Piemontese PDO cheese(large cheese, long-ripened type) during ripening. Lane 1: 4 days; lane 2: 19 days; lane 3: 29 days; lane4: 39 days; lane 5: 49 days; lane 6: 64 days.

SDS-PAGE showed a different degreeof proteolysis of the commercial samples,in agreement with the R.I. data (Fig. 3,Table 4). Different R.I. values could berelated mainly to the residual coagulantused in cheesemaking, but also to theaction of the bacterial flora (starterstrains and typical non-lactic bacterialflora). Storage conditions of the cheese(refrigeration temperature, size of the

cheese portions) and maintenance of re-frigeration in retail shops or supermar-kets are other significant parametersthat influence the proteolysis of commer-cial samples.

All data obtained for the referencecheese samples and the commercialsamples were submitted to PCA analy-sis. The first principal component(score: 52.6%), for both types of Toma

1a 1b


(short-ripened and long-ripened refer-ence series) increased with the age ofthe sample and can thus be used as aripening index. The main variables con-tributing to the first principal compo-nent were ripening index (correlationcoefficient: 0.3334), % soluble N (corre-lation coefficient: 0.3306), phenyla-lanine (correlation coefficient: 0.3231)and αs1-casein (correlation coefficient:0.3211).

The commercial samples were locat-ed in lines created using the three ref-erence samples (Fig. 2). Firstly, the two

Fig. 2 - Linear representation of the first principal component (PCA analysis using all parameters, seethe text for explanation) showing ripening trends of different types of Toma. ■ : small type, short-ripened reference Toma; ▲: large type, long-ripened Toma winter-produced; ◆ : large type, long-ripenedToma spring-produced; ● : commercial samples. Numbers above all symbols represent the age of thecheese; numbers below symbols (1-9) represent numbers of commercial samples (Tables 4-5).

Table 5 - Concentration (mg kg-1) of biogenic amines and related precursor amino acids in some com-mercial samples of Toma Piemontese PDO cheese. Tr: traces; Nd: not detected.

Commercial sample n.: Tyrosine Histidine Phenylalanine Tryptophan Tyramine Histamine

1 81±0.8 83±8 278±12 Tr 39±1 Nd2 196±2 26±2 823±25 87±4 178±7 Nd3 Tr 97±34 312±119 19± 5 95±28 40±124 107±4 503±12 459±77 54±8 58±9 67±35 50±2 241±18 471± 93 25±1 79±6 29± 16 358±18 677±10 965±14 57± 20 Nd Nd7 19±8 Tr 508± 21 12±4 146±29 Nd8 90±7 Tr 673±16 75±6 165±1 54±29 341±50 355±50 708±35 37±3 31±2 27±6

series of long-ripened Toma (winter andspring) showed different ripeningtrends, confirming different proteolysisfor spring and winter Toma. Consider-ing the three reference series, the com-mercial samples were correctly clus-tered as “long-ripened” Toma. Usingthis approach, it is obvious that somecommercial samples were aged for lessthan the minimum ripening time re-quired by the CE regulation for long-ripened Toma (60 days). Apparently, theuse of PCA could be usefully employedto assess the age of unknown samples


of Toma. It is suggested that more ref-erence samples be analyzed to stand-ardize and validate a PCA-based meth-od for assessing cheese age. This tech-nique could be a useful tool for “Con-sorzio di Tutela” to monitor the correctripening-time before sale. In conclu-sion, the assessment of proteolysis(particularly SDS-PAGE, free aminoacids and biogenic amines, R.I.) con-firmed different ripening trends for thetwo types of Toma (large long-ripenedand small short-ripened cheeses) aswell as for the season of cheesemak-ing. In conclusion, PCA can be used toclassify and assess the age of unknowncommercial samples.

ACKNOWLEDGEMENTS

This work was financed by Regione Piemonte –Assessorato Agricoltura (Project: “Caratterizza-zione chimico-nutrizionale di alcuni formaggiDOP Piemontesi“) and Università del PiemonteOrientale “A. Avogadro”(ex-60% FAR funds).

The authors also thank Dr. R. Arru (AssociazioneRegionale Produttori Latte Piemonte - Asso DOC

- Consorzio di Tutela Toma Piemontese D.O.P.)and Dr. Gianni Comba (INOQ) for their contri-bution and for supplying the reference series ofcheeses during ripening.

REFERENCES

Ambrosoli R., Gerbi V., Zeppa G., Terrone S. andTallone G. 1998. Aspetti tecnologici, microbio-logici e sensoriali. In “Toma Piemontese”, M.Soster (Ed.), p. 53. Regione Piemonte Assesso-rato Agricoltura, Torino, Italy.

Antila P., Antila F., Mattila J. and Hakkarainen H.1984. Biogenic amines in cheese. II. Factorsinfluencing the formation of biogenic amines,with particular reference to the quality of themilk used in cheese making. Milchwissenschaft39: 400.

AOAC 1990. In “Official Methods of Analysis (15th

ed.)”, K. Helrich (Ed.). Association of OfficialAnalytical Chemists Publ. Arlington, VA, USA.

Arlorio M., Coïsson J.D. and Martelli A. 1998. Ion-pair HPLC determination of biogenic amines andprecursor amino acids. Application of a meth-od based on simultaneous use of heptanesul-phonate and octylamine to some foods. Chro-matographia 48: 763.

Battaglini L.M. and Fortina R. 1992. La Tomagrassa e il Murianengo: due formaggi della Valdi Susa. Sci. Tecn. Latt.-cas. 43: 44.

Bican P. and Spahni A. 1993. Proteolysis in Swiss-

Fig. 3 - SDS-PAGE profiles of some commercial samples of Toma Piemontese PDO cheese. Lane 1: 4days long-ripened type; lane 2: 64 days long-ripened type; lane 3-11 commercial samples n. 1-9; lane12 molecular weight marker.


type cheeses: a High-Performance Liquid Chro-matography study. Int. Dairy J. 3: 73.

Brink B., Damink C., Joosten H.M.L.J. and Huisin ‘t Veld J.H.J. 1990. Occurrence and forma-tion of biologically active amines in foods. Int.J. Food Microbiol. 11: 73.

Bütikofer U. and Fuchs D. 1997. Development offree amino acids in Appenzeller, Emmentaler,Gruyère, Raclette, Sbrinz and Tilsiter cheese.Lait 77: 91.

Christensen J.E., Dudley E.G., Pederson J.A. andSteele J.L. 1999. Peptidases and amino acidcatabolism in lactic acid bacteria. Antonie vanLeeuwenhoek 76: 217.

Delforno G. 1966. Studio sulla composizione chimi-ca del formaggio Toma. L’Industria del Latte 1: 1.

Delforno G. 1977. Indagine sulla composizionechimica del formaggio Toma. Il Latte 5: 332.

Dewettinck K., Dierckx S., Eichwalder P. and Huy-ghebaert A. 1997. Comparison of SDS-PAGEprofiles of four Belgian cheeses by multivariatestatistics. Lait 77: 77.

Fox P.F., Singh T.K. and McSweeney P.L.H. 1994.Proteolysis in cheese during ripening. In “Bio-chemistry of Milk Products”, A.T. Andrews andJ. Varley (Eds.), p. 1. Royal Society of Chemis-try, Cambridge, U. K.

Fox P.F., O’Connor T.P., McSweeney P.L.H., GuineeT.P. and O’Brien N.M. 1996. Cheese: physical,biochemical and nutritional aspects. Adv. FoodNutr. Res. 39: 163.

Garde S., Tomillo J., Gaya P., Medina M. and NuñezM. 2002. Proteolysis in Hispanico cheese man-ufactured using a mesophilic starter, a ther-mophilic starter, and bacteriocin-producing Lac-tococcus lactis subsp. lactis INIA 415 adjunctculture. J. Agric. Food Chem. 50: 3479.

Gobbetti M., Lowney S., Smacchi E., Battistoni B.,Damiani P. and Fox P.F. 1997. Microbiology andbiochemistry of Taleggio cheese during ripen-ing. Int. Dairy J. 7: 509.

Haasnoot W., Stouten P. and Venema D.P. 1989.High-Performance Liquid Chromatography de-termination of the extent of proteolysis in Gou-da cheese. J. Chromatogr. 483: 319.

Halasz A., Barath A., Simon-Sarkadi L. and Hol-zapfel W. 1994. Biogenic amines and their pro-duction by microorganisms in food. Trends FoodSci. Technol. 5: 42.

Innocente N. 1997. Free amino acids and water-

soluble nitrogen as ripening indices in Monta-sio cheese. Lait 77: 359.

Joosten H.M.L.J. 1987. Conditions allowing theformation of biogenic amines in cheese. Part 3:Factors influencing the amounts formed. Neth.Milk Dairy J. 41: 329.

Joosten H.M.L.J. and Stadhouders J. 1987. Con-ditions allowing the formation of biogenicamines in cheese. 1-Decarboxylative propertiesof starters bacteria. Neth. Milk Diary J. 41: 247.

Laemmli U.K. 1970. Cleavage of the structural pro-teins during the assembly of the head of bacte-riophage T4. Nature 227: 680.

Martelli A., Arlorio M. and Tourn M.L. 1993. De-termination of amines and precursor aminoacids in Gorgonzola cheese by ion-pair HPLCwithout derivatization. Riv. Sci. Alim. 22: 261.

Madkor S., Fox P.F., Shalabi S.I. and Metwalli N.H.1987. Studies on the ripening of Stilton cheese:proteolysis. Food Chem. 25: 13.

Mayer H.K. 1996. Electrophoretic ripening indexfor the evaluation of proteolysis and the deduc-tion of the age of Parmesan cheese. Z. Lebens.Unters. Forsch. A 202: 465.

Mooney J.S., Fox P.F., Healy A. and Leaver J. 1998.Identification of the principal water-insolublepeptides in Cheddar cheese. Int. Dairy J. 8: 813.

Official Italian Journal, n. 229. 1986. Metodi Uffi-ciali di analisi dei formaggi. D.M. n. 88 del21.4.1986.

Singh T.K., Fox P.F., Højrup P. and Healy A. 1993.A scheme for the fractionation of cheese nitro-gen and identification of principal peptides. Int.Dairy J. 4: 111.

Singh T.K., Fox P.F. and Healy A. 1995. Water-soluble peptides in Cheddar cheese: isolationand identification of peptides in the diafiltra-tion retentate of the water-soluble fraction. J.Dairy Res. 62: 629.

Smith T.A. 1980. Amines in food. Food Chem. 6:169.

Soster M. 1998. Il progetto di caratterizzazione dellaToma Piemontese. In “Toma Piemontese”, M.Soster (Ed.), p. 7. Regione Piemonte Assessora-to Agricoltura, Torino, Italy.

Vale S. and Gloria M.B.A. 1998. Biogenic aminesin Brazilian cheeses. Food Chem. 63: 343.

Voigt M.N. and Eitenmiller R.R. 1977. Productionof tyrosine and histidine decarboxylase by dairy-related bacteria. J. Food Protect. 40: 241.

Revised paper received December 23, 2002 Accepted February 27, 2003


PAPER

- Key words: disinfection, leafy vegetable quality, sensory attributes, storage -

DIP TREATMENTSFOR FRESH ROMAINE LETTUCE

TRATTAMENTI PER IMMERSIONEDI LATTUGA FRESCA ROMANA

S.I. ROURA1, M. DEL R. MOREIRA2, A. PONCE3 and C.E. DEL VALLE2

Grupo de Investigación en Ingeniería en Alimentos.Facultad de Ingeniería de la Universidad Nacional de Mar del Plata (UNMDP),

Juan B. Justo 4302, 7600 Mar del Plata, Argentina,Tel. +54 223 481 6600, Fax +54 223 481 0046

1 CONICET (Consejo Nacional de Investigaciones Científicas y Técnicas)2 Facultad de Ingeniería de la Universidad Nacional de Mar del Plata

3 CIC-BA (Comisión de Investigaciones Científicas de la Provincia de Buenos Aires)

ABSTRACT

The effects of dip treatments withchlorine, citric acid, ascorbic acid andCaCl2 on the storage life of Romaine let-tuce leaves were investigated. While thepresence of chlorine delayed degrada-tion related to the activity of the nativeand exogenous microflora, the othercompounds did not improve the stor-age life of the product when used alone.A combination of citric and ascorbicacid had a beneficial effect on the over-all visual quality (color, texture, bright-ness). The ratio of ascorbic acid con-

RIASSUNTO

È stato studiato l’effetto del tratta-mento con cloro, acido citrico, acidoascorbico e CaCl2 sulla vita di scaffaledella lattuga Romana. Mentre la pre-senza di cloro è fondamentale per ri-tardare la degradazione collegata all’at-tività della flora nativa ed esogena, glialtri additivi, quando applicati da soli,non hanno migliorato la vita di scaffaledel prodotto. La combinazione di acidocitrico e acido ascorbico ha avuto uneffetto vantaggioso sulle caratteristicheorganolettiche (colore, tessitura e lucen-


tents in samples treated and not treat-ed with ascorbic acid as additive wasabout 4.2:1 immediately after the diptreatment but dropped to about 2.1:1after 10 days of storage.

tezza). Il rapporto tra il contenuto inacido ascorbico nei campioni trattati enon trattati con questi composti, è sta-to 4.2:1 immediatamente dopo il trat-tamento, ma è scesa a 2.1:1 dopo 10giorni di conservazione.

INTRODUCTION

Lettuce is a very perishable product,extremely sensitive to abuse such astemperature or mishandling. One of themost common processing-induced dis-orders is discoloration. Techniques thatreduce surface and edge browning in-clude low storage temperatures (BOLINand HUXSOLL, 1991), modified or con-trolled atmospheres with low oxygen (2%)and/or relatively high carbon dioxidelevels (7%) (KE and SALTVEIT, 1989a;MATEOS et al., 1993a,b; HEIMDAL et al.1995) and chemical additives (McEVILYet al., 1992).

Several chemical compounds are re-ported to have inhibitory effects on en-zymatic browning of various commodi-ties. SAPERS et al. (1990) reduced dark-ening of “Russet Burbank” and “Katah-din” potatoes by using a dip treatmentwith ascorbic acid, CaCl2 and citric acid.CaCl2 dips have been reported to delayripening of pears and bananas (WILLSand TIRMAZI, 1982, POOVAIAH, 1986), re-tard flesh softening (BANGERTH et al.,1972) and retain the vitamin C contentof apple fruit (DRAKE and SPAYD, 1983).Dip treatments with ascorbic acid andcitric acid alone have been shown to re-tard enzymatic browning in freshly pre-pared vegetables (BAUERNFEIND andPINKERT, 1970; ANONYMOUS, 1983).However, PONTING et al. (1972) foundthat ascorbic acid or CaCl2 alone did notprevent discoloration of refrigerated ap-ple slices, but when used in a combina-tion treatment they were very effective.Calcium used alone or combined with

other substances such as ascorbic acidor sulfur dioxide has been shown tomaintain firmness of pear and strawber-ry slices (ROSEN and KADER, 1989), andcarrot sticks (BRUEMMER, 1987).

Several authors have studied the qual-ity of head and shredded iceberg lettuceas affected by processing conditions. Al-though iceberg is the predominant let-tuce used for prepared salad, other typesof lettuce are now used in salad mixes.Information on the postharvest behav-ior of these other types of lettuce is lim-ited, especially with regard to their phys-iology and quality as a minimally proc-essed product (LOPEZ-GALVEZ et al.,1996). Lettuce quality is reduced and itsshelf life shortened by tissue browning.Some of the most common post-harvestbrowning disorders of iceberg lettuce tis-sue are russet spotting (RS), brown stain(BS) (SALTVEIT, 1997), leaf surfacebrowning (LSB), leaf edge browning (LEB)(LÓPEZ-GÁLVEZ et al., 1996) and stembrowning (SB).

BOLIN et al. (1977) stated that refrig-eration temperature is used to extend theshelf life of shredded lettuce. Lettucestored at 5° or 10°C had a statisticallyshorter shelf life (10 days) than samplesheld at 2°C (26 days) (BOLIN et al., 1977).

BOLIN and HUXSOLL (1991) found thatdipping solutions of CaCl2, citric acid orZnCl2 were ineffective in retaining thequality of salad-cut lettuce, but dippingin 0.5% ascorbic acid increased the shelflife by about 10%. KRAHN (1977) report-ed that sodium bisulfite and CaCl2 wereof little or no value in extending the shelflife of cut head lettuce. IZUMI and WATA-


DA (1995) reported that the addition ofchorine to a CaCl2 dip enhanced the de-sirable effect of calcium on zucchinisquash.

Taking into account that fresh horti-cultural crops differ in morphologicalstructure, in composition and in gener-al physiology, it is clear that require-ments and recommendations for maxi-mum post-harvest shelf life will differ.KE and SALTVEIT (1989b) and COUTUREet al. (1993) reported differences in thebrowning activities of various cultivarsof lettuce when shredded and kept indifferent storage conditions.

The effectiveness of chemical dip treat-ments to increase post-harvest storagelife is affected by pH, temperature, wa-ter activity, initial microbial load andstorage atmosphere. Moreover, when dif-ferent additives are used together theymay cause either synergistic, additive orantagonistic effects (WILEY, 1997).

The purpose of the present work wasto determine the response of whole Ro-maine lettuce leaves to various chemi-cal treatments in order to extend its shelflife.


Sample preparation

Heads of Romaine lettuce (Lactucasativa, type Cos, variety Logifolia) wereharvested at optimal maturity when theyhad reached a marketable size (approx-imately 20-24 leaves per head). Theywere immediately transported to the lab-oratory and analyzed by a sensory pan-el. Outer leaves were discarded and onlyphotosynthetic leaves (green leaves) wereincluded in the samples. Lettuce leaveswere dipped in the various baths for 4minutes at room temperature (ca. 20°C)at a weight ratio 1:10. The dipping bathswere: 1) unchlorinated water (tap waterwith the following cation compositions:calcium 20 mg/L, magnesium 10 mg/L,

sodium 250 mg/L, potassium 15 mg/L,total iron 0.05 mg/L, manganese 0.05mg/L, and a water pH of 6.8-7.0); 2) chlo-rinated water (Control). The source ofchlorine was liquid sodium hypochlorite(NaOCl) and to achieve the best balanceof activity and stability of the hypochlo-rous acid, the pH of the water was main-tained between 6.5 and 7.5; 3) chlorineplus ascorbic acid (Chl-AA); 4) chlorineplus citric acid (Chl-CI); 5) chlorine plusCaCl2 (Chl-CA); 6) chlorine plus ascor-bic acid plus citric acid (Chl-AA-CI); 7)chlorine plus ascorbic acid plus CaCl2(Chl-AA-CA); 8) chlorine plus citric acidplus CaCl2 (Chl-CI-CA); 9) chlorine plusascorbic acid plus citric acid plus CaCl2(Chl-AA-CI-CA). The chlorine concentra-tion was 0.1% sodium hypochlorite, theascorbic acid concentration was 0.5%,the citric acid concentration was 0.1%and the CaCl2 concentration was 1% inthe baths that included each of thesecompounds. Lettuce leaves were re-moved from the baths without rinsingand were centrifuged for 30 s at 200 rpmin a salad spinner to eliminate excesssolution adhering to the tissue surface.Leaves were piled up in 100 g stacks,placed in open plastic containers andcovered with a 15 m polyvinyl chloridefilm, with an O2 permeability of 620-2,248 cm3.m-2.day-1 and a CO2 permea-bility of 4,263-8,138 cm3.m-2.day-1 (1atm-22°-25°C). Each container of 100 gstacks corresponded to a duplicate ofindependent lots. These containers wereplaced in holding boxes at a relative hu-midity (RH) of 97-99% and a tempera-ture of 4±1°C for 10 days. Beakers withsaturated K2SO4 solutions were placedin the boxes to maintain RH ca. 98%.

Microbiological studies

Lettuce leaves (25 g) were maceratedin PO4H2K buffer solution (1.68x10-4 g/L adjusted to pH = 7.2 with 1 mol/LNaOH), with a homogenizer (Stomacher400 Circulator Homogenizer, London,


UK). The enumeration and differentia-tion of microorganisms were performedusing the following culture media andculture conditions: mesophilic aerobicbacteria on PCA (Plate Count Agar) in-cubated at 35°C for 48 hours (ICMSF,1983; MOSSEL and MORENO GARCÍA,1985); psychrotrophic bacteria on PCAincubated at 5°C for 3-4 days (ICMSF,1983); molds and yeasts in YGC (yeast-glucose-chloramphenicol) medium andincubated at 25°C for 5 days (ICMSF,1983). Microbial counts were performedin duplicate on three independent lots,therefore count values are the mean ofsix determinations.

Determinationof reduced ascorbic acid

Ascorbic acid content was determinedby the titrimetric assay described byPELLETIER (1985). Ground lettuce (20g) was torn manually by hand and add-ed to 100 mL 6% metaphosphoric acidand homogenized for 3 min, it was madeup to 250 mL with 3% metaphosphoricacid and filtered through Whatman # 42filter paper. Aliquots (5 mL each) of thefiltrate were titrated with 2,6-dichloroin-dophenol. Ascorbic acid concentrations(mg/100 g) are reported on a wet basis.All assays were performed in duplicateon three independent lots, therefore eachvalue represents the mean of six deter-minations.

Calcium content

The calcium concentration of samplestreated with CaCl2 was calculated usingthe method described by ROSEN andKADER (1989). Samples were filtered us-ing Whatman N. 12 filter paper and an-alyzed on a Varian AA-375 Series Atom-ic Absortion Spectrophotometer (VarianTechtron Pty. Ltd, Springuale, Austral-ia) and recorded on a dry-weight basis.All assays were performed in duplicateon three independent lots and each val-

ue represents the mean of six determi-nations.

Sensory evaluation

Six judges, aged 30-45 years (4 fe-males and 2 males, all members of theUNMDP Food Engineering ResearchGroup), with sensory evaluation experi-ence with leafy vegetables, were trainedin the quality evaluation of Romaine let-tuce. At each sampling time (0, 2, 4, 6and 10 days of storage), lettuce leaveswere removed from the plastic contain-ers 20 min prior to evaluation to reachroom temperature. Lettuce leaves fromthree independent lots were subjected tosensory evaluation in duplicate. The cod-ed (3 digit) samples were presented oneat a time in random order to the judgeswho sat at around a table and made in-dependent evaluations. Samples fromeach treatment were evaluated for per-centage and severity of injury and sub-sequent decay. Evaluated indexes wereoverall visual quality (OVQ: color, tex-ture and brightness) (KADER et al., 1973),leaf surface browning (LSB) (LÓPEZ-GÁLVEZ et al., 1996), stem browning (SB)and russet spotting (RS). OVQ was rat-ed on a scale from 5 to 1, where 5 = ex-cellent and 1 = unusable. An OVQ rat-ing of 3 was considered the limit of sale-ability. LSB, SB and RS were scored ona scale from 1 to 5, where 1 = lack ofdefects and 5 = severe defects. A scoreof 3 was considered the limit for salea-bility. For each of these defects, an in-dex was calculated by multiplying thescores by the percentage of pieces affect-ed.


ANOVA was used to establish the lev-els of significance of the differencesamong the mean values from differentassays. Regression analysis was used todraw least square lines representing thedegradation of sensory attributes over


time. Differences among slopes were test-ed as indicated by VOLK (1958).


