InternationalJournalofFoodMicrobiology62(2000)123131 www.elsevier.nl/locate/ijfoodmicroSurvivalandgrowthof Salmonellabaildon inshreddedlettuceand dicedtomatoes,andeffectivenessofchlorinatedwaterasasanitizer* W.R.Weissinger,W.Chantarapanont,L.R.BeuchatCenterforFoodSafetyandQualityEnhancement,DepartmentofFoodScienceandTechnology,UniversityofGeorgia, 1109ExperimentStreet,Grifn,GA30223-1797,USA Received3April2000;receivedinrevisedform19June2000;accepted29June2000 Abstract AnoutbreakofsalmonellosisassociatedwithdicedtomatoesoccurredintheUnitedStatesin1999.Experimentswere donetodeterminetheefcacyofchlorineinkilling Salmonellabaildon ,thecausativeserotype,inoculatedontoshredded lettuceanddicedtomatoes,andtodeterminesurvivalcharacteristicsoftheorganismontheseproduceitemsstoredat4 8 Cfor upto12daysandontomatoesstoredat21or30 8 Cforupto72h.Populationsof S.baildon inlettuceandtomatoes(pH 4.51 6 0.02)inoculatedwith3.60logand3.86logcfu/g,respectively,werereducedbylessthan1logwhentheproduce1010wasimmersedfor40sina120or200 m g/mlfreechlorinesolution.Produceinoculatedwith0.600.86logcfu/gwas10positiveforthepathogenaftertreatmentwith200 m g/mlchlorine.Initialpopulationsof3.28and3.40logcfu/goflettuce10andtomatoes,respectively,decreasedbyabout2logcfu/gduringstoragefor12daysat4 8 C.Oneofsixsamplesoflettuce10initiallycontaining0.28logcfuof S.baildon pergramwaspositiveafterstoragefor12days,butthepathogenwasnot10detectedintomatoesanalyzedwithin15minofinoculationwith0.40logcfu/g.Whilethenumberofviablecells10decreasedduringstorageat4 8 C,initialpopulationsof0.28logcfu/gofshreddedlettuceand3.40logcfu/gofdiced1010tomatoesarenotreducedtoundetectablelevelsduringstorageat4 8 Cfor12days.Toleranceof S.baildon toanacidicpH (4.5)wasnotinuencedbythepH(4.5,5.8,or7.2)ofthemediuminwhichitwasgrown,suggestingthatthisstrain possessesunusualresistancetoacidpH.Thepathogengrewindicedtomatoes(pH4.40 6 0.01)fromaninitialpopulationof 0.79logcfu/gto5.32and7.00logcfu/gwithin24hat21and30 8 C,respectively. â€ 2000ElsevierScienceB.V.All1010rightsreserved.Keywords:Salmonellabaildon ;Chlorination;Salmonellosis;Tomato;Lettuce 1.Introduction Consumptionoffreshsaladvegetablesandfruits hasincreasedinquantityandvarietyindeveloped*Correspondingauthor.Tel.: 1 1-770-412-4740;fax: 1 1-770-countriesinrecentyears.Alargenumberofmini-229-3216.mallyprocessedproductsarecommerciallyavailableE mailaddress:firstname.lastname@example.org(L.R. Beuchat).insupermarketsandinfoodservicefacilities.With0168-1605/00/$seefrontmatter â€ 2000ElsevierScienceB.V.Allrightsreserved. PII:S0168-1605(00)00415-3
124 W.R.Weissingeretal./ InternationalJournalofFoodMicrobiology62(2000)123131anincreaseinconsumptionhascomeanincreasedinduciblebyvariousacidsandisinuencedby frequencyofoutbreaksofillnessassociatedwithrawnutrientavailability,temperature,andotherfactors. fruitsandvegetables(Nguyen-theandCarlin,1994;Whetherthe S.baildon strainisolatedfrominfected DeRoever,1998;Francisetal.,1999).Salmonellaepatientsintheoutbreakassociatedwithdiced havebeenisolatedfromseveraltypesofrawveget-tomatoesundergoeschangesinacidtoleranceupon ablesfrommanycountries(HedbergandOsterholm,exposuretoacidicenvironments,e.g.indiced 1993;Beuchat,1996).Amongtherawvegetablestomatoesortomatojuice,wasnotknownatthe implicatedinoutbreaksofsalmonellosisintheconclusionoftheepidemiologicinvestigation. UnitedStatesarelettuce(CentersforDiseaseCon-Thepurposeoftheworkreportedherewasto trolandPrevention,1999)andtomatoes(Hedbergetdetermineifacidtoleranceof S.baildon changed al.,1999;Woodetal.,1991).InarecentoutbreakuponexposuretoreducedpHofgrowthmediaand associatedwithdicedtomatoes, Salmonellabaildon ,theefcacyoftreatingshreddediceberglettuceand aserotyperarelyimplicatedinhumansalmonellosis,dicedtomatoeswithchlorinatedwaterinkilling S.wasisolatedfrompatientsingeographicallyseparate baildon inoculatedontothesesaladvegetables.SurareasoftheU.S.(CentersforDiseaseandPreven-vivalandgrowthcharacteristicsoftheorganism tion,1999),suggestingthatcontaminationoccurredinoculatedontoshreddedlettuceanddicedtomatoes atacommonproductionorprocessingsite.wasalsoinvestigated. Chlorinatedwaterisusedinsomecountriesto washvegetablesshortlyafterharvestingandat variousstagesofhandlingandprocessing(Beuchat, 2.Materialsandmethods 1998;Seymour,1999).Theeffectivenessoftreatmentwithwatercontainingupto200 m g/mlchlor-2.1.Bacterialculture ineinreducingnumbersofnaturallyoccurring microorganismsandpathogenicbacteriaisminimal,Acultureof S.baildon (strain61-99)wasobtained oftennotexceeding2logsonlettuce(Adamsetal.,fromtheCentersforDiseaseControlandPrevention, 1989;BeuchatandBrackett,1990;ZhangandAtlanta,GA,USA.Thisstrainisastoolisolatefrom Farber,1996;Beuchatetal.,1998;Beuchat,1999)apatientinanoutbreakofsalmonellosisassociated andtomatoes(Weietal.,1995;Zhuangetal.,1995;withdicedtomatoes.Theorganismwasgrownin ZhuangandBeuchat,1996;Beuchatetal.,1998).trypticsoybroth(TSB,pH7.2 6 0.1,Difco,Detroit, LittleisknownaboutfactorsinuencingthesurvivalMI,USA)supplementedwith50 m g/mlnalidixic andgrowthof Salmonella onshreddedlettuce.acid(Sigma,St.Louis,MO,USA)(TSBN).Stock Severalserotypes,however,havebeenreportedtoculturesontrypticsoyagar(TSA,pH7.2 6 0.1, growindicedripetomatoes(AsplundandNurmi,Difco)supplementedwith50 m g/mlnalidixicacid 1991;Zhuangetal.,1995).ThepHofthese(TSAN)werestoredat4 8 C. tomatoesrangedfrom3.99to4.30atthetimeof inoculation.Survivalandsubsequentgrowthof2.2.Studiestodeterminechangeinacidtolerance Salmonella andotherpathogensonproducebetween thetimeofminimalprocessingandconsumptionFilter-sterilized2Mcitricacidwasaddedto mayresultinincreasedriskofillness.sterilemolten(50 8 C)TSA(pH7.2)toachievepH Acid-adaptedoracid-shockedbacterialcellsarevaluesof5.8,4.8,and4.5.Mediawerepouredinto knowntohaveenhancedprotectionagainstsub-Petridishesandheld1dayat22 8 C,thenat5 8 Cfor sequentexposuretoacidicorotherstressenviron-upto4daysbeforeuse. ments.Acidtoleranceresponse(ATR)of Salmonella ThepHofaportionofcommerciallysterilized typhimurium hasbeenmostextensivelystudiedtomatojuice(pH4.8)obtainedfromalocalsuper(FosterandSpector,1995).Studieshaveshownthatmarketwasadjustedto5.8byaddingsterile0.1M whenshiftedtoanormallylethalpH3.3,exponentialNaOH.JuiceatpH4.8and5.8wassupplemented phasecellsgrownatpH5.8havea100-to1000-foldwith50 m g/mlnalidixicacid. greatersurvivalratethandocellsgrownatpH7.6 S.baildon growninTSBN(10mlin16 3 150 (FosterandHall,1990).Acidtoleranceresponseismmscrew-cappedtesttubes)at37 8 Cfor48hwas
W.R.Weissingeretal./ InternationalJournalofFoodMicrobiology62(2000)1231311252 4usedtoinoculateTSBNandnalidixicacid-sup-asdescribedabove,were,respectively,diluted102 7 2 3 2 6plementedtomatojuiceatpH4.8and5.8.Culturesand10or10and10insteriledeionized incubatedat37 8 Cweretransferred(1ml)to10mlofwater.CellsgrowninTSBNwereusedtoinoculate TSBNandtomatojuicetwiceat24-hintervals.lettuce,whereascellsgrownintomatojuicewere Theseculturesservedasinoculaforstudiestousedtoinoculatetomatoes.Dilutedcellsuspensions determinechangesinacidtoleranceofcellsplated(10ml)wereaddedtolettuceortomatoesand onacidiedTSA(pH4.5,4.8,5.8,and7.2)andthoroughlymixedtogivedesiredpopulationsof3401growthpatternsinTSBN(pH7.2)andtomatojuice1010cfu/g(highinoculum)and1010cfu/g (pH4.8and5.8).(lowinoculum).Producewassubjectedtotreatment withchlorinatedwaterandanalyzedwithin20min ofinoculation.2.2.1.StudiesusingacidiedTSAN CellsgrowninTSBN(pH7.2)andnalidixic2.3.3.Treatmentoflettuceandtomatoeswith acid-supplementedtomatojuice(pH4.8and5.8) chlorinatedwater wereseriallydilutedinsterile0.1%peptoneand Treatmentsolutionsforshreddedlettuceconsisted surfaceplated(0.1mlinduplicate)onTSANatpH of120and200 m g/mlfreechlorinepreparedby 4.5,4.8,5.8,and7.2.Plateswereincubatedat37 8 C combiningsodiumhypochlorite(Aldrich,Milfor24hbeforecolonieswerecounted. waukee,WI,USA)with0.05Mpotassiumphosphate buffer(pH6.8,4 8 C);deionizedwater(4 8 C)served2.2.2.StudiesusingTSBNandtomatojuice asacontrol.Treatmentsolutionsfordicedtomatoes S.baildon growninTSBNandnalidixicacidconsistedof120and200 m g/mlfreechlorinein0.05 supplementedtomatojuice(pH4.8and5.8)was Mpotassiumphosphatebuffercontaining0.21%2 3inoculated[1mlofundilutedordiluted(10)24-h calciumchloride(pH6.8,4 8 C);deionizedwater culture]into10mlofthesamemediumandthetwo containing0.21%calciumchloride(4 8 C)servedasa othermedia,i.e.usingathree-mediumfull-factorial control.Calciumchloridewasaddedtochlorinated design.Culturesincubated0,12,24,36,and48hat watertosimulatecommercialapplicationprocedures. 37 8 Cwereseriallydilutedinsterile0.1%peptone Texturalqualityofdicedtomatoesispreservedby andsurfaceplated(0.1mlinduplicate)onTSAN thispractice.Freechlorinewasmeasuredwitha (pH7.2).Plateswereincubated24hat37 8 Cbefore chlorinetestkit(Hach,Ames,IA,USA). colonieswerecounted. Inoculatedshreddedlettuce(25g)ordiced tomatoes(25g)wereplacedinastomacherbagand2.3.Efcacyofchlorineasadisinfectant 225mlofchlorinatedwaterorchlorinatedwater containing0.21%calciumchloride,respectively,was2.3.1.Sourcesanddescriptionoflettuceand added.Vegetableswerevigorouslyshakenfor40s, tomatoes treatmentsolutionsweredecanted,and225mlof Shreddediceberglettuce( Lactucasativa )(0.125 lactosebroth(Difco,Detroit,MI,USA)supinch,ca.00.32cm)in5-lb(ca.11-kg)bagsanddiced plementedwith50 m gofnalidixicacidperml Romacv.tomatoes( Lycopersciumesculentum ) (LBN)wereimmediatelyadded.Themixtureof [0.375inch(ca.0.95cm)cut]in5-lbtrayswere vegetablesandLBNwashomogenizedina obtainedfromcommercialfresh-cutproduceprostomacherfor30satmediumspeed. cessors.Seedshadbeenremovedfromthediced Quadruplicate0.25-mlsamplesandduplicate0.1tomatoes.Experimentsusinglettuceandtomatoes mlsamplesofhomogenatecontaininghighinocula wereinitiatedwithin2hofreceiptfromprocessors. weresurfaceplatedonTSANandXLDagar(Difco); duplicatesamples(0.1ml)seriallydilutedinsterile 0.1%peptonewaterwerealsoplatedonTSANand2.3.2.Inoculationoflettuceandtomatoes XLDagar.Plateswereincubatedat37 8 Cfor24h Shreddedlettuce(300g)ordicedtomatoes(300 beforepresumptive S.baildon colonieswere g)at4 8 Cwereplacedinastainlesssteelpan.To counted. prepareinocula,cellsgrowninTSBN(pH7.2)or Homogenatesofshreddedlettuceordiced nalidixicacid-supplementedtomatojuice(pH4.8),
126 W.R.Weissingeretal./ InternationalJournalofFoodMicrobiology62(2000)123131tomatoesinLBNwereincubatedat37 8 Cfor24h,water(1:3ratio)for10min.Rinsewaterwas thentransferred(1ml)toselenitecysteinebrothdecantedandtomatoeswerediced[ca.0.375inch (Difco)containing50 m gofnalidixicacidperml.(0.95cm)cut]andwashedinsterile0.05MpotasAfterincubatingtubesat37 8 Cfor24h,culturessiumphosphatebuffer(pH6.8)containing0.21% werestreakedonbismuthsulteagar(BSA,Difco)calciumchloride(1:3ratio)bygentlymixingfor40 andXLDagar.Plateswereincubatedat37 8 Cfor24s.Dicedtomatoeswerethenseparatedfromthe h,thenexaminedforpresumptive S.baildon calciumchloridesolutionandmostoftheseeds,then colonies.Coloniesofpresumptive S.baildon formedallowedtodrainfor5min. onBSAandXLDagarwererandomlyselectedforTwo600-gquantitiesofdicedtomatoeswere conrmationusingbiochemicaltests.placedinseparatestainlesssteelbowls,inoculated with20mloflowinoculumorhighinoculum(see2.4.SurvivalandgrowthofS.baildoninlettuce Section2.3.2),andthoroughlymixed.Tomatoes andtomatoes (100g)weresealedinplasticbagsandstoredat21 or30 8 C.Afterstoringfor24,48,or72hat21or2.4.1.Storageofinoculatedlettuceandtomatoes 30 8 C,thoroughlymixedsamples(25g)werecomat4 8C binedwith225mlofLBNinastomacherbagand Commerciallypreparedicebergshreddedlettucehomogenizedfor30s.Analysisforpresenceand/or (2270g)anddicedRomacv.tomatoes(2270g)atpopulationsof S.baildon wasasdescribedfor 4 8 Cwereseparatelyplacedinlargesterilestainlesssamplestreatedwithchlorinatedwater. steelbowls.Cellsusedtoinoculatelettuceand tomatoesweregrowninTSBNandnalidixicacid-2.5.Statisticalanalysis supplementedtomatojuice(pH4.8),respectively. Thirtymillilitersofdilutedcellsuspensionscon-Allexperimentswerereplicatedthreetimes.In taininghighinoculumorlowinoculum(seeSectionexperimentsinvolvingtestsforacidtolerance,one 2.3.2)werecombinedwithlettuceandtomatoes,andsampleofeachcombinationoftestparameterswas thoroughlymixed.Inoculatedlettuce(450g)andanalyzed;triplicatesampleswereanalyzedinchlortomatoes(450g)wereplacedin2-lb(ca.4.4-kg)inetreatmentstudiesandduplicatesampleswere bagsandsealedtoachieveasurface/weightratioanalyzedinstoragestudies.Dataweresubjectedto similartothatof5-lb(ca.11-kg)bagswhicharetheStatisticalAnalysisSystem(SAS;SASInstitute, morecommonlyusedincommercialfoodserviceCary,NC,USA)foranalysisofvariance(ANOVA) operations.Samples(25g)wereanalyzedforpres-andDuncan'smultiplerangeteststodetermine enceand/orpopulationsof S.baildon initially(daysignicantdifferences(a5 0.05)betweenmean 0)andafterstoragefor2,5,8,and12daysat4 8 C.values. Onday0,sampleswereanalyzedwithin15minof inoculatingwith S.baildon .Lettuceandtomatoes werethoroughlymixedimmediatelybeforesamples 3.Resultsanddiscussion (25g)wereremoved,combinedwith225mlofLBN inastomacherbag,andhomogenizedfor30sat3.1.Studiesonacidtolerance mediumspeed.Analysisforpresenceand/orpopulationsof S.baildon wasasdescribedforsamplesThepopulationsof S.baildon in24-hTSBN(pH treatedwithchlorinatedwater(seeSection2.3.3).7.2),tomatojuice(pH5.8),andtomatojuice(pH 4.8)cultureswere8.90,8.68,and7.84logcfu/ml,102.4.2.Storageofinoculatedtomatoesat21and respectively.Whilethenumberofcellsintomato30 8C juiceatpH4.8wassignicantlylessthannumbersin Romacv.tomatoespurchasedfromalocalsuper-TSBNortomatojuiceatpH5.8,cellsgrownineach marketwerewashedbyagitatinginchlorinatedmediumdidnotdifferintheirabilitytoform water(200 m g/mlchlorine;1:3,wt/vol,tomatoes/coloniesonTSANatpH7.2,5.8,4.8,and4.5. water)at22 6 2 8 Cfor10min.ChlorinatedwaterwasColoniesformedonTSANatpH4.8and,pardecantedandtomatoeswererinsedinsteriletapticularly,pH4.5weresmallerthanthoseformedon
W.R.Weissingeretal./ InternationalJournalofFoodMicrobiology62(2000)123131127TSANatpH7.2and5.8but,withineachbroth mediuminwhichcellsweregrownbeforeplatingon TSAN,thenumberofcoloniesformedwasnot signicantly(a5 0.05)affectedbythepHof acidiedTSAN.Acidadaptationof Salmonella typhimurium uponexposuretoacidicenvironments hasbeenreportedbyseveralresearchers(Leyerand Johnson,1992,1993;FosterandHall,1990;Leeet al.,1994).Inthosestudies,cellsexposedtoenvironmentsinthepHrangeof5.55.8producedacid shockproteinswhichrenderedthemcross-protected againstsubsequentexposuretostressenvironments. Inourstudy,toleranceof S.baildon topH4.5was notinuencedbythepH(4.5,5.8,or7.2)ofthe mediuminwhichtheyhadgrown,suggestingthat thisstrainpossessesinnatetolerancetopHaslowas 4.5.Therewasnoevidencethatanincreasedacid toleranceresponsewasinducedin S.baildon culturedintomatojuiceatpH4.8. ShowninFig.1aregrowthcurvesfor S.baildon culturedinTSBN(pH7.2)andtomatojuice(pH5.8 and4.8),theninoculated(1ml),usingafull-factorialdesign,into10mlofeachmediumandincubated upto48hat37 8 C.TherateofgrowthinTSBNwas morerapidthanintomatojuiceatpH5.8which,in turn,wasmorerapidthanintomatojuiceatpH4.8 duringtherst24h,regardlessofthemediumin whichcellswereoriginallycultured.Between24and 48h,populationsremainedstable.However,cellsFig.1.Populationsof S.baildon grownfor24hat37 8 CinoriginallygrownintomatojuiceatpH5.8,thenTSBN,pH7.2(A);tomatojuice,pH5.8(B);andtomatojuice,pHinoculatedintoTSBNortomatojuice(pH4.8or4.8(C),theninoculated(1ml)into10mlofTSBN,pH7.2( s ); tomatojuice,pH5.8( h );ortomatojuice,pH4.8( n ),and5.8),declinedby2.12.5logcfu/mlbetween2410incubatedupto48hat37 8 C.and48h;cellsoriginallygrownintomatojuiceat pH4.8followedasimilarpopulationdecreasepatternwheninoculatedintoTSBNbutnottomatowerenotevidentthroughoutthe48-hincubation juice.Itappearsthatoncemaximumpopulationsofperiod.Thetimesrequiredtoreachmaximumpopucellsarereached,decreasesoccurintomatojuice,lationsinthethreemediawerelongercomparedwith regardlessoftheoriginalculturemedium,althoughthetimesrequiredinstudiesusinghigherinocula thistrendisdelayedintheorderofjuiceatpH(Fig.1).Incontrasttoourstudy,Weietal.(1995) 4.8 . juiceatpH5.8 . TSBN.reportedthat Salmonellamontevideo growninTSB Fig.2showsgrowthcurvesfor S.baildon culturedat37 8 Cfor24hunderwentanapproximate2-log10inTSBN(pH7.2)andtomatojuice(pH5.8andreductioninpopulationwheninoculatedintojuice2 34.8),thendiluted10beforeinoculating1mlinto(pH4.2)preparedfromfreshtomatoes. 10mlofeachmedium,usingafull-factorialdesign. TrendsaresimilartothoseshowninFig.1,usinga3.2.Efcacyofchlorinationtreatmentoflettuce higherinoculum.However,decreasedpopulationsof andtomatoes cellsoriginallygrownintomatojuiceatpH5.8were delayeduntilafter36handdecreasesinpopulationsTheefcacyoftreatmentwith120and200 m g/ml ofcellsoriginallygrownintomatojuiceatpH4.8chlorineinkilling S.baildon inoculatedintoshred-
128 W.R.Weissingeretal./ InternationalJournalofFoodMicrobiology62(2000)123131greateronlettucethanontomatoes.Acomparisonof thenumberofcellsrecoveredfromhigh-inoculum controlsamplesrelativetothenumberinoculated ontolettuce(3.60logcfu/g)andtomatoes(3.8610logcfu/g)revealsthatahigherpercentageofcells10diedorwerenotrecoveredfromtomatoes(86%) comparedtolettuce(49%)betweenthetimeof inoculationandanalysis,whichdidnotexceed15 min.Inbothinstances,someviablecellswouldbe removedwithwater(control)aswellasthechlorine solutionsusedtotreattheproduce.However,the greaterreductioninnumberofviablecellsrecovered fromtomatoesmaybeattributed,inpart,toacidpH stress.ThepHofdicedtomatoesusedinthisstudy was4.51 6 0.02.ThepHofshreddedlettucewas 6.08 6 0.04.Thelargestandarddeviationsinmean valuesshouldnotbeconsideredexceptional.Despite attemptstoevenlydistributeinoculaandcollect homogenoussamples,microbialcellsareknownto lodgedisproportionallyindamagedandcuttissues, resultinginsubstantialdifferencesinnumbersof cellspresentindifferentsamplesand/orexposedto treatmentsolutions.Thisphenomenonhasbeen observedinotherstudiesinvolvingvariousdisinfectantsandawiderangeofrawfruitsandvegetables (Beuchat,1998). Othershaveobservedlessthan1logreductions10in Salmonella oniceberglettucetreatedwith200 m g/mlchlorine(Beuchatetal.,1998).TreatmentofFig.2.Populationsof S.baildon grownfor24hat37 8 Cinwholetomatoeswithupto310 m g/mlchlorinekilledTSBN,pH7.2(A);tomatojuice,pH5.8(B);andtomatojuice,pH22 31.36logcfuof Salmonellamontevideo percmof4.8(C);diluted10inrespectiveuninoculatedmedia,then10inoculated(1ml)into10mlofTSBN,pH7.2( s );tomatojuice,skinsurface(Zhuangetal.,1995).TreatmentwaspH5.8( h );ortomatojuice,pH4.8( n ),andincubatedupto48hlesseffectiveinkillingthepathogenafteritsinltra-at37 8 C.tionintothestemscartissueoftomatoes.Weietal. (1995)reportedthattreatmentoftomatoeswithupto dedlettuceanddicedtomatoesisshowninTable1.250 m g/mlchlorinewasconsiderablymoreeffective Thepathogenwasdetectedbyenrichmentinallinkilling S.montevideo .Removalofthisserotype treatedsamplesoflettuceinitiallyinoculatedwithfromthesurfaceofwholetomatoesbytreatment low(0.60logcfu/g)orhigh(3.60logcfu/g)with412%trisodiumphosphatehasalsoshown1010populationofcells.Indicedtomatoesinoculatedpromise(ZhuangandBeuchat,1996). withlow(0.86logcfu/g)orhigh(3.86log1010cfu/g)numbersof S.baildon ,allsamplesexcept3.3.Survivalandgrowthasaffectedby ninelow-inoculumtomatosamplestreatedwith200 temperature m g/mlchlorinewerepositiveforthepathogenafter treatment. S.baildon wasnotdetectedinthree.ShowninTable2arepopulationsof S.baildon in Populationsof S.baildon inhigh-inoculumlettuceshreddedlettuceinoculatedwithlow(0.28log10andtomatoeswerereducedbylessthan1logasacfu/g)andhigh(3.28logcfu/g)inoculaandin10resultoftreatmentwith120or200 m g/mlchlorine.dicedtomatowithlow(0.40logcfu/g)andhigh10Thepercentageofcellsinactivatedbychlorinewas(3.40logcfu/g)inocula,andstoredat4 8 Cforup10
W.R.Weissingeretal./ InternationalJournalofFoodMicrobiology62(2000)123131129 Table1 Presenceandpopulationsof Salmonellabaildon detectedoninoculatedshreddedlettuceanddicedtomatoestreatedwithchlorinatedwater ProductTreatmentNo.of25-gsamples(outPopulationReductionbcofnine)positivebyenrich-(logcfu/g)(%)10 amentaftermeasurement Lowinoc.Highinoc. dShreddedControl993.31(2.59)0 lettuceChlorine(120ppm)992.53(2.48)83 Chlorine(200ppm)992.49(2.09)85dDicedControl993.00(2.12)0 tomatoChlorine(120ppm)992.73(1.95)45 Chlorine(200ppm)692.69(2.15)51 aPopulationsof S.baildon onlettucebeforetreatmentfor40sat4 8 C(ascalculated,basedoncfu/mlofinoculum)were0.60logcfu/g10(lowinoculum)and3.60logcfu/g(highinoculum).Populationsof S.baildon ontomatoesbeforetreatmentwere0.86logcfu/g(low10 10inoculum)and3.86logcfu/g(highinoculum).10 bPopulationsof S.baildon detectedinhigh-inoculumlettuceandtomatoesaftertreatment.Standarddeviationsareshowninparentheses. Thelowerlimitofdetectionof S.baildon is10cfu/g.cPercentreductioninnumberof S.baildon detectedinhigh-inoculumlettuceandtomatoescomparedtorespectivecontrols.dControlsforlettuceandtomatoesconsistedofdeionizedwateranddeionizedwatercontaining0.21%calciumchloride,respectively. Table2 Presenceandpopulationsof Salmonellabaildon detectedoninoculatedshreddedlettuceanddicedtomatoesstoredat4 8 Cforupto12days ProductStorageNo.of25-gsamples(outofPopulationReductionabctimesix)positivebyenrichment(logcfu/g)(%)10 (days) Lowinoc.Highinoc. Shredded0663.16(2.54)0 lettuce2663.24(3.04) 1 20 5663.07(3.01)52 8662.69(2.43)66 12161.81(1.54)96 Diced0063.17(2.41)0 tomato2063.00(2.21)33 5062.73(1.73)65 8032.65(2.41)70 12011.80(1.92)96 aPopulationsof S.baildon atday0(ascalculated,basedoncfu/mlofinoculum)were0.28and3.28logcfu/gonlettuceinoculated10withlowandhighpopulations,respectively.Populationsof S.baildon atday0were0.40and3.40logcfu/gontomatoesinoculatedwith10lowandhighpopulations,respectively.bPopulationsof S.baildon detectedinhigh-inoculumlettuceandtomatoes.Standarddeviationsareshowninparentheses.Thelowerlimit ofdetectionof S.baildon is10cfu/g.cPercentreductioninnumberof S.baildon detectedinhigh-inoculumlettuceandtomatoescomparedtonumbersdetectedonday0.to12days.Regardlessofinoculumlevel,thepatho-nocellsweredetectedinsix25-gsamples.Aswith genwasdetectedbyenrichmentinalllettucesam-disinfectionstudies,deathof S.baildon isattributed, plesanalyzedondays0,2,5,and8ofstorage.Afterinpart,toacidstressimposedoncellsinoculated 12daysofstorage, S.baildon wasrecoveredfromintotomatoes.Allhigh-inoculumsamples(sixofsix) oneofsixsamplesinitiallycontainingalowin-oftomatoesstoredat4 8 Cfor0,2,or5days oculumandsixofsixsamplescontaininghighcontainedviable S.baildon .Graduallossofviability inoculum.Thepathogenwasnotdetectedinofcellsinlettuceandtomatoesinitiallycontaininga tomatoesreceivingalowinoculum,regardlessofthehighinoculawasevidentasstoragetimeprogressed. storagetime;within15minofinoculation(day0),Populationsof S.baildon inhigh-inoculumlettuce
130 W.R.Weissingeretal./ InternationalJournalofFoodMicrobiology62(2000)123131andtomatoeswerereducedby1.37log(96%) References10duringthe12-daystorageperiod.Aswithdisinfectantstudies,largestandarddeviationsinvaluesAdams,M.R.,Hartley,A.D.,Cox,L.J.,1989.Factorsaffectingthe efcacyofwashingproceduresusedintheproductionofoccurred.Thelownumbersofhigh-inoculumtomatopreparedsalads.FoodMicrobiol.6,6977.samples(threeofsixat8daysandoneofsixat12Asplund,K.,Nurmi,E.,1991.Thegrowthofsalmonellaeindays)showntocontain S.baildon whensubjectedtotomatoes.Int.J.FoodMicrobiol.13,177182.enrichmentdoesnotcorrelatewithlogcfu/g10Beuchat,L.R.,1996.Pathogenicmicroorganismsassociatedwithrecoveredbydirectplating.Anexplanationforthisfreshproduce.J.FoodProt.59,204316.phenomenonisnotevident.SurvivalofsomeofanBeuchat,L.R.,1998.Surfacedecontaminationoffruitsand vegetableseatenraw:areview.WHO/FSF/FOS/98.2,42pp.initialpopulationof3.40logcfuof S.baildon per10Beuchat,L.R.,1999.Survivalofenterohemorrhagic Escherichiagramofsomeofthedicedtomatosamplesstoredatcoli O157:H7inbovinefecesappliedtolettuceandthe4 8 Cfor12daysdid,nevertheless,occur.effectivenessofchlorinatedwaterasadisinfectant.J.FoodExperimentstodeterminesurvivalandgrowthProt.62,845849.characteristicsof S.baildon indicedtomatoes(pHBeuchat,L.R.,Brackett,R.E.,1990.Survivalandgrowthof Listeriamonocytogenes onlettuceasinuencedbyshredding,4.40 6 0.01)storedat21 8 Crevealedthataninitialchlorinetreatment,modiedatmospherepackagingandtem-populationof0.79logcfu/gincreasedto5.32,10perature.J.FoodSci.55,755758,870.7.60,and8.10logcfu/gwithin24,48,and72h,10Beuchat,L.R.,Nail,B.V.,Alder,B.B.,Clavero,M.R.S.,1998.respectively;respectivepopulationsintomatoesEfcacyofsprayapplicationofchlorinatedwaterinkillingstoredat30 8 Cwere7.0,8.73,and8.55logcfu/g.pathogenicbacteriaonrawapples,tomatoes,andlettuce.J.10Aninitialpopulationof3.46logcfu/gincreasedtoFoodProt.61,13051311.10CentersforDiseaseControlandPrevention,1999.Personal7.30,8.39,and8.06logcfu/goftomatoesstored10communication.for24,48,and72hat21 8 C;respectivepopulationsDeRoever,C.,1998.Microbiologicalsafetyevaluationsandintomatoesstoredat30 8 Cwere7.30,7.90,and7.94recommendationsonfreshproduce.FoodControl9,321347.logcfu/g.ThetremendousincreaseintheinitialFoster,J.W.,Hall,H.K.,1990.Adaptiveacidicationresponseof10lowpopulation(0.79logcfu/g)by4.53and6.21Salmonellatyphimurium .J.Bacteriol.172,771778.10Foster,J.W.,Spector,M.P.,1995.How Salmonella surviveagainstlogswithin24hat21and30 8 C,respectively,clearlytheodds.Annu.Rev.Microbiol.49,145174.demonstratestheabilityof S.baildon togrowwellinFrancis,G.A.,Thomas,C.,O'Beirne,D.,1999.Themicrobiologi-dicedtomatoes.Othershavereportedtheabilityofcalsafetyofminimallyprocessedvegetables.Int.J.FoodSci.Salmonella togrowinripecuttomatoes.AsplundTechnol.34,122.andNurmi(1991)observedthatabout2logcfuofHedberg,C.W.,Angulo,F.J.,White,K.E.,Langkop,C.W.,Schell,10W.L.,Stobierski,M.G.,Schuchat,A.,Besser,J.M.,Dietrich,S.,Salmonellainfantis pergramoftomatoes(pH4.11Helsel,L.,Grifn,P.M.,McFarland,J.W.,Osterholm,M.T.,4.22)increasedbyabout6logcfu/gafter24hat101999.Outbreaksofsalmonellosisassociatedwitheatingun-22 8 C.At30 8 C,theincreasewasabout1log10cookedtomatoes:implicationsforpublichealth.Epidemiol.higher. S.montevideo wasreportedtogrowinInfect.122,385393.choppedtomatoes(pH4.1)storedat20or30 8 CHedberg,C.W.,Osterholm,M.T.,1993.Outbreaksoffoodborne andwaterborneviralgastroenteritis.Clin.Microbiol.Rev.6,(Zhuangetal.,1995)andinwoundedareasofwhole199210.ripetomatoesatpH4.24.4(Weietal.,1995).Lee,I.S.,Slonczewski,J.L.,Foster,J.W.,1994.AlowpHInsummary,resultsfromstudiesonthebehaviorinducible,stationary-phaseacidtoleranceresponsein Sal -of S.baildon inoculatedontoshreddedlettuceandmonellatyphimurium .J.Bacteriol.176,14221426.dicedtomatoesconrmobservationsmadebyothersLeyer,G.J.,Johnson,E.A.,1992.Acidadaptationpromotes survivalof Salmonella spp.incheese.Appl.Environ.Micro-thatseveralserotypesofsalmonellaecangrowonbiol.58,20752080.thesesaladvegetablesandaredifculttoeliminateLeyer,G.J.,Johnson,E.A.,1993.Acidadaptationinducescross-bytreatmentwithupto200 m g/mlchlorine.Out-protectionagainstenvironmentalstressesin Salmonellabreaksofsalmonellosisassociatedwithrawfruitstyphimurium .Appl.Environ.Microbiol.59,18421847.andvegetablesarelikelytooccurinthefuturebutNguyen-the,C.,Carlin,F.,1994.Themicrobiologyofminimally processedfreshfruitsandvegetables.Crit.Rev.FoodSci.thenumbercanbeminimizedbyapplicationofNutr.34,371401.agronomic,distribution,processing,andstorageSeymour,I.J.,1999.Reviewofcurrentindustrypracticeonfruitpracticesdesignedtopreventcontaminationwithandvegetabledecontamination.ReviewNo.14,CampdenandSalmonella andmaximizetheefcacyofdisinfectionChorleywoodFoodResearchAssociation,ChippingCampden,treatments.UK,36pp.