Sensory attributes

Scores for leaf superficial browning,stem browning and overall visual quali-ty of Romaine lettuce samples dipped inchlorinated water over a 10-day storageperiod at 4ºC and 98% RH are present-ed in Fig. 1. The regression equationsfor the different quality parameters of thesamples from all the dip treatments arereported in Table 1. Lower slopes repre-sent lower rates of degradation. In com-paring samples dipped in unchlorinatedand chlorinated water, there were high-er rates of degradation for OVQ indexes(p<0.01) for the unchlorinated samples,but there were no differences in theslopes for the LSB and SB indexes. Thiswas due to the fact that, in the unchlo-rinated samples, the native and exoge-nous flora were still active and couldhave caused necrosis and other delete-rious effects (Fig. 2A). On the other hand,the fact that chlorine did not affect thedevelopment of LSB and SB indicatesthat they reflect mostly enzymatic phe-nomena and are not directly affected bythe presence of chlorine. The activity ofphenylalanine ammonia lyase is involvedin the phenylpropanoid pathway in phe-nolic metabolism, leading to lettucebrowning (PEISER et al., 1998).

The addition of ascorbic acid, citricacid or CaCl2 to the chlorine dipping so-lution did not make any improvements(P<0.10) in the sensory indexes as com-pared to lettuce treated with chlorinat-ed water (control) (Table 1).

Samples treated with ascorbic acid(Chl-AA) had higher slopes for OVQ, LSBand SB (P<0.10) than with chlorinealone. BOLIN and HUXSOLL (1991) re-ported that 0.5% ascorbic acid increasedthe shelf life of iceberg lettuce by 10%. A

Fig. 1 - Evolution of LSB, SB and OVQ in lettuceleaves dipped in water with 100 ppm of chlorine(Chlorinated water) during 10 days of storage at4°C. Means of six determinations. Vertical linesrepresent SD. Some SD bars are masked by thegraph symbol.

possible explanation could be a differ-ent response of the varieties (Iceberg andRomaine) and the different storage con-ditions used.

Samples treated with chlorine and cit-ric acid (Chl-CI) had a faster decline inOVQ (P<0.10), than with chlorine alone.This is contradictory to the reports thatcitric acid can benefit the storage life of


lettuce by lowering the pH (BOLIN et al.,1977) or by preventing enzymatic brown-ing by chelation of Cu in polyphenol ox-idase (WILEY, 1997).

Samples treated with chlorine andCaCl2 (Chl-CA) had a higher slope for LSB(p<0.05), with respect to chlorine alone.KRAHN (1977) also reported that the useof CaCl2 did not produce any benefits inextending the shelf life of cut head let-tuce. Also, the storage life of shreddedlettuce (BOLIN et al., 1977) was not ex-tended by calcium treatment. The lackof response of lettuce to calcium com-pared to other products such as apples,strawberries, pears and carrots, may bedue to differences in its form, the con-centration of the solution and storagetemperature (IZUMI and WATADA, 1994).The concentration of the remaining CaCl2

Table 1 - Regression equations for changes in the quality parameters (indexes) over 10 days of storagetime at 4±1°C.

Quality parameters

DIP treatments OVQ LSB SB

Unchlorinated water y=-0.461x + 4.808 y=0.249x + 0.700 y=0.280x + 0.825b n.s. n.s.

Chlorinated water (Chl) y=-0.321x + 4.859 y=0.183x + 0.870 y= 0.206x + 0.898Chl + ascorbic acid y=-0.491x + 4.925 y=0.293x + 0.792 y=0.358x + 0.776

b b bChl + citric acid y=-0.441x + 4.601 y=0.235x + 0.878 y=0.217x + 0.927

b n.s. n.s.Chl + CaCl2 y=-0.326x + 4.509 y=0.328x + 0.737 y=0.195x + 1.018

n.s. c n.s.Chl + ascorbic acid + citric acid y=-0.232x + 4.995 y=0.170x + 1.003 y=0.280x + 0.834

a n.s. n.s.Chl + ascorbic acid + CaCl2 y=-0.414x + 4.881 y=0.160x + 0.887 y=0.310x + 0.822

b n.s. n.s.Chl + citric acid + CaCl2 y=-0.512x + 5.028 y=0.241x + 0.739 y=0.142x + 0.988

b n.s. n.s.Chl + ascorbic acid + citric acid + CaCl2 y=-0.262x + 4.431 y=0.412x + 0.898 y=0.326x + 0.844

a c n.s.

a Slope significantly lower than slope of dip treatment with chlorinated water (Chl) at P<0.10.b Slope significantly higher than slope of dip treatment with chlorinated water (Chl) at P<0.10.c Slope significantly higher than slope of dip treatment with chlorinated water (Chl) at P<0.05.n.s. No significant differences.OVQ = overall visual quality, LSB = leaf superficial browning, SB = stem browning.

on the lettuce was determined. Aftertreatment, the concentration was 4 timeshigher (1,344 µg/g) than in non treatedsamples (330 µg/g). This concentrationmight be too high and produce negativeeffects on Romaine lettuce. SAMS et al.(1993), studying the effectiveness of post-harvest CaCl2 treatments to maintainfirmness and reduce decay of whole ap-ples, found that excessive calcium couldresult in fruit injury.

The use of solutions containing a com-bination of two additives with chlorinat-ed water produced mixed results. The si-multaneous addition of citric and ascor-bic acid (Chl-AA-CI) produced a lowerslope for the OVQ index (p<0.10) com-pared with lettuce treated with chlorin-ated water (control), without significant-ly affecting indexes that show enzymatic


degradation such as LSB and SB. Thiscould be due both to the lowering of thepH and to citric acid strengthening theaction of ascorbic acid (WILEY, 1997).

The combination of ascorbic acid andCaCl2 (Chl-AA-CA) and the combinationof citric acid and CaCl2 (Chl-CI-CA) pro-duced greater declines in the OVQ in-dex (p<0.10).

The simultaneous addition of ascor-bic and citric acids and CaCl2 (Chl-AA-CI-CA) gave beneficial effects on the OVQindex, but deleterious effects on the LSBindex. Probably the presence of a highconcentration of remaining calcium onthe lettuce leaves after the CaCl2 treat-ment, produced this negative effect.

Only the simultaneous addition ofascorbic and citric acids to chlorinatedwater (Chl-AA-CI) gave an improvementover chlorinated water alone (control).

Russet spotting (RS) is a major phys-iological disorder of harvested Iceberglettuce. It is induced by exposure to eth-ylene and is more severe at a tempera-ture of 5°C (RITENOUR et al., 1995). Theseauthors indicated that when harvestedIceberg lettuce was exposed to 10 ppmof ethylene at 5°C, symptoms appearedafter approximately 3 days of storage. Inall the samples, the development of rus-set spotting showed no significant in-crease up to 6 days of storage. By day10 of storage, the indexes increased slow-ly regardless of the bath treatment,reaching final values of 1.6 to 2.0.

Microbiological counts

A fundamental problem in extendingthe shelf life of minimally processed fruitsand vegetables is microbial proliferationwhich must be retarded. Growth of dis-ease-causing microorganisms is a foodsafety concern, especially with higher pHvegetables (in contrast to lower pH fruits).Visible growth and off-odors caused bymicroorganisms are aesthetically unac-ceptable. The ecological niche caused bythe composition of the food will deter-

mine the kind of microorganisms thatwill grow.

PONCE et al. (2002) reported large dif-ferences in the initial microbial countsamong different lots of fresh leaves ofSwiss chard obtained by conventionalmethods that could not be entirely ex-plained by the variability of the assess-ment methods. Numerous factors, suchas ambient conditions during harvest,amount of soil accompanying the prod-uct, time elapsed between last irrigationand harvest or post-harvest handling,could be responsible for these differenc-es. Since some of these factors can notbe controlled, initial counts would benecessary whenever the effect of post-harvest processes on the micro flora ofthis type of product is evaluated.

Fig. 2 shows the trend of a) mesophilicbacteria, b) psychrotrophic bacteria andc) yeast and mold populations on lettuceleaves dipped in the various baths. Ini-tial counts for mesophilic aerobic micro-organisms were in the range of 1.3x105

to 2.5x106 CFU/g for all sample treat-ments. The initial counts correspondingto control samples (chlorinated water)were 7-8x105 CFU/g, the highest countscorresponded to samples with unchlo-rinated water (tap water) with 2-3x106

CFU/g and the lowest counts corre-sponded to samples with chlorine plusascorbic acid plus citric acid (Chl-AA-CI)with 1-2x105 CFU/g (Fig 2A). These sam-ples maintained the lowest counts up to2 days of storage. Chlorine treatmentreduced (P<0.01) the initial populationof natural contaminants. Hypochloritedips are commonly used in postharvestpractices for sanitizing fruits and vege-tables, particularly in the fresh cut in-dustry (BEHRSING et al., 2000). Chlorineacts on the microbial flora of vegetables(WILEY, 1997; WATADA, 1997). Microbialactivity has been shown to be associat-ed with the development of necrosis dur-ing storage of lettuce leaves (NGUYEN-THE and PRUNIER, 1989; JACQUES andMORRIS, 1994). By 10 days no differenc-


es (P<0.01) in the populations of mes-ophilic aerobic microorganisms werenoted on lettuce regardless of type ofbath (1.5-8x107 CFU/g). For minimallyprocessed vegetables, French legislationallows a maximum of 5x107 CFU/g at

Fig. 2 - Mean microbial Log10 counts of lettuceleaves during storage: Mesophilic bacteria (A), Psy-crotrophic bacteria (B) and Yeasts and molds (C).Values correspond to a mean of duplicates, ob-tained from three independent lots, therefore val-ues are the means of six determinations. Verticallines represent SD. Some SD bars are masked bythe graph symbol.� - Unchlorinated water treatment, - Chlorinat-ed water treatment, � - Chl-AA-CI.

the stage of consumption (BARRIGA et al.,1991), hence the microbiological qualitycould be considered satisfactory during10 days of storage.

The levels and trend of psychrotroph-ic microorganisms were similar to thoseof mesophilic aerobic microorganisms(Fig. 2B). The population increased froman initial level of 0.25-6x105 to 2-6x107

CFU/g. In general, the numbers of psy-chrotrophic microorganisms were ac-ceptable. With the various chemicaltreatments investigated, no differencesin the initial counts (P<0.01) were found.The lowest initial counts correspondedto the sample treated with chlorinatedwater.

The lowest initial counts for yeasts andmolds (3-9x103 CFU/g) were for lettucedipped in chlorine plus ascorbic acid(Chl-AA), chlorine plus ascorbic acid pluscitric acid (Chl-AA-CI), chlorine plusascorbic acid plus calcium (Chl-AA-CA),chlorine plus citric acid plus calcium(Chl-CI-CA) and chlorine plus ascorbicacid plus citric acid plus calcium (Chl-AA-CI-CA). This could probably be dueto the low pH of these baths (pH 4-5). Byday 10 of storage no differences (p<0.01)were found among dip treatments (Fig.2C). BEUCHAT and GOLDEN (1989) stud-ied antimicrobials that occur naturallyin foods, and suggested that some or-ganic acids can exert fungicidal and fun-gistatic action which is related direct-ly to lowering the pH of the substrate.

The ability of both mesophilic and psy-chrotrophic bacteria and yeast andmolds to multiply at the low tempera-tures assayed indicates their potentialability to cause lettuce spoilage.

Ascorbic acid contents

The initial ascorbic acid content offresh leaf samples was 8.3 mg/100g offresh weight. The endogenous ascorbicacid concentrations in the fresh lettuceused in this study were higher than thosereported by ALBRECHT (1993) for Ro-


maine lettuce. The ascorbic acid concen-trations of samples dipped in treatmentsfree of ascorbic acid were not significant-ly different (P<0.05). On the other handsamples treated with ascorbic acid hadabout a 4 times higher initial ascorbicacid content without significant differ-ences among them. Fig. 3A presents themean values obtained for the ascorbicacid concentration of samples treatedwithout ascorbic acid (treatments withunchlorinated water, chlorinated water,Chl-CI, Chl-CA and Chl-CI-CA) and Fig.3B presents the mean values obtainedfor the ascorbic acid concentration ofsamples treated with ascorbic acid (treat-ments with Chl-AA, Chl-AA-CI, Chl-AA-CA and Chl-AA-CI-CA). The differencebetween the means of both groups dur-ing storage was always significant(p<0.01).

The rate of ascorbic acid loss was lowerin samples without ascorbic acid treat-ments than in samples with ascorbic acid

treatments. For example, by day 4, sam-ples without ascorbic acid had lost 50%of their initial ascorbic acid content,while samples with ascorbic acid had lost77%. This difference could be attributedto the ascorbic acid being on the sur-face and more exposed to oxidation.

Nevertheless, in spite of the higherrates of ascorbic acid degradation inascorbic acid-treated samples, their ab-solute concentrations remained higherthan in the other samples during stor-age.

CONCLUSIONS

The effect of dip treatments with chlo-rine, CaCl2, citric acid and ascorbic acidon the quality of Romaine lettuce leavesduring storage was investigated. Chlo-rine may be required to inhibit thegrowth of bacteria responsible for necro-sis. Neither citric acid nor ascorbic acid

Fig. 3 - Ascorbic acid content (mg per 100g of fresh vegetable) in samples treated (A) without ascorbicacid (Chlorinated water, Unchlorinated water, Chl-CI, Chl-CA and Chl-CI-CA) and (B) samples treatedwith ascorbic acid (Chl-AA, Chl-AA-CI, Chl-AA-CA, Chl-AA-CI-CA) during 10 days of storage at 4°C.Means of six determinations. Vertical lines represent SD.


in chlorinated solutions resulted in im-provements in LSB, SB, OVQ indexes,which is in contrast with published re-sults on Iceberg lettuce. However, thecombination of both acids resulted in aslower decline of the overall visual qual-ity of the Romaine lettuce leaves. Theaddition of calcium chloride did not re-sult in improvements in the visual qual-ity and could be responsible for fasterleaf surface browning. Samples treatedwith ascorbic acid maintained highercontents of ascorbic acid throughoutstorage, although the rate of ascorbicacid loss was higher than in samples notdipped in ascorbic acid.

REFERENCES

Albrecht J.A. 1993. Ascorbic acid content and re-tention in lettuce. J. Food Quality 16:311.

Anonymous. 1983. Prevention of browning in fresh,prepared vegetables. Miles Laboratories, Inc.,BioTech Division, Elkhart, IN

Bangerth F., Dilley D.R. and Dewey D.H. 1972.Effect of post-harvest calcium treatment on in-ternal breakdown and respiration of apple fruits.J. Amer. Soc. Hort. Sci. 97:679.

Barriga M.I., Trachy G., Willemot C. and SimardR.E. 1991. Microbial changes in shredded Ice-berg lettuce stored under controlled atmos-pheres. J. Food Sci. 56:1586.

Bauernfeind J.C. and Pinkert D.M. 1970. Foodprocessing with added ascorbic acid. Adv. FoodRes. 18:219.

Behrsing J., Winkler S., Franz P. and Premier R.2000. Efficacy of chlorine for inactivation ofEscherichia coli on vegetables. Postharvest Biol.Technol. 19:187.

Beuchat L. R. and Golden D.A. 1989. Antimicrobi-als occurring naturally in foods. Food Technol.43:134.

Bolin H.R. and Huxsoll C.C. 1991. Effect of prepa-ration procedures and storage parameters onquality retention of salad-cut lettuce. J. FoodSci. 56:60.

Bolin H.R., Stafford A.E., King A. D Jr. and Hux-soll C.C. 1977. Factors affecting the storage sta-bility of shredded lettuce. J. Food Sci. 42:1319.

Bruemmer J.H. 1987. Stability of prepared carrotsticks in storage. Proc. Fla. State Hort. Soc.100:36.

Couture R., Cantwell M.I. and Saltveit M.E. Jr.1993.Physiological attributes related to quality

attributes and storage life of minimally proc-essed lettuce. HortSci. 28:723.

Drake S.R. and Spayd S.E. 1983. Influence of cal-cium treatment on “Golden Delicious” applequality. J. Food Sci. 48:403.

Heimdal H., Kuhn B.F., Poll L. and Larsen L.M.1995. Biochemical changes and sensory quali-ty of shredded and MA-packaged iceberg let-tuce. J. Food Sci. 60:1265.

ICMSF. 1983. Métodos recomendados para el análi-sis microbiológico en alimentos. In “Microorgan-ismos de los alimentos I. Técnicas de análisismicrobiológicos” (2da), p. 105. Acribia S.A.Zaragoza, España.

Izumi H and Watada A.E. 1994. Calcium treat-ments affect storage quality of shredded car-rots. J. Food Sci. 59:106.

Izumi H. and Watada A.E. 1995. Calcium treat-ment to maintain quality of zucchini squashslices. J. Food Sci. 60:789.

Jacques M.A. and Morris C.E. 1994. Diversity ofbacteria contributing to the decay of ready-to-use salads. In “Proceeding of the 8th Interna-tional Conference of Plant Pathogenic Bacteria”.M. Lemattre, F. Freigoun, K. Rudolph and J.G.Swing (Eds) p. 165. 9-12 June, Versailles.

Kader A.A., Lipton W.J. and Morris L.L. 1973. Sys-tems for scoring quality of harvested lettuce.HortSci. 8:408.

Ke D. and Saltveit M.E. Jr. 1989a. Regulation ofrusset spotting, phenolic metabolism and IAAoxidase by low oxygen in iceberg lettuce. J.Amer. Soc. Hort. Sci. 114:638.

Ke D. and Saltveit M.E. Jr. 1989b. Developmentalcontrol of russet spotting, phenolic enzymes,and IAA oxidase in cultivars of iceberg lettuce.J. Amer. Soc. Hort. Sci. 114:472.

Krahn T.R. 1977. Improving the keeping quality ofcut head lettuce. Acta Hort. 62:70.

Lopez-Galvez G., Saltveit M.E. and Cantwell M.1996. Wound-induced phenylalanine ammonialyase activity: factors affecting its induction andcorrelation with the quality of minimally proc-essed lettuce. Postharvest Biol. Technol. 9:223.

Mateos M., Ke D., Cantwell M. and Kader A. 1993a.Phenolic metabolism and ethanolic fermenta-tion of intact and cut lettuce exposed to CO2-enriched atmospheres. Postharvest Biol. Tech-nol. 3:225.

Mateos M., Ke D., Kader A.A. and Cantwell M.1993b. Differential responses of intact and min-imally processed lettuce to high carbon dioxideatmospheres. Acta Hort. 342:171.

McEvily A.J., Iyengar R. and Otwell, W.S. 1992.Inhibition of enzymatic browning in foods andbeverages. Crit. Rev. Food Sci. Nutr. 32:253.

Mossel D.A. and Moreno García R. 1985. “Micro-biología de los Alimentos”. p. 214. Acribia S.A.Zaragoza, España.

Nguyen-The C. and Prunier J.P. 1989. Involvement


of Pseudomonas in “ready-to-use salads dete-rioration”. Int. J. Food Sci. Tech. 24:47.

Peiser G., Lopez-Galvez G., Cantwell M. and Salt-veit M.E. 1998. Phenylalanine ammonia lyaseinhibitors control browning of cut lettuce. Post-harvest Biol. and Technol. 14:171.

Pelletier O. 1985. Vitamin C: L-ascorbic and dehy-dro-L-ascorbic acids. In “Methods of VitaminAssay”, J. Augustin, B.P. Klein, and D. Becker(Ed), 4th ed. John Wiley & Sons, New York.

Ponce A., Roura S., del Valle C. and Fritz, R. 2002.Characterization of native microbial populationof Swiss chard (Beta vulgaris, type cicla). Leb-ensm.-Wiss U.-Technol. 35:331

Ponting J.D., Jackson R. and Walters G. 1972.Refrigerated apple slices: preservative effects ofascorbic acid, calcium and sulfites. J. Food Sci.37:434.

Poovaiah B.W. 1986. Role of calcium in prolong-ing storage life of fruits and vegetables. FoodTechnol. 40:86.

Ritenour M.A., Ahrens M.J. and Saltveit M.E. 1995.Effects of temperature on ethylene-inducedphenylalanine ammonia lyase activity and rus-set spotting in harvested Iceberg lettuce. J.Amer. Soc. Hort. Sci. 120:84

Rosen J. C. and Kader A.A. 1989. Postharvest phys-

iology and quality maintenance of sliced pearand strawberry fruits. J. Food Sci. 54:656.

Saltveit M.E. 1997. Postharvest diseases. In “Com-pendium of Lettuce Diseases” Davies, R.M.,Subbarao, K.V., Raid, R.N., Kurtz, E.A. (Eds).p. 57. American Phytopathological Society Press,St. Paul, MN, USA.

Sams C.E., Conway W.S., Abbot J.A., Lewis R.J.and Ben-Shalom N. 1993. Firmness and decayof apples following postharvest pressure infil-tration of calcium and heat treatment. J. Amer.Soc. Hort. Sci.. 118:623.

Sapers G.M., Garzarella L. and Pilizota V. 1990.Application of browning inhibitors to cut appleand potato by vacuum and pressure infiltra-tion. J. Food Sci. 55:1049.

Volk W. (Ed), 1958. “Applied Statistics for Engi-neers”. Mc Graw-Hill, Inc. New York.

Watada A.E. 1997. Quality maintenance of fresh-cut fruits and vegetables. Foods and Biotech-nol. 6:229.

Wiley R.C. 1997. “Minimally Processed Refrigerat-ed Fruits and Vegetables”. p. 221. Chapman &Hall, Inc. One Penn Plaza, New York, NY.

Wills R.B. and Tirmazi S.H. 1982. Effect of post-harvest application of calcium or ripening ratesof pears and bananas. J. Hort. Sci. 57:431.

Revised paper received March 4, 2003 Accepted May 30, 2003


PAPER

- Key words: lectins, muscle regeneration, myofibers, wheat germ -

EFFECTS OF A WHEAT GERM-ENRICHED DIETON SKELETAL MUSCLE REGENERATION

EFFETTI DI UNA DIETA ARRICCHITA IN GERME DI GRANOSULLA RIGENERAZIONE DEL MUSCOLO SCHELETRICO

C. RIZZI*, R. CHIGNOLA, G. ZOCCATELLI, M. DONÀ1,A.D.B. PERUFFO, U. CARRARO1 and K. ROSSINI1

Dipartimento Scientifico e Tecnologico, Università di Verona,Strada le Grazie, Ca’ Vignal 1, 37134 Verona, Italy

1 C.N.R. Istituto di Neuroscienze, Unità di Biologia e Fisiopatologia Neuromuscolare,Laboratorio di Miologia Applicata del Dipartimento di Scienze Biomediche,

Università di Padova, Via Trieste 75, 35121 Padova, Italy* Corresponding author:

Tel. +39 045 8027960; Fax +39 045 8098929; E-mail: [email protected]

ABSTRACT

One of the common ingredients usedin dietary supplements is wheat germand its derivatives. In this study theeffect of a wheat germ-enriched diet(WGED) on tissue repair was investi-gated in a model of skeletal muscledamage induced by the anesthetic bupi-vacaine. Rats were fed normal feedstuffor a WGED; injections of bupivacainecaused skeletal muscle necrosis fol-lowed by tissue regeneration. Morpho-metric analyses were performed on themuscles in order to quantify cell size

RIASSUNTO

Uno dei costituenti più utilizzati ne-gli integratori è il germe di grano ed isuoi estratti. In questo studio abbiamovoluto saggiare gli effetti di una dietaarricchita in germe di grano (WGED)sulla rigenerazione muscolare in un bennoto modello sperimentale di dannoindotto da un anestetico locale di usocomune (la bupivacaina). Muscoli diratti alimentati con normale mangimeo con WGED sono stati iniettati con bu-pivacaina che causa necrosi seguita darigenerazione del tessuto. Sono state


and nuclei migration toward the cellperiphery because these are indices ofmuscle regeneration. Statistical analy-sis shows that myoblast maturation,determined by nuclei migration to aperipheral position, was significantlydelayed in rats fed WGED compared torats fed the control diet. No statistical-ly significant differences were observedin the myocellular areas. This studyprovides evidence that dietary wheatgerm can interfere with muscle regen-eration.

eseguite quindi analisi morfometricheconsiderando quali indici di rigenera-zione le dimensioni delle cellule rigene-ranti e la migrazione dei nuclei alla pe-riferia. L’analisi statistica dei datimostra che la maturazione dei miobla-sti, valutata misurando la migrazionedei nuclei, era significativamente ritar-data nel tempo nei ratti alimentati conWGED rispetto ai controlli. Differenzestatisticamente non significative sonostate invece osservate per quanto con-cerne le dimensioni cellulari. Si puòquindi ipotizzare che il germe di grano,assunto per via orale, abbia effetti sul-la rigenerazione muscolare.

INTRODUCTION

Increasing importance is being givento natural foods and nutritional supple-ments. For example, nutritional supple-ments are currently being promoted forathletes due to their supposed benefi-cial effects on muscle growth and repair(MAUGHAN, 1999). Similar supplementsare also used in the paediatric field. Al-though there is some experimental evi-dence to support the use of some of theseproducts, the claims made for others arenot based on controlled experiments.

Many dietary supplements containwheat germ derivatives that may include,among other proteins, wheat germ ag-glutinin (WGA). WGA is a lectin, a mem-ber of the class of carbohydrate-bindingproteins whose effects on several physi-ological functions have been addressedin numerous studies (MANAGE et al.,1972; MITCHELL and CLARKE, 1979; LIV-INGSTONE and PURVIS, 1980; NOAH etal., 1980; PUSZTAI, 1991; PUSZTAI et al.1993; VAN DAMME et al., 1998). Lectins,including WGA, bind glycoconjugatesexpressed at the cell surface level includ-ing those of the gastrointestinal tract,where they may cause functional disor-

ders (nausea, vomiting, diarrhoea) andanatomical alterations (hyperplasia andhypertrophy) (VAN DAMME et al., 1998;PUSZTAI et al., 1993). Some lectins areremarkably toxic upon ingestion as is thecase for lectins from Ricinus communis(PUSZTAI, 1991) and Phaseolus vulgaris(MANAGE et al., 1972). Direct cytotoxicactivity has also been demonstrated forWGA in vitro (VAN DAMME et al., 1998),although this lectin is structurally andfunctionally unrelated to both R. commu-nis and P. vulgaris lectins. Due to theircapacity to cross the gastrointestinalbarrier in a biologically active form,lectins can also have systemic effects,such as hypersensitivity reactions(MITCHELL and CLARKE, 1979; PUSZTAI,1991), increased lipogenesis and pancre-atic hypertrophy (LIVINGSTONE and PUR-VIS, 1980). In particular, WGA has beenshown to interfere with several functionsof the immune system in both in vitroand in vivo assays (GREENE et al., 1985;TORIBIO et al., 1985; MURAILLE et al.,1999). It is important to note that, alongwith other lectins, WGA has the abilityto promote the secretion of pro-inflam-matory cytokines (HAAS et al., 1999).However, little is known about the effects


of these proteins on the possible accu-mulation resulting from their long-termintake and on their possible interactionswith other dietary components when giv-en in amounts comparable to thosepresent in a common diet. These effectswould be shown more clearly in in vivoassays with complex mixtures that aresimilar to natural foodstuffs.

Bupivacaine is a well-studied drug thatcauses extensive skeletal muscle necro-sis followed by rapid muscle regenerationwith established kinetics and histologiccharacteristics (JIRMANOVA andTHESLEFF, 1972; CARRARO et al., 1983;MUSSINI et al., 1987). An injection ofbupivacaine causes sarcolemmal lysis ofmost of the fibers with extensive dilationof the sarcoplasmic reticulum accompa-nied by overcontraction and overstrech-ing of the myofilaments (BRADLEY, 1979).Most fibers are totally disrupted, leavingempty basal lamina bags (BRADLEY,1979). These effects can be observed 1 hafter injection when sector lesions appear;they are similar to those which occur inDuchenne muscular dystrophy as report-ed by MOKRY and ENGEL (1975) andCRAWFORD et al. (2000). Two days afterinjection, phagocytes begin to infiltratethe muscle and satellite cells becomeprominent below the basement mem-brane (BRADLEY, 1979). By day 4, regen-eration of the muscle is prominent evenoccurring side-by-side with necrotic fib-ers (BRADLEY, 1979). On day 6, fiber re-generation is extensive but fibers are stillsmaller and rounder than normal, non-injured ones. The nuclei occupy a cen-tral position in the fiber. In the rat model,myoblasts progressively increase theircytoplasm and finally show peripheralmigration of their nuclei (BRADLEY, 1979).

The above model of tissue damage bynecrosis/inflammation and of tissue re-pair was used to test the possible inter-actions of a wheat germ-enriched diet(WGED) with a complex series of patho-physiological pathways. The myocellu-lar areas and the fraction of central my-

onuclei were selected as parameters cor-related with muscle regeneration. WGEDdelayed tissue repair and this delay ap-peared to be related to a specific altera-tion of the molecular mechanism(s) thatdrives the localisation of the myonucleiin muscle fibers.


Animals, diet compositionand feeding procedures

Three-month-old Wistar rats, rearedat the Università di Padova, were used.They were fed with normal feedstuff (NF)for rodents (Harlan Italy, Udine, Italy)containing 18% protein or with the wheatgerm-enriched diet (WGED). The WGEDwas obtained by adding 1:1 (w:w) wheatgerm to NF (see Table 1 for diet compo-sition). The WGED was continuouslyadministered during the experimentstarting 4 days before muscle injury.Animals were housed individually. Theexperiments were carried out accordingto current Italian laws (D.L. 116, 1992)in agreement with the directive 86/609/CEE on animal experimentation.

WGA determination in wheatgerm and in the dietary formulations

The WGA content in the normal ratfeedstuff, wheat germ and WGED wasdetermined by ELISA assays as de-scribed by VINCENZI et al. (2002). Brief-ly, samples were solubilised (1:10 w/v)with phosphate-buffered saline (PBS)and centrifuged for 15 min at 10,000 g.Samples were loaded onto Ovalbumin-coated microtiter plates. Bound WGAwas detected using specific antibodiesand its concentration in unknown sam-ples was determined by comparing ODvalues with those obtained in parallel forstandard dose-response assays carriedout with purified WGA (Sigma-Aldrich,Milano, Italy).


Tissue damage/regenerationprocedures

Six control NF-fed and six WGED-fedrats were anaesthetised with 0.1 mg/gketamine chloridrate. Soleus musclesfrom the right hind limbs were identi-fied, raised and then a strip of parafilmwas passed below the muscle belly. Themuscles were injected with 200 µL of0.5% bupivacaine in 0.9% NaCl andsoaked in the same solution. After 5min, the muscles were replaced in thenormal anatomic position and the skinwas sutured. Fifteen days after surgery,the animals were sacrificed under an-aesthesia and the muscles were explant-

ed and immediately frozen in liquid ni-trogen at resting length. The muscleswere stored at -80°C until used for themorphologic and morphometric analy-ses.

Morphologic, morphometricand statistical analyses

The 12 frozen muscles were sec-tioned at the mid-belly using a cryo-stat. Three 10 µm thick sections wereput on glass slides and stained withhematoxylin and eosin (H&E) usingconventional techniques. Images wereobtained using an Olympus IX70 mi-croscope (Olympus Italia, Milano, Ita-

Table 1 - The average composition of normal feedstuff (NF) and wheat germ-enriched diet (WGED)(main components only).

Component 1NF 2WGED Ratio (WGED/NF)

Protein % 18.90 22.80 1.20Fat % 5.70 7.45 1.30Carbohydrate % 57.30 51.00 0.89Na % 0.23 0.11 0.47K % 0.68 0.81 1.19Ca % 1.01 0.53 0.52Mg % 0.20 0.23 1.15Fe mg/kg 226.00 155.50 0.68Cl % 0.40 0.24 0.60P % 0.65 0.85 1.30Cu mg/kg 15.21 12.10 0.79Zn mg/kg 77.00 123.50 1.60Vitamin A mg/kg 4.65 2.32 0.49Vitamin B1 mg/kg 16.50 14.25 0.86Vitamin B2 mg/kg 14.90 11.05 0.74Vitamin B6 mg/kg 18.50 25.75 1.39Vitamin B12 mg/kg 0.08 0.04 0.50Biotin mg/kg 0.30 0.27 0.90Vitamin C mg/kg 0.00 0.00 —Vitamin E mg/kg 101.00 160.00 1.58Energy value MJ/kg 13.70 14.40 1.053WGA mg/100 g <0.03 20 >600

1 Composition of NF diet was taken from Harlan Teklad (18% Protein Rodent Diet, see www.teklad.com/global/data/2018.htm).2 Composition of WGED was calculated by taking the average compositions of wheat germ (McCANCE and WID-DOWSON, 1997) and of NF and by considering the 1:1 mixture.3 Determination of WGA was from our own experiments (see the text for details).


ly) connected to a 2-CDD camera (mod-el KYF58, JVC Professional Products,Milano, Italy). The Image-Pro Plus soft-ware (Media Cybernetics, Silver Spring,USA) was used for the morphometricanalysis.

Six control NF-fed and six WGED-fed rats were used in these assays.The morphometric measurements ofthe myocellular areas were performedat 20X magnification (in a range of 3to 6 fields for each muscle), while thedetermination of the myonuclei posi-tion was performed at 40X (in a rangeof 4 to 8 images). In both analyses mi-croscopic fields were randomly select-ed.

Myocellular area determinations

Tissue preparations from one controland one WGED-fed rat were excludedfrom the analysis because of an incor-rect injection procedure which was re-vealed by histologic inspection. On theaverage 624 cells per rat were analysedfor the control NF-fed group and 1,004cells per rat for the WGED-fed groupfor a total of 3,120 and of 5,020 cellsanalysed, respectively. Data from eachrat were collected and the mean±SD ar-eas calculated. The mean areas werethen compared using the Student t-test.

Determination of myonuclei position

Tissue preparations from one con-trol and one WGED-fed rat were ex-cluded from the analysis as mentionedabove. On the average 435 cells perrat were analysed for the control NF-fed group and 810 cells per rat for theWGED-fed group for a total of 2,175and 4,050 cells analysed, respective-ly. The fraction of cells with nucleiclearly in a central position were cal-culated for each rat and the data weresubjected to analysis of variance us-ing the F-test.

RESULTS

WGA determination in wheat germ,in NF diet and in WGED

LIS and SHARON (1981) reported a30±20 mg /100 g WGA/WG ratio infood-store available products. Due toa high variability in the concentrationsof WGA lectin in the wheat germ usedin our assays and in the NF (largelybased on raw cereal derivatives) theseconcentrations should be directly de-termined. Using a sensitive and specif-ic ELISA method (VINCENZI et al., 2002),the WGA concentration in the wheat germwas found to be 40±1 mg /100 g where-as in NF it was found to be below theELISA sensitivity threshold (30 ng/mLcorresponding to 0.03 mg/100 g). West-ern Blot analysis of the NF revealedthe presence of a faint WGA band (datanot shown). The concentration of WGAin WGED (1:1 wheat germ:NF) wasfound to be 20±0.2 mg /100 g as ex-pected. The WGA content constitutedthe main difference in the compositionof NF and WGED feedstuffs, whereasno significant differences were ob-served for the other components (Ta-ble 1).

Effects of wheat germ on muscledamage/regeneration responses

No differences in the dietary intake(approximately 20 g/day) or of commonphysiological parameters (e.g. modifica-tion of body weight, fur status) were ob-served during the course of the experi-ments carried out with rats on NF andWGED. The histological profiles at day15 of the regeneration kinetics are shownin Fig. 1. In both control NF-fed andWGED-fed groups, the rat skeletal mus-cles still exhibited a great variability inmyofiber diameter with a certain propor-tion of the nuclei in the central position.These morphometric parameters are cor-related with an incomplete recovery of


Fig. 1 - Transversal mid-belly section of normalrat soleus muscle (Panel A), of recovering soleusmuscle after bupivacaine-induced injury (day 15)in control NF-fed rats (Panel B) or in WGED-fedrats (Panel C). H&E stain at 20X magnification.

the muscles (panels B and C). A sectionof tissue from a rat muscle not subject-ed to tissue injury is also shown for com-parison purposes (panel A). The regen-erating muscles in panels B and C show

an incompletely restored fascicle archi-tecture with sheaths of fibrotic tissueenclosing groups of fibers.

In comparison to normal tissue (pan-el A), the mean myofiber diameters were23.3±16.6% and 21.3±12.8% in musclesfrom the control NF-fed and from theWGED-fed rats, respectively. The meanmyocellular areas±SD measured in tis-sue preparations from each rat in thetwo analysed groups are shown in Fig.2. Great variability characterised thisparameter even within each group ofrats. Overall, the areas in the NF-fedgroups appear slightly greater than inthe WGED-fed group as expected for ahigher degree of myofiber recovery. Nosignificant difference was found betweenthe control NF-fed and WGED-fed rats(t-test, p>0.05).

The determination of the myonucleiposition in the control NF-fed and in theWGED-fed rats is shown in Table 2. Alarger fraction of the myonuclei occupieda central position in WGED-fed rats com-pared to tissue preparations from the NF-fed rats. The differences were statisticallysignificant (F-test, p<0.01).

DISCUSSION

Knowledge that bioactive moleculesthat enter the body via the alimentarytract can influence a number of patho-physiological responses is not new (KIL-PATRICK, 1999; MAUGHAN, 1999). Forexample, some dietary plant lectins havebeen shown to modulate the activity ofthe immune system (TORIBIO et al., 1985;HAAS et al., 1999; KILPATRICK, 1999).Bupivacaine-induced skeletal muscleinjury is a well established model ofmuscle necrosis and regeneration thatincludes a phase of active inflammation(JIRMANOVA and THESLEFF, 1972; CAR-RARO et al., 1983; MUSSINI et al., 1987).It is a way to analyse the possible inter-actions between dietary signals and an-imal patho-physiological pathways with-


Fig. 2 - Measures of mean±SD myocellular areas in tissue sections of skeletal muscles from control NF-fed (white bars) and WGED-fed (black bars) rats.

in a complex framework of regulatorycells and signals, including immunecells, cytokine production which regu-lates the extent of inflammation and anumber of signals involved in the regen-

eration of myofibers. Included in the lat-ter are growth factors (basic fibroblastgrowth factors), insulin-like growth fac-tors I and II and transforming factor(BORNEMANN and ANDERSON, 2000).

Table 2 - Determination of myonuclei position in recovering skeletal muscles of control NF-fed andWGED-fed rats.

WGED-fed rats NF-fed rats

Animals Total cells Cells with 1 Ratio Total cells Cells with Ratiocentral nuclei central nuclei

1 748 215 0.287 323 73 0.2262 633 278 0.439 438 67 0.1533 967 321 0.322 388 56 0.1444 791 299 0.378 583 82 0.1415 911 303 0.333 443 110 0.248

Total 4,050 1,416 2,175 388Mean 810 283.2 0.352 435 77.6 0.182SD 132.6 41.07 0.058 95.82 20.42 0.05

1 Differences in the ratio values between the two treatment groups were evaluated using the F-test. F**=23.92(F0.01=11.26).


Consequently, the general nutritionalcondition of the organism is importantfor muscle recovery, as well as any oth-er factor affecting local circulation (e.g.infections or the presence of a foreignbody leading to local proliferation of con-nective tissue) (FIELD, 1960).

A diet is a complex mixture of mole-cules that can potentially act at differ-ent levels in the body. Among the die-tary molecules, plant lectins attract at-tention because they can interfere withseveral physiological pathways in theanimal (LIVINGSTONE and PURVIS, 1980;KILPATRICK, 1999). In particular, WGAcan reduce the digestibility of dietaryproteins and induce anatomical modifi-cations of the gut that alter animalgrowth (PUSZTAI, 1991; PUSZTAI et al.,1993). WGA can alter mammalian me-tabolism at the endocrine level bymimicking the effects of insulin on adi-pocytes, thereby stimulating the synthe-sis of triglycerides and the transport andoxidation of glucose in fat cells (LIVING-STONE and PURVIS, 1980). A WG-en-riched diet, thus containing WGA, wouldtherefore be expected to interfere withskeletal muscle regeneration.

It should be stressed that studies thatinvestigate the role of dietary signals (e.g.lectins) on some functional responsesusually analyse the effects of artificial-ly-introduced high doses of purified mol-ecules in experimental animals (PUSZTAIet al., 1993). The effective role of dietarymolecules under physiological conditionswould therefore not be included in suchstudies. Moreover, the possible interac-tions of the candidate molecule with oth-er components of the diet and/or thecumulative effects resulting from thecontinuous intake of low doses of thesecompounds would not be explored. Forthese reasons the present work usedcomplex diet formulations, the NF diet,that had undetectable levels of WGA, andthe WGED, that contained WGA at a con-centration of 20 mg/100 g. The poten-tial effect of WGA, as well as the other

components of the diet, on muscle re-generation were therefore studied undernutritional conditions that resemblednatural ones (e.g., the diet of a wild ratliving in a mill). In fact, we failed to ob-serve the typical systemic effects of WGAin WGED-fed rats that have been report-ed in other studies that used much high-er doses of WGA (PUSZTAI et al., 1993).

A significant difference in the regenera-tion of skeletal muscle myofibers was ob-served between control NF-fed and WGED-fed rats with respect to the localisation ofmyonuclei, whereas no differences wereobserved for the measured myocellularareas. The position of myonuclei is a pa-rameter correlated with the later phasesof muscle regeneration. Hence, the ob-served differences do not appear to be re-lated to a delay in the general maturationof myofibers but rather to some specificalteration of the mechanisms by whichnuclei migrate to the periphery at a latertime. Unfortunately, little is known aboutthe nuclear movement at the molecularlevel. The modified regeneration inducedby WG may be of help in future studiesaimed at clarifying these mechanisms inmore detailed time-course assays.

In conclusion, this study provides evi-dence that dietary wheat germ can inter-fere selectively with complex patho-phys-iological pathways under standard non-limiting nutritional conditions. Dietarywheat germ altered skeletal muscle re-generation. This fact should be consid-ered when claims are made about thebeneficial effects of natural foods and,above all, of nutritional supplements.Many supplements are promoted as hav-ing beneficial effects on muscle growthand regeneration and are therefore rec-ommended for athletes (MAUGHAN, 1999).Similar supplements are largely used inthe paediatric area and in the diet of re-covering patients. Since food supplementsare not subjected to regulation(MAUGHAN, 1999), as are drugs, and sincetheir use is frequently self-determined bythe consumer, the results of the present


work suggest that these products shouldbe evaluated more critically.

ACKNOWLEDGEMENT

This work was supported in part by grants fromthe Ministero dell’Istruzione dell’Università edella Ricerca (Cofin02, MURST 40%, Studio delleproteine delle farine di frumento e degli alimentiderivati in relazione alla loro attività citotossicae alle interazioni con il sistema immunitario)and from the Fondazione Cariverona (Ambientee Sviluppo Sostenibile 2002, Aspetti fisiopato-logici dell’interazione tra alimenti derivati daicereali e sistema immunitario).

REFERENCES

Bornemann A. and Anderson L.V. 2000. Diagnos-tic protein expression in human muscle biop-sies. Brain Path. 10: 193.

Bradley W.G. 1979. Muscle fiber splitting. In “Mus-cle Regeneration” A. Mauro, R. Bischoff, B.M.Carlson, S.A. Shafiq, I. Konigsberg and B. Lip-ton (Eds.). p. 215. Raven, New York.

Carraro U., Dalla Libera L. and Catani C. 1983.Myosin light and heavy chains in muscle re-generating in absence of the nerve: transientappearance of the embryonic light chains. Exp.Neur. 79: 106.

Crawford G.E., Faulkner J.A., Crosbie R.H., Camp-bell K.P., Froehner S.C. and Chamberlain J.S.2000. Assembly of the dystrophin-associatedprotein complex does not require the dystrophinCOOH-terminal domain. J. Cell Biol. 150: 1399.

Field E.J. 1960. Muscle regeneration and repair.In “The Structure and Function of Muscle” G.H.Bourne (Ed.). p. 139. Academic, New York.

Greene W.C., Goldman C.K., Marshall S.T., Fleish-er T.A. and Waldmann T.A. 1985. Stimulation ofimmunoglobulin biosynthesis in human B cellsby wheat germ agglutinin. J. Immunol. 127: 799.

Haas H., Falcone F.H., Schramm G., Haisch K.,Gibbs B.F., Klauche J., Poppelmann M., Beck-er W.M., Gabius H.J. and Schlaak M. 1999.Dietary lectins can induce in vitro release of IL-4 and IL-13 from human basophils. Eur. J. Im-munol. 29: 918.

Jirmanova I. and Thesleff S. 1972. Ultrastructuralstudy of experimental muscle degeneration andregeneration in the adult rat. Zeitschrift fur Zell-forschung und Mikroskopische Anatomie 131: 77.

Kilpatrick D.C. 1999. Immunological aspects of thepotential role of dietary carbohydrates andlectins in human health. Eur. J. Nutr. 38: 107.

Lis H. and Sharon N. 1981. Lectins in higher plant.In “The Biochemistry of Plants” A. Marcus (Ed.).p. 371. Academic, New York.

Livingstone J.N. and Purvis B.J. 1980. Effects ofwheat germ agglutinin on insulin binding andinsulin sensitivity of fat cells. Am. J. Physiol.238: 267.

Manage L., Joshi A. and Sohonie K. 1972. Toxicityto rats and mice of purified phytohaemaggluti-nins from four Indian legumes. Toxicon 10: 89.

Maughan R.J. 1999. Nutritional strategies to promotemuscle hypertrophy. Basic Appl. Myol. 9: 279.

Mc Cance R.A. and Widdowson E.M. 1997. “TheComposition of Foods”, The Royal Society ofChemistry and Ministry of Agriculture, Fisher-ies and Foods, Redwood Books Ltd., Trowbridge,Wiltshire, United Kingdom.

Mitchell G.F. and Clarke A.E. 1979. Allergenicityof concanavalin A in mice. Int. Arch. AllergyAppl. Immunol. 58: 391.

Mokry B. and Engel W.K. 1975. Duchenne dystro-phy: electron microscopic findings pointing toa basic or early abnormality in the plasma mem-brane of the muscle fiber. Neurology 25: 1111.

Muraille E., Pajak B., Urbain J. and Leo O. 1999.Carbohydrate-bearing cell surface receptorsinvolved in innate immunity: interleukin-12induction by mitogenic and nonmitogeniclectins. Cell. Immunol. 191: 1.

Mussini I., Favaro G. and Carraro U. 1987. Matu-ration, dystrophic changes and the continuousproduction of fibers in skeletal muscle regener-ating in the absence of nerve. J. Neuropath. Exp.Neurol. 46: 315.

Noah N.D., Bender A.E., Reidai G.B. and GilbertR.J. 1980. Food poisoning from raw kidneybeans. Br. Med. J. 281: 236.

Pusztai A., Ewen S.W., Grant G., Brown D.S., Stew-art J.C. Peumans W.J., Van Damme E.J. andBardocz, S. 1993. Antinutritive effects of wheat-germ agglutinin and other N-acetylglucosamine-specific lectins. Br. J. Nutr. 70: 313.