W.R.Weissingeretal./ InternationalJournalofFoodMicrobiology62(2000)123131131 Wei,C.I.,Huang,T.S.,Kim,J.M.,Lin,W.F.,Tamplin,M.L.,against Listeriamonocytogenes onfresh-cutvegetables.Food Bartz,J.A.,1995.Growthandsurvivalof Salmonellamon -Microbiol.13,311321. tevideo ontomatoesanddisinfectionwithchlorinatedwater.J.Zhuang,R.-Y.,Beuchat,L.R.,1996.Effectivenessoftrisodium FoodProt.58,829836.phosphateforkilling Salmonellamontevideo ontomatoes.Lett. Wood,R.C.,Hedberg,C.,White,K.,1991.AmultistateoutbreakAppl.Microbiol.22,97100. of Salmonellajaviana infectionsassociatedwithrawtomatoes.Zhuang,R.-Y.,Beuchat,L.R.,Angulo,F.J.,1995.Fateof Sal In:CDCEpidemicIntelligenceService,40thAnnualConfermonellamontevideo onandinrawtomatoesasaffectedby ence,Atlanta.U.S.DepartmentofHealthandHumanServices,temperatureandtreatmentwithchlorine.Appl.Environ.MicroPublicHealthService,p.69,abstracts.biol.61,21272131. Zhang,S.,Farber,J.M.,1996.Theeffectsofvariousdisinfectants
INFLUENCE OF TEMPERATURE DIFFERENTIAL BETWEEN TOMATO FRUITS AND POSTHARVEST WATER ON THE INTERNALIZATION OF SALMONELLA AT AMBIENT AND REFRIGERATION TEMPERATURES By ASHLEY NICOLE TURNER A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR TH E DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2014
Â© 2014 Ashley Nicole Turner
To my mom and dad, who have always supported me and believed in me. You have been the driving force behind my goals and dreams and have allowed me to follow them. I attribute everything I have accomplished in life thus far to you. To John, without you I truly do not believe I would have been able to complete this journey; thank you for always seeing the light in me, especially when I was not able to.
4 ACKNOWLEDGMENTS I would like to first thank my major advisor Dr. Michelle Danyluk; I cannot express my gratitude for the opportunity to be a graduate research assistant in her lab. She has offered more than just guidance in research and laboratory techniqu es, she has been a mentor as well. She was always willing to answer any questions I had and offer advice and wisdom on many topics. Dr. Danyluk has always made herself available to me and her other students, which I am very thankful for. She has opened man y doors for me by encouraging my attendance to educational luncheons, conferences, and other and I truly would not have wanted to learn and work with anyone else. I w ant to also thank my supervisory committee members, Dr. Keith Schneider and Dr. Jerry Bartz. I would like to thank Dr. Schneider for his passion for food microbiology. In his classes, he discusses real life experiences and scenarios which help to make the topics more applicable, understandable, and interesting. I would like to thank Dr. Jerry Bartz for his vast amount of knowledge of tomato physiology and the topic of internalization in tomatoes. I am very grateful for his expertise in this field and his in sights into different ways to approach my research. I thank you all for your time, knowledge, advice, and commitment to me and my research. Finally I would like to thank everyone in the lab for always being willing to hel p: Lorrie Friedrich, Travis Chap in , Pardeepinder Brar, Zeynal Toplacengiz, Gwen Lundy, Karen Plant, and Luis Martinez. I honestly would not have been able to complete this work without the efforts of every one of you and I owe you more than I could ever repay for your assistance and moral support.
5 TABLE OF CONTENTS page ACKNOWLEDGMENT S ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 ABSTRACT ................................ ................................ ................................ ................. 10 0 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 12 2 LITERATURE REVIEW ................................ ................................ .......................... 18 Foodborn e Illness ................................ ................................ ................................ ... 18 Salmonella ................................ ................................ ................................ .............. 20 Importance in Produce Related Foodborne Illness Outbreaks ......................... 22 Potential Sources of Contamination of Produce ................................ ............... 24 Tomatoes ................................ ................................ ................................ ................ 26 Postharvest Processing of Tomatoes ................................ ............................... 29 Significance of Tomato Related Outbreaks ................................ ...................... 34 Internalization ................................ ................................ ................................ ......... 38 Preharvest Internalization ................................ ................................ ................. 39 Postharvest Internalization ................................ ................................ ............... 41 Temperature Differential and Internalization ................................ .............. 43 Submersion Depth/Time ................................ ................................ ............ 46 Time after Harvest ................................ ................................ ..................... 48 Internalization of Salmonella in Tomatoes ................................ ........................ 50 3 MATERIALS AND METHODS ................................ ................................ ................ 54 Tomatoes ................................ ................................ ................................ ................ 54 Salmonella Strains ................................ ................................ ................................ .. 54 Inoculum P reparation ................................ ................................ .............................. 54 Submersion Media ................................ ................................ ................................ .. 55 Tomato Submersion ................................ ................................ ................................ 56 Tissue Ext raction ................................ ................................ ................................ .... 56 Enumeration of Pathogens ................................ ................................ ..................... 57 Water Ingress ................................ ................................ ................................ ......... 58 Statistical An alysis ................................ ................................ ................................ .. 58 4 RESULTS ................................ ................................ ................................ ............... 59 pH ................................ ................................ ................................ ........................... 59
6 Oxidation Reduction Potential ................................ ................................ ................ 59 Difference in Tomato Weight ................................ ................................ .................. 59 Pathogen Recovery ................................ ................................ ................................ 60 Internalizatio n into Green Tomatoes (21Â°C) ................................ ............................ 61 Internalization into Red Tomatoes (21Â°C) ................................ ............................... 63 Internalization in 4Â°C Red Tomatoes ................................ ................................ ...... 65 Comparison of Internalization Among all Maturities and Pulp Temperatures .......... 66 5 DISCUSSION ................................ ................................ ................................ ......... 98 6 CONCLUSIONS AND FUTURE WORK ................................ ............................... 106 LIST OF REFERENCES ................................ ................................ ............................. 109 BIOGRAPH ICAL SKETCH ................................ ................................ .......................... 117
7 LIST OF TABLES Table page 4 1 pH of water in submersion media. ................................ ................................ ...... 68 4 2 Oxidation Reduction Potential (ORP) measurements of water in submersion media. ................................ ................................ ............................. 69 4 3 Difference in tomato weight (g) of green tomatoes (21Â°C) before and after submersion. ................................ ................................ ................................ ........ 70 4 4 Difference in tomato weight (g) of red tomatoes (21Â°C) before and after submersion. ................................ ................................ ................................ ........ 70 4 5 Difference in tomato weight (g) of red tomatoes (4Â°C) before and after submersion. ................................ ................................ ................................ ........ 70 4 6 Salmonella populations (log MPN/segment) recovered from segment A, just below the stem scar. ................................ ................................ ........................... 71 4 7 Salmonella populations (log MPN/segment) recovered from segment B, middle core segment. ................................ ................................ ......................... 72 4 8 Salmonella populations (log MPN/segment) recovered from segment C, just above the blossom end. ................................ ................................ ..................... 73
8 LIST OF FIGURES Figure page 4 1 Average Salmonella populations plus or minus standard deviation recovered from Segment A ( n=6 12) in green tomatoes at ambient temperatures ( 21Â°C) . . 74 4 2 Average Salmonella populations plus or minus standard deviation recovered from Segment B (n=6) in green tomatoes at ambient temperatures (21Â°C) . ....... 75 4 3 Average Salmonella populations plus or minus standard deviation recovered from Segment C ( n=6) in green tomatoes at ambient temperatures (21Â°C) . ...... 76 4 4 Internalized Salmonella populations recovered from (i) segment A, (ii) segment B , and (iii) segment C of green tomatoes (21Â°C) in submersion media at th e 5Â°C temperature differential . ................................ ......................... 77 4 5 Internalized Salmonella populations reco vered from (i) segment A, (ii) segment B , and (iii) segment C of green tomatoes (21Â°C) in submersion media at th e 3Â°C temperature differential . ................................ ......................... 78 4 6 Internalized Salmonella populations recovered from (i) segment A, (ii) segment B , and (iii) segment C of green tomatoes (21Â°C) in submersion media at th e 0Â°C tempe rature differential . ................................ .......................... 79 4 7 Internalized Salmonella populations recovered fr om (i) segment A , (ii) segment B, and (iii ) segment C of green tomatoes (21Â°C) in submersion media at th e 3Â°C temperature differential . ................................ ......................... 80 4 8 Internalized Salmonella populations recovered fro m (i) segment A , (ii) segment B , an d (iii) segment C of green tomatoes (21Â°C) in submersion media at th e +5Â°C temperature differential . ................................ ........................ 81 4 9 Average Salmonella populations plus or minus standard deviation recovered from Segment A ( n=6 12) in red tomatoes at ambient temperatures (21Â°C) . ..... 82 4 10 Average Salmonella populations plus or minus standard deviation recovered from Segment B ( n=6) in red tomatoes at ambient temperatures (21Â°C). .......... 83 4 11 Average Salmonella populations plus or minus standard deviation recovered from Segment C ( n=6) in red tomatoes at ambient temperatures (21Â°C). .......... 84 4 12 Internalized Salmonella populations recovered from (i) segment A , (ii) segment B, and (ii i) segment C of red tomatoes (21Â°C) in submersion media at the 5Â°C temperature differential. ................................ ................................ .... 85
9 4 13 Internalized Salmonella pop ulations recovered fro m (i) segment A , (ii) segment B, and (iii) segment C of red tomatoes (21Â°C) in submersion media at the 3Â°C temperature differential. ................................ ................................ .... 86 4 14 Internalized Salmonella populations recovered fro m (i) segment A , (ii) segment B, and (iii) segment C of red tomatoes (21Â°C) in submersion media at the 0 Â°C temperature differential. . ................................ ................................ .... 87 4 15 Internalized Salmonella populations recovered fr om (i) segment A , (ii) segment B, and (iii) segment C of red tomatoes (21Â°C) in submersion media at the +3Â°C temperature differential. ................................ ................................ ... 88 4 16 Internalized Salmonella populations recovered fro m (i) segment A , (ii) segment B, and (iii) segment C of red tomatoes (21Â°C) in submersion media at the + 5Â°C temperature differential. ................................ ................................ ... 89 4 17 Average Salmonella populations plus or minus standard deviation recovered from Segment A ( n=6 12) in red tomatoes at refrigeration temperatures (4Â°C).. ................................ ................................ ................................ ................. 90 4 18 Average Salmonella populations plus or minus standard deviation recovered from Segment B ( n=6) in red tomatoes at r efrigeration temperatures (4Â°C) . ...... 91 4 19 Average Salmonella populations plus or minus standard deviation recovered from Segment C ( n=6) in red tomatoes at r efrige ration temperatures (4Â°C) . ...... 92 4 20 Internalized Salmonella populations recovered fro m (i) segment A , (ii) segment B, and (iii) segment C of red tomatoes (4Â°C) in submersion media at th e 5 Â°C temperature differential ................................ ................................ ......... 93 4 21 Internalized Salmonella populations recovered fro m (i) segment A, (ii) segment B, and (iii) s egment C of red tomatoes (4Â°C) in submersion media at th e 3Â°C temperature differential . ................................ ................................ ........ 94 4 22 Internalized Salmonella populations recovered fro m (i) segment A , (ii) segment B, and (iii) segment C of red tomatoes (4Â°C) in submersion media at the 0Â°C temper ature differential . ................................ ................................ ......... 95 4 23 Internalized Salmonella populations recovered fr om (i) segment A , (ii) segment B, and (iii) segment C of red tomatoes (4Â°C) in submersion media at the +3Â°C temperature differential. ................................ ................................ ....... 96 4 24 Internalized Salmonella populations recovered fro m (i) segment A , (ii) segment B , and (iii) segment C of red tomatoes (4Â°C) in submersion media at the +5Â°C temperature differential. ................................ ................................ ....... 97
10 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for th e Degree of Master of Science INFLU E NCE OF TEMPERAT URE DIFFERENTIAL BETWEEN TOMATO FRUITS AND POSTHARVEST WATER ON THE INTERNALIZATION OF SALMONELLA AT AMBIENT AND REFRIGERATION TEMPERATURES By Ashley Nicole Turner August 2014 Chair: Michelle Danyluk Major: Food Science and Human Nutrition Washing steps beyond the packinghouse can provide a route for internalization of Salmonella spp. into tomatoes ; the temperature differential between water and packed tomatoes may affect Salmonella internalization. Salmonella internalization into mature green and red tomatoes at ambient (21Â°C) and refrigeration (4Â°C) pulp temperatures submerged into water at various temperature differentials, simulating repacking and fresh cut operations was evaluated. Red (4Â°C and 21Â°C) and m ature green (21Â°C) tomatoes were submerged (6 cm) into a six strain Salmonella cocktail (6 log CFU /ml), maintained at Â±5, Â±3, 0Â°C temperature differentials for up to 5 min. Following submersion, tomatoes were surface sanitized , the stem scar and blossom en d skin removed, and cores recovered. Cores were cut into three segments: (A) upper segment below stem scar; (B) middle segment; and (C) lower segment above blossom end. Salmonella populations (n=6) in each segment were enumerated by (MPN). The effect of te mperature differential and maturity on Salmonella population internalization was analyzed significance at P Salmonella populations were highest in segment A and higher in green tomatoes (21Â°C) than in red
11 tomatoes (4 a nd 21Â°C). Salmonella populations in segments B and C were higher in red tomatoes at 4Â°C than in green and red tomatoes at 21Â°C. Internalized populations were not significantly different ( P Salmonella internalized in to tomatoes under all treatment conditions, populations were low and varied greatly between tomatoes. The temperature differential between tomatoes and water beyond the initial packinghouse may not play a significant role in Salmonella internalization.