Pusztai A. (Ed.). 1991. “Plant Lectins”, CambridgeUniversity, Cambridge.

Toribio M.L., de la Hera A., Pereira P. and de Lan-dazuri M.C. 1985. Modulation of cytolitic T cellfunction by lectins. J. Immunol. 134: 2179.

Van Damme E.J., Peumans W.J., Pusztai A. andBardocz S. (Eds.). 1998. “Handbook of PlantLectins: Properties and Biomedical Applica-tions”. Wiley, New York.

Vincenzi S., Zoccatelli G., Perbellini F., Rizzi C.,Chignola R., Curioni A. and Peruffo A.D.B.2002. Quantitative determination of dietarylectin activities by enzyme-linked immunosorb-ent assay using specific glycoproteins immo-bilized on microtiter plates. J. Agr. Food Chem.55: 6266.

Revised paper received February 27, 2003 Accepted June 20, 2003


SHORT COMMUNICATION

- Key words: fungicides, grapes, marc, must, wine -

GAS CHROMATOGRAPHICDETERMINATION OF AZOXYSTROBIN,

DINOCAP, FENARIMOL, PENCONAZOLEAND QUINOXYFEN DURING WINE MAKING

DETERMINAZIONE DI FUNGICIDIIN PRODOTTI VITIVINICOLI MEDIANTE HRGC-ECD

G. DI BELLA, M. SAITTA1, F. SALVO1, M. NICOTINA2 and G. DUGO1*Dipartimento Farmaco Chimico, Facoltà di Farmacia, Università di Messina,

V.le Annunziata, 98165 Messina, Italy1 Dipartimento di Chimica Organica e Biologica,

Università degli Studi di Messina,Salita Sperone 31, 98166 Messina, Italy2 Dipartimento di Entomologia e Zoologia Agraria, Facoltà di Agraria,Università degli Studi di Napoli Federico II, 80055 Portici (Na), Italy

* Corresponding author: e-mail [email protected]

ABSTRACT

A quick gas chromatographic meth-od is described to determine residuesof azoxystrobin, dinocap, fenarimol,penconazole and quinoxyfen in grapes,must, marc and wine. The samples wereextracted with hexane-ethyl acetate anddirectly analyzed using a high resolu-tion gas chromatograph fitted with anelectron-capture detection system(ECD). No clean-up was necessary be-cause there were no interferences in thechromatogram. Recoveries from un-treated samples spiked with known

RIASSUNTO

In questo lavoro viene descritto unmetodo rapido per la determinazionesimultanea di azoxystrobin, dinocap,fenarimol, penconazolo e quinoxyfen inuve, mosti, vinacce e vini medianteHRGC-ECD. I campioni, estratti conesano:acetato di etile sono analizzati di-rettamente, senza nessuna purificazio-ne. I valori di recupero, determinati atre diversi livelli di fortificazione in tri-plo, oscillano per l’azoxystrobin da80,4% (vinacce) a 100,7% (uve); per ildinocap da 81,8% (vinacce) a 110,4%


amounts of the pesticides (0.01, 0.1 and1.0 mg/kg) ranged from 80.4% (marc)to 100.7% (grapes) for azoxystrobin;from 81.8% (marc) to 110.4% (wine) fordinocap; from 82.8% (grapes) to 98.2%(wine) for fenarimol; from 80.9% (marc)to 97.3% (wine) for penconazole; from87.0% (wine) to 95.8% (wine) for qui-noxyfen. The limits of quantitation were0.010 mg/kg for dinocap and 0.005mg/kg for the other fungicides; the de-tection limits were 6x10-4 ng for dinoc-ap and 3x10-4 ng for the others.

(vini); per il fenarimol da 82,8% (uve) a98,2% (vino); per il penconazolo da80,9% (vinacce) a 97,3% (vino); per ilquinoxyfen da 87,0% (vino) ta 95,8%(vino). I limiti di quantificazione sonopari a 0,010 mg/kg per il dinocap e0,005 mg/kg per gli altri fungicidi; i li-miti di rivelabilità sono pari a 6x10-4 ngper il dinocap and 3x10-4 ng per gli altriprodotti.

INTRODUCTION

The grapevine is subject to attack bynumerous parasites. The most frequentdiseases caused by fungi are downy mil-dew, powdery mildew and gray mold. Thepesticides commonly used to control pow-dery mildew (Uncinula necator) belong tothe chemical classes of acylanilines, tria-zols and dicarboximides. Among these,dinocap, fenarimol and penconazole arewidely used. In the 1990s azoxystrobinwas used and, more recently, quinoxyfen(belonging to the family of the quinolines)is used as a fungicide.

Numerous methods aimed at isolatingand extracting fungicides in grapes, mustand wine appear in the literature, but of-ten the extraction procedure is complexand expensive. SCARPONI et al. (1992)used an opportune solvent and a separa-tory funnel to extract the wine sample;OLIVA et al. (1999) used a micro on-lineextraction method and NAVARRO et al.(1999) homogenized the wine sample withsolvents and then filtered it through celiteand a NaCl support. Because no datahave been reported on the determinationof dinocap residues in vines or wine, thisstudy was designed to standardize a quickanalytical method for their extraction inthe wine products with a single and sen-sitive procedure.


Field trials

The trial was carried out on grapesfrom three Italian vineyards: one locat-ed in Tuscany (Franci of Magliano farm),one in Campania (Vadiaperti of Monte-frondone farm) and one in Sicily (Giuffr-ida of S. Venerina farm). Wines fromTuscany were produced from 25-year-old plants, cultivated on a calcareous soil(100 m elevation); vines were grafted withthe Sangiovese variety (Morellino clone).Wines from Campania were producedfrom 25-year-old plants, cultivated on aclay soil (700 m elevation); vines weregrafted with the Fiano d’Avellino varie-ty. Sicilian wines were produced from 15to 20-year-old plants, cultivated on avolcanic soil (300 m elevation); vines weregrafted with the Inzolia and Carricantevarieties in a ratio of 1:1.

Pesticides were used at the doses rec-ommended by the manufacturer andwere sprayed with a Fox manual sprayer(Fox Motori, Poviglio – Re, Italy). Eachfungicide treatment was applied every 7-12 days during blossoming and matu-ration (from April to August). The lasttreatment was carried out, on average,28 days before the grape harvest. Thefungicide treatments were: 1,000 mg/kg


of azoxystrobin, 400 mg/kg of dinocap,300 mg/kg of fenarimol, 300 mg/kg ofpenconazole and 350 mg/kg of quinox-yfen. Only dinocap was applied 3 times,because, it is usually used to start thedefence strategy. A random-block designwas used with three replications for eachtest, and each block contained 60 plants.

Wine making process

The grapes were placed in crates, re-frigerated at 4°C and transported quick-ly to the Genna Cellar (Marsala – TP)where vinification took place.

White Vinification. Newly – croppedVitis vinifera from Sicily and Campaniawere crushed and destemmed; theywere then soft pressed using a pneu-matic press. The first must obtained wastreated with SO2 (30 g/hL), pectolyticenzymes (1.5 g/hL) and vitamin C (5 g/hL) to favour clarification before fermen-tation; the temperature was maintainedat 8°C for 24 hours. Clear must wasspiked with 20 g/hL of thiamin andammonium phosphate as fermentationcoadjuvant and 30 g/hL of selectedyeasts were added for fermentationwhich was kept at 15°C. To remove thelees, after fermentation the wine wasdecanted into a tank and spiked withSO2 (5 g/hL). After 10 days, the winewas decanted again and treated withSO2 (5 g/hL). Subsequently the winewas filtered twice through 1 µm and0.45 µm cardboard filters, spiked withSO2 (2-3 g/hL) and bottled in dark bot-tles at 4°C for the duration of the ex-periment. Each sample from Sicily andCampania were separately vinified fol-lowing the protocol mentioned above.

Red vinification. Newly – cropped Vi-tis vinifera from Tuscany were crushedand destemmed and then spiked withSO2 (5 g/hL) and selected yeasts wereadded (30 g/hL) and let to ferment for10 days at 28°C, effecting three fullingsa day. The wine was drawn from the vatand the marc was pressed with a hydrau-

lic press. After 4 weeks the lees were re-moved and the wine was spiked with SO2(2-3 g/hL). Subsequently the wine wasspiked again with SO2 (2-3 g/hL) andbottled in dark bottles and kept at 4°Cfor the duration of the experiment.

Reagents

Analytical standards of the pesticideswere purchased from Dr. Ehrenstorfer(Angsburg, Germany) (azoxystrobin, di-nocap, penconazole, and bromophos-methyl) and from Dow Agrosciences(Rotterdam, The Netherlands) (fenari-mol and quinoxyfen). Bromophos-me-thyl was used as the internal stand-ard.

N-hexane (Carlo Erba, Milano, Italy)and ethyl-acetate (Baker, Deventer, TheNetherlands) were used for pesticide res-idue determination. N-hexane was usedto solubilize the pesticide standards andthe solutions were stored at +4°C in arefrigerator.

Apparatus and chromatography

An HRGC 5300 Mega Series gas chro-matograph (Carlo Erba Instruments) wasused to determine the five fungicides. Itwas fitted with an electron capture de-tection (ECD) system and split-splitlessinjector, connected to an HP 3394 A re-porting integrator (Hewlett Packard,Avanlade, PA, USA).

An SPB-5 fused – silica column (30 mx 0.25 mm i.d.) was employed, with 5%diphenyl 95% dimethyl siloxane as liq-uid phase (film thickness 0.25 µm) (Su-pelco, Bellefonte, PA, USA). The injectorand detector operated at 250° and 280°C,respectively.

The sample (1 µL) was injected in thesplit mode (1:3), and the oven tempera-ture was programmed as follows: 200°C(1 min) raised to 270°C (5°C/min), heldfor 20 min. Helium was the carrier gasand nitrogen the make-up gas both at150 KPa.


Extraction procedure

Ten grams of the sample (grapes,must, marc and wine ) were collected ina 100 mL Erlenmeyer flask with a te-flon-lined screw-cap. Twenty millilitersof hexane-ethyl acetate (9:1, v/v) wereadded and the mixture was sonicated(Transsonic T420, Elma) for 30 min. Thesupernatant (2 mL) was evaporated un-der nitrogen flow; 0.2 mL of the internalstandard solution (bromophos-methyl1.00 mg/kg) was added, and the mix-ture was directly analyzed by gas chro-matography. No clean-up was necessarybecause there was no interference in thechromatogram.

Recovery trials

Previously, a blank assay was used tocheck for the absence of residues in thegrapes, must, marc and wine. Untreat-ed samples of grapes, must, marc andwine from the three regions were spikedwith 0.01, 0.1 and 1.0 mg/kg of the fun-gicides. The samples were allowed toequilibrate for 60 min prior to extraction,and were processed according to the pro-cedure described earlier. The recoveryassays were replicated 3 times. Since theextraction procedure of the active ingre-dients was not influenced by the placeof origin of the samples, and the amountrecovered did not depend on the culti-var, soil or climate, the recovery percent-age was calculated from the data of thethree regions.


Azoxystrobin, fenarimol, penconazole,and quinoxyfen were confirmed using theretention times of the standard solutions,ranging from 0.05 to 1 mg/kg. Dinocapwas identified using the retention timesof the six isomers.

The active ingredients were quantita-tively determined by the internal stand-

ard method; the calibration curves wereobtained by plotting peak heights ver-sus concentrations. Good linearity wasachieved in the range 0.05 to 1 mg/kg;the correlation coefficients were 0.9986for azoxystrobin, 0.9977 for dinocap,0.9980 for fenarimol, 0.9987 for penco-nazole, and 0.9980 for quinoxyfen.

The results of the recoveries reportedin Table 1 show that the method is sim-ple, fast and repeatable. Natural compo-nents of skins, grapes, etc. did not inter-fere with the analyses. The recovery val-ues ranged from 80.4% (marc) to 100.7%(grapes) for azoxystrobin; from 81.8%(marc) to 110.4% (wine) for dinocap; from82.8% (grapes) to 98.2% (wine) for fenar-imol; from 80.9% (marc) to 97.3% (wine)for penconazole; from 87.0% (wine) to95.8% (wine) for quinoxyfen. These val-ues show that the extraction method de-scribed is appropriate for the determina-tion of residues of these compounds.

The quantitation limit (QL) for azoxys-trobin, fenarimol, penconazole and qui-noxyfen was 0.005 mg/kg and 0.010mg/kg for dinocap. The determinationof the QL was based on the standarddeviation of the response and the slopeof the calibration curve (PHARMEUROPA,1999). The detection limits (setting a sig-nal to noise ratio of 2) (PHARMEUROPA,1999) were 3x10-4 ng for azoxystrobin,fenarimol, penconazole and quinoxyfenand 6x10-4 ng for dinocap. The resultsare reported in Table 2.

The azoxystrobin residues in grapesranged from 0.005 mg/kg (Campania) to0.077 mg/kg (Sicily); in marc from 0.016mg/kg (Campania) to 0.078 mg/kg (Sic-ily); in must from 0.014 mg/kg (Campa-nia) to 0.034 mg/kg (Sicily); in winesfrom 0.005 mg/kg (Campania) to 0.021mg/kg (Sicily). The residue levels in thewines were nearly 75% lower than in themarcs, and nearly 51% lower with re-spect to the must. CABRAS et al. (1998)reported that azoxystrobin residues arepreferably adsorbed on the skins.

In the grapes treated with dinocap, all


Table 2 - Fungicide residues (mg/kg ±s.d. x 10-3) in grapes, marc, must and wine. Each value is themean of three determinations.

azoxystrobin dinocap fenarimol penconazole quinoxyfen

grapes s 0.077±9 ≤0.010 0.046±6 0.065±6 ≤0.005marc s 0.078±12 0.023±3 0.282±17 0.093±9 0.030±3must s 0.034±4 ≤0.010 0.051±5 0.039±3 0.006±1wine s 0.021±6 ≤0.010 0.013±2 0.010±1 ≤0.005

grapes c 0.005±1 ≤0.010 0.011±2 0.042±4 0.007±1marc c 0.016±3 ≤0.010 0.051±6 0.104±5 0.098±8must c 0.014±2 ≤0.010 0.043±5 0.032±2 0.006±1wine c 0.005±1 ≤0.010 0.006±2 0.008±2 ≤0.005

grapes t 0.038±5 ≤0.010 0.146±2 0.030±3 ≤0.005marc t 0.076±9 0.024±3 0.293±15 0.109±7 0.028±3must t 0.024±3 ≤0.010 0.165±13 0.050±5 ≤0.005wine t 0.012±2 ≤0.010 0.070±6 0.012±1 ≤0.005

s Sicily; c Campania; t Tuscany.

Table 1 - Recovery of azoxystrobin, dinocap, fenarimol, penconazole and quinoxyfen (% ±s.d.) fromgrapes, marc, must and wine.

Fortification azoxystrobin dinocap fenarimol penconazole quinoxyfenlevel (mg/kg)

grapes 1 100.7±4.9 89.4±3.7 87.9±2.8 88.7±2.8 95.1±3.60.1 87.6±3.2 90.5±3.8 92.8±2.5 91.4±3.9 93.3±3.90.01 82.9±2.1 94.1±4.8 82.8±2.6 85.5±2.3 87.2±2.8

marc 1 80.4±2.2 89.8±3.0 91.6±3.7 88.7±3.1 87.2±2.90.1 84.4±3.6 93.8±3.9 92.6±3.7 90.9±3.9 87.8±2.30.01 87.3±2.8 81.8±2.9 92.7±4.1 80.9±2.5 93.3± 3.7

must 1 93.8±4.0 101.3±4.7 90.3±3.6 89.8±3.9 87.6±2.40.1 94.7±4.5 85.3±2.8 88.6±4.2 89.9±3.9 88.8±3.30.01 87.6±2.5 87.7±2.9 88.2±3.6 86.4±3.0 87.2±2.2

wine 1 86.6±3.5 110.4±5.0 98.2±5.4 97.3±4.5 93.4±4.30.1 92.5±3.3 96.5±4.4 88.9±3.5 93.9±3.8 87.0±3.00.01 90.2±4.1 89.5±2.6 90.6±3.7 94.7±3.9 95.8±4.7

the values were below the QL; in Sicilianmarc there were 0.023 mg/kg of dinocapand in Tuscan marc 0.024 mg/kg; in mustand wine all the values were below the QL.

Fenarimol residues in grapes rangedfrom 0.011 mg/kg (Campania) to 0.146mg/kg (Tuscany); in marc from 0.051mg/kg (Campania) to 0.293 mg/kg(Campania); in must from 0.043 mg/kg(Campania) to 0.165 mg/kg (Tuscany);in wine from 0.006 mg/kg (Campania)to 0.070 mg/kg (Tuscany). The decrease

in pesticide residues in wine and marcwas very high (almost 86%). These val-ues are comparable to results of otherstudies (NAVARRO et al., 1999).

Penconazole residues in grapes rangedfrom 0.030 mg/kg (Tuscany) to 0.065mg/kg (Sicily); in marc from 0.093 mg/kg (Sicily) to 0.109 mg/kg (Tuscany); inmust from 0.032 mg/kg (Campania) to0.050 mg/kg (Tuscany); in wines from0.008 mg/kg (Campania) to 0.012 mg/kg (Tuscany). There was a 75% decrease


in penconazole residues in wine withrespect to the must. NAVARRO et al.(1999) reported that a 85% reduction ofpenconazole residue is usually observedfrom must to wine.

Quinoxyfen residues in grapes rangedfrom 0.005 to 0.007 mg/kg (Campania);in marc from 0.028 mg/kg (Tuscany) to0.098 mg/kg (Campania); in must from0.005 to 0.006 mg/kg (Campania); allthe values were below the QL in wine.

The residues of quinoxyfen tend toaccumulate completely in the solid frac-tion, leaving no residues in the wine (CA-BRAS et al., 2000; DUGO et al., 2001).

As reported in Table 2, quinoxyfendecreased more rapidly with respect tofenarimol and penconazole. The levelsobserved after the same time in the samesamples were much lower than the fe-narimol and penconazole residues. Thesedata showed that this compound couldalso be used to produce wine free fromdetectable amount of residues.

In conclusion, the residues found inall wines were below the limits that theItalian law (D.M. 19 May 2000) has es-tablished for grapes (2 mg/kg for azoxys-trobin, 1 mg/kg for dinocap, 0.3 mg/kgfor fenarimol, 0.1 mg/kg for pencona-zole and 0.5 mg/kg for quinoxyfen) andwines (0.5 mg/kg for azoxystrobin and0.01 mg/kg for quinoxyfen).

ACKNOWLEDGEMENTS

This work was financially supported by Universitàdegli Studi of Messina in the PRA 2001 – Pro-getti ordinari Area n 03.

REFERENCES

Cabras P., Angioni A., Garau V.L., Pirisi F.M., Es-pinoza J., Mendoza A., Cabitza F., Pala M. andBrandolini V. 1998. Fate of azoxystrobin, flu-azinam, kresoxim-methyl, mepanipyrim, andtetraconazole from vine to wine. J. Agric. FoodChem. 46:3249.

Cabras P., Angioni A., Garau V.L., Pirisi F.M., Ca-bitza F., Pala M. and Farris A.G. 2000. Fate ofquinoxyfen in grapes, wine, and their process-ing products. J. Agric. Food Chem. 46:6128.

Dugo G., Visco A., Saitta M., Vinci V. and Di BellaG. 2001. Dosaggio di quinoxyfen su prodottivitivinicoli siciliani. Vignevini. 7-8:91.

Navarro S., Barba A., Oliva J., Navarro G. andPardo F. 1999. Evolution of residual levels ofsix pesticides during elaboration of red wines.Effect of wine-making procedures in their de-seapperance. J. Agric. Food Chem. 47:264.

Oliva J., Navarro S., Barba A., Navarro G. 1999.Determination of chlorpyrifos, penconazole, fe-narimol, vinclozolin and metalaxyl in grapes,must and wine by on-line microextraction andgas chromatography. J. Chromatogr. A. 833:43.

Pharmeuropa. Technical guide for the elaborationof monographs. 3Rd edition. December 1999.Published by Council of Europe, StrasbourgCedex France.

Scarponi L. and Martinetti L. 1992. Simultaneousdetermination of metalaxyl and penconazoleresidues in wine. Ital. J. Food Sci. 3:171.

Revised paper received March 21, 2003 Accepted May 6, 2003


SHORT COMMUNICATION

- Key words: beef, dry, salted, smoked -

EFFECT OF SALT DOSEON THE QUALITY OF DRY SMOKED BEEF

EFFETTO DI DIVERSE CONCENTRAZIONI DI SALESULLA QUALITÀ DELLA CARNE BOVINA ESSICCATA

T. DZUDIE*, M. BOUBA, C. M. MBOFUNG and J. SCHER1

University of Ngaoundéré, ENSAI, B.P. 455, Ngaoundéré, Cameroon1 Laboratoire de physico-chimie et Génie Alimentaire, Nancy, France

* corresponding author

ABSTRACT

Longissimus dorsi muscles from beefcarcasses were dry-salted at doses of0, 25, 50, 75 and 100 g salt/kg meat.Salted samples were kept in a cooler at2°C for 5 days after which time onetreatment group of the cured meat waspre-dried at 35°C for 2 h, smoked at88°C for 6 h and then dried at 95°C toa constant weight for about 52 h. Prox-imate analysis showed that the mois-ture and protein contents of the sam-ples decreased with increased salt con-tent of the meat, while the ash content

RIASSUNTO

Le carni di Longissimus dorsi, prove-nienti da carcasse bovine, sono statesalate a secco con dosi pari a 0, 25, 50 e100 g di sale/kg di carne. I campionisalati sono stati mantenuti a 2°C per 5giorni per poi essere pre-essiccati a 35°Cper 2 ore, affumicati a 88°C per 6 ore efinalmente essiccati a 95°C fino a pesocostante per circa 52 ore. L’analisi pros-simata ha mostrato che l’umidità ed ilcontenuto in proteine dei campioni di-minuiva con l’aumentare del contenutosalino, mentre il contenuto in ceneri


increased. The pH of the fresh meat in-creased with increasing salt. In gener-al, the addition of salt significantly(P<0.05) increased the degree of oxida-tion, while smoking produced antioxi-dant activity. Iron and phosphorus de-creased with increasing salt content.With salting, smoking and drying of themeat the microbial loads generally de-creased but the halophilic microorgan-isms increased.

aumentava. Il pH della carne fresca au-mentava all’aumentare del sale. In ge-nerale, l’aggiunta di sale aumentava si-gnificativamente (p<0.05) il grado di os-sidazione, mentre l’affumicamento favo-riva l’attività antiossidante. Il contenu-to in ferro e fosforo diminuiva all’aumen-tare del contenuto in sale. Con la sala-tura, l’affumicamento e l’essiccamentodella carne, la carica microbica totaletendeva a diminuire, permettendo, peròun aumento della concentrazione deimicrorganismi alofili.

INTRODUCTION

At high ambient temperatures, meatspoils within a few hours after the onsetof rigor mortis due to its high moistureand protein contents. In tropical Africancountries, and particularly in Cameroon,the major meat producing areas are farfrom large consumer centres. The princi-pal constraints for the distribution of meatand meat parts are the lack of distribu-tion and refrigeration facilities. In ruralareas, the conservation of meat by rela-tively unsophisticated techniques is ofparamount importance. Among the manypreservation methods, drying and smok-ing were two of the earliest and most ef-fective methods developed in rural areaswhere refrigeration was not available (MU-SONGUE et al., 1994; DZUDIE and OKU-BANJO, 1992; OCKERMAN et al., 2000).