12 CHAPTER 1 INTRODUCTION Production and consumption of fresh fruits and vegetables in the United States continue to increase (USDA ERS, 2008). The variety and availability of fresh fruits and vegetables have increased as a result of new packaging technologies, increased import from other countries, and more efficient storage and transportation conditions. In addition to the increase in fresh fruit and vegetable consumption, produce related foodborne illness outbreaks have a lso increased (Scallan et al . , 2011). Both raw and minimally processed, or fresh cut produce does not undergo a pathogen reduction step prior to being sold to the consumer. Once contaminated with a pathogen, removal is difficult and the risk of further cro ss contamination can be increased (Zhuang et al., 1995; Allen et al., 2005; Stine et al., 2005; Yuk et al., 2006). Contamination of produce can occur during plant growth, harvesting, transportation, packing, processing, distribution, or at the retail and foodservice level. Steps to ensure produce safety and reduce the risk of contamination should be taken at all levels of the fruit and vegetable supply chain. Foodborne illness comes with a heavy economic burden, including medical expenses, lost wages, and product losses from recalls; the annual estimated cost of foodborne illness is between $51 and 77.7 billion (Scharff, 2012). The Center for million domestically acquired cases of foodborne illness occur every year (CDC, 2011 a ). The majority of these cases (80%) are caused by unspecified pathogens. The remaining 20% of cases are caused by one or more of several known human pathogens (CDC, 2011 a ). Of these known pathogens, t he top five contributing to foodborne illness,
13 in order, are Norovirus, nontyphoidal Salmonella , Clostridium perfringens , Campylobacter spp., Staphylococcus aureus . Many foodborne illness cases are mild and symptoms subside without medical treatment, howe ver, 128,000 hospitalizations and 3,000 deaths occur from foodborne illness annually (Scallan et al . , 2011). Salmonella causes one million cases of foodborne illness annually (Scallan et al . , 2011). Salmonella enterica is divided into nearly 2,600 serovar s many of which cause gastroenteritis. These are referred to as nontyphoidal Salmonella ( Ellermeir and Slauch, 2006; Waldner et al . , 2012). Although previously linked to products of animal origin, produce related outbreaks of Salmonella are becoming more common. A variety of fruits and vegetables have caused outbreaks of salmonellosis in the United States including cantaloupe, tomatoes, peppers, watermelon, lettuce, sprouts, and mangoes (Brackett, 1999; Sivapalasingam et al . , 2004; Scallan et al . , 2011). T omatoes are a popular commodity in the United States accounting for over $2 billion in annual farm cash receipts (USDA ERS, 2012). Tomato production is divided between fresh market and processed. Processing tomatoes accounted for 89% of all tomato producti on in 2008; however, fresh market field grown tomato production has trended higher over the past several decades (USDA ERS, 2012). Tomatoes grown for fresh market consumption are produced primarily in California and Florida with minor production existing i n all states. Supermarkets carry many tomato types including standard rounds, plum s (Roma) , , grape s, cherr ies , and a wide variety of greenhouse and hydroponic types (USDA ERS, 2012). Tomatoes specifically have been linked to 12 cases of foodborne illness since 1990, 11 of which were caused by Salmonella (Hedberg et al . , 1999; Cummings et al . ,
14 2001; CDC 2005, CDC 2007; Gupta et al . , 2007). Contamination of tomatoes with Salmonella can o ccur pre harvest and postharvest. In 2002 and 2005, surface water used for irrigation and preparation of pesticide sprays was contaminated with Salmonella Newport and led to two multistate foodborne illness outbreaks. Trace back investigations of other tom ato related outbreaks were not able to uncover a specific contamination point in the supply chain; however, it is thought that contamination occurred postharvest. This was the case in an outbreak of Salmonella Balidon infections linked to round tomatoes in 1999. Distribution sources included restaurants in several states and investigations of packing facilities and one diced tomato processor were conducted (Cumming et al . , 2001). Once contaminated, pathogen removal from tomatoes is difficult. Salmonella can attach to the skin of tomatoes and subsequently contaminate other fruits when combined in flume systems or washing tanks. Salmonella may also internalize within the tomato when contaminated water infiltrates the fruit through cuts in the cuticle and natur al apertures in the stem scar. Once inside, Salmonella may proliferate and reduction can become impossible without a heat step (Zhuang et al . , 1995). Absorption of water into tomatoes through the stem scar has been studied previously and can be explained b y the general ideal gas law, which states that any change in pressure of a gas in a container of constant volume is directly proportional to a change in tempera ture (Bartz and Showalter, 1981). Tomatoes are a type of container and are relatively constant i n volume; however, they are not completely closed. As a tomato cools, gases in numerous airspaces within the tomato will exert a reduced pressure, creating a series of partial vacuums. The gas exchange ports for mature tomato fruit are located in the
15 stem scar (Showalter, 1979). Internal partial vacuums interact with the external gas pressures at these locations ; i f the exchange ports are in water or covered with water, internal partial vacuums will draw water into the ports. If tomatoes cool while submerge d due to a negative temperature differential, water and any suspended particulate matter including viable bacteria will pass through the portals into the interior of the fruit (Studer and Kader, 1976; Showalter, 1979; Bartz and Showalter, 1981; Bartz, 1982 ). If this occurs, bacteria and other suspended particulates are considered internalized. Numerous studies have been done on the effect of negative temperature differential between tomato fruit pulp and wash water and the implications for tomato quality an d safety. These studies primarily evaluated the internalization of plant pathogens such as Pectobacterium carotovorum , Psuedomonas marginalis , and Psuedomonas aeruginosa and the resulting rot incidence. Tomatoes subjected to the largest negative temperatur e differentials reported the highest incidence of rot while those exposed to positive temperature differentials had the lowest rot incidence. These studies were performed to test extreme conditions including large temperature differentials, submersion time s, and submersion depths. The variables tested in these experiments , while common at the time of writing, are not indicative of current practices in the industry . Salmonella internalization into tomatoes was studied by Xia et al . (2011) as influenced by te mperature differential, tomato variety, and post harvest stem removal time. Tomato variety and delay between stem removal and immersion significantly affected internalization frequency while temperature differential had no significant effect on internalizat ion frequency or the population of internalized cells. A major finding of this
16 study was that out of a variety of internal tissues recovered, internalized Salmonella populations were only present in tissues directly below the stem scar which reconfirms pre vious results stating that the stem scar is the major point of entry for bacteria into tomatoes (Bartz 1982; Xia et al . , 2011). The studies previously done on internalization of bacteria into tomatoes have highlighted that time between stem removal and sub mersion, submersion depth, submersion time, and temperature differential may all be factors in the uptake of bacteria into tomatoes through the stem scar. In 2008, the Florida Department of Agriculture and Consumer Services (FDACS) made the implementation of tomato Good Agricultural Practices and tomato Best Management Practices (TGAPs and T BMPs) a requirement. Although currently in place in several tomato o perations due to the number of cases of foodborne illness linked to tomatoes in the early 2000s, these programs were mandated as of 2008 (FDACS, 2008). This program requires all postharvest water, including water in dump tanks and flume systems, to have a free chlorine level of 150ppm, have a pH level of 6.5 7.5 and be maintained at a temperature 5Â°C/10Â°F higher than the incoming fruit pulp temperature. There is also a maximum 2 minute time limit for tomatoes to be present in dump tanks (FDACS, 2008). Typically this is followed, but times could inadvertently exceed 2 minutes during breaks and equipment failure (Bartz, 1982). These recommendations regarding water quality were made for the primary packinghouse where freshly harvested tomato fruits could be much warmer than the packinghouse water supply ; however they are being carried through to the tomato supply chain to subsequent re packing, fresh cut, retail, and foodservice opera tions . The first edition of the Commodity Specific Food Safety Guidelines for the Fresh Tomato Supply Chain
17 recommend that water used in washing steps for repacking of tomatoes be maintained at a temperature 5Â° C/10 Â° F warmer than the incoming fruit temperat ure (FDA, 2006) . The second edition of the same guidance document specifies a positive temperature differential for water used in fresh cut operations. Prior to slicing, it is recommended that tomatoes are cooled in cold rooms and then washed. The water us ed for the whole tomato wash should be maintained at a temperature 5Â° C/10 Â° F warmer than the incoming fruit temperature (FDA, 2008) . Several washing steps may follow the washing and packing of tomatoes in the primary packinghouse. Maintaining warm water temperatures in these steps requires a large amount of energy . Determination of whether a positive temperature differential should be used in repacking, fresh cut, and retail/foodservice kitchens could prove to be beneficial, as the energy cost associated with a dditional washing steps could potentially be lowered . Further research in this field is necessary to determine whether this temperature differential is needed. Due to the extreme temperature differentials and submersion times used in some studies tha t have evaluated the internalization of bacteria caused by a negative temperature differential, there could be a lack of correlation between the two. Additionally, previous studies have shown that time between stem removal and immersion play a significant role in water and bacterial infiltration. In order to determine if a positive temperature differential is required in further processing steps in the tomato supply chain, research needs to be conducted on washed and waxed tomatoes where the stem scar has h ad sufficient time to dry such that its surfaces are relatively hydrophobic.
18 CHAPTER 2 LITERATURE REVIEW Foodborne Illness An outbreak of foodborne illness is defined as two or more persons experiencing a similar illness following ingestion of a common fo od which epidemiological analysis implicates as the source of the illness (CDC, 2014). Foodborne illness outbreaks have a devastating cost associated with them, from the lives of those affected to the losses of product due to recalls. It is estimated that every year in the United States, 48 million people become sick from foodborne illnesses, over 128,000 people are hospitalized, and 3,000 people die (CDC 2011 a ;Scallan et al., 2011;). This astonishing number is also coupled with the economic cost of foodb orne illness which is in the range of $10 83 billion each year (FDA, 2009). Foodborne illness outbreaks can result from bacterial, viral, and parasitic infections as well as intoxications and toxin mediated infections. The Centers for Disease Control and P revention (CDC) report estimates for two major groups of foodborne illnesses, those associated with known pathogens and those from unspecified agents. Annually, 80% of the reported illness outbreaks are related to unspecified agents while only 20% are from known pathogens. There are 31 known pathogens that cause the 20% of foodborne illnesses in the United States (CDC 2011 a ; Scallan et al., 2011). The 31 known pathogens, whose illnesses are reportable to the CDC, cause 9.4 million cases of foodborne illnes s annually. Scallan et al. (2011) estimated that 5.5 million of this the total number of foodborne illness cases were viral, 3.6 million bacterial, and 0.2 million parasitic. Viral foodborne illnesses are largely attributed to norovirus, which accounts for 58% of domestic foodborne illness cases (Scallan et al.,
19 2011). Although foodborne illness caused by viruses is most prevalent in the US, bacterial foodborne illness cause millions of cases every year; salmonellosis, infections with Salmonella spp account s for 11% of foodborne illness cases. Clostridium prefrigens and Camplyobacter spp cause the second and third largest number of foodborne illnesses, resulting in 10% and 9% of cases respectively (Scallan et al., 2011). The pathogens that cause the highest number of illness cases are not necessarily those that cause the most harm in terms of severity. Bacterial pathogens cause the highest number of hospitalizations and deaths due to foodborne illness. Salmonella spp infections result in the highest number of hospitalizations, about 19,633 annually (CDC, 2011 a ). Norovirus, Camplyobacter spp, and Toxoplasma gondii infections result in the highest number of hospitalizations out of the 31 known pathogens (Scallan et al., 2011). Domestic foodborne illnesses result in 1,351 deaths annually. Salmonella spp, Listeria monocytogenes , n orovirus, and Toxoplasma gondii are responsible for the largest number of deaths (Scallan et al., 2011). Surveillance of foodborne dise ase outbreaks in the US during 2009 2010 implicated four commodities which were most often linke d to foodborne illness outbreaks during that time. The four commodities, in order of largest number of outbreaks, are beef, dairy, fish, and poultry (CDC, 2013) . Of the dairy related outbreaks, 81% involved unpasteurized products. Sprouts and vine stalk vegetables, including tomatoes, were also implicated in outbreaks during this time. Salmonella infections from both sprouts and vine stalk vegetables , including t omatoes, resulted in the highest
20 number of hospitalizations during 2009 2010 along with E. coli O157:H7 in ground beef (CDC, 2013). The escalation of foodborne illness outbreaks linked to produce has been apparent for several years. Historically, most foo dborne illnesses have been caused by poultry, beef, and other animal products. In recent years, however, there has been a steady increase in the number of outbreaks where produce is the vehicle for infection. According to Sivapalasingam and others (2004) t here were a total of 190 produce associated outbreaks between 1973 and 1997. More recent data indicates that from 1996 2008 there were 82 produce related outbreaks (FDA, 2009). In the 1970s only 0.7% of the foodborne illness outbreaks were caused by consum ing produce (Yuk et al., 2006). This increasing number of outbreaks attributed to produce; however, has caused an increase to 12% of all foodborne illness outbreaks being produce related (FDA, 2007b). Many factors may be affecting the rise of produce relat ed outbreaks. Increased globalization of the food supply, a heightened surveillance system for reporting foodborne illness outbreaks, and a more health conscious, diet oriented population are just a few of these factors. This noted rise in produce associat ed foodborne illnesses has been placed on the forefront of food safety research. Determining ways to prevent contamination of produce and better methods for sanitation are critical since most produce is consumed as a fresh, raw product with little to no he ating or processing. Salmonella Salmonella are facultatively anaerobic gram negative, motile, rod shaped bacteria of the family Enterobacteriaceae. The bacterial cells are oxidase negative and catalase positive, ferment glucose, and utilize citrate as a s ole carbon source (Cabral, 2010). Salmonella has been documented causing illness in humans for over 200
21 hundred years ( Ellermeir and Slauch, 2006) . Two species make up the genus Salmonella , S almonella bongori and Salmonella enterica . S almonella enterica is further divided into six subspecies including enterica , salamae , arizonae , diarizonae , indica , and houtenae . Strains of Salmonella enterica subspecies enterica (further referred to as Salmonella enterica ) are responsible for 95% of human salmonellosis cases (Tindall et al., 2005; Waldner et al., 2012). Serotyping based on the White Kauffmann Le Minor scheme is currently used to classify strains Salmonella enterica . This strategy classifies strains isolated from human patients based on the combination o f flagellar (H), lipopolysaccharide (Vi), and somatic (O) antigen reactivity (Grimont and Weill, 2007; Fatica and Schneider, 2011). Biochemical assays of these antigens are used to characterize current serovars and designate new ones as well (Waldner et al ., 2012). More than 2600 serovars exist with approximately 60% being members of S almonella . enterica (Jacobsen et al., 2011). Many serovars of Salmonella . enterica cause gastroenteritis; these are referred to as nontyphoidal Salmonella e whereas typhoidal S almonella e include serovars that cause blood stream invasion in the absence of gastroenteritis, the disease commonly known as typhoid fever (Waldner et al., 2012). While Salmonella is a natural part of the intestinal tract of many warm blooded animals, it still poses a threat to humans (Callaway et al., 2008). Salmonella causes illness in humans by traversing the gut lumen and entering the epithelium of the small intestine which causes inflammation and systemic releases of an enterotoxin (FDA, 2012). The resulting gastroenteritis is termed nontyphoidal salmonellosis (further referred to as salmonellosis). Human infection with Salmonella enterica can be
22 attributed to the ability of the microorganism to survive the harsh conditions of the stomach and ultimat ely reach a site of colonization (Fatica and Schneider, 2011). Salmonella grows at an optimum pH of 6.5 7.5; however, the bacteria have evolved mechanisms for survival in low pH environments through acid shock proteins associated with the acid tolerance re sponse system ( Foster, 1991 ). The acid tolerance response system allows Salmonella to grow in a broad pH range of 4.05 9.5 (Foster, 1991). Salmonella enterica is also proliferates in a wide range of temperatures, from 7 48Â°C with an optimum growing tempera ture of 37Â°C ( Ellermeir and Slauch, 2006) . The adaptability of Salmonella has allowed it to live outside of an animal host and subsequently cause infection once ingested by humans (Waldner et al., 2012). Importance in Produce Related Foodborne Illness Out breaks Salmonella is estimated to cause over one million cases of foodborne illness each year, and while this is not the most common pathogen, it is one of the leading causes of foodborne gastroenteritis in the US (Scallan et al., 2011). Common symptoms i nclude vomiting, diarrhea, cramps, dehydration, and fever. These symptoms are usually mild and the disease is self limiting (FDA, 2012). For those who are classified as part of the high risk population, young children, the elderly, and immune compromised i ndividuals, the disease can be much more severe and result in death (Reller et al., 2008). Salmonella causes 1 million cases of infection, 19,533 hospitalizations, and 378 deaths every year (CDC 2011 a ; Scallan et al., 2011). Of these one million cases, an estimated 95% of them are caused by foodborne sources and have been linked to both animal and produce based outbreaks (Tauxe et al., 1991). In 2009, over 6,000 Salmonella enterica isolates were serotyped by the CDC and state health departments. The five m ost commonly found Salmonella enterica serovars
23 resulting from foodborne illness patients were Enteritidis, Typhimurium, Newport, Javiana, and Heidelberg (CDC, 2010). Prior to 1990, most cases of salmonellosis were directly related to contaminated poultry and poultry products. Starting in the early 2000s, however, there has been a significant increase in the number of non animal based products, including produce (CDC, 2 008). Salmonella outbreaks have now been linked to tomatoes, sprouts, melons, peppers, lettuce, mangoes, chocolate, powdered infant formula, raw almonds, dry seasonings, cereals, and peanut butter (Waldner et al., 2012). In 2002 2003, Salmonella spp. was r eported to cause 31 outbreaks of produce origin while the number of outbreaks related to poultry was only 29 (CDC, 2008). While salmonellosis outbreaks have been linked to a wide variety of produce, the largest outbreaks are associated with tomatoes, canta loupes, alfalfa sprouts, and peppers (Sivapalasingam et al., 2004). The wide pH and temperature tolerances of Salmonella allow the bacteria to proliferate in a variety of environments including those of various food matrixes (Fatica and Schneider, 2011). T his has made it possible for Salmonella to utilize these food products as a vehicle for transmission to different hosts. When Salmonella is shed by an infected animal, the bacterium is able to persist in the plant environment until it is ingested by anothe r host (Waldner et al., 2012). Studies have demonstrated the ability of Salmonella to survive in the plant environment, including on the surface of plants as well as soil samples. Stine and others (2005) inoculated the surfaces of cantaloupes, lettuce, and bell peppers with a cocktail containing several bacterial and viral microorganisms, including Salmonella enterica . The inoculated produce was held under different temperature and humidity conditions in order to simulate preharvest conditions.