Another method of preserving meat isby curing it with salt followed by dryingand / or smoking (ALONGE and AGBA,1981; ZAPATA et al., 1990). In temperatedeveloped countries, a wet-curing proc-ess has been developed which involvesthe use of a solution of salt, sodium ni-trate and / or nitrite. In addition, lightsmoking is often used to impart a desir-able flavour to the product. However, inorder to limit undesirable microbiologi-cal activity, a low ambient temperature

is necessary while the salt solution ispenetrating into the meat. The depend-ence on the availability of refrigerationfacilities has prevented widespread ap-plication of this technique in developingcountries. Though salted, smoked anddried beef has been accepted for manyyears in Cameroon, little is known re-garding the chemical, nutritional andmicrobial quality of such products.

With the aim of increasing and diver-sifying the supply of locally-producedand processed meat in urban areas, thepresent study was carried out to deter-mine the effects of curing, smoking anddrying on the physico-chemical and mi-crobial properties of beef.


Materials

Beef filets (musculature surroundingthe dorsal part of the spinal column,mainly longissimus dorsi) were obtainedfrom a local retail butcher. All visible fatand connective tissue were trimmed fromthe filets. Each filet was then divided intoportions of approximately 400 g each.For each replication, ten portions wereobtained and randomly assigned to twotreatment groups of five portions each.


Salting

In each treatment group, dry curingsalt was applied uniformly to the 5 por-tions at 0, 25, 50, 75 and 100 g per kgmeat, respectively, which was applied attwo-day intervals. The total curing timewas five days at 2°C. Cuts were weighedbefore and after curing to determineweight loss.

After curing, portions of meat from onetreatment group were packed in cellu-lose film and stored at 2°C for furtheranalysis. Portions from the other groupwere smoked and dried. The smoke-driedmeats were packed in cellulose film andstored at room temperature (ca 25°C)until required for analysis.

Smoking and drying

The smoking and drying processeswere carried out in a Matindex air-con-ditioned smokehouse (model 74560 La-murax, Halmaln, Germany) with a mat-index smoke generator (model 80, Lamu-rax, Halmaln, Gremany). The meat por-tions were placed on wire racks such thatno portion touched another during thesmoking process. They were pre-dried at35°C for 2 h (HR=90%), smoked at 90°Cfor 6 h (HR=80%) and dried at 95°C for52 h (HR=60%). The meat portions wereweighed at 2 h intervals during eachprocessing step to determine weight loss.During processing, shelf temperatureswere monitored using copper thermocou-ples with a digital temperature recorder(Didalab-model 918, Gesvrine, France).

Analytical techniques

All analyses were carried out on sam-ples that were ground twice through a 4mm plate (Moulinex Model 133, Alençon,France). Proximate analysis, mineral com-position, pH and thiobarbituric acid (TBA)measurements were carried out on freshand smoke-dried meat, while the micro-biological analysis were carried out on

smoke-dried samples. The moisture con-tent was determined by oven drying, theprotein content by using the Kjedahl ni-trogen estimation, the fat content by soxh-let extraction with petroleum ether and ashcontent by incineration at 525°C as de-scribed in the AOAC methods (1980). So-dium chloride as chloride was determinedby dissolving the ash in boiling water andtitrating against nitrate (AOAC, 1980).

To measure pH, 10g of ground meatsample was homogenized in 100 mL ofdistilled deionised water. The pH wasdetermined using a Minisi pH meter(Model 8000, Nantes, France). Thiobar-bituric acid (TBA) reactive substanceswere measured as described by WITTEet al.(1970). The results are expressedas mg malonaldehyde / kg of meat.

Samples for mineral analysis were pre-pared according to AOAC (1980) proce-dures, and the minerals (except phos-phorus) were assayed spectroscopicallyusing an atomic absorption spectropho-tometer (Model 1100, Perkin-Elmer, Nor-walk, USA)). Phosphorus was determinedfollowing the colorimetric molybdenum(Mo)-blue method (AOAC, 1980).


To count the viable aerobic microor-ganisms and fungal growth, 30 g ofground samples from each treatmentgroup were homogenised in a sterile labblender stomacher (Model 400, London,UK) in 90 mL of sterile peptone (0.1 g/100 mL) water. Subsequent dilutionswere also made in peptone water. Difcostandard plate count agar was used tocount the viable, aerobic microorgan-isms. These plates were incubated at32°C for 48 h. To enumerate fungalgrowth, 10% tartaric acid was used. Theplates were incubated at 25°C for 120 h.Sulfite reducing bacteria were deter-mined by transferring 10 mL of initialhomogenate to a 20 mL test tube, whichwas heated to 80°C for 10 min beforeplating anaerobically on trypticase sulfite


(TSN) agar containing neomycin and in-cubated at 37°C for 24 h.

For Salmonella, 25 g of minced sam-ple were blended with 75 mL of sterilepeptone water in a stomacher bag. A 5mL aliquot of the homogenate was incu-bated in 10 mL selenite cystine broth at37°C for 18 h, after which 1 mL wasspread on the surface of Salmonella agarwhich was incubated at 37°C for 24 h.Colonies were counted after incubation.


The experiments were performed in trip-licate and the means ± standard devia-tion of three values are reported. Data weresubjected to one way analysis of variance,and Ducan’s multiple range test (STEELand TORRIE, 1980) was used to determinethe significant differences among means.


Fig. 1 shows weight changes duringpre-drying, smoking and drying of sam-ples. During the first 12 h of processing,all the samples dried at the same rate,after which, the unsalted meat dried fast-er than the salted meat. As the dose ofsalt in the meat increased, the salt servedas a great obstacle to the diffusion ofwater from within the muscle to the sur-face of the meat by acting as a barrieragainst the establishment of an appro-priate osmotic gradient. At the end of thedrying period, the unsalted meat had lostmore weight than the salted cuts. DZUD-IE and OKUBANJO (1992) reported thatat the end of the ageing period (27°C for4 weeks), cured rabbit cuts lost lessweight than uncured ones. However,KEMP et al. (1986) observed that injectedhams absorbed salt faster but lost moreweight than non-injected hams. The dis-crepancy in the published data could beattributed to the differences in the cur-ing methods and drying temperatures.

The means for the proximate analysis

are shown in Table 1. The moisture con-tent of the fresh meat significantly(P<0.05) decreased as the salt concen-tration in the meat increased. The high-est percent weight loss recorded for thesalted samples over the unsalted onesduring curing could account for the ob-served results. During smoking and dry-ing of the meat the salt formed a driedlayer on the surface of the meat, reduc-ing the rate of diffusion of water fromwithin the muscle cells to the surface.Consequently, the moisture content ofthe dried meat increased with theamount of salt applied to the meat.

A slight decrease in the protein con-tent of the meat was observed as the saltdose in the meat increased. During os-motic withdrawal of water from the inte-rior of the meat cuts, part of the salt-soluble protein is lost as drip.

The lipid content of the meat was notaffected by the salt treatments. Increasein the lipid content of the dried meat overthe fresh meat was due to the dehydra-tion effect of smoking and drying. Theash content in the two types of meat sig-nificantly (p<0.05) increased as theamount of salt applied to the surfaceincreased. This was expected and can beattributed to the increasing salt concen-trations in the different samples. OWENet al. (1986) found that the mean ashcontent of processed cured chicken wassignificantly higher than that of freshuncured chicken.

The pH of the fresh meat significantly(p<0.05) increased as the amount of saltadded to the meat increased (Fig. 2). Thesame trend was observed with thesmoke-dried samples, except at 50 g add-ed salt/kg meat, the pH of the samplesstarted to decrease. The increase in thepH of the fresh and salted meat can beattributed to the effect of salting on mus-cle proteins. BABJI et al. (1982) found thatthe pH of salted turkey breast muscleswas higher than that of unsalted sam-ples. In the present study, the variationobserved in the pH changes of the sam-


Fig. 1 - Weight changes during pre-drying, smoking and curing of meat with different concentrationsof salt.

Table 1 - Proximate composition* of fresh and smoke-dried meat subjected to salt treatments.

Salt treatment (g/kg)

Meat 0 25 50 75 100

Moisture Fresh 76.95±0.57a 74.24±0.47b 73.63±0.6±1b 71.68±0.50c 69.68±0.93d

Smoke-dried 7.61±0.25d 12.85±0.15c 12.88±0.14c 13.80±0.22b 15.45±0.39a

Proteins Fresh 19.10±0.95a 18.372±2.21b 18.30±0.89b 17.52±0.43c 16.62±0.45c

Smoke-dried 65.27±1.20a 56.12±0.10b 54.11±0.10b 52.45±1.46b 45.26±0.24c

Lipids Fresh 4.02±0.09a 4.26±0.99a 4.43±0.79a 4.50±0.17a 5.09±0.26a

Smoke-dried 7.73±0.63a 8.19±0.39a 7.83±0.67b 8.50±0.05b 8.63±0.14c

Ash Fresh 1.14±0.01d 3.12±0.06c 5.27±0.42b 5.45±0.03b 7.90±0.01a

Smoke-dried 7.80±0.17d 12.37±0.22c 18.24±1.32b 20.99±0.46b 22.43±0.93a

* values expressed as % standard deviation;abcd values in the same row with different superscripts are significantly different (p<0.05).

ples during smoking and drying of themeat can be attributed to the change inthe ionic strength surrounding the mus-

cle proteins. DZUDIE and OKUBANJO(1992) reported that the pH of smoke-dried rabbit meat was related to the salt


Fig. 2 - Effect of salt doses on the pH of fresh and smoke-dried meat.

content of the meat. In the present study,the pH values of the smoke-dried meatwere higher than those of the fresh andsalted samples. In contrast, VISHWANATHet al. (1998) found that the pH of smokedmud eel fish was more acidic than that ofthe fresh fish. The discrepancies betweenthe results of these tests could be due todifferences in processing methods be-cause the above-mentioned investigatorsdid not treat their samples with salt as inthe present study.

The stability of lipids in the fresh andsmoke-dried meat cured at different saltlevels was evaluated by the TBA proce-dure (Fig. 3). Overall, the results indi-cated that each salt level had a pro-oxi-dant effect on the lipids in the fresh sam-ples. These results confirm those ofWHEELER et al. (1990) who found thatTBA values in restructured beef steakswere significantly (P <0.05) higher in salt-ed samples than in unsalted ones.

In a second study, concerned with the

effect of smoke and the level of salt onthe stability of lipids in smoke-dried beef,the antioxidant activity of the smoke wasevident. The TBA values of the smoke-dried meat were significantly lower (P<0.05) than those of the non-smokedcounterparts. The preservative effect ofwood smoke on food is the result of par-tial surface dehydration and depositionof anti-oxidant compounds from thesmoke on the surface on the meat. Asobserved in the present study, there wasfirst an apparent decrease followed by anincrease in the TBA values of the smoke-dried meat as the level salt added to themeat increased. On the other hand, CUP-PETT et al. (1989) found that adding saltsignificantly (P<0.05) increased the TBAvalues of smoked great lakes white fish.Obviously TBA values are not good indi-ces of the quality for these dried meats.ZAPATA et al. (1990) reported thatmalonaldehyde is not an end product ofthe oxidative changes in meat products.


Fig. 3 - Effect of salt doses on TBA of fresh and smoke-dried meat.

Table 2 - Mineral content and microbiological count* of fresh and smoke-dried meat.

Salt treatment (g/kg)

Meat 0 25 50 75 100

NaCl (%) Fresh ND 4.01±0.18d 5.47±0.17c 5.77±0.19b 8.80±0.12a

Smoke-dried ND 5.69±0.16d 11.82±0.57c 13.03±0.61b 14.600.14a

Calcium Fresh 17.06±2.14b 18.0±41.87c 17.54±1.97c 20.48±1.12b 26.821.79a

(mg/100g) Smoke-dried 29.63±1.17d 33.9±11.82c 51.31±4.30b 78.92±2.01a 78.581.86a

Phosphorus Fresh 446.57±119.78a 253.22±74.29b 236.60±16.11b 229.00±38.16b 180.71±38.81c

mg/100g) Smoke-dried 1114.81±173.11a 748.98±62.96b 691.06±82.22b 484.14±62.33c 431.89±37.12c

Magnesium Fresh 22.28±11.56c 59.75±18.1b 49.03±12.51b 67.73±18.75a 73.28±11.73a

(mg/100g) Smoke-dried 292.66±20.69c 314.04±13.12c 324.24±16.74d 341.54±12.16a 348.64±10.77a

Iron Fresh 0.94±0.15 0.56±0.10 0.21±0.17 0.25±0.16 0.28±0.10(mg/100g) Smoke-dried 2.96±0.10a 1.97±0.31a 0.55±0.16c 0.55±0.18c 0.56±0.11c

Aerobicmesophilic(CFU/g) Smoke-dried 1.03.103 2.33.103 1.80.103 1.43.103 1.05.103

Fungalcounts(CFU/g) Smoke-dried 5.1.103 6.2.103 7.0.103 9.8.103 9.0.103

* values standard deviation; abcd values in the same row with different superscripts are significantly different (p<0.05);ND: Not determined.


Regardless of the dose of salt added tothe meat, the mineral content of thesmoked-dried meats was higher than thatof the fresh-salted samples (Table 2).MBOFUNG (1993) found that the mineralcontent of Kilishi was higher than that offresh beef. Sodium chloride, calcium andmagnesium values of fresh, as well assmoked-dried meat, increased with thedose of salt added. This increase is re-flected in the increased ash content. Ageneral decrease in phosphorus and ironvalues could be attributed to the loss ofthese minerals through drip during drycuring. Because this product is a typicalfood in very restricted areas, no specificdata have been found in the literatureabout its mineral composition. Table 2shows that aerobic plate counts de-creased as the dose of salt added to themeat increased. These results were ex-pected due to the effect of salt on micro-bial growth. An inverse trend was ob-served with fungal counts. OTENG-GYANG(1984) reported that halophilic microor-ganisms constituted the principal micro-organisms of salted and smoked meatproducts. Salmonella and sulfite-reduc-ing bacteria were not present in any ofthe samples analysed

Data from this study show that thegreatest weight loss was recorded for theunsalted meat. Salting of the meat bothincreased the pH and TBA values of themeat. The smoke-dried meat underwentless oxidation than the fresh samples.The high salt content of the meat shouldnot be a problem because it could beused for stewing or cooking without us-ing any additional salt or after desaltingit with water.

REFERENCES

Alonge D.O. and Agba A. H. 1981. Traditionalmethods of meat preservation and prepara-

tion in Nigeria. West African Farming March/April.

AOAC. 1980. “Official methods of analysis”, 13th

ed., Association of Official Analytical ChemistsWashington, D.C.

Babji A.S., Froning B.W. and Nogka D.A. 1982.The effect of short term tumbling and saltingon the quality of turkey breast muscle. Poultry.Sci. 61: 300.

Cuppett S.L., Gray J.I., Booren A.M., Price J.F. andStachiw M.A. 1989. Effect of processing varia-bles on lipid stability in smoked great lakeswhitefish. J. Food Sci. 54: 52.

Dzudie T. and Okubanjo A. 1992. Effects of twoskinning and curing methods on the quality ofrabbit meat. J. Food Sci. Technol. 29: 162.

Kemp J.D., FOX J.D. and Lauglosis B.E. 1986.Characteristics of dry cured hams as affectedby needle tenderization and vacuum packag-ing. J. Food. Sci. 51: 9.

Mbofung C.M.F. 1993. Effect of a traditional afri-can method of meat processing on the availa-bility of iron and other minerals from the fin-ished product (Kilishi) following in vitro enzy-molysis. Bioavailability 93: 169.

Musongue P., Eric N. and Njolai E.N. 1994. Dryingand infusion during the traditional processingof Kilishi. J. Food Engin. 23 : 159.

Ockerman H.W., Li C.T. and Marriott N.G. 2000.Sensory and microbial attributes of a dehydrat-ed pork product (shredded pork). J. MuscleFoods. 11:143.

Oteng-Gyang K. 1984. “Introduction à la Micro-biologie Alimentaire dans les Pays Chauds”. Tec.et Doc. Lavoisier, Paris.

Owen J.E., Cutierrez F.J. and Carbajal E.P. 1986.The preparation of smoked cooked chicken byconventional and accelerated curing methods.Poultry Sci. 65: 314.

Steel R.G.D. and Torrie J.N. 1980. “Principles andProcedures of Statistics”, 2nd Edition. McGraw-Hill, New York.

Vishwanath W., Lilabati H. and Bijen M. 1998.Biochemical, nutritional and microbiologicalquality of fresh and smoked mud eel fish. FoodChem. 61: 153.

Wheeler T.L., Seideman S.C., Davis G.W. and Ro-lan T.L. 1990. Effect of chloride salts and anti-oxidants on sensory and storage traits of re-structured beef steaks. J. Food Sci. 55: 1274.

Witte V.C., Kause G.F. and Bailey M.E. 1970. Anew extraction method for determining 2-thio-barbituric acid value of pork and beef duringstorage. J. Food Sci. 325: 582.

Zapata J.F.F., Ledward D.A. and Lawrie R. A. 1990.Preparation and storage stability of dried salt-ed mutton. Meat Sci. 27: 109.

Revised paper received January 13, 2003 Accepted June 13, 2003


SHORT COMMUNICATION

- Key words: antimicrobial effect, hypochlorite, Listeria monocytogenes -

INFLUENCEOF HYPOCHLORITE CONCENTRATION,

pH AND TEMPERATURE ON INACTIVATIONOF LISTERIA MONOCYTOGENES

INFLUENZA DELLA CONCENTRAZIONE DI IPOCLORITO, DEL pH E DELLATEMPERATURA SULL’INATTIVAZIONE DI LISTERIA MONOCYTOGENES

O. ERKMENDepartment of Food Engineering, Faculty of Engineering,

University of Gaziantep, 27310 Gaziantep, Turkey*Corresponding author: Fax +342 3601105; E-mail: [email protected]

ABSTRACT

The effect of hypochlorite on Listeriamonocytogenes at varying concentra-tions, pH, temperature, and type of so-lution (with or without peptone in phos-phate buffer) was studied. L. monocy-togenes was reduced by about 3.55,4.17, 4.56 and 5.59 log units, respec-tively, after 30 sec of exposure to 5, 10,50 and 100 mg L-1 hypochlorite in phos-phate buffer (pH 6.5) at 30°C. There wasan increase in the rate of disinfection(p<0.05) with an increase in tempera-ture from 20° to 40°C. Hypochlorite had

RIASSUNTO

È stato studiato l’effetto di diverseconcentrazioni di ipoclorito sull’inatti-vazione di Listeria monocytogenes alvariare del pH, della temperatura e didiversi tipi di soluzione (con e senzapeptone in tampone fosfato). La concen-trazione iniziale di L. monocytogenes siè ridotta di 3,55, 4,17, 4,56 e 5,59 uni-tà logaritmiche dopo esposizione per 30sec a 5, 10, 50 e 100 mg L-1 di ipoclori-to in tampone fosfato (pH 6,5) a 30°C.Si è notato un incremento del grado diinattivazione (p<0,05) con incrementi di


a greater antimicrobial effect on L.monocytogenes at pH 4.5 than at 5.5,6.5 and 8.5. Addition of 0.1% peptoneprovided a marked neutralization of thechlorine in the phosphate buffer. Sur-vival curves showed two distinct phas-es in the destruction of L. monocy-togenes. At all hypochlorite concentra-tions, the first phase (about 30 sec) wascharacterized by a higher rate of inac-tivation, shown by a steeper inactiva-tion slope, which was followed a moregradual slope.

temperatura tra 20° e 40°C. È stato os-servato un maggior effetto antimicro-bico dell’ipoclorito su L. monocytogenesa pH 4,5 rispetto a pH di 5,5, 6,5 e 8,5.L’aggiunta di peptone allo 0,1% ha por-tato ad una marcata neutralizzazionedell’ipoclorito nel tampone fosfato. Lecurve di sopravvivenza hanno mostra-to due distinte fasi di inattivazione diL. monocytogenes. A tutte le concentra-zioni di ipoclorito, la prima fase (di ca.30 sec) è caratterizzata da un maggiorgrado di inattivazione, evidenziato dauna maggior pendenza della curva diinattivazione, seguita poi da una pen-denza più dolce nella seconda fase.

INTRODUCTION

Food, such as milk, milk products,meat, frankfurters, poultry and seafood,are usually preserved by refrigeration.Refrigeration temperatures, however, of-fer an appropriate environment for mul-tiplication of Listeria monocytogenes(SWAMINATHAN, 2001). L. monocytogenesis a small Gram-positive bacteriumwhich can be found in food equipment,soil and water (KIM et al., 2000; SWAMI-NATHAN, 2001). This organism has beenimplicated in several outbreaks of foodand water-borne diseases (ERKMEN,2000; KIM et al., 2000; RAY, 1996; SWA-MINATHAN, 2001). Listeriosis can betransmitted through consumption of Lis-teria contaminated foods or from a widerange of infected animals (ADAMS andMOSS, 1995; RAY, 1996). L. monocytogenesis ubiquitous, moderately heat resistantand psychrotrophic and can grow un-der refrigeration in the absence of oxy-gen (ADAMS and MOSS, 1995; FOEGED-ING and LEASOR, 1990; RAY, 1996; SWA-MINATHAN, 2001).

Sanitizing practices are necessary toimprove the microbiological safety of foodproducts (FAVIER et al., 2001; RAY, 1996).

Chlorine is widely used as a disinfectantin the food industry and for the treat-ment of drinking water in communitydistribution systems and reservoirs(ODLAUG, 1981; HAN et al., 2002). Chlo-rine compounds are effective againstbacteria, yeasts, molds, spores and vi-ruses and require short to moderate con-tact time. Chlorine is readily availablein several forms (liquid chlorine, hy-pochlorites, inorganic and organic chlo-ramines, and chlorine dioxide).

Unlike some of the other disinfectionmethods, several simple test kits existto measure the availability of chlorine(residual chlorine). The antimicrobial(germicidal) action of the chlorine com-pounds is due to the oxidizing effect ofchlorine on the –SH group in many en-zymes and structural proteins. In addi-tion, damage to membranes, destructionof protein synthesis, reactions with nu-cleic acids, and interference with metab-olism have been suggested (RAY, 1996).Its efficacy against a broad spectrum ofmicroorganisms is well documented(TRUEMAN, 1971; ODLAUG, 1981).

The overall objective of this study wasto evaluate the behavior of L. monocy-togenes against different hypochlorite


concentrations at different pHs, temper-atures and types of solutions.


Bacterial strain

L. monocytogenes type 4a KUEN 136was obtained from the MicroorganismsCollection Center, Faculty of Medicine,University of Istanbul (Istanbul, Turkey).The culture was maintained on brainheart infusion agar (BHIA; Difco, Detroit)slants and stored at 4°C. The cultureswere subcultured twice by inoculatingthem into 10 mL of brain heart infusionbroth (BHIB; Difco, Detroit) and incubat-ing at 35°C for 18 h in a bench-type wa-ter bath shaker ST-402 (NÜVE; SanayiMalzemeleri Imalat ve Ticaret A.S., Istan-bul Turkey) with shaking (120 strokes permin). After centrifugation supernatantfluids were discarded and cell pellets werewashed twice with a 0.1 M phosphatebuffer (pH 6.5) solution. They were thensuspended in phosphate buffer solutionto yield about 2.2x108 colony forming unit(cfu) mL-1 and used in the experiments.Cell suspensions were prepared immedi-ately before each treatment.