24 Pathogens we re still detectable after 14 days, with Salmonella enterica displaying one of the highest bacterial recoveries (Stine et al., 2005). This study demonstrates the ability of Salmonella to survive in the plant environment and to be a threat to food safety pri or to harvest. The persistence of Salmonella enterica in non animal environments, especially raw produce commodities, makes prevention of Salmonella contamination a vital task for all those in the food industry. Potential Sources of Contamination of Produ ce Salmonella contamination in produce can occur through a variety of pathways. Contamination can occur prior to harvest, during harvest, during processing and distribution, and at retail and foodservice facilities (Brackett, 1999). Preharvest contaminati on is most often due to contaminated agricultural water, improperly composted manure, feces, and animal contact (Beuchat and Ryu, 1997). Many animals, warm and cold blooded have been shown to shed Salmonella in their feces. Introduction of Salmonella ente rica to the environment can occur through shedding of animals throughout grazing areas which allows for direct deposition of Salmonella into soils (Baloda et al., 2001). Fertilizers composed of manure from cattle and pigs also introduce Salmonella into agr icultural areas used for growing produce and other non animal commodities. Studies have shown that fertilizers and soils onto which it is spread recovered Salmonella 160 230 days after addition to farmlands (Islam et al., 2004). Shedding of warm blooded animals is not the only source of fecal contamination of produce farms. Defecation by farm animals is not the only vehicle for Salmonella contamination of soils. Birds and insects such as flies are also vectors for widespread Salmonella distribution in the environment (Davies and Wray, 1996). Infection of wild bird and fly populations with Salmonella has been correlated with proximity to farms
25 and/or the incidence of human salmonellosis (Davies and Wray, 1996). Poorly treated irrigation may also facilitate the contamination of fruits and vegetables with Salmonella . Groundwater, wastewater, and surface water are commonly used for irrigation of produce farms (Steele and Odumeru, 2004). Salmonella has a high survival rate in aquatic environments, and can outliv e Staphylococcus aureus and waterborne Vibrio chlorae in both groundwater and heavily eutrophied river water (DiRita, 2001). Postharvest sources of contamination will vary depending on the amount of handling and processing each commodity undergoes prior t o final arrival in the consumers home. Human handling of fresh produce can occur at harvest, during processing, at the retail, and foodservice levels. Poor personal hygiene at these levels can result in contamination of produce (Beuchat and Ryu, 1997). Lar ge quantities of produce come together during processing in packinghouse facilities; cross contamination at this stage is a major concern (Rushing et al., 1996). Proper sanitation of processing equipment including dump tanks, flume systems, buckets, bins, cutting utensils, and trucks used for distribution must be carried out to prevent contamination of fruits and vegetables (Brackett, 1999). Microbial quality of water used in postharvest processing is critical to reduce contamination of harvested produce. Contaminated water in dump tanks and flume systems can spread contamination throughout the packinghouse facility (Fatica and Schneider, 2011). Also, a single contaminated fruit or vegetable can cross contaminate entire batches of produce once it enters the packinghouse. Many packinghouses add a sanitizer such as chlorine to their dump tanks and flume systems in order to prevent cross contamination between produce coming in from the field, however, the efficacy of the sanitizer is affected by pH, organic
26 mat ter, and concentration (Beuchat and Ryu, 1997). The potential for Salmonella to survive in a packinghouse setting was determined by Allen and others (2005). A five strain Salmonella cocktail was inoculated onto various surfaces within the packinghouse inc luding stainless steel, polyvinyl chloride, unfinished oak, conveyor belts, and sponge rollers. Temperature and relative humidity combinations were altered to mimic conditions in Florida during late summer and fall/winter. Salmonella was recovered at detec table levels on all surfaces regardless of environmental conditions for 7 days. Stainless steel, wood surfaces, and polyvinyl chloride harbored Salmonella at detectable levels throughout the duration of the sampling period, 28 days (Allen et al., 2005). Tomatoes The tomato is a member of the Solanaceae family, along with potatoes, eggplant, and peppers (Knapp, 2002). Species within this family range from tropical to temperate regions and most likely originated in Central or South America (Daives et al., 1981). The cultivated tomato is a member of the genus Lycopersicon . Lycopersicon is subdivided into Eulycopersicon and Eriopersicon . Fruit belonging to the first subgroup are red , yellow , pink, black (a combination of lycopene and chlorophyll), or white wh en ripe; the tomato L. esculentum is a member of this subgroup. The latter is comprised of fruits that are green or purple throughout development (Daives et al., 1981). The tomato plant is perennial, but is normally grown as an annual plant. L ycoperscion esculentum is readily hybridized with other species within the Lycopersicon genus. The fruit is botanically referred to as a berry because the seeds are contained within the mesocarp (Daives et al., 1981). The mature tomato fruit is a complex organ, contai ning two or more carpels radially separated by tissue referred to
27 as the septum. The septum is a result of the fusion of the two adjacent carpel walls known as the pericarp (Lemaire Chamley et al., 2003). The pericarp evolves during fruit development from endocarp, mesocarp, and exocarp (Gillaspy et al., 1993). The septum which separates two carpels divides the fruit into two locular cavities contained within the pericarp. The locules c ontain the seeds of the fruit attached to a placenta of parenchymatous axis, or columella (Lemaire Chamley et al., 2003). As the fruit matures, cell expansion occurs in the pericarp and placenta as it extends to fill the locular cavity and a jelly like app earance becomes evident (Daives et al., 1981). Cells within the pericarp contain most of the chloroplasts which give unripe tomatoes their green color (Gillaspy et al., 1993). Ripening is a highly coordinated phenomenon in the tomato plant. Tomatoes are classified, along with fruits such as bananas and avocadoes, as climacteric fruits (Kader, 2002). Climacteric fruits are characterized by an increase in respiration and a concomitant increase in ethylene production upon initiating ripening. Due to their ph ysiology, unripe tomatoes have an absolute requirement for ethylene because they will not ripen without production from the fruit or exposure following harvest (Klee and Giovannoni, 2011). Several stages have been identified by the USDA for the ripening pr ocess. These ripeness stages are based largely on color and include the following designations: green, breaker, turning, pink, light red, and red (USDA AMS, 1991). The change in tomato flesh color is caused by the synthesis of the carotenoid lycopene (Bran dt et al., 2006). Lycopene is a pigment with highly efficient antioxidant power. Carotenoids accumulate and chlorophylls are broken down as part of the ripening
28 process (Brandt et al., 2006). Along with an increase in lycopene and a loss of chlorophyll, in creased respiration and ethylene production are signs of the progression of the ripening process (Klee and Giovannoni, 2011). The first signs of ripening are helpful indications of maturity levels when harvesting fruit (Kader and Morris, 1976). Tomatoes are harvested at various stages of maturity ranging from physiologically mature known as mature green, to red ripe (Sargent et al., 1992). The appearance of the locular tissue is a good tool for determination of maturity and the onset of ripening. During t he beginnings of ripening, the locular tissue, which contains the developing seeds and placental tissue, will transform from a firm pericarp like tissue to an amorphous gel (Saltveit, 1991). Differences in the appearance of the locular tissue is a basis fo r differentiating maturity levels; designations of maturity levels include M 1 through M 4 with M 1 characterizing those fruit with white (immature seeds) and no gel in the locular cavity and M 4 characterizing fruits with a red appearance in gel and peric arp tissue (Kader and Morris, 1976). Tomatoes harvested at either the M 3 or M 4 stage will ripen properly with careful handling. Fruits harvested at the M 2 stage will ripen to a moderate quality and those harvested at the M 1 stage will not ripen to an a cceptable quality (Sargent et al., 1992). Tomatoes are allowed to ripen following packing in ripening rooms where temperature, humidity, and ethylene concentrations are controlled. Tomatoes are typically stored under controlled atmosphere conditions in ord er to extend shelf life prior to distribution (Saltviet, 1991) Tomato production accounts for nearly two billion dollars of farm cash receipts every year (USDA ERS, 2012). Tomato production is separated into fresh and processed industries. The processed tomato industry accounts for about 89% of all
29 tomatoes produced in the US. Fresh market tomatoes make up the remainder of production and are hand picked, while the tomatoes picked for processing are done so with machines. Fresh market tomatoes are grown in every state in the Nation and commercial scale production is present in 20 states (USDA ERS, 2012). The majority of fresh tomato acreage is divided between Florida and California, accounting for two thirds of overall fresh tomato production (USDA ERS, 201 2). California produces the source of processed tomato products in the US (USDA ERS, 2012). Imports account for approximately one third of US tomato consumption, the majorit y of which is imported from Mexico (USDA ERS, 2012). Mexico and Florida generally compete for fresh tomato production during the winter and early spring season. The majority of tomatoes imported from Mexico are greenhouse varieties which are gaining popula rity in the US. As of 2008, Mexico makes up 71% of the imported greenhouse tomato market (USDA ERS, 2012). An array of greenhouse and hydroponic varieties can be found in local supermarkets along with round tomatoes, cluster tomatoes (tomatoes on the vine) , Roma (plum) tomatoes, grape, and cherry tomatoes (USDA ERS, 2012). Postharvest Processing of Tomatoes In 2008 there was a major foodborne illness outbreak that involved Salmonella Saintpaul infections. There were 1500 identified cases, of which 21% of victims were hospitalized, and two died. The food most commonly associated with the outbreak was tomatoes. Upon further investigation, however, the illnesses were officially linked to jalapeno and Serrano peppers (Behravesh et al., 2008). This false implic ation led to the required implementation of the Florida Tomato Good Agriculture Practices (TGAPs)
30 (FDACS, 2008). While TGAPs had been adopted by some previously, FDACS mandated their adoption in 2008. TGAPs contain guidelines for sanitary harvesting and p ackinghouse conditions for tomatoes. Proper sanitation and worker hygiene are key elements in this document as they are vital elements in prevention contamination. Topics included in the manual are hand washing facilities, proper equipment sanitation, sani tization of dump tanks and flume systems, sanitation of ripening areas, and documentation (FDACS, 2008). Tomatoes for fresh market distribution are typically harvested by hand. These fruits are then collected in bins or gondolas. Subsequently they are tra nsported to a packinghouse for washing , sorting, sizing, and waxing (FDA, 2009). Upon arrival at the packinghouse, tomatoes are emptied into water filled receiving tanks (dump tanks) and flumes in order to allow the tomatoes to float to the packing line wi thout sustaining mechanical injury. Contamination at this step in postharvest processing can be very likely as the water may infiltrate the fruit through the stem scar or through cracks in/on the cuticle (Bartz and Showalter, 1981). To prevent cross contam ination between tomato fruits in dump tanks and flume systems and as well as an increase of microorganisms in these recirculated water systems, managers add sanitizer s to the water. The current sanitation method outlined by Florida TGAPs requires water in dump tanks and flume systems to contain a minimum of 150ppm of free chlorine which is effective in reducing bacteria l populations on the surfaces of tomato fruit. The water must also be maintained at a pH of 6.5 7.5 , so that most of the free chlorine is i n the highly oxidative HOCl form and the temperature must be 5Â°C/10Â°F above the pulp temperature of the tomatoes. F ruit dwell time is limited to 2 minutes (FDACS, 2008).
31 These chlorine concentration, pH, time, and temperature requirements must be documente d at the start up of production and must be monitored and recorded every hour (FDA, 2009). After fruit have been washed and lifted up on the packing line by mechanical belts, workers sort out culls and color. The fruit then progress through wax ing and siz ing . The sorted tomatoes are then packaged, palletized, and cooled for ripening and/or storage (Saltveit, 1991). Mature green tomatoes may be ripened immediately, ripened slowly, or stored for ripening at a later time. The optimum temperature range for no rmal ripening is 18 21Â° C/65 70 Â° F while the range for slow ripening is 14 16 Â° C/57 61 Â° F. Storage at 13Â°C/55Â°F for 2 weeks prior to ripening is acceptable for mature green tomatoes; however, storage longer than this can increase the likelihood for chill injury, decay, and cause a lack of color development (LeStrange et al., 2000). To ensure uniform color ripening, fruits are commonly treated with 100 150ppm ethylene in gas rooms for 24 48 h prior to shipment. If uniform color development does not occur, r epacking may be necessary in order to create shipments of uniformly ripe fruits (LeStrange et al., 2000). Distributors frequently repack tomatoes to meet the needs of commercial customers that demand boxes containing tomatoes of similar size and ripeness (LeStrange et al., 2000). The FDA Commodity Specific Food Safety Guidelines for the Fresh Tomato Supply Chain highlights repacking operations as a potential source of contamination as the diversity of handling practices at repack operations make a single, universally applicable food safety approach unrealistic (FDA, 2006). Lapses in facility sanitation can amplify localized contamination and/or promote internalization of
32 pathogens into healthy fruits. If a washing step is required in repack line, chlorine s hould be used at a concentration of 90 200 ppm and pH should be between 6 7; water must also be 5Â°C/10Â°F warmer than incoming fruit temperature. Traceability can become difficult when tomatoes enter repacking operations as tomatoes from different suppliers may become mixed despite recommendations to avoid this practice (FDA, 2006). Tomatoes may be sliced prior to delivery to foodservice or retail establishments or sliced at the location itself. Sliced tomatoes are routinely seen in the foodservice industry and on a smaller scale, in supermarkets and grocery stores; however, sliced tomatoes tend to deteriorate quickly leading to a short shelf life (Hong and Gross, 2001). Prior to being sli ced , tomatoes may be cooled by forced air or cold water (hydrocooling) ; however the latter has been associated with contamination due to infiltration of the water into fruit (FDA, 2008). Once cooled, tomatoes undergo a whole tomato wash where fruits are typically immersed in water or spray rinsed, however, this is less commo n (FDA, 2008). Water used in immersion tanks must contain sufficient sanitizer to prevent cross contamination, pH must be maintained at 6.5 7.5 to ensure chlorine efficacy, and water temperature must be 5Â°C/10Â°F warmer than the incoming fruit pulp temperat ure (FDA, 2008). Pre slicing of tomatoes can also increase the risk of bacterial contamination as the natural barrier of the fruit is broken (FDA, 2008). During the cutting process, bacteria on the surface of the tomato can be transferred to the interior o f the fruit where water and nutrients can provide a stable environment for proliferation (Pan and Schaffner, 2010) Cut fruit is then exposed to several surfaces including human hands, knives, tables, etc. Pan and Schaffner (2010) inoculated tomatoes with a four strain Salmonella cocktail, sliced the tomatoes, and then incubated
33 them in a range of temperatures in order to determine a mathematical model to describe Salmonella growth on cut tomatoes at a variety of different temperatures. Results showed that t he relationship between the growth rate of Salmonella and temperature is linear (Pan and Schaffner, 2010).Safety and sanitation practices should be followed in fresh cut facilities in order to prevent cross contamination (FDA, 2006; FDA, 2008). Physical d eterioration of plant tissue is also common in fresh cut tomatoes, which can lead to a reduced shelf life. The 2007 FDA Federal Food Code has defined cut be refrigerated after cutting or processing in any way (FDA, 2007 a ). While shelf life can be increased by refrigeration, tomatoes are sensitive to chilling injury. Chilling injury is a physiological disorder resulting from plants being subjected to low, non freezing temp eratures (Hobson, 1987). Chilling injury symptoms in tomatoes include uneven ripening, surface pitting, increased susceptibility to microbial infections, loss of aroma volatiles, and water soaked areas in red fruits (Hobson, 1987). C hilling injury can be p revented to an extent by combining cold temperature storage with modified atmosphere packaging (Hong and Gross, 2001) or the application of ethylene (Hong and Gross, 2000). Postharvest handling of tomatoes will differ slightly depending on the variety, ar ea harvested, and purchaser. Similarly, packinghouses will vary in their location, production volume, environment, and tomato cultivars processed, etc. however, it is crucial that these facilities maintain consistent food safety and sanitation practices in order to ensure a safe product is delivered to consumers (FDA, 2009). No foodborne illness outbreaks have been linked to tomatoes since the required implementation of
34 Florida TGAPs; however, maintaining vigilance in food safety practices for tomato produc ers and handlers is paramount. Significance of Tomato Related Outbreaks Tomatoes are a very popular commodity worldwide; in the US tomatoes are the fourth popular fresh market vegetable behind the potato, lettuce, and onions (USDA ERS, 2012). S ince 1990 tomatoes have been linked to twelve major foodborne illness outbreaks (Hedberg et al., 1999; FDA, 2009). The first time tomatoes were implicated as a vehicle for foodborne illness outbreaks occurred in 1990 and since then the number and severity of outbrea ks associated with tomatoes has increased over the past couple decades (Gupta et al., 2007; Hedberg et al., 1999). Between 1990 and 2004, a total of 1616 illnesses have been caused by tomato associated Salmonella enterica outbreaks (Gupta et al., 2007). T he outbreaks described below are all multistate outbreaks, indicating that contamination of the implicated tomatoes most likely occurred prior to distribution to restaurants and supermarkets. The first large scale foodborne illness outbreak associated w ith tomatoes occurred in 1990, where the consumption of whole round tomatoes was linked to 176 cases of Salmonella Javiana infections (Hedberg et al., 1999). The investigation that followed this outbreak led to the initial recognition that tomatoes could b e a vehicle for foodborne illnesses. A similar outbreak occurred in 1993. In this outbreak 100 cases of Salmonella Montevideo were reported in multiple states. The outbreak was linked to fresh, uncooked tomatoes. Hedberg et al. (1999) noted that both the 1 990 and 1993 outbreaks were caused by tomatoes originating from the same tomato packer in South Carolina.
35 Another multistate salmonellosis outbreak occurred in 1999, which linked raw, round tomato consumption to 86 cases of Salmonella Baildon infections i n eight states. Three deaths were reported (Cummings et al., 2001). Traceback investigations discovered distribution sources of tomatoes including three Virginia and two California restaurants, six branches of a Mexican fast food restaurant chain in Calif ornia, and two nursing homes in Virginia. Investigations of a grower/packer cooperative, five packing facilities, and a diced tomato processor were conducted during traceback; however, the source of contamination was not found (Cummings et al., 2001). Sa lmonella Newport infections were also associated with tomatoes in 2002 (Greene et al., 2008). A case control study was conducted during this outbreak investigation and it was found that the illness was associated with raw tomatoes eaten at restaurants. Th is multistate outbreak caused 510 reported cases of illness in 26 states. No single restaurant or chain was implicated, however, it was determined that the tomatoes were from two farms on the eastern shore of Virginia. A strain of Salmonella Newport was is olated from pond water on the farm was linked to those isolated from patients. Water from the pond had been used to irrigate the fields and for the preparation of pesticide sprays (CDC 2007; Greene et al., 2008). In the summer of 2004, multiple outbreaks o f salmonellosis were reportedly linked to consuming tomatoes. A multi serotype Salmonella outbreak was identified in nine states; 429 cases were reported. No deaths were reported as a result of this outbreak, but 30% of patients were hospitalized (CDC, 200 5). The serotypes identified during the outbreak included Salmonella Javiana, Salmonella Typhimurium, Salmonella Anatum, Salmonella Thompson, Salmonella Muenchen, and six untypable strains.
36 Illness was reported among patients who ate at an unspecified restaurant chain with a strong association to consuming sliced Roma tomatoes. The restaurant chain purchased pre sliced Roma tomatoes from a single distributor. A Good Management Practices (GMP) violation noted at the fresh cut operation was the soaking of tomatoes (41 65Â° F) in cold water (33 35 Â° F) for firming prior to slicing. Soaking time was not given, however, if prolonged it could have been possible for substantial infiltration of the soak water into the tomato tissues (Bartz, 2009). Salmonella Anatum with a pulse field gel electrophoresis (PFGE) pattern matching that of the outbreak isolated strain was recovered from pre sliced Roma tomatoes upon sampling of the restaurant chain in July 2004 (CDC, 2005). A second tomato associated multistate outbreak o ccurred in the summer of 2004, where 125 cases of Salmonella Braenderup were identified in 16 states (CDC, 2005; Gupta et al., 2007). Of those who fell ill, 20% were hospitalized. A case control study was utilized to investigate the outbreak. Initially che ese, lettuce, and tomato eaten at a restaurant appeared to be associated with illness. Further investigation discovered that Roma tomatoes were the only exposure significantly linked to illness. Roma tomatoes from two restaurants implicated in the outbreak were linked through two distributors to a single packing facility in Florida. Environmental sampling of the packing house in December 2004 did not uncover a source of contamination (Gupta et al., 2007). Several more outbreaks of Salmonella infections ass ociated with the consumption of raw tomatoes were reported between 2005 and 2006. In 2005, Salmonella Newport isolates were recovered in 72 culture confirmed stool samples from patients in 16 states (Greene et al., 2008). No deaths occurred and 11% of pati ents
37 were hospitalized (CDC, 2007). Cases were associated with consuming raw, red, round tomatoes at restaurants. The implicated tomatoes had been purchased whole and sliced in the restaurant. Trace back investigation revealed that tomatoes associated with the outbreak had been grown on two farms on the eastern shore of Virginia. Salmonella Newport was recovered from irrigation pond water near the farms (Greene et al., 2008). This outbreak strain of S almonella Newport had an identical PFGE pattern as the st rain in the 2002 outbreak (CDC, 2007; Greene et al., 2008). Another outbreak of Salmonella Newport occurred in 2006 where 115 cases of illness were identified in 19 states. The PFGE pattern of this outbreak strain was again indistinguishable from the strai n from the 2002 and 2005 outbreaks; however, no source of implicated tomatoes was discovered and no restaurant was found to be associated with the outbreak (CDC, 2007). A second major tomato linked multistate outbreak occurred in 2005. A total of 82 cases of Salmonella Braenderup infections were reported in eight states (CDC, 2007). No deaths were reported, but 35% of patients were hospitalized. Consumption of contaminated pre diced Roma tomatoes from an unspecified restaurant chain was identified as the so urce of exposure. Trace back discovered that implicated tomatoes were grown in one of two farms in Florida and were pre diced and packaged at a firm in Kentucky prior to shipment to the restaurant chain. Multiple potential animal reservoirs of Salmonella w ere present in close proximity to drainage ditches; however, environmental sampling of these ditches yielded Salmonella isolates of different serotypes than Braenderup (CDC, 2007).
38 The most recent multistate foodborne illness outbreak linking tomatoes a nd salmonellosis occurred in 2006 when 190 cases of Salmonella Typhimurium infections were reported (CDC, 2007; CDC, 2011 b ). Cases were reported in 21 states between September and October 2006. No deaths were reported; 22% of patients were hospitalized. E ating raw, red, round tomatoes at a restaurant was associated with illness. Contaminated tomatoes were traced to a packinghouse in Ohio which was supplied with tomatoes from three growers from 25 fields (CDC, 2007). Contamination is often introduced or am plified in the packinghouse were thousands of tomatoes pass through a common water bath (Rushing et al., 1996). Amplification of localized contamination or internalization of pathogens can occur at the packinghouse and at repacking facilities (FDA, 2006). Slicing can amplify contamination as Salmonella can multiply on cut surfaces of tomatoes at ambient temperatures (Lin and Wei, 1997). When one contaminated fruit is cut, contamination can spread through cross contamination to other uncontaminated fruits an d proliferate (Zhuang et al., 1995). Since tomatoes are generally consumed as a fresh, raw commodity, there is an increased risk for foodborne illness for the consumer. The lack of any heat treatment and minimal processing causes little microbial reductio n (Sivapalasingam et al., 2004). Due to this, prevention of microbial contamination on tomatoes is imperative during pre and post harvesting of the fruit. Internalization Pathogen internalization is defined as the process by which a pathogen traverses p ast the surface of a mass and is then incorporated into the mass ( Bartz, 2006) . Internalization of bacteria can happen through many avenues, including the roots, leaves, stems, fruit, etc. In tomatoes, internalization of bacteria is most commonly seen
39 in t he fruit (Erickson, 2012). While bacteria present on the smooth, undamaged surfaces of tomatoes pose little or no threat, bacteria that become internalized are nearly impossible to eliminate (Zhuang et al., 1995). Internalization of bacteria into tomatoes can occur during preharvest through roots, stems, and blossoms; however, this is unlikely. During postharvest submersion of tomatoes into water in dump tanks and flume systems has been shown to cause internalization of bacteria through the stem scar (Bart z and Showalter, 1981; Zheng et al., 2013). Preharvest Internalization Colonization and internalization of Salmonella in tomato fruits/plants, however, has not been investigated in great detail (Gu et al., 2011). Many of these studies have focused on the route Salmonella may take to colonize a tomato plant. These avenues include agricultural water, direct contact with leaves, stems, and leaves, and soil (Guo et al., 2002; Garcia et al., 2010; Gu et al., 2011). While evidence for contamination via many of these routes still remains inconsistent, it is important to consider them each as a critical point of entry for bacterial pathogens. Salmonella may be introduced to tomato plants via pathways other than contaminated soil and irrigation water. Internalizat ion through roots, leaves, and blossoms has been studied in laboratory settings; however, these conditions would not be likely to occur in the field. Zheng et al. (2013) demonstrated a contamination rate of 70.4% when inoculation occurred through flowers. Salmonella was recovered from tomato surfaces and internal tomato fruits when inoculated at the blossom. In blossom inoculated tomatoes, Salmonella persisted in internal fruit tissue for at least 35 days of fruit development (Zheng et al., 2013). Guo et al . (2001) was also able to verify the ability of Salmonella to survive both in tomato tissue and on tomato surfaces when
40 stems and blossoms were inoculated. Plants were inoculated either by brushing of the flower or application to the stem before or after f ruit set. Of tomatoes harvested from inoculated plants, 37% were positive for Salmonella . Overall, tomato pulp harbored less Salmonella than did tomato surfaces and stem scar tissue suggesting that internalization is possible, however, colonization of stem scar and surface tissue is more likely (Guo et al., 2001). Both studies used high inoculum concentrations (10 9 10 10 CFU /ml) and inoculation was performed by either injecting or brushing onto the blossoms or leaves. Contamination of this sort is neither lik ely to occur nor are such hi gh Salmonella populations. The Microbiological Data Program (MDP) is a voluntary data gathering program under the guidance of the USDA Agricultural Marketing Service. In 2009, the MDP conducted microbiological testing on a variety of commodities. Over 2000 round tomatoes were sample d for the presence of Salmonella. Of these, only 14 presumptive positive samples were found and none of these were confirmed as Salmonella isolates (USDA AMS, 2009). Enumeration of E. coli O157:H7 resulted in 96% of samples being below 10 MPN/g. This data suggests that populations of pathogenic bacteria in the field are well below those used in the studies described previously (USDA AMS, 2009). Deposition of Salmonella onto tomato leaves has been shown to cause internalization and subsequent translocatio n of the pathogen within tomato plants. Gu et al. (2011) dipped leaflets of growing tomato plants into a suspension of Salmonella Typhimuirum (10 9 CFU /ml) in a two year study where inoculated leaves, non inoculated leaves, and ripe fruits were tested for t he presence of Salmonella . Salmonella was recovered from both inoculated leaves and adjacent non inoculated leaves on the same
41 plant. Fruit pulp was also analyzed for the presence of Salmonella from plants with inoculated leaves. Internalization was seen; however, the rate of internal pulp contamination was low (Gu et al., 2011). While internalization of fruit pulp at low levels may be possible through contamination of leaves, high Salmonella populations used in this study are not likely to be found in natu re and the method of inoculation used in this study is not a practical image of what would occur in the field. Studies examining preharvest sources of Salmonella internalization have also shown that certain are serotypes more fit for survival inside tomato fruit. Zheng et al. (2013) utilized a five strain cocktail of Salmonella when inoculating soil and plants. Serotyping was performed in order to determine which strains were able to persist best in various environments. Salmonella Newport and Salmonella Javiana were more prevalent in soil samples while Salmonella Montevideo and Salmonella Newport were dominant on tomato leaves and blossoms. This is consistent with research performed by Guo et al. (2001, 2002) which demonstrated a higher prevalence of Salmonella Montevideo during analysis. Salmonella Typhimurium was shown by Zheng et al. (2013) and Garcia et al. (2010) to survive poorly in tomato plants, suggesting that contamination of tomatoes with Salmonella Typhimurium most likely occurs during postharvest (Zheng et al., 2013). Postharvest Internalization Internalization of bacteria into tomatoes during postharvest processing has been studied extensively. The st em scar has been identified as the most likely entry point for water and bacterial infiltration during postharvest handling. Since the tomato fruit has no stomatal openings, almost all exchange of water and gases occurs at the stem scar. This impermeabilit y of the undamaged skin, or rather the permeability at the stem scar,