Preparation of test solutions

All glassware was thoroughly cleanedand rinsed with water free of residual chlo-rine before use. A stock hypochlorite so-lution was prepared from householdbleach (sodium hypochlorite with 92.66 gL-1 of available chlorine) with phosphatebuffer to provide 5, 10, 50, and 100 mg L-1

hypochlorite concentrations when mixedwith the inoculum. A sodium thiosulphatesolution (0.1 mol L-1) was used to neutral-ize the antimicrobial activity of hypochlo-rite. All solutions were prepared fresh.

Available chlorine concentration inthe original and prepared solutions wasdetermined by the standard idiometrictitration method (ANONYMOUS, 1985).

Procedure

The system was designed to determinerates of bacterial inactivation but notintended to simulate natural environ-ments. Thus, hypochlorite concentra-tions employed here may not be directlycomparable with those reported else-where.

Hypochlorite solutions (5, 10, 50,and 100 mg L-1) were prepared fromhousehold bleach and phosphate bufferat pH 4,5, 5,5, 6,5 and 8,5; in sterile100 mL Erlenmeyer flasks, 49 mL ofeach was mixed with 1 mL of bacterialsuspension (final concentration about2.2x108 cfu mL-1) and placed in a waterbath at 20, 30 and 40°C. Ten mg L-1 ofhypochlorite solution in phosphatebuffer containing 0.1% peptone wasalso prepared at pH 4.5. The pH of thephosphate buffer was prepared by mix-ing 0.1 M K2HPO4 and 0.1 M KH2PO4solutions.

After exposure of L. monocytogenes,1 mL of 0.1 mol L-1 of sodium thiosul-fate was added to neutralize the hy-pochlorite in solution. Each sample fromthe initial inoculum and after hypochlo-rite treatment was serially diluted with0.1% peptone (Difco, Detroit) solution.Initial counts and the surviving L. mono-cytogenes were determined by spreadplating of 0.5 mL of diluted (or non di-luted) samples on duplicate plates ofBHIA. The plates were incubated at35°C for 48 h, after which all the char-acteristic visible colonies on BHIA werecounted (ERKMEN, 2000). Each experi-ment was carried out in duplicate andrepeated two times on separate days.The average number of colonies from theduplicate plates was then recorded foreach sample.

Analysis of variance was performed ondata obtained at different rates of L.monocytogenes inactivation of hypochlo-rite treatment using the computer pro-gram, Statgraphics 2.0 (Stsc., Inc., Rock-ville, MD).



Effect of hypochlorite concentrations

The antimicrobial effect of varioushypochlorite concentrations on L.monocytogenes in phosphate buffer at pH6.5 and at 30°C is shown in Fig. 1. After30 sec exposure to 5, 10, 50, and 100 mgL-1 hypochlorite, initially inoculated Lmonocytogenes (about 8.34 log units)dropped by 3.35, 4.17, 4.56 and 5.59 logunits, respectively. The sterilization ef-fects (i.e., no survival cells in plate of 1mL of treated sample) of 5, 10, 50 and100 mg L-1 hypochlorite at pH 6.5 at 30°Con L. monocytogenes were observed af-ter 210, 150, 120 and 100 sec, respec-tively. Inactivation increased with in-creasing hypohlorite concentration from5 to 100 mg L-1. The antimicrobial effectof hypochlorides has been previouslyobserved on other microorganisms;about 8 mg L-1 chlorine was effective inkilling viable Vibrio cholerae in brothculture (SOUSA et al., 2001). KIM et al.(2000) reported that application of 13 mgL-1 of chlorine water (pH 3.90) achievedmore than 7.0 log reductions on L. mono-cytogenes in 30 sec and complete elimi-nation in one min. ZHAO et al. (2001)reported that chlorine water is even moreeffective (0.25 mg L-1 of free chlorine) andachieved more than 7.0 log reductionson E. coli 0157:H7 in one min.

Effect of pH

The effect of 10 mg L-1 hypochlorite inphosphate buffer at various pHs at 30°Cis shown in Fig. 2. Decreasing the pH ofhypochlorite from 8.5 to 4.5 increasedthe disinfection of L. monocytogenes dueto the synergistic effect of chlorine andthe acidic environment. Hypochloriteshowed maximum activity in acid solu-tions. All initially inoculated (about 8.34log units) L. monocytogenes were com-pletely inactivated after 80 sec exposureto 10 mg L-1 hypochlorite at pH 4.5, while

it was reduced about 3.59, 5.94 and 7.59log units were at pH 8.5, 6.5 and 4.5,respectively. The least antimicrobial ef-fect of hypochlorite was observed at al-kaline pH (8.5). BLOOMFIELD and MILES(1979) suggested that the bactericidalaction of chlorine was produced by hy-pochlorous acid (HOCl). It probably en-ters the cell and reacts with the –SHgroup of proteins. HOCl is stable at acidpH and thus more effective; at alkalinepH, it dissociates to H+ and OCl- (hy-pochlorite ions), which reduces its bac-tericidal effectiveness. It is also lesseffective on microorganisms in the pres-ence of organic matter. There were sig-nificant (P<0.05) differences among thevarious pH treatments in the number ofL. monocytogenes. This suggests the im-portance of the H+ ion concentration be-sides HOCl on the survival of L. monocy-togenes.

Effect of temperature

Effects of 10 mg L-1 hypochlorite inphosphate buffer (pH 6.5) at 20, 30 and40°C is shown in Fig. 3. Reduction inthe number of L. monocytogenes after 30sec was about 3.81, 4.17 and 5.32 logunits at 20°, 30° and 40°C, respectively.The hypochlorite was found to be moreeffective (P<0.05) at high temperature.The inactivation of L. monocytogenes wassignificantly different (P<0.05) at the var-ious temperatures, indicating that thesensitivity of L. monocytogenes to tem-perature changes was not the same.

Effect of composition of solution

The effect of 10 mg L-1 hypochlorite inphosphate buffer (pH 6.5) containing0.1% peptone at 30°C is shown in Fig. 4.A relatively low peptone concentration(0.1%) caused a large decrease in bacte-ricidal activity of hypochlorite against L.monocytogenes. All of inoculated L. mono-cytogenes (8.34 log units) were inactivat-ed after 35 min with 10 mg L-1 hypochlo-


Fig. 1 - Effect of 5, 10, 50, and 100 mg L-1 hypochlorite on L. monocytogenes in phosphate buffer at pH6.5 at 30°C.

Fig. 2 - Effect of 10 mg L-1 hypochlorite at pH 4.5, 5.5, 6.5, and 8.5 on L. monocytogenes at 30°C.


Fig. 3 - Effect of 10 mg L-1 hypochlorite at 20°, 30° and 40°C on L. monocytogenes in phosphate bufferat pH 6.5.

Fig. 4 - Effect of 5, 10, 50, and 100 mg L-1 hypochlorite in phosphate buffer containing 0.1% peptone onL. monocytogenes at pH 6.5 at 30°C.


rite in phosphate buffer containing 0.1%peptone at pH 4.5 at 30°C, while L. mono-cytogenes was completely inactivated af-ter 80 sec in chlorinated phosphate buff-er without peptone (Fig. 2). This suggeststhat the composition of the solution pro-tected L. monocytogenes in the treatmentwith hypochlorite. Chlorination and oxi-dation reactions between the aminogroups and chlorine are most likely re-sponsible for this antimicrobial effect(RAY, 1996). Peptone containing nitroge-nous compounds such as proteins shouldbe thoroughly removed before using hy-pochlorite as a disinfectant (EL-KEST andMARTH, 1988; PAZ et al., 1993). Washingsalad leaves with 0.01% hypochlorite de-creased mesophilic aerobic bacteria by2.5 logs after 5 min of contact (PIERNASand GUIRAUD, 1997). Yersinia enterocol-itica inoculated onto eggshell surface, wasexposed to 100 mg L-1 hypochlorite, whichresulted in a reduction of about 1.28 logunits (FAVIER et al., 2001).

Survival curves (Figs. 1-4) revealed twodifferent rates of L. monocytogenes in-activation for each concentration of hy-pochlorite. This suggests a complexmechanism in the inactivation of bacte-ria with hypochlorite. At any hypochlo-rite concentration, the earlier phase(about 30 sec) was characterized by ahigher inactivation, which continuedslowly thereafter with a straight-lineinactivation. This suggests that the dif-fusivity of hypochlorite into cells occursin a very short time (within 30 sec) andcells could then be stressed or sensitizedand inactivated by hypochlorite. TRUE-MAN (1971) indicated that the hypochlo-rite antimicrobial activity correlates withthe concentration of undissociated hy-pochlorite (HOCl) molecules. An in-creased microbicidal HOCl effect may bedue to its diffusion through microbialwalls (PAZ et al., 1993) followed by oxi-dation of cellular components (PIERNASand GUIRAUD, 1997).

Hypohlorite concentration, pH, tem-perature, physical and chemical prop-

erties of suspending solution and expo-sure time were effective factors in theinactivation of L. monocytogenes. Therewere significant differences (P<0.05)among inactivation rates of L. monocy-togenes with respect to hypochlorite con-centrations at constant temperature. Thehigher hypochlorite concentrations andlonger treatment times were more effec-tive in reducing the bacterial numbers.Particulate matter associated with L.monocytogenes can decrease the bacte-ricidal activity of hypochlorite. Theseconsiderations are particularly impor-tant in determining the efficiency of chlo-rination used for disinfection of equip-ment and water. Food is usually not con-taminated with large numbers of L.monocytogenes (8.34 log units). There-fore hypochlorite can be considered asan effective antimicrobial agent in thedisinfection of L. monocytogenes in termsof food production equipment. This studyindicates that hypochlorite appears toadequately control this pathogen.

REFERENCES

Adams M.R. and Moss M.O. 1995. “Food Microbi-ology”. The Royal Society of Chemistry, p. 192-203, Cambridge.

Anonymous. 1985. “Standard Methods for theExamination of Water and Waste Water”. Amer-ican Public Health Association, American Wa-ter Works Association and Water Pollution Con-trol Federation, 16th edn., American PublicHealth Association, Washington.

Bloomfield S.F. and Miles G.A. 1979. The antibac-terial properties of sodium dichloroisocyanurateand sodium hypochlorite formulations. J. App.Bacteriol. 46: 65.

El-Kest S.E. and Marth E.H. 1988. Inactivation ofListeria monocytogenes by chlorine. J. Food Pro-tec. 51: 520.

Erkmen O. 2000. “Basic Methods for the Micro-biological Analysis of Foods”. p. 37. Universityof Gaziantep Press, Gaziantep, Turkey.

Favier G.I. Escudero M.E. and de Guzman A.M.S.2001. Effect of chlorine, sodium chloride, triso-dium phosphate and ultraviolet radiation on thereduction of Yersinia enterocolitica and mes-ophilic aerobic bacteria from eggshell surface.J. Food Protec. 64: 1621.


Foegeding P.M. and Leasor S.B. 1990. Heat resist-ance and growth of Listeria monocytogenes inliquid whole egg. J. Food Protec. 53: 9.

Han Y., Linton R.H., Nielsen S.S. and Nelson P.E.2002. A comparison of methods for recovery ofchlorine dioxide-injured Escherichia coliO157:H7 and Listeria monocytogenes. FoodMicrobiol. 19: 201.

Kim C., Hung Y.-C. and Brackett R.E. 2000. Effi-ciency of electrolyzed oxidizing (EO) and chem-ically modified water on different types of food-borne pathogens. Intern. J. Food Microbiol. 61:199.

Odlaug T.E. 1981. Antimicrobial activity of halo-gens. J. Food Prot. 44: 608.

Paz M.L., Duaigues M.V.G., Hanashiro A., D’AquinoM. and Santini P. 1993. Antimicrobial effect ofchlorine on Yersinia enterocolitica. J. App. Bac-teriol. 75: 220.

Piernas V. and Guiraud J.P. 1997. Disinfection ofrice seeds prior to sprouting. J. Food Sci. 62:611.

Ray B. 1996. “Fundamental Food Microbiology”.p. 296, CRC Press, Inc., Florida.

Sousa O.V., Vieira R.H.S.F., Patel T.R., Hofer E.and Mesquita V. P. 2001. Effects of chlorine oncells of Vibrio cholerae. Food Microbiol. 18: 355.

Swaminathan B. 2001. Listeria monocytogenes. In“Food Microbiology: Fundamentals and Fron-tiers”. Doyle M.P., Beuchat L.R. and MontvilleT.J. 2nd edn., ASM Press, Washington.

Trueman J.R. 1971. The halogens. In “Inhibitionand Destruction of the Microbial Cell”. Hugo,W.B. Ed. p. 137, Academic Press, London.

Zhao T., Doyle M.P., Zhao D., Blake P. and WuF.-M. 2001. Chlorine inactivation of Es-cherichia coli O157:H7 in water. J. Food Pro-tec. 64:1607.

Paper received February 5, 2003 Accepted May 6, 2003


SURVEY

- Key words: melons, pesticide monitoring, residues -

PESTICIDE RESIDUES IN EPRW´98 MELONS

RESIDUI DEI PESTICIDI NEI MELONI DELL´EPRW´98

A. VALVERDE GARCIA*, A. AGUILERA DEL REAL,M. RODRíGUEZ SÁNCHEZ and F. CAMACHO FERRE

Pesticide Residue Research Group. Facultad de Ciencias ExperimentalesUniversidad de Almería, 04071 Almería, Spain

Tel. +34 950 015309; Fax +34 950 015008; E-mail: [email protected]* Corresponding author

ABSTRACT

The objective of this study was tocompare the results of pesticide resi-due analysis carried out by differentofficial laboratories in Europe, partic-ularly for recently introduced pesti-cides. At the closing session of EPRW´98(2nd European Pesticide Residue Work-shop), 200 delegates from 35 countrieseach received two melons from a ho-mogenous lot. Delegates working inpesticide residue control laboratorieswere invited to analyze the melons andsend the results to the EPRW´98 Sec-

RIASSUNTO

L’esperimento è stato progettato pervalutare quali fossero i risultati di la-boratori ufficiali di diverse nazioni pervalutare residui di pesticidi, soprattut-to per quelli di recente introduzione.Nell’ultima sessione dell’EPRW’98 (2nd

European Pesticide Residue Workshop),200 delegati di 35 Paesi hanno ricevu-to una scatola contenente due meloni.I delegati dei laboratori di controllo deiresidui di pesticidi hanno spedito i ri-sultati al Segretariato dell’EPRW’98. Suimeloni sono stati effettuati trattamenti


retariat. Melons had been treated with16 pesticides, including some relative-ly new active ingredients. No residuesof the pesticides used were determinedin the EPRW´98 melons by the sevenparticipating laboratories. Surprising-ly, low levels of two pesticides not usedduring the production of the melons(endosulfan and procymidone) werefound by a number of laboratories.Some hypothesis about the causes ofthese findings are proposed.

con 16 pesticidi, alcuni dei quali “rela-tivamente nuovi”. Nessun residuo de-gli insetticidi usati è stato trovato neimeloni dai sette laboratori che parteci-pavano all’esperimento. Sorprendente-mente, bassi livelli di due insetticidi,non usati durante la produzione deimeloni (endosulfan e procimidone),sono stati trovati da alcuni laboratori.Su questo aspetto sono state avanzatealcune ipotesi.

INTRODUCTION

One of the major tasks of the foodcontrol agencies in most of the devel-oped countries is monitoring for pesti-cide residues in foods. The main ob-jective of the pesticide residue moni-toring programs is to enforce compli-ance with the legislated national (or in-ternational) maximun residue limits(MRLs) (ANDERSEN and POULSEN,2001; EU COMMISSION, 2001; MAPA,2001). Monitoring data are also usedto assess the dietary exposure of pes-ticide residues from food (ANDERSSONet al., 2001; CAMONI et al., 2001; MAFF/PSD/HSE, 2000; WHO, 1997). Thesemonitoring programmes are usuallyexecuted on an annual basis and en-compass the sampling and analysis ofa pre-determined number of samplesof different foods, mainly fruits andvegetables. In general, samples includ-ed in the monitoring programs are “sur-veillance” samples (there is no priorknowledge about the possible presenceof pesticide residues). This is the rea-son for which the analytical methodsroutinely used in almost all the foodcontrol laboratories to determine pes-ticide residues in foods are multiresi-due methods (CAIRNS and SHERMA,1992), i.e., methods capable of detect-

ing a large number of pesticides in oneprocedure.

According to the laboratory manualspublished by different agencies (MPHWS,1996; NFA, 1998; FDA, 1994) and recentreviews on this general area of analysis(SCHENZLER and THIER, 2001; LEHOTAY,1998), most pesticide residue analysesare performed using multiresidue meth-ods involving ethyl acetate, acetone oracetonitrile extraction; liquid-liquid par-titioning, gel permeation chromatogra-phy (GPC) or solid phase extraction (SPE)clean-up; and gas chromatography (GC)or liquid chromatography (LC) determi-nation. Without doubt, capillary GC withdifferent detection systems, includingmass spectrometry (MS), is by far thepredominant method of analysis in mul-tiresidue determination. However, LCcombined with UV, diode-array, or fluo-rescence detection is also used in manypesticide residue laboratories as a com-plementary technique to determine manyof the new pesticides developed duringthe last decade. Also, some importantlaboratories are just now starting to usevarious LC-MS multiresidue methods inroutine analyses for most of the non-GC-amenable pesticides (VALVERDE, 2000).

The total number of pesticides cov-ered by the multiresidue methodologiesused in the food control laboratories


may be very different depending on theanalytical capabilities of each labora-tory. The number can vary from 40 to300. This is one of the reasons (but notthe only one) for which the results ofthe monitoring programs of differentcountries, or of different regions in acountry, are not totally comparable.However, it is interesting to note thatthe percentages of fruit and vegetablesamples found to contain pesticide res-idues above the MRLs in the differentnational monitoring programs are verysimilar, and usually vary from 2 to 4%(EU COMMISSION, 2001).

One general and important limitationof all the pesticide residue monitoringprograms is that the introduction of anew pesticide in the routine multiresi-due analytical methodology of the con-trol laboratories is usually carried outseveral years after the pesticide has beenintroduced on the market. For this rea-son, it is not unusual that residues ofsome pesticides widely used by the grow-ers in different countries are not foundfor several years in the respective moni-toring programs.

The objective of this paper is to reportthe results obtained from a unique ex-periment carried out during the celebra-tion of the “2nd European Pesticide Resi-due Workshop -EPRW´98” (Almería,Spain, 24-27 May, 1998), which wasnamed “EPRW´98 Melons Survey”. At theend of the closing session of EPRW´98,about 200 delegates from 35 countriesfrom all over the world received two mel-ons, cultivar “Galia” and cultivar “Can-taloupe”. The melons were produced intwo different plastic greenhouses in Alm-ería (Spain) under rigorous supervisionby the organisers. Delegates working inpesticide residue laboratories were invit-ed by the organisers to take the melonshome and analyse them in their labora-tories. They were then asked to send thepesticide residue levels found in eachmelon to the EPRW´98 Secretariat.

This experiment was initially planned

to evaluate the variability of the resultsobtained by different pesticide residuecontrol laboratories in the analysis of in-dividual pieces of melons from a “homo-geneous lot” (same plantation, same pes-ticide treatments, and same harvest date)after having been handled and storedunder different and unknown conditionsbefore analysis. It is important to notethat the experiment was not planned tobe an interlaboratory test, but a “survey”on the pesticide residues which, from a“formal” point of view, consumers of dif-ferent countries (or regions) would ingestby eating the “same” melons.

The EPRW´98 melons were producedusing the typical horticultural practicesof the Almería plastic-greenhouse plan-tations, which include the use of manypesticides recently introduced on the Eu-ropean market. Therefore, the experimentwas also planned to evaluate the real ca-pability of the current pesticide residuemonitoring programmes to control “rela-tively new” pesticides. Almería is a Span-ish province that has the highest concen-tration of plastic greenhouses in theworld. More than 3 million metric tons offruit and vegetables are produced yearly,and about half of the total production isexported to different European countries.For this reason, Almería is one of the plac-es in Europe where many new pesticides,mainly insecticides, acaricides and fun-gicides, start to be marketed.

The results of this experiment werepresented as an “informal poster” at the3rd European Pesticide Residue Work-shop (VALVERDE et al., 2000). However,the interest generated by this presenta-tion, and the importance of some of theresults obtained, led to this publication.


“Galia” and “Cantaloupe” melons werecultivated in two different commercialgreenhouses of 5,000 m2 surface areaeach (CampoNix S.L., Nijar-Almería,


Spain), using state-of-the-art good agri-cultural practices. Melons were plantedon the 25th (Galia) and 27th (Cantaloupe)of February, 1998 at a planting densityof 1 plant/m2 in both cases. From eachplantation, about 1% of the total produc-tion (400 melons; 700-800 g/each) werecollected on the 26th of May, 1998 to bepacked and distributed as “EPRW´98Melons”. Pesticide treatments given to themelons (Tables 1 and 2) were rigorouslycontrolled and recorded by the organis-ers throughout the production process,but they were unknown to the partici-pants in the survey. The pre-harvest in-tervals for all the pesticides were respect-ed since they were, in all the cases, lessthan 15 days. For those compounds ap-plied in the last treatment (on the 6th or7th of May), the pre-harvest intervals were3 days, except for methomyl (7 days) andcarbendazim (15 days) (DE LIÑAN, 1997).

Melons were packed in specially de-signed boxes on the 27th of May, 1998,after receiving a routine industrialbrushing. Each EPRW´98 box containedone “Galia” melon and one “Cantaloupe”melon, with written instructions for theparticipants. In total, 200 boxes ofEPRW´98 melons were distributed to theEPRW´98 delegates on the 27th of May,1998. According to the enclosed instruc-tions, delegates working in pesticide res-idue laboratories, and interested in par-ticipating in the survey were to take themelons to their laboratories, analysethem, and send the pesticide residue lev-els found in each melon to the EPRW´98Secretariat. The two melons were to beanalysed separately, and the analyseswere to be done on the whole, fresh mel-ons, using the analytical methodologiesroutinely used by their laboratories intheir monitoring programmes.


Seven laboratories analysed the mel-ons according to the instructions given

and sent the results to the organisers.Table 3 shows the names and addressesof the participating laboratories. The to-tal number and type of pesticides cov-ered by the analyses carried out in eachlaboratory are given in Table 4, and theresults obtained are reported in Table 5.In Tables 4 and 5 the names of the labo-ratories are not specified as they wereidentified with a randomly assignednumber. After receiving the results fromthe seven laboratories, melons were alsoanalysed by the organisers at the Uni-versity of Almeria (Ualm) by the super-critical fluid extraction method developedby VALVERDE et al. (1996) and GC de-termination of 43 GC-amenable pesti-cides, including endosulfans and pro-cymidone. Results obtained in theseanalyses are also included in Table 5.

Results in Table 5 indicate that no res-idues of the pesticides used in the twoplantations of melons were found by theparticipating laboratories. Out of the sev-enteen different pesticides used, elevencould not be determined because they werenot included in the scope of the analyticalmethods routinely applied by all the labo-ratories at the time of making the analy-ses. These eleven pesticides are: abamec-tin, azadirachtin, Bacillus thuringiensis,copper oxychloride, cymoxanil, imidaclo-prid, kasugamycin, kresoxim-methyl, py-rifenox, pyriproxyfen, and tetraconazol.Except for copper oxychloride (an inorganicfungicide not included in the pesticide res-idue monitoring programmes) all thesepesticides can be considered as “new pes-ticides” recently introduced on the Euro-pean market. Regarding the other six pes-ticides used on the melons, only chloroth-alonil was analysed by all the laboratories(including the University of Almeria),whereas methomyl, carbendazim, thi-ophanate-methyl, mancozeb and manebwere only analysed by five (four for meth-omyl) of the seven laboratories participat-ing in the experiment. For these six pesti-cides, it could be assumed that the resi-due levels in the melons were below the


Table 1 - Pesticide treatments given to the “Galia” melons.

Date of Days before Active Type ofTreatment harvest ingredients pesticide

March 29, 1998 58 Abamectin Insecticide/AcaricideImidacloprid Insecticide

Thiophanate-methyl FungicideMancozeb FungicideCymoxanil Fungicide

April 9, 1998 47 Copper oxychloride FungicideKasugamycin Bactericide/FungicideAzadirachtin Insecticide

May 6, 1998 20 Chlorothalonil FungicideCarbendazim FungicideImidacloprid Insecticide

Kresoxim-methyl FungicideMethomyl Insecticide

Table 2 - Pesticide treatments given to the “Cantaloupe” melons.

Date of Days before Active Type ofTreatment harvest ingredients pesticide

March 25, 1998 62 Pyrifenox FungicideThiophanate-methyl Fungicide

Mancozeb FungicideCymoxanil Fungicide

April 18, 1998 38 Abamectin Insecticide/AcaricideMethomyl Insecticide

Maneb Fungicide

April 28, 1998 28 Cymoxanil FungicideMancozeb Fungicide

Tetraconazol FungicidePyriproxyfen Insecticide

B. Thuringiensis Insecticide

May 7, 1998 19 Abamectin Insecticide/AcaricidePyriproxyfen Insecticide

Table 3 - Names and addresses of the laboratories that participated in the EPRW´98 Melons Experiment.

Kantonales Laboratorium, Sanitätsdepartement des Kantons Basel-Stadt: Kannenfeldstrasse 2, Postfach, 4012 Basel,Switzerland.Gestione Laboratori di Analisi, Progetto Natura: Via Giovanni Marradi 41, 59100 Prato, Italy.Laboratoire Interregional de la Represión des Fraudes: 369 Rue Jules Guesde, 59651 Villeneuve D´Ascq, France.Swedish National Food Administration-AnalyCen Nordic AB: Livsmedelsverket, Box 622, 75126 Uppsala, Sweden.Laboratorio Agrario, Conselleria Agricultura-Valencia: Pintor Goya 8, 46100 Burjassot, Valencia, Spain.Laboratorio SOIVRE-Almería, Ministerio Economía y Hacienda: General Tamayo 11, 04004 Almería, Spain.Laboratorio Sanidad Vegetal-Almería, Consejería Agricultura-Andalucía: Hermanos Machado 4, 3ª planta, 04004Almería, Spain.


respective analytical limits of determina-tion, which were 0.01 mg/kg for almostall the pesticide/laboratory combinations.

Surprisingly, residues of two pesticidesnot used during the production of themelons were determined in the EPRW´98melons by a number of laboratories. Thesepesticides are endosulfan and procymi-done, which were included in the scope ofthe analytical methods used by all the lab-oratories. Specifically, residues of endo-sulfan were determined by three labora-tories in the “Galia” melons (0.04–0.11 mg/kg) and by five laboratories in the “Canta-loupe” melons (0.01-0.04 mg/kg), where-as residues of procymidone were deter-mined by only three laboratories in the“Cantaloupe” melons (0.01-0.03 mg/kg).Residues of endosulfan (0.01 mg/kg) andprocymidone (0.02 mg/kg) were also foundin the “Cantaloupe” melon in the analy-ses carried out at the University of Alme-ria. The only explanation to justify theseresults is that earlier crops of eggplants in

the two greenhouses used to produce theEPRW´98 melons, received two treatmentswith endosulfan in August, 1997 and Oc-tober, 1997, and one treatment with pro-cymidone in December-1997. After remov-ing the eggplants and cleaning the green-houses (at end of January, 1998), someresidues of these two pesticides remainedin the greenhouses as contaminants,mainly in the soil and/or on the plasticcover. The presence of residues of bothpesticides in the melons could have origi-nated from the natural process of watercondensing on the inside plastic cover ofthe greenhouse (which takes place everynight throughout the production cycle) andthe subsequent dripping of the water con-taining the pesticides on the melons. Inthe case of procymidone, which is a sys-temic fungicide, the contamination of themelons could also have originated from ab-sorption from the soil through the roots ofthe melon plants and translocation to thefruit (TOMLIN, 1994).

Table 5 - Pesticide residues levels (in mg/kg) determined in the EPRW´98 Melons.

Lab 1 Lab 2 Lab 3 Lab 4 Lab 5 Lab 6 Lab 7 Ualm

“Galia” Melon Endosulfan 0.08 - - 0.04 - - 0.11 -“Cantaloupe” Melon Endosulfan 0.03 - 0.04 0.04 - 0.01 0.03 0.01 Procymidone 0.01 - 0.03 - - 0.02 - 0.02

“–” indicates below limit of determination.

Table 4 - Type and number of pesticides covered by the analyses carried out by the participatinglaboratories (Numbers were randomly assigned to the various laboratories).

Lab 1 Lab 2 Lab 3 Lab 4 Lab 5 Lab 6 Lab 7

GG-Multiresidue Method 62 65 18 48 123 50 198(Organophosphorous, Organo-chlorine, Pyrethroids, others)HPLC Methods 11 20 - - 25 4 49(Benzimidazols, Carbamates,Benzoyl/Phenyl-ureas, others)Dithiocarbamate Method Yes Yes - - Yes Yes Yes(as CS2)

”–” indicates that these methods were not used.


The results obtained for endosulfan areespecially important, since this pesticidehas been recently prohibited in Europefor use on almost all vegetable crops (cu-cumber, eggplant, lettuce, green beans,etc.). However, it is still authorised to beused on melon, pepper and tomato (EU-ROPEAN COMMUNITIES, 2000). Accord-ing to the results obtained in this exper-iment, some routine (and “good”) agricul-tural practices in greenhouse cropping,such as making crop rotation withouttaking into account if the earlier crop wastreated with endosulfan, should be ques-tioned in some cases. For example, themelon-cucumber rotation, which is a veryusual crop rotation in the Almería green-houses, should be avoided if the meloncrop had been treated with endosulfan.

ACKNOWLEDGEMENTS

We thank I.V.M. Zehringer, F. Romani, P. Pucci,M. Olek, A. Andersson, H. Sifverson, J. Garcíade la Cuadra, M. Gamón, L. Ortega, B. Abel-lanas, V. Aparicio, and J.C. Herrera for theirparticipation in this experiment.

REFERENCES

Andersen J.H. and Poulsen M.E. 2001. Resultsfrom the monitoring of pesticide residues in fruitand vegetables on the Danish market, 1998-1999. Food Add. Cont. 18:906.

Andersson A., Jansson A. and Jahrl S. 2001. TheSwedish Monitoring of Pesticide Residues inFood of Plant Origin: 2000. LivsmedelsverketsRapport 16/2001. National Food Administra-tion. Uppsala, Sweden.

Cairns, T. and Sherma J. (Eds). 1992. “EmergingStrategies for Pesticide Analysis”, Part II: Mul-tiresidue Approaches, p. 71. CRC Press. BocaRaton, Florida.

Camoni I., Fabbrini R., Attias L., Di Muccio A.,Cecere E., Consolino A. and Roberti F. 2001.Estimation of dietary intake of pesticide resi-dues by the Italian population during 1997.Food Add. Cont. 18: 932.

De Liñan C. 1997. “Vademécum de Productos Fit-osanitarios y Nutricionales 1998”. EdicionesAgrotécnicas. Madrid, Spain.

EU Commission. 2001. Monitoring for Pesticide

Residues in Products of Plant Origin in the Eu-ropean Union and Norway-Report 1999. Foodand Veterinary Office of the European Commis-sion, Brussels. (http://europa.eu.int/comm/food/fs/ inspections/fnaoi/reports/pesticides/mon_rep/index_en.html).

European Communities. 2000. Commission Direc-tive 2000/42/EC of 22 June 2000 amendingthe Annexes to Council Directives 86/362/EEC,86/363/EEC and 90/642/EEC on the fixing ofmaximun levels for pesticide residues in andon cereals, foodstuffs of animal origin and cer-tain products of plant origin, including fruitsand vegetables respectively. Official Journal ofthe European Communities L158: 51.

FDA. 1994. “Pesticide Analytical Manual”, Vol. I.US Food and Drug Administration, Departmentof Health and Human Services. Washington, DC.

Lehotay S.J. 1998. New Technologies Applied inPesticide Residue Analysis. In “Residuos dePlaguicidas´96/Pesticide Residues´96”. A. Val-verde (Ed.), p. 53. Instituto de Estudios Alm-erienses/Universidad de Almería. Almería, Spain.

MAFF/PSD/HSE. 2000. Annual Report of theWorking Party on Pesticide Residues: 1999.Ministry of Agriculture Fisheries and Food. Lon-don, UK.

MAPA. 2001. Programa de Control de LMRs-1999:Informe de los Resultados del Programa Nacion-al de Vigilancia de Residuos de Productos Fit-osanitarios en Origen. Ministerio de Agricultu-ra, Pesca y Alimentación. Madrid, Spain.

MPHWS. 1996. “Analytical Methods for PesticideResidues in Foodstuffs”. Inspectorate for HealthProtection, Duch Ministry of Public Health,Welfare and Sport. Den Haag, The Netherlands.

NFA. 1998. Pesticide Analytical Methods in Swe-den, Part 1. Livsmedelsverkets Rapport 17/98.National Food Administration. Uppsala, Sweden.

Schenzler C. and Thier H.P. 2001. European stand-ardisation of methods for pesticide residue anal-ysis in foods - current status. Food Add. Cont.18: 875.

Tomlin C. (Ed.). 1994. “The Pesticide Manual” 10th

ed. British Crop Protection Council/The RoyalSociety of Chemistry. Surrey, UK/Cambridge, UK.

Valverde A., Fernandez-Alba A., Contreras M. andAgüera, A. 1996. Supercritical fluid extractionof pesticides from vegetables using magnesiumsulphate for sample preparation. J. Agric. FoodChem. 44: 1780.

Valverde A. 2000. Chromatographic pesticide res-idue analysis. J. AOAC Int. 83: 679.

Valverde A., Aguilera A., Rodríguez M. and Cama-cho F. 2000. The EPRW´98 Melons Experiment.Paper No. 17, presented at 3rd European Pesti-cide Residue Workshop. York, UK. July 3-5.

WHO. 1997. Guidelines for predicting dietary in-take of pesticide residues (revised). WHO/FSF/FOS Report No. 97.7. World Health Organisa-tion. Geneva, Switzerland.

Paper received October 3, 2002 Accepted January 20, 2003

FLAIR-FLOW EUROPE


ITALIAN NETWORK LEADERDipartimento di Scienze degli AlimentiUniversità di Udine, Via Marangoni 97, Udine, ITALYTel. +39 0432 590711 - Fax +39 0432 590719e-mail: [email protected] site: www.uniud.it/ffe/welcomeflair.htm

F-FE PROJECT LEADER

Mr. Jean-Francois QuillienINRA-CRIAA (France)[email protected]

RISK ASSESSMENT IN FOOD AND DIETFFE 575/03/SME65

Consumers are greatly concerned by the risk of chemicals in food, in particu-lar of pollutants, chemical additives, agrochemicals and veterinary drugs. TheEU introduced its new and improved food safety policy a few years ago throughthe EC White Paper on Food Safety (12 January 2000 Comm (1999) 719 Final),where a reduced load of risky chemicals and an improved risk assessment proce-dure were announced, together with the set up of the European Food SafetyAuthority.

Further, the World Trade Organization (WTO) is also dealing with risk assess-ment, and in the UN, the Codex Alimentarius has developed standards, guide-lines and recommendations in order to protect human health from food bornehazards.

However, the area of risk assessment is complex. Scientific progress has beenmade but a lack of consensus and of European harmonization exists. For thesereasons, a European Concerted Action has now focused this field.

A risk assessment comprises: 1) hazard identification, 2) hazard characterisa-tion, 3) exposure assessment, and 4) risk characterisation. The Concerted Actionfollows this structure, setting up individual theme groups of experts, critically ex-amining and developing each area.

The project explored means of improving the principles applied to, and scientificbasis of, risk assessment with respect to food additives and contaminants, micro-nutrients and nutritional supplements, macronutrients and whole foods.

The first report from this Action has now been published in Food and ChemicalToxicology 40 (2002), 137-428, (Pergamon, ISSN 0278-6915), and the final reporton risk characterisation is expected at the beginning of 2003.

Project No: QLK1-1999-00156 (FOSIE) http://www.ilsi.org/misc/fosie/Project Co-ordinator: Dr. Juliane Kleiner, ILSI Europe, 83, Avenue E. Mounier,

B-1200 Brussels, Belgium, Tel. +32 (0)2 7710014, Fax +32 (0)2 7620044, E-mail:[email protected]


DECODING THE CAUSESOF EATING DISORDERS

FFE 578/03/HP65

Obesity is considered by many health professionals to be an eating disorder, andthere is additional concern that inappropriate and unnecessary attention to bodyweight could trigger diseases such as Anorexia nervosa and Bulimia nervosa.

European scientists have embarked on a project aimed at furthering the under-standing of the aetiology of these eating disorders. The environment, genome andbrains of individuals from hundred of families are all being investigated in a verycomplex study in order to identify and measure psychosocial variables and traitswhich are risk factors for these diseases.

Researchers are investigating:- Personal, environmental and developmental risk factors, by studying pairs of

sisters, one of whom has a disorder.- The role of dieting in triggering eating disorders, by evaluating obese adoles-

cents in 1000 twin families.- Candidate genes to define genetic risk factors, by studying 600 families and

600 cases and controls with the three disorders.- Neuronal responses to food in the brain, by using scanning techniques (mag-

netic resonance imaging, PET).Males and females from across Europe will be studied to account for gender and

cultural differences. The researchers hope to use the results to improve the level ofhealthy eating in the population by reducing the risk factors for eating disordersand obesity; to educate experts and the general public about the reasons thesedisorders develop, thus removing stigma and discrimination, and introduce newpharmacological treatments.

Project Reference: QLK1-1999-00916 (FACTORS IN HEALTHY EATING)www.eatingresearch.com

Project contact: Prof Janet Treasure and Dr David A Collier, The Institute ofPsychiatry; Division of Psychiatry, Eating Disorders Unit, PO Box 053; De Cres-pigny Park, Denmark Hill, London SE5 8AF, UK, Tel. +44 20 78480180/0631, Fax+44 20 78480560/0051, E-mail: [email protected]; [email protected]

DETECTING GLUTEN IN FOODS:A RELIEVE FOR COELIAC PATIENTS

FFE 580/03/CG64

An easy-to-use tool for the detection of gluten in foods will be developed in anEU-funded project coordinated by Prof. Ludvig Sollid, which will help people withcoeliac disease to avoid problem foods.


An every day torment for coeliac disease patients is having to scrutinise theingredients of all foods they are going to eat. Gluten, which causes the symptomsof the disease, is common in foodstuffs, even if the product itself is not a cerealproduct: due to its increasing use as a food additive its presence in many productsis not immediately obvious.

Coeliac disease affects approximately one million Europeans. The intolerance togluten results causes a damage to the small intestine mucosa which is reversiblewith avoidance of dietary gluten. The damaged mucosa disturbs the absorption ofwater and nutrients. This can cause malnutrition, which can be severe, especiallyfor small children and adolescents. Coeliac disease is different from allergy: somepeople with wheat allergies are not gluten intolerant and can eat rye and barley.According to present knowledge, oat is allowed for coeliac patients, under supervi-sion of a physician.

The symptoms of coeliac disease can vary with each individual and they do notalways involve the digestive system. They can range from no symptoms at all, tosevere ones, such as gas, bloating, diarrhoea, abdominal pain, irritability, musclecramps or fatigue. Because of the broad range of symptoms that coeliac diseasepresents, it can be difficult to diagnose.

Scientists in this project will use new knowledge on the development of reliable andeasy to handle test systems for the detection of gluten in food products. You can findmore information of the project from their web page http://immunology.no/cd/eu/

Project No: QLK1-2000-00657 (GLUTEN EPITOPES IN CD)Project Co-ordinator: Prof Dr Ludvig Sollid, University of Oslo, Institute of Im-

munology, Rikshospitalet, 0027 Oslo, Norway, Tel. +47-23073811, Fax +47-23073822/3510, E-mail: [email protected]

CHEESE PACKED TO LAST LONGERFFE 581/03/CG65

The shelf-life of cheeses may triple from the present 2-3 months up to 9 months,with the help of a new concept in food packaging. This improvement does notrequire more food additives in the cheese itself, but in the packaging material.

Participants in a project named Biopack are developing a new approach to theuse of oxygen scavengers, and other preservatives, as active, protective agents in anew biobased packaging material. In the first instance, this new active packagingwill be targeted at improving the quality and safety of cheese, but the project isalso expected to improve the shelf-life after opening the package at home, reducingthe growth of moulds and development of rancid taste.

The importance of foods being properly packaged is seldom noticed in everyday life.However, an excellent product can be totally wasted if the packaging of it is not appro-priate. Packages are not only an important communication channel to the consumers,they also protect food from microbiological spoilage and mechanical damage.

The new technology developed, and knowledge obtained from this project, willalso be applicable in other packaging concepts, to extend the shelf-life of a broadvariety of food products from meat to fruits and vegetables, and improve theirquality for the consumer.


Further information is available at the project web site http://www.biopack.org/

Project No: QLK5-2000-00799 (BIOPACK)Project Co-ordinator: Prof Grete Bertelsen, Centre for Advanced Food Studies

(LMC-KVL), The Royal Veterinary and Agricultural University, Rolighedsvej 30, DK-1958 Frederiksberg C, Denmark, Tel. +45 3528 3212, Fax +45 3528 3190, E-mail:[email protected]

BRAIN FOOD AND FATSFFE 586/03/HP67

Work has just started to investigate the role of lipids in preventing age-relatedbrain neurodegeneration and Alzheimer’s disease (AD). AD is the most commonneurodegenerative disorder amongst EU citizens, affecting about one third of thepopulation, so treatment and prevention are key challenges.

The cause of the AD is not yet fully understood, although certain key processesinvolved in its development have been established over the last 20 years. In mostcases, the disease depends on several environmental and genetic factors, but inabout 5% of cases the disease is genetic, resulting from mutations in 3 differentgenes. A protein called Amyloid Precursor Protein (APP) is central to the diseasecausing process. The protein occurs normally in most cells, and sections of theprotein (known as A-beta) may be ‘cut out’ by protease enzymes. The A-beta sec-tion accumulates over time in the brain and forms a dense amyloid cluster, orplaque. All people produce these plaques to some extent, but not all develop AD. Itappears that a specific form of A-beta called A-beta 42 specifically increases thelikelihood of developing AD at an earlier age. Efforts to treat or prevent AD havebeen directed towards decreasing production of A-beta 42 or helping the brain toclear it.

Recently, research on AD has found that cholesterol increases the activity of theprotease enzymes responsible for A-beta production. Treatment of patients withAD using cholesterol-lowering drugs (statins) has shown promising results, withdecreased production of A-beta 42. Cholesterol levels can also be influenced bydiet.

The Lipidiet project began in 2002 and it is expected to end in 2005. It willdevelop and evaluate the role of dietary lipids (fats) and related food additives inthe prevention and treatment of AD. Three main strands of investigation will becarried out:

1) Cellular and molecular studies on the role of lipids in brain neurodegenera-tion

2) Potential lipid-related treatments for brain neurodegeneration and AD will beformulated and assessed using in vivo models

3) Modified lipid diets to slow down, and eventually prevent, AD in humans willbe studied in vivo.

Project Reference: QLK1-2002-00172 (LIPIDIET) http://www.lipidiet.org/Project contact: Dr Tobias Hartmann, Center for Molecular Biology Heidelberg


(ZMBH), University of Heidelberg, Im Neuenheimer Feld 282, D-69120 Heidelberg,Germany, Tel. +49 6221 546844, Fax +49 6221 545891, E-mail:[email protected]

MAKING MOVESON HEART DISEASE PREVENTION

FFE 587/03/HP68

Does moving to a different country and changing your diet alter your risk ofdeveloping cardiovascular disease (CVD)? Researchers across Europe are trying tofind out the answer to this question using populations in Italy, Belgium and theUnited Kingdom.

Cardiovascular diseases are the major cause of death in Europeans, butrisk is not evenly distributed across Europe. Compared to the Italian popula-tion, the rate of CVD is three times higher in the UK and twice as high inBelgium. Ecological studies showed that a low rate of CVD in Mediterraneanpopulations was associated with, for example, low consumption of saturatedfatty acids (such as those found in butter, cream, pies and pastries) and con-sumption of red wine with meals. Unfortunately in the last 20 years, there hasbeen a tendency for Italians and other southern Europeans to adopt morenorthern European food habits, and this has been reflected in an increase inthe prevalence of obesity. Paradoxically, populations in Northern Europe, withhigher prevalence of CVD, are moving towards a more Mediterranean-stylediet, for example by using more olive oil. Therefore the impact of diet on theCVD risk profile amongst European populations needs to be re-evaluated.Clearly, genetic risk factors may be implicated too, and genes also interactwith environmental factors.

The Immidiet project will evaluate the present dietary habits of three Europe-an communities (UK, Belgium and Italy) at different risk of myocardial infarc-tion (MI). The impact of migration on risk factors for MI will be evaluated fromItalian migration to Belgium – the integration of the two populations will beconsidered as a model of gene-environment interaction. Data are now beingcollected, including measurements of diet composition, biomarkers of food in-take (fatty acids, selenium and zinc), anthropometric variables, and many oth-ers.

It is anticipated that the study will facilitate prevention campaigns in Belgium,Italy and the UK specifically by involving general practitioners (GPs) through whomprimary prevention can be implemented.

Project Reference: QLK1-2001-00100 (IMMIDIET) http://www.negrisud.it/im-midiet/

Project contact: Dr Licia Iacoviello, “Angela Valenti” Laboratory of Genetic andEnvironmental Risk Factors for Thrombotic Disease, Department of Vascular Med-icine and Pharmacology, Consorzio Mario Negri Sud, Via Nazionale, 66030 SantaMaria Imbaro (Ch) Italy, Tel. +39 0872 570298/302, Fax +39 0872 570299, E-mail: [email protected]


NEWS

Announcement

10th SEMINAR OF THE EUROPEANNUTRITION LEADERSHIP

PROGRAMMELuxembourg, 17-25 March 2004

The aim of the programme is to as-sist in the development of future lead-ers in the field of human nutrition inEurope. Emphasis is given to under-standing the qualities and skills ofleaders, team building, communica-tion of nutrition information in abroader context, and to understand-ing the role of nutrition science in so-ciety. The participants of the 2003Seminar aptly expressed the vision ofthe seminars as follows: “To provide aEuropean platform where young sci-entists meet in harmony, realise theirpotential and leave as leaders”. Semi-nars have been held annually since1994, except in 1999. The programmeis organized by a group of leading Eu-ropean nutritionists in close collabo-ration with and support from the Eu-ropean Commission and the Europe-an food industry.

The programme is designed for finalyear PhD students and postdoctoral fel-lows in human nutritional science inEurope. Preference will be given to can-didates under the age of 35 years. Fromthe applications received, thirty candi-dates will be selected by an internation-al selection committee to attend the sem-inar.