42 can be verified by submerging a tomato in water of a higher temperature and observing the expanded internal air bubbling which occur exclusively at the stem scar and corky ring. Bacteri a may become internalized into tomatoes if contaminated water is absorbed into the fruit through wounds in the cuticle or the stem scar. Water absorption through the stem scar can be explained by general gas law principles. As the fruit cools, gases in th e intercellular space will exert a reduced pressure, allowing the combined atmospheric and hydrostatic forces on the submerged fruit to force water and other particles in the external environment into pores on the stem scar (Bartz, 1982). Showalter noted i n his 1993 report that water uptake by tomatoes is increased when the tomato is submerged in water that is cooler than the fruit temperature. Previous research done on this topic has also shown an increase infiltration of water into warm fruits that are su bmerged in water of cooler temperatures (Studer and Kader , 1976 ; Showalter, 1979; Bartz and Showalter, 1981). Those fruits contaminated with spoilage organisms such as Pectobacterium carotovorum will begin to decay rapidly (Bartz, 1991). Pectobacterium car otovorum is known to incite bacterial soft rot (BSR) in a variety of produce, including tomatoes. This disease causes significant postharvest losses of tomatoes annually (Vigneault et al., 2000; Smith et al., 2007). The decay from BSR in tomatoes can be se en as little as 24 h after inoculation. These tomatoes will begin to leak a bacterial secretion which is able to infect other nearby tomatoes, compounding the losses associated with this disease (Bartz, 1991). Water may also be contaminated with Salmonella spp. allowing this pathogen to infiltrate the fruit; however, once internalized, no external sign of contamination may be apparent. Salmonella will be able to multiple within the fruit because once bacteria have
43 become internalized; they are protected by the pulp of the tomato and are able to proliferate throughout (Xia et al., 2011). These contaminated tomatoes may then be sold to restaurants, markets, and consumers as there is no effective sanitizing method in place to remove internalization (Fatica and Schneider, 2011). In terms of bacterial soft rot, contamination by infiltration has proven more devastating in terms of postharvest losses than when inoculation is caused by surface attachment (Bartz, 1982). One reason for this is that all tomatoes can bec ome inoculated by infiltration whereas only fruits with small wounds, around 2mm 3 in volume, which is roughly the size of a puncture made by a grain of sand, would be involved in inoculation via surface contamination (Bartz et al., 1975). By this same prin ciple, significantly more contaminated water may be able to infiltrate tomatoes at the stem scar than through a 2 mm 3 wound. A third and final reason for the devastating effects of internalization of bacteria via water infiltration is that when inoculum is deposited into the fruit, it is deposited in unwounded tissue. With surface contamination, the inoculum may be deposited in an area that has become acidic due to damaged cells, an area that will be prone to drying out, or an area that may be affected by a wound response from the fruit (Bartz and Showalter, 1981). Temperature Differential and Internalization Numerous studies have been done on the effect of negative temperature differential between tomato fruit pulp and wash water and the implications for t omato quality and safety. A study performed by Bartz and Showalter (1981) examined this difference in temperature using three different experiments. The experiments were divided into two main studies, one that tested the infiltration of tomato fruit by wat er and one that tested the infiltration of tomatoes by bacteria. The water infiltration was tested
44 by submerging fruit to 8 cm below the water surface for 30 m in with water temperatures ranging from 10 43Â°C and fruit temperatures ranging from 4 43Â°C, makin g the water/fruit temperature differentials range from +33 to 35Â°C. Results were determined by weighing the fruits after submersion and calculating the percent weight gained. Visual results were also included as visible cracking of the skin and water soak ing about the stem scar were noted as indications of water uptake. The results showed an increase in weight due to the ingress of water in all tomatoes. Subsequently the tomatoes that were subjected to the largest negative temperature differential displaye d the largest weight increase while those subjected to the largest positive temperature differential displayed the smallest weight increase. (Bartz and Showalter, 1981) This experiment also included a bacterial infiltration study in which a variety of test ing methods were used. In one method, a suspension of red marker bacteria, Serratia marcesens Bizio, was used in place of water. The tomatoes in this method were heated to 34Â°C or cooled to 18Â°C then submerged for 30 m in in the bacterial suspension which w as maintained at 18Â°C. Another method using the same bacterial suspension involved heating the tomatoes to 36Â°C and 20Â°C and immersing them in a bacterial suspension at 22Â°C for 15 m in . The third and final method involved the use of suspensions of organism s known to cause decay in tomatoes. These species were Pectobacterium carotovorum , Psuedomonas marginalis , and Psuedomonas aeruginosa . For this method, tomatoes were either 20Â°C or 40Â°C and were submerged in fruit rot suspensions that were 20, 22, and 40Â°C . These parameters allowed positive, zero, and negative temperature differentials to be analyzed in their relationship to water ingress. Results from the bacterial infiltration experiments showed a positive correlation between
45 post immersion decay and nega tive temperature differentials. For all tomatoes submerged in water that was cooler than the pulp ( 5 to 35Â°C temperature differentials), substantial bacterial soft rot resulted, regardless of the time for which they were submerged. Positive or zero diffe rential treatments resulted in little to no decay (Bartz and Showalter, 1981). These studies were performed to examine worst case scenarios rather than to evaluate typical conditions in a packinghouse. As mentioned above, the Florida TGAPs state that tomat oes should be in contact with the wash water for a maximum of two minutes. For each of the experiments performed by Bartz and Showalter (1981) the time of submersion is much higher than 2 min, submersion times were 10, 15, and 30 m in . The water infiltratio n experiment also utilizes a submersion depth of 8 cm while i n a packinghouse, tomatoes will be floating on the surface of the water as they travel to the packaging line. These distinctions between experimental procedures and those routinely seen in packin ghouses were meant to prove the principle of internalization. These studies were not meant to simulate conditions within packinghouses. When temperature differentials were reduced to slightly negative, zero, and slightly positive (0, Â±2Â°C) little or no dec ay was seen. Based on the se results , it does not sound reasonable for there to be a mandatory 5Â°C/10Â°F difference between tomato pulp and wash water. A second study performed by Bartz (1982), was done involving temperature differential between fruit pul p and water. This study focused primarily on tomatoes held at a constant temperature being suspended in bacterial suspensions containing Pectobacterium car o tovorum of two different temperatures for varying times and at
46 varying depths. The fruits were subme rged to the specified depths by direct or simulated methods. The simulated method involved the use of a pressure cooker where compressed air was added until the desired pressure was attained. Tomatoes were kept at 37Â°C and were suspended in Pectobacterium car o tovorum suspensions at 37 and 20Â°C which created zero and 17Â°C temperature differentials, respectively. The original method dictated that the tomatoes be held for 10 min then stored after immersion to examine the development of BSR. The results from t his study showed that for fruits held just under the surface of the Pectobacterium car o tovorum suspension for 10 m in at the 17Â°C temperature differential, a significant weight increase resulted, followed by BSR. When the submersion time was decreased to 5 m in , only a slight weight increase was recorded; reducing the time to 2 min resulted in a weight increase and BSR development of fruit similar to those in the control group (zero differential) (Bartz, 1982). These results show the importance of time on t emperature differential and infiltration rate of bacteria. According to these results, for infiltration of water and bacteria to occur, the tomato must be initially warmer than the water and that fruit must be in contact with the water long enough to cool (Bartz, 1982). This conclusion is of great importance to the processing of fresh market tomatoes. Since the time constraint placed on dump tanks and flume systems is a maximum of 2 mi n , it is unlikely that the fruit will have sufficient time to cool while in dump tanks/flumes in repacking and fresh cut facilities. Submersion Depth/Time As discussed by Bartz (1982), tomatoes may be infiltrated when immersed in water by two main phenomena. The first is the temperature differential phenomenon which is discuss ed above and has been shown to rely heavily on the time of
47 submersion. The second phenomenon is attributed to hydrostatic forces. These forces are created as the fruits are immersed. The effect of hydrostatic pressure on bacterial infiltration was tested b y Bartz (1982) where tomatoes were submerged to a variety of depths using both direct and simulated methods. This is the same study discussed above where the simulated submersion depth was achieved using a pressure cooker. In this study, several experiment s were performed to determine the effects of submersion depth on internalization of bacteria which was observed through the incidence of BSR in tomatoes. One experiment tested the effect of depth and temperature differential for tomatoes immersed in Pectob acterium car o tovorum suspensions. Weight increases and disease prevalence was not greater in tomatoes submerged to 15 cm than those submerged to 1 cm in either the zero or negative temperature differential groups (Bartz, 1982). Since that experiment was in conclusive, an experiment was performed in which the temperature differential between fruit pulp and bacterial suspension were zero. The variables of this experiment were depth and the time of submersion. Tomatoes were directly submerged to 31 cm and were held there for 0, 2, 5, 10, and 10 m in . Only fruits exposed to the Pectobacterium car o tovorum suspension for 10 m in or longer resulted in BSR. Those immersed for 0 5 m in showed no weight gain and did not decay during storage. A third experiment was perform ed where tomatoes were subjected to a simulated depth of 68 cm in both water and bacterial suspension. In this experiment a surfactant was added to the water and suspension which did result in water ingress and bacterial infiltration at time intervals as l ow as 2 min . The fruits in the zero time intervals, however, still did not show any signs of weight gain or BSR. The final study showed
48 evidence of instantaneous water ingress. Fruits in this experiment were exposed to bacterial suspensions via simulated i mmersion at 1, 31, 61, 122, and 244 cm for 0, 2, and 10 m in . Tomatoes submerged to 122 and 244 cm at time zero developed BSR in 30 and 70% of fruits, respectively, after 48 h of storage. This experiment was the only test to give positive results for tomato es subjected to bacterial suspensions for less than 2 m in (Bartz, 1982). In commercial packinghouses, tomatoes in the dump tanks and flume systems are floating along the water as they are being carried to the packaging line. Tomatoes may become submerged w hen an overloading of fruits are added at one time. The fruits that are submerged well below other fruits as unloading occurs could be more susceptible for water infiltration, however, it is unlikely in commercial settings that depths will reach 122 124 cm . . m in are also not relevant to conditions in packinghouses as the maximum time for fruits in the wash water is 2 m in . Time after Harvest Aside from submersion depth, the time after harvest may have an effect on water infiltration and subsequent bacterial internalization of tomatoes. Once harvested the stem scars on tomatoes begin to dry out. In commercial tomato production, field bins can be shaded under ambient temperatures and not unloaded until th e day after harvest; this extended time period after harvest results in drier stem scars (Smith et al., 2007). It has been suggested that older stem scars will absorb less water because they become congested with air as time elapses after harvest, hinderin g the ability of water to move into tissues. This theory was tested by Smith et al. (2007).The method for this study involved immersing tomatoes in water at time intervals of 2, 8, 14, and 26 h after
49 harvest while keeping all other variables, temperature d ifferential between fruit pulp and water, hydrostatic pressure, and submersion depth, constant. The only other variable in this study was the variety of tomato used. Two cultivars of tomato were used, Florida 47 and Sebring. This study was also completed o ver two seasons, one in the fall and one in the spring. In the spring season, the tomatoes were immersed in water at time intervals of 2, 4, 6, 8, and 14 h after harvest. (Smith et al., 2007) The overall results of this study showed that Sebring tomatoes had less water uptake than Florida 47 cultivars. The more relevant results of this study are those pertaining to the water uptake at various time intervals after harvest. In the fall study there was a large amount of variability between results, most like ly attributed to the larger difference in time intervals and the conditions inside the pressure cooker. Despite this variability, the water absorbed tended to be greatest in the 2 h group as opposed to the other time intervals. Overall the difference in wa ter uptake seemed to dissipate between the 2 and 8 h time intervals; because of this the spring experiment used more time intervals between 2 and 8 h to determine where differences in water uptake were most prevalent. The spring experiment showed less vari ability and thus more consistent results. In the spring, both cultivars showed the greatest water absorption at two hours and the biggest decrease in water ingress between the two and four hour time intervals. The recommendation made by this study was that fruits should be held for at least four hours after harvest before being introduced to the dump tank in order to sufficient stem scar drying (Smith et al., 2007). The results from this study have major implications for tomato packinghouse facilities, how ever, the conditions outlined in this laboratory are different from those
50 seen in commercial packinghouses. In this study, the tomatoes were arranged in a single layer with stem scars facing upward and were kept at a constant relative humidity of (46%) (Sm ith et al., 2007). In commercial settings, fruits are collected in bins or gondolas at a depth of up to 1.5 m. This stacking of tomatoes will cause the stem scars of the tomatoes to dry out faster than those in the middle or on the bottom layers. The relat ive humidity would also be impossible to keep constant under these conditions, thus preventing stem scars from drying as well. Due to the difference in these conditions, it may be necessary for tomatoes to be held for more than 4 h after harvest in order t o prevent water ingress and possible bacterial internalization. Internalization of Salmonella in Tomatoes All of the studies discussed above have tested the rate of water ingress associated bacterial soft rot incidence in tomatoes. These studies have conducted experiments with variables including temperature differential, submersion depth, situations were investigated. A temperature differential of 35 to +33Â°C was us ed in extreme submersion times such as 10 30 m in and extraordinary submersion depths such as 122 and 244 cm. In commercial settings, the water in dump tanks and flume systems are maintained at 5Â°C above the temperature of the incoming fruit. The fruit is also not in contact with the wash water for more than 2 min. Tomatoes are also never supposed to be submerged further than the depth of two layers of tomatoes in the dump tanks and flumes. Since the previous studies do not accurately reflect the conditions in a commercial packinghouse, a study was performed by Xia et al. (2011) to better model
51 the commercial dump tank and flume systems. This study also determines the effects t hese variables have on Salmonella spp . internalization as opposed to the prevalence of BSR. This study involved the testing of temperature differential between fruit pulp and bacterial suspension, tomato variety, and post stem removal time on the internalization of Salmonella Thompson. In orde r to determine the distribution of Salmonella Thompson in the tomato fruit, tomatoes were subjected to extreme conditions so that complete infiltration of the pathogen could occur. The tomatoes were maintained at 32.2Â°C and were immersed in a suspension of Salmonella Thompson at 2.3Â°C to a depth of 12 cm for 30 m in . After 30 m in , tomatoes were removed and allowed to dry. The fruits were then sliced so that internal tissues from six locations were excised to microbial analysis. The results from this part of the study showed the highly variable distribution of Salmonella enterica inside the tomatoes. While the internalization varied significantly among different locations and tissue types in the fruit, Salmonella enterica was only found in core tissue location s. The populations in the core segments of the fruit declined with distance from the stem scar in all tomatoes tested. The distribution pattern of the pathogen further verifies the fact that the stem scar is the major point of entry for Salmonella internal ization. The remainder of the study was performed to determine the effect of a variety of variables on Salmonella enterica internalization. Temperature differential between fruit pulp and Salmonella suspension was adjusted to 5.6, 0, and +5.6Â°C. Three d ifferent tomato varieties were also used; these included Applause, BHN961, and Mountain Spring. The final variable included was post stem removal time. The time intervals tested were 0, 2, and 16 h after harvest. Groups of five tomatoes underwent each
52 comb ination of temperature differential, variety, and post stem removal times to give a total of 24 trials. Of these three variables, time between harvest and immersion and tomato variety had the most significant impact on the prevalence of internalization of Salmonella Thompson. Those tomatoes with stems removed immediately before immersion (time 0) had a significantly higher rate of internalization as opposed to those fruits whose stems were removed 2 and 16 h prior to immersion. This observation complies wit h the findings in Smith et al. (2007) where water infiltration decreased significantly as time between harvest and submersion increased. Of the three tomato varieties used, Applause and BHN961 showed similar results in regard to post stem removal times, wh ile Mountain Spring tomatoes showed no significant difference in Salmonella enterica internalization among the three post stem removal times. Under these testing conditions, temperature differential revealed no significant effect on the internalization of Salmonella Thompson and limited effect on the population of internalized cells. The lack of correlation between temperature differential and internalization of bacteria is contrary to results obtained from the variety of studies discussed above. A major difference in those studies and the one performed by Xia et al. (2011) is the use of Salmonella Thompson rather than Pectobacterium caro tovorum . It is difficult to compare these studies as internalization of Salmonella was quantified by enumeration of the bacteria while visual inspections of rot were used to determine the internalization of Pectobacterium carotovorum ; temperature differentials used between the studies also differed . The most congruent results from all of these studies is the relationship b etween post stem removal time and internalization of bacteria. The longer the time
53 period between harvest and water immersion, the less water that will infiltrate the fruit. A recommended holding time of 4 h has been outlined in Smith et al., (2007); the r esearch performed by Xia et al., (2011) supports these findings. The ability of phyto and foodborne pathogens to internalize into tomato fruit pulp has been studied previously; the majority of these studies have evaluated extreme temperature differentia ls, time intervals, and submersion depths. Studies performed to determine Salmonella internalization have been conducted on mature green tomatoes under conditions typical of a packinghouse operation. The objectives of this study were to determine: (i) the effect of Â±5, Â±3, and 0Â°C temperature differentials between tomato pulp and water on the internalization of Salmonella into the core of tomatoes; (ii) the effect of maturity on the internalization of Salmonella into the core segments of mature red and matu re green tomatoes; (iii) the location of Salmonella internalization within the core segment following submersion into inoculated postharvest water, simulating repacking and fresh cut operations.