The fee for attending the seminar is €1400. This includes full board and lodg-ing from the evening of 17 March to themorning of 25 March 2004 at the semi-

nar hotel. Travel and personal costs suchas drinks and phone calls are not includ-ed.

The ENLP organization may providegrants of € 600 towards the seminar feeand a small number of travel grants upto € 500. Those interested in applyingfor such grants, should send, togetherwith the application form, a letter includ-ing a budget justifying the applicationfor a grant towards the seminar fee andcost of travel.

Deadline for applications: 15 November 2003Further information and application forms can

be obtained from:Mrs Lous Duym-Brookman - Division of Hu-

man Nutrition, Wageningen University, P.O. Box8129, 6700 EV Wageningen, the Netherlands - Tel.+31 317 483054, Fax +31 317 483342 - Email:[email protected] Internet: www.enlp.info

Final Announcement

IDF WORLD DAIRY SUMMIT7-12 September 2003

Bruges, Belgium

100 years of collaboration and net-working between the dairy communitiesof so many countries all over the worldis really an extraordinary achievement.IDF and Belgium are very happy to in-vite you to join them in celebrating IDF’sfirst one hundred years of existence, inBruges from September 7-12. This willbe 100 years to the day of the First In-ternational Dairy Congress, held inBrussels, which saw the creation of IDF.Of course IDF has seen many changesin the course of its first century. Nowwe have brought together modern com-munication technology and traditional


networking, always a major factor inIDF’s effectiveness for the dairy sector.

Conferences will address:1) The Dairy (R)evolution – 100 Years

of Change2) Advances in Fractionation and Sep-

aration: Processes for Novel Dairy Ap-plications

3) Quality Management at Farm Level4) Effects of Pro- & Prebiotics on

Health Maintenance5) Energy Consumption & Life Cycle

Assessment towards Sustainability6) 100 Years with Liners and Pulsa-

tors in Machine Milking.

Further information is available atwww.idf2003.com or

IDF Secretariat - Attn. Mrs. Caroline Brooks -Boulevard Auguste Reyers - 80 Diamant Building –B-1030 Brussels, Belgium - Tel. +32 2 733 98 88,Fax +32 2 733 0413 - e-mail: [email protected]

Confectionery Manufacturing Expo27 to 29 April 2004

Wiesbaden, Germany

Confectionery Manufacturing Expo isa biennial event that takes place everyother year. It is organised by SimplyEvents, a leading international tradeshows organiser and publisher. It is theonly pan-European major exhibition ex-clusively featuring suppliers to the con-fectionery industry.

A seminar programme is currently thesubject of discussion and negotiationwith the industry and will be announcedlater in the year. Themes include:

- an innovations theatre- seminars on how East European

companies can compete beyond their tra-ditional marketplaces

- workshops designed to provide in-gredient buyers with the tools to success-fully forecast market supply and de-mand.

For further information contact:Gillian Pattison, Show press office - Tel: +44

(0)20 8542 9090 - E-mail: [email protected] - www.cm-expo.com

REPORT

Ethical Corporation Magazine“Does the Structure of Europe’s Food

Supply Chain Work For or AgainstOur Environmental Objectives?”

The report reviews Europe-wide foodproduction, processing and retail net-works, focusing on the effects of large-scale consolidation, food clustering andagricultural intensification on marketsand the environment.

Data from over 320 accredited pan-European sources has been collated toillustrate the report’s key thrust – theshift from a food supply chain to a fooddemand chain. The report provides in-dustry stakeholders with a clear expla-nation of the trends that are shaping thefuture of retail, processing, production,policy, environmental issues and insti-tutional investment in the food system.

Risk exposure is becoming increasinglyconcerning for companies involved in large-scale diversification throughout the foodsupply chain – the report outlines thethreats and opportunities presented by ver-tical integration from seed to supermarketfor enterprises contemplating such a move.

The report also evaluates the strengthsand weaknesses of Europe’s retailers,processors and growers, as well as trendsin branding and consumer spending. Adetailed analysis of the environmentalimpact of agricultural intensification andbiodiversity issues is also provided.

This groundbreaking report will be re-leased to coincide with the announcementof the speaker roster for the “EuropeanFood Supply Chains 2003 Conference” tobe held in Brussels on November 5-6,2003. The event will be Europe’s first-everindependent forum to address the con-cerns of retailers, processors, growers andenvironmentalists about the risks to thefood supply chain in the EU and beyond.

To receive your free pre-release copyof the report, please contact KathrynHingston on +44 (0) 20 73757201 /[email protected] information on speaking at, or attending the

conference, contact John Bodenham on +44 (0)20 73757500 / [email protected].


BOOKS

IDF Publications

HEALTH EFFECTS OF PROBIOTICSAND CULTURE-CONTAINING

DAIRY PRODUCTS IN HUMANS

A.C. Ouwehand, B. Bianchi Salvadori,R. Fondén, G. Mogensen, S. Salminen,R. Sellars

This scientific summary reviews thelatest research related to the wide arrayof health benefits attributed to probiot-ics and culture-containing dairy prod-ucts.

17 pp - English only

DETECTIONOF FERMENTATION-PRODUCED

CHYMOSIN (FPC) IN RENNET

J.-C. Collin, C. Repelius, M.K. Harboe

This study deals with the use of en-zymes in cheese-making. It describes themethodology for determining the pres-ence of certain milk-clotting enzymes incheese.

5 pp - English only

LABORATORIES IN TRANSITION

S. Coates, R. López-Fandiño, G. Wil-son, H. van den Bijgaart, S. Holroyd

A symposium session organized dur-ing the IDF/ISO/AOAC International An-alytical Week 2002 (Toledo, Spain, April2002) dealt with the accreditation of lab-

oratories, notably the validation of meth-ods, the new ISO Standard on accredita-tion and the application of modern infor-mation technology to the handling andcommunication of analytical data.

32 pp - English onlyTotal: 57 pp

Bulletin N° 380/2003-60 Euro

REVIEW OF PRACTICESFOR CLEANING

AND SANITATION OF MILKINGMACHINES

D.J. Reinemann, G.M.V.H. Wolters, P.Billon, O. Lind, M.D. Rasmussen

This paper gives an overview of clean-ing and sanitation practices for milkingmachines with an update of recent re-search. Moreover, the paper provides avery practical section for trouble-shoot-ing cleaning problems and guidance onstrategic sampling.


MEASURING VACUUMIN MILKING MACHINES

M.D. Rasmussen, D.J. Reinemann,G.A. Mein

The technical condition of milkingequipment may have a direct or indirectinfluence on udder health and milk qual-ity. This document explains how meas-urements carried out in the milking unitcan be interpreted and why faults are


likely to affect the outcome of the milk-ing.

14 pp - English onlyTotal: 32 pp

Bulletin N°381/2003-60 Euro

GUIDE FOR DAIRY MANAGERSON WASTAGE PREVENTION

IN DAIRY PLANTS

Increasingly stringent environmentalrequirements demand a new revised ap-proach to dairy wastage management.This guide covers process description,

types of waste, sources of loss, and wastereduction practices together with adviceon minimizing environmental impact andcost. Technical aspects and systematicwastage management make up an over-all environmental management system.Methods of measuring flow, samplingand methods of analysis are also cov-ered.


Bulletin N°382/2003-60 Euro

FIL-IDF - Building Diamant, Boulevard AugusteReyers 80, B-1030 Brussels, Belgium - Tel. +322 7339888- Fax +32 2 7330413 - E-mail:[email protected] - Web Site: http://www.fil-idf.org


GUIDE FOR AUTHORSITALIAN JOURNAL OF FOOD SCIENCE - IJFS

1. Manuscript Preparation

(1) Manuscripts must be typed, double-spaced and two copies submitted along withthe computer disk. There should be liberal margins on top, bottom and sides (2.5 cm).English is the official language. Authors who are not fluent in written English shouldseek help from a fluent person before the final version is typed. The Assistant Editorreserves the right to make literary corrections and to make suggestions to improve brevity,but the paper should be revised by a native English speaker before submission.

The paper must also be submitted on a Macintosh or Windows floppy disk. Indicatewhich word processor was used to generate the file and save the file also in format“Text only”, DCA-RTF or ASCII, if you do not have programs for Macintosh; graphics,pictures and diagrams must be saved in TIF, JPEG, EPS, CGM or PICT formats (notincluded in MsWord documents).

(2) Every paper should be divided under the following headings in this order:Title. Informative of the content of the article (<50 characters + spaces). Author(s).

Initials and Surname, omit professional and official titles. The Institute and addresswhere the research was carried out and the current address of each author should begiven as a footnote on the title page.

Abstract. Clearly state the objective of the study, give a concise description ofexperiment(s), observations, results and conclusions. No references should be cited.DO NOT EXCEED 100 WORDS. An abstract and title in Italian (corresponding tothe English) must also be included.

Keywords. Up to six words, in alphabetical order, which describe the documentmust be given to aid data retrieval and indexing.

Introduction. Review pertinent previous work and cite appropriate references. Statethe purpose of the investigation.

Materials and Methods. Indicate apparatus, instruments, reagents, etc., givingsufficient detail to allow the work to be repeated.

Results and Conclusions. Results and Conclusions may be presented together orseparately. Concisely present results using tables and figures to help justify conclusions(do not present the same information in both forms). Use statistical analysis whenappropriate. Unsupported hypotheses should be avoided. Conclusions should pointout the significance of the findings and, if possible, relate the new findings to someproblem in Food Science and Technology.

Acknowledgments. Acknowledgments of assistance are appropriate provided theyare not related to analyses, or other services performed for a fee. Financial support,thanks for assistance, article number or thesis fulfillment may be included.

Units. A list of units particular to the paper may be included.References. References should be arranged alphabetically, and for the same author

should be arranged consecutively by year, typed double-spaced. Each individual citationshould begin flush left (no indentation). Refer to attached examples taken from “StyleGuide for Research Papers” by the Institute of Food Technologists (Chicago - Illinois -USA). Literature citations in the text should be referred to by name and year inparentheses (only the initials in capital letters). If there are more than two authors,mention the first author and add et al.


(3) Lines on all pages, including those pages for “References” and figure legends, mustbe numbered (by pen) in the left margin, beginning with number one at the top of the page.

(4) Tables should be as few and as simple as possible and include only essentialdata. Each table must be on a separate sheet and saved on floppy disk, and have anArabic number, e.g. Table 4 NOT Tab. 4. Legends must be self-explanatory and on aseparate sheet. Use lower-case letters for footnotes in tables and explain below thetable in the order in which they appear in the table.

(5) Figures must be drawn on separate sheets of paper and saved on floppy disk inTIF, JPEG, EPS, CGM or PICT formats. They should be drawn so that on 50% reduction,lines, figures and symbols will be clearly legible and not overcrowded. A photocopy ofhow the figure should appear must be included. Photographs must be unmounted,glossy prints or slides. All figures must be given Arabic numbers, e.g. Fig. 3, in the textand in the final copy only on the back where the title of the paper, the senior author’ssurname and the top of the illustration must also be marked; for reviewing procedures,do not include this information in the first submitted copies. Legends for figures mustbe self-explanatory and should be typed on a separate sheet under “Legends to Figures”.

(6) Standard Usage, Abbreviations and Units. The Concise Oxford and Webster’sEnglish Dictionaries are the references for spelling and hyphenation. Statistics andmeasurements should always be given in figures, e.g. 10 min, except when the numberbegins a sentence. When the number does not refer to a unit of measurement it isspelled out unless it is 100 or greater. Abbreviations should be used sparingly, onlywhen long or unwieldy names occur frequently, and never in the title; they should begiven at the first mention of the name. International Standard abbreviations shouldgenerally be used except where they conflict with current practice or are confusing. Forexample, 3 mm rather than 3x10-3m. Abbreviations should be defined the first timethat they are used in the text and they should be used consistently thereafter.Temperatures should be expressed in the Celsius (centigrade) scale. Chemical formulaeand solutions must specify the form used, e.g. anhydrous or hydrated, and theconcentration must be in clearly defined units. Common species names should befollowed by the Latin binomial (italics) at the first mention. For subsequent use, thegeneric name should be contracted to a single letter if it is unambiguous.

2. Review Policy

Scientific contributions in one of the following forms may be submitted:Opinions and Reviews - Papers may be sent directly to the Editor-in-Chief who will

decide upon publication or articles will be requested directly from the authors by theEditor-in-Chief.

Short Communications and Surveys - They do not need to have the formal organizationof a research paper; they will receive priority in publication;

Papers - The paper must follow manuscript preparation.Short Communications, Surveys and Papers will be subjected to critical review by

the referees. Upon receiving papers from authors, the Advisory Board with the Editor-in-Chief will select papers in relationship to innovation and originality and send copiesto the referees. A letter stating that the paper has been accepted for refereeing will besent to the authors. Papers needing revision will be returned to the author, and theauthor must return the revised manuscript to the Editor-in-Chief within 1 month,otherwise the paper will be considered as withdrawn. Papers not suitable for publicationwill be returned to the author with a statement of reasons for rejection.


3. Editorial Policy

Referees may not be from the same institution as the author. Referees should maketheir comments and questions in detail and return the paper to the Editor-in-Chief assoon as possible, usually within 4 weeks. The identity and the report of the referees aremade known to the Editor-in-Chief, but only the anonymous report is routinely sent tothe author. If all referees recommend acceptance or rejection, the decision stands. Ifthe opinions of the referees tie, the Editor-in-Chief has the freedom to decide uponacceptance or rejection of the paper. Manuscripts will be edited in the order receivedand accepted papers will be published as closely as possible in this order. A letterannouncing the issue of publication will be sent to the author after the manuscript hasbeen accepted by the Editor-in-Chief. Each paper is accepted with the understandingthat it is the sole document under active consideration for publication covering thework reported (it has NOT been previously published, accepted or submitted for publi-cation elsewhere). Upon acceptance of the paper for publication, the author agrees topay the page charges as published on the first page of each issue. Authors take fullresponsibility for all opinions stated in their papers and published in this journal.

4. Mailing Instructions

Papers for publication and communications regarding editorial mattersshould be sent to:Prof. Paolo Fantozzi or Dr. Mary F. Traynor, F.S.E.Dipartimento di Scienza degli Alimenti, Università di Perugia, S. Costanzo,I - 06126 Perugia, ItalyE-mail: [email protected] or [email protected]

(Anonymous)Anonymous. 1982. Tomato product invention merits CTRI

Award. Food Technol. 36(9): 23.

(Book)AOAC. 1980. “Official Methods of Analysis” 13th ed. Associ-

ation of Official Analytical Chemists, Washington, DC.Weast, R.C. (Ed.). 1981 “Handbook of Chemistry and Phys-

ics” 62nd ed. The Chemical Ruber Co. Cleveland, OH.

(Bulletin, circular)Willets C.O. and Hill, C.H. 1976. Maple syrup producers

manual Agric. Handbook No. 134, U.S. Dept. of Agricul-ture, Washington, DC.

(Chapter of book)Hood L.F. 1982. Current concepts of starch structure. Ch.

13. In “Food Carbohydrates”. D.R. Lineback and G.E.Inglett (Ed.), p. 217. AVI Publishing Co., Westport, CT.

(Journal)Cardello A.V. and Maller O. 1982. Acceptability of water,

selected beverages and foods as a function of servingtemperature. J. Food Sci. 47: 1549.

IFT Sensory Evaluation Div. 1981a. Sensory evaluation guidefor testing food and beverage products. Food Technol.35 (11): 50.

IFT Sensory Evaluation Div. 1981b. Guidelines for the prep-aration and review of papers reporting sensory evalua-tion data. Food Technol. 35(4): 16.

(Non-English reference)Minguez-Mosquera M.I., Franquelo Camacho A, and Fern-

andez Diez M.J. 1981. Pastas de pimiento. I. Normaliza-cion de la medida del color. Grasas y Aceites 33 (1): 1.

(Paper accepted)Bhowmik S.R. and Hayakawa, K. 1983. Influence of select-

ed thermal processing conditions on steam consump-tion and on mass average sterilizing values. J. Food Sci.In press.

(Paper presented)Takeguchi C.A. 1982. Regulatory aspects of food irradia-

tion. Paper No. 8, presented at 42nd Annual Meeting ofInst. of Food Technologists, Las Vegas, NV, June 22-25.

(Patent)Nezbed R.I. 1974. Amorphous beta lactose for tableting U.S.

patent 3,802,911, April 9.

(Secondary source)Sakata R., Ohso M. and Nagata Y. 1981. Effect of porcine

muscle conditions on the color of cooked cured meat.Agric. & Biol. Chem. 45 (9): 2077. (In Food Sci. Technol.Abstr. (1982) 14 (5): 5S877).

Wehrmann K.H. 1961. Apple flavor. Ph. D. thesis. MichiganState Univ., East Lansing. Quoted in Wehrmann, K.H.(1966). “Newer Knowledge of Apple Constitution”, p. 141,Academic Press, New York.

(Thesis)Gejl-Hansen F. 1977. Microstructure and stability of freeze-

dried solute containing oil-in-water emulsions. Sc. D.Thesis, Massachusetts Inst. of Technology, Cambridge.

(Unpublished data/letter)Peleg M. 1982. Unpublished data. Dept. of Food Engineer -

ing., Univ. of Massachusetts, Amherst.Bills D.D. 1982. Private communication. USDA-ARS. East-

ern Regional Research Center, Philadelphia, PA.

REFERENCE EXAMPLESEXAMPLES of use in a Reference list are given below. The bold-faced parenthetical type of citation above the example isindicated ONLY for information and is NOT to be included in the reference list.


CONTRIBUTORS

Gratitude is expressed to the following entities for contributing to

the realization of the Journal by being supporting subscribers for 2003.

Si ringraziano i seguenti Enti, Ditte ed Istituti per aver voluto

contribuire fattivamente alla realizzazione della Rivista, sottoscrivendo

un abbonamento sostenitore per il 2003.

ASSOCIATIONS

Associazione Italiana di Tecnologia Alimentare (A.I.T.A.) - Milano Fax +39-02-2365015

INDUSTRIES

Besana spa - San Gennaro Vesuviano (NA) Fax +39-081-8657651

Birra Peroni Industriale spa - Roma Fax +39-06-22544313

Corial scpa - Foggia Fax +39-0881-680077

Kraft Foods Italia spa - Milano Fax +39-02-41337595

Soremartec Italia srl - Alba Fax +39-0173-313966

Tecnoalimenti scpa - Milano Fax +39-02-67077405


RESEARCH INSTITUTES

Dipartimento di Scienze Tecnologie Agroalimentari (D.I.S.T.A.), Facoltà di Agraria Università degli Studi della Tuscia, Viterbo Fax +39-0761-357498

Dipartimento di Ingegneria e Tecnologie Agro-Forestali Università di Palermo, Palermo Fax +39-091-484035

Dipartimento di Scienze degli Alimenti, Università di Udine, Udine Fax +39-0432-501637

Dipartimento di Scienze e Tecnologie Agroalimentari e Microbiologiche (DI.S.T.A.A.M.), Università del Molise, Campobasso Fax +39-0874-404652

Dipartimento di Scienze e Tecnologie Alimentari e Microbiologiche (DI.S.T.A.M.), Università di Milano, Milano Fax +39-02-50316601

Dipartimento di Valorizzazione e Protezione delle Risorse Agroforestali (DI.VA.P.R.A.), Sezione Microbiologia ed Industrie Agrarie Università di Torino, Grugliasco Fax +39-011-6708549


ITALIAN JOURNAL OF FOOD SCIENCERivista Italiana di Scienza degli Alimenti

DIRETTORE RESPONSABILE: Giovanni ChiriottiAUTORIZZAZIONE: n. 3/89 in data 31/1/1989

del Tribunale di PerugiaProprietà dell’Università di PerugiaTIPOGRAFIA Giuseppini - Pinerolo

Una copia € 6.00

ISSN 1120-1770 © 2003

CHIRIOTTI EDITORI spa - 10064 Pinerolo - Italy

publishes the technical magazines:

VOLUME XV No. 3, 2003

CONTENTS

PAPERS

Electochemical Method for the Evaluation of the Number of Eggs in Egg PastaM.S. Cosio, S. Buratti and S. Mannino ........................................................................................................... 321Chemometric Determination of the Geographical Origin of Cow Milk using ICP-OES Dataand Isotopic Ratios: A Preliminary StudyM.A. Brescia, V. Caldarola, G. Buccolieri, A. Dell’Atti and A. Sacco ........................................................ 329A Method for Anthocyanin Extraction from Fresh Grape SkinG. Spagna, R.N. Barbagallo, A. Todaro, M.J. Durante and P.G. Pifferi ..................................................... 337Phenolic Compounds in Red-Berry Skins of Uva di Troia and Bombino Nero Grapes(Vitis vinifera L.)P. Tamborra, M. Esti, M. Minafra and F. Sinesio ........................................................................................... 347Effect of Sodium Lactate and Potassium Sorbate on Quality Characteristics and Shelf Lifeof Mediterranean Mussel (Mytilus galloprovincialis) Meat during Chilled Storage in Pouches with WaterA. Vasakou, K. Vareltzis and J.G. Bloukas ..................................................................................................... 359Effect of Smoking on Quality Characteristics and Shelf Life of Mediterranean Mussel(Mytilus galloprovincialis) Meat under Vacuum in Chilled StorageA. Kyriazi-Papadopoulou, K. Vareltzis, J.G. Bloukas and S. Georgakis ................................................... 371Shelf Life Extension of Durum Wheat BreadM.A. Del Nobile, T. Martoriello, S. Cavella, P. Giudici and P. Masi .......................................................... 383Proteolysis and Production of Biogenic Amines in Toma Piemontese PDO Cheese during RipeningM. Arlorio, J.D. Coïsson, F. Travaglia, M. Capasso, M. Rinaldi and A. Martelli ..................................... 395Dip Treatments for Fresh Romaine LettuceS.I. Roura, M. del R. Moreira, A. Ponce and C.E. del Valle .......................................................................... 405Effects of a Wheat Germ-Enriched Diet on Skeletal Muscle RegenerationC. Rizzi, R. Chignola, G. Zoccatelli, M. Donà, A.D.B. Peruffo, U. Carraro and K. Rossini ................... 417SHORT COMMUNICATIONS

Gas Chromatographic Determination of Azoxystrobin, Dinocap, Fenarimol,Penconazole and Quinoxyfen during Wine MakingG. Di Bella, M. Saitta, F. Salvo, M. Nicotina and G. Dugo ......................................................................... 427Effect of Salt Dose on the Quality of Dry Smoked BeefT. Dzudie, M. Bouba, C.M. Mbofund and J. Scher ......................................................................................... 433Influence of Hypochlorite Concentration, pH and Temperature on Inactivation of Listeria monocytogenesO. Erkmen ............................................................................................................................................................ 441SURVEY

Pesticide Residues in EPRW’98 MelonsA. Valverde Garcia, A. Aguilera del Real, M. Rodríguez Sánchez and F. Camacho Ferre ....................... 449

FLAIR FLOW ...................................................................................................................................................... 456NEWS ................................................................................................................................................................... 461BOOKS ................................................................................................................................................................. 463GUIDE FOR AUTHORS .................................................................................................................................... 465

Volume XV Number 3 2003 - Chiriotti Editori archivio... · 2020. 5. 15. · J.C. Cheftel...

Documents

Transcript of Volume XV Number 3 2003 - Chiriotti Editori archivio... · 2020. 5. 15. · J.C. Cheftel...