54 CHAPTER 3 MATERIALS AND METHODS Tomatoes Previously washed and waxed m ature green and red round tomatoes were purchased, on the day of delivery, from a local supermarket (Winter Haven, Florida). Tomatoes originated from either Blue Ribbon Farms or H.D. Budd Farms, Inc., in Plant City, Florida. Tomatoes were stored at 4Â°C for up to one week prior to use. To obtain the required internal pulp temperature, tomatoes were transferred to an incubator set to either 4 (red) or 21Â°C (red and green), for 18 Â± 2 h before experimentation. Only tomatoes void of defects an d with undamaged stem scars were used. Salmonella Strains Six strains of Salmonella enterica were obtained from the Danyluk laboratory culture collection before each experiment. These include Salmonella Typhimurium LT 2 (MDD 14; ATCC 700720), Salmonella M ontevideo (MDD 22; LJH 0519; tomato outbreak; human isolate), Salmonella Anatum (MDD 33; K2669; tomato outbreak; clinical isolate), Salmonella Javiana (MDD 226; ATCC BAA 1593; tomato outbreak; clinical isolate), Salmonella Branderup (MDD 227; 04E61553; tom ato outbreak; clinical isolate), and Salmonella Newport (MDD 314; tomato outbreak; environmental isolate). All strains are resistant to rifampicin (Fisher Scientific International, Inc., Hampton, New Hampshire) at 80Âµg/ml and were stored in glycerol bead s tock cryogenic vials at 80Â°C. Inoculum Preparation A frozen stock of each strain of Salmonella was streaked onto tryptic soy agar supplemented with 80 Âµg/ml rifampicin (TSAR; Difco, Benton Dickinson, Sparks, MD) and incubated overnight (24 Â± 2 h) at 37Â° C. A single isolated colony was transferred to
55 10 ml of tryptic soy broth supplemented with rifampicin (TSBR; Difco, Benton Dickson, Sparks, MD) and incubated overnight (24 Â± 2 h) at 37Â°C. Following incubation, a 10 Âµl loop full was transferred into fresh TSBR and incubated for 24 Â± 2 h at 37Â°C. After incubation, each strain was subjected to centrifugation (Allegra X 12, Beckman Coulter, Fullerton, CA) at 2,095 xg for 10 m to collect the cultures. Cells were washed three times by removing the supernatant a nd then suspending the cell pellet in 10 ml of 0.1% peptone (Difco, Benton Dickson, Sparks MD) then centrifuging. Washed cells were suspended in 0.1% peptone at half the original culture volume. Each Salmonella strain was added in equal parts to produce 60 ml of a Salmonella cocktail at a concentration of 10 8 CFU /ml. The cocktail concentration was verified by preparing serial dilutions of the cocktail in 0.1% peptone then enumerating the bacterial populations on TSAR following i ncubation at 37Â°C for 24 Â± 2 h. Submersion Media Ground water, meeting the microbial standards for potable water, was collected from a well at the Citrus Research and Education Center (ca. 518 m, Lake Alfred, FL). Six liters of water was added to a 32 x 29 x 13.5 cm Nalgene Â® bin whic h was then placed into a water bath (ecoline RE120 Lauda Brinkman, LP, Delran, NJ) in order to achieve the correct experimental water temperature ( 1, 1, 4, 7, 9, 16, 18, 21, 24, and 26Â°C) to create 0, Â±3, Â±5Â°C temperature differentials between water and t omato pulp. For cold temperatures ( 1 and 1Â°C) ice was added to the water bath in order to achieve and maintain these temperatures. Once the temperature was attained, 60ml of the Salmonella cocktail was added to the water to achieve a submersion media with an inoculum level of 10 6 CFU/ml; initial inoculum levels were confirmed by serial dilutions of the submersion media in 0.1% peptone followed by enumeration of Salmonella
56 populations on TSAR. Prior to experimentation, pH (Fisher Scientific International, I nc., Hampton, New Hampshire) and ORP (Ohaus Corporation, Parsippany, NJ) m easurements of the submersion media were taken in triplicate. Tomato Submersion Tomatoes were submerged to a depth of 12 cm with a sterilized wire rack and were held in place by two weighted rings. This depth allowed for submersion of the stem scar below the surface of the water to be 6 cm. Trials were run in which batches of tomatoes (n=6) were removed after submersion for 0 (negative control), 30, 60, 120, and 300 s. During each tr ial, an additional tomato was submerged for 30 m in (1800 s) to act as a positive control. When the time interval had elapsed, tomatoes were removed from the submersion media and surface sanitized by spraying with 95% ethanol and wiping with a paper towel soaked in 95% ethanol. Once surface sterilized, tomatoes were allowed to dry for up to 10 m in on sterile Petri plates. Tissue Extraction On a sterilized cutting board, the blossom end skin, 3 5 mm, of each tomato was removed using a sterilized serrated kn ife. A sterilized corer (1.9 cm in diameter) was pushed from the blossom end toward the stem scar in order to remove the entire core from the tomato. A sterile cotton swab was used to remove the core segment from the corer. The stem scar, 2 3 mm was then r emoved from the core segment using a sterile scalpel blade. Each core segment was then cut into three smaller segments of equal size using a sterile scalpel blade. Scalpel blades were only used once; additional cuts were made using new, sterile scalpel bla des. The core segments were labeled according to their location: an upper segment located just below the stem scar (A), a middle segment (B), and a lower segment located above the blossom end (C).
57 Depending on tomato size, individual core segments ranged f rom approximately 1.5 2 cm in length . Each segment was placed into a sterile 207 ml whirl pak Â® bag with filter (Nasco, Fort Atkinson, Wisconsin) using sterilized tongs. Enumeration of Pathogens Following preparation of core segments, 25 ml of lactose br oth (Remel Products, Thermo Fisher Scientific, Lenexa, KS) was added to each 207 ml bag and samples were macerated for 60 s on the highest setting (Smasher, AES Chemunex [Biomerieux], broth was added and the bag was shaken by hand for 30 s. A 3x3 Most Probably Number (MPN) analysis was then set up for the homogenized tissue mixture by extracting 3 x 10 ml, 1 ml, and 0.1 ml of each sample and adding each to a sterile test tube, 9 ml of l actose broth, and 9.9 ml of lactose broth, respectively. Tubes were then incubated for 24 Â± 2 h at 35 Â± 2Â°C. Following incubation, 1 ml from each tube was transferred to 9 ml of tetrathionate broth (TT; Difco, Benson Dickson, Sparks, MD) and 0.1 ml transfe rred to 9.9 ml of Rappaport Vassiliadis R10 broth (RV; Difco, Benson Dickson, Sparks, MD). TT broth tubes were incubated overnight (24 Â± 2 h) at 35 Â± 2Â°C and RV tubes were incubated for 48 Â± 2 h at 42 Â± 2Â°C. Following incubation, TT and RV tubes were strea ked onto Hektoen enteric agar (HE; Difco. Benson Dickson, Sparks MD) and xylose lysine desoxycholate agar (XLD; Remel Products, Thermo Fisher Scientific, Lenexa, KS) agars supplemented with 80 Âµg/ml rifampicin (HER and XLDR). Plates were stored for 24 Â± 2 hours at 35 Â± 2Â°C. Following incubation, presumptive positive Salmonella colonies were confirmed by stabbing and streaking onto triple sugar iron agar (TSI; Remel Products, Thermo Fisher Scientific, Lenexa, KS) and lysine iron agar (LIA; Difco, Benson Dick son, Sparks, MD) slants which were incubated at 35 Â± 2Â°C for
58 24 Â± 2 h. One replication (n=6 tomatoes) was completed for all experimental time intervals (0, 30, 60, 120, 300 s) for all temperature differentials (0, Â± 3, Â± 5Â°C). A second replication (n=6 tom atoes) was performed for all experimental time intervals for 0 and Â± 5 temperature differentials. During this replication, only the core segment just below the stem scar (segment A) was recovered and used for MPN analysis. Water Ingress In order to quant ify water infiltration into tomatoes during submersion, tomatoes used in the second replication were weighed prior to submersion once they had reached the appropriate experimental temperature. Tomatoes (n=6) were then submerged by the same procedure descri bed above; however, only the 5, 0, or +5Â°C temperature differentials were evaluated. Once the time interval elapsed tomatoes were removed. Prior to surface sterilization, tomatoes were gently wiped with a paper towel to remove any excess surface water, an d were weighed again. Statistical Analysis The effects of segment location, submersion time, temperature differential, and maturity levels will be analyzed for tomatoes using the Wilcoxon Signed Rank Test (JMP, software version 9; SAS Institute Inc., Car y, NC, USA.). Differences will be considered significant at 0.05. If differences were present, the Steel Dwass All Pairs Test was performed to see which groups were different. This statistical analysis was chosen to be more robust than a one way ANOVA large variations within data sets causing the data not to follow a normal distribution.
59 CHAPTER 4 RESULTS pH Ground water was obtained prior to experimentation and pH measurements were taken. To improve precision, pH measu rements were taken in triplicate. Prior to beginning the study, initial groundwater pH was 7.5 Â± 0.2. Average pH readings (n=3) for water used in each experiment are listed in Table 4 1.The pH remained constant throughout the time experiments were conducte d and the pH of water in the submersion media did not differ from the initial groundwater pH. Overall average pH for all water used in experiments was 7.6 Â± 0.1. Oxidation Reduction Potential Oxidation reduction potential (ORP) measurements were taken for water used in experiments. ORP measurements were also taken in triplicate to improve precision. Average ORP readings (mV; n=3) are displayed in Table 4 2 for water used in each experiment. Ini tial OPR of groundwater before beginning experiments was 152 Â± 2. The ORP of the water used in all experiments remained constant and did not change from the initial water reading. The overall average ORP of submersion media was 158 Â± 7 mV. Difference in T omato Weight In order to quantify water ingress, tomatoes were weighed before and after submersion into inoculum. The difference in weight (g) was calculated by subtracting the weight before submersion from the weight after submersion. The average (n=6) di fference in weight of tomatoes before and after submersion for each maturity treatment, temperature differential, and time interval are listed in Tables 4.3 4.5. All
60 tomato maturity treatments (Green 21Â°C, Red 21Â°C, and Red 4Â°C) were weighed for the 5, 0, and +5Â°C temperature differentials at each time interval (30, 60, 120, 300, and 1800 s) except the 0 s time interval as these tomatoes had not entered the submersion media. For all tomato treatments/temperature differentials/time intervals examined, weigh t loss and/or gain did not exceed 0.1 g and difference in weight before and after submersion was not significant ( P > 0.05) for any treatment. Water infiltration was higher in green tomatoes than in red. In the green tomatoes, water ingress was highest at the 0Â°C temperature differential and did not increase with submersion time. Red tomatoes (21 and 4Â°C) had negative weight gain following submersion for at least two time intervals at all temperature differentials. At all temperature differentials/time inte rvals examined, red tomatoes at 21Â°C had an overall average weight loss of 0.01Â± 0.02 g, while red tomatoes at 4Â°C had no difference between weight before and after submersion (0.0 Â± 0.07 g). Pathogen Recovery Internalized Salmonella populations in each segment were reported in log MPN/segment. Results were calculated following confirmation of presumptive positive Salmonella colonies from selective agars onto TSI and LIA agar slants. Combinations of positive and negative tubes were compared to the appropr iate MPN Index table for a 3 Ã— 3 MPN analysis using MPN dilutions of 10, 1, and 0.1 ml (USDA, 2013). The detection limit for this methodology is 0.079 log MPN/segment. Average internalized Salmonella populations (n=6 12) are listed in Table 4 6 for segment A, the core segment just below the stem scar. One replication was performed for 3 and +3Â°C temperature differentials (n=6); two replications were performed for 5, 0, and +5Â°C temperature differentials (n=12). Tables 4 7 and 4 8 include average internali zed Salmonella populations (n=6)
61 for segments B and C, the middle core segment and lowest core segment above the blossom scar, respectively. Within each maturity and pulp temperature examined, there were no significant differences ( P > 0.05) in internalize d Salmonella populations between water temperature differentials. Internalization into Green Tomatoes (21Â°C) In mature green tomatoes, the highest populations of internalized Salmonella were recovered from segment A, just below the stem scar. Average inte rnalized Salmonella populations (n=6 12) are displayed in Table 4 6 for segment A. Figure 4 1 depicts average internalized Salmonella populations in segment A across temperature differentials based on submersion time. None of the tomatoes in the negative c ontrol trials (0 s) were positive for Salmonella internalization. Significantly larger ( P < 0.05) populations internalized into the 1800 s time interval, positive control, than the 30, 120, and 300 s time intervals at the 5Â°C temperature differential and the 30, 60, and 120 s time intervals at the 0Â°C temperature differentials. No Salmonella populations were recovered (< 0.079 log MPN/segment) from any tomatoes at the 120 s time interval and the +3Â°C temperature differential; however this was not significantly different ( P > 0.05) from other time intervals at the +3Â°C temperature differential. The highest average internali zed Salmonella populations, not including the positive control, was 0.948Â± 0.960 log MPN/segment, recovered from the 300 s time interval and the 0Â°C temperature differential. This internalization was 0.862 log MPN/segment higher ( P < 0.05) than the 30 s ti me interval at the same 0Â°C temperature differential. For all temperature differentials examined, internalized Salmonella populations did not increase with time. Average internalized Salmonella populations (n=6) for segment B, the middle core segment, ar e reported in Table 4 7 and Figure 4 2. Although not significant ( P >
62 0.05), populations in segment B are lower than in segment A. No Salmonella populations were recovered (<0.079 log MPN/segment) from any tomatoes at the 30 s time interval at the 5Â°C tem perature differential, at the 60 and 120 s time intervals at the 3Â°C temperature differential, at the 30 s time interval at the 0Â°C temperature differential, and at the 60 and 120 s time intervals at the +5Â°C temperature differential. Internalized Salmone lla populations were not significantly different ( P > 0.05) based on submersion time; increased submersion time did not lead to increased Salmonella internalization. Although not statistically significant ( P > 0.05) internalized Salmonella populations were highest, 0.839 Â± 0.181 log MPN/segment, excluding the positive control, following a 300 s submersion at the +5Â° C temperature differential . Internalized Salmonella populations in core segment C, directly above the blossom end, did not significantly chang e across submersion times ( P > 0.05). Average Salmonella populations (n=6) are displayed in Table 4 8 and Figure 4 3 for segment C. No Salmonella populations were recovered (<0.079 log MPN/segment) from any tomatoes at the 30 and 120 s time intervals at th e 5Â°C temperature differential, at the 120 s time interval at the 3Â°C temperature differential, at the 30 s time interval at the 0Â°C temperature differential, at the 120 s time interval at the +3Â°C temperature differential, and at the 30, 60, and 120 s t ime intervals at the +5Â°C temperature differential. Excluding the positive control submersion time, the highest internalized Salmonella populations into core segment C, 0.945 Â± 0.956 log MPN/segment, were recovered following a 300 s submersion at the +5Â°C temperature differential. For all core segments, internalized Salmonella populations varied greatly on a tomato to tomato basis. Individual internalized Salmonella populations in each core
63 segment at each temperature differential are displayed in Figures 4 4 through 4 8. Internalized Salmonella populations were not significantly different ( P > 0.05) based on segment location. Internalization into Red Tomatoes (21Â°C) Internalization of Salmonella populations was highest in segment A, just below the stem s car in 21 Â°C red tomatoes. Average internalized Salmonella populations (n=6 12) in segment A can be found in Table 4 6. The distribution of internalized Salmonella populations across temperature differentials based on submersion time can be viewed in Figur e 4 9. None of the tomatoes in the negative control trials (0 s) were positive for Salmonella internalization. Although populations did not always increase with submersion time, significantly larger ( P < 0.05) populations internalized into the positive con trol 1800 s time interval than the 30 and 120 s time intervals at the 5Â°C temperature differential, 30 and 60 s time intervals at the 3Â°C temperature differential, and 30, 60, and 120 s time intervals at the +5Â°C temperature differential. No Salmonella p opulations (<0.079 log MPN/segment) were recovered from any tomatoes at the 30s time interval at the 5Â°C temperature differential and the 30 and 60 s time intervals at the 3Â°C temperature differential. The highest average internalized Salmonella populati ons in segment A, 0.588 Â± 0.717 log MPN/segment, was recovered at the 300 s time interval at the 5Â°C temperature differential. Average internalized Salmonella populations were lower in segment B than in segment A, although not significantly ( P > 0.05). Average internalized Salmonella populations (n=6) are displayed in Table 4 7 and Figure 4 10. Internalized Salmonella populations were not significantly different ( P > 0.05) at any time interval tested. No Salmonella populations (<0.079 log MPN/segment) we re recovered from any tomatoes
64 at the 30 and 60 s time intervals at the 5Â°C temperature differential, 30, 60, 120, and 300 s time intervals at the 3Â°C temperature differential, 120 and 300 s time interval at the 0Â°C temperature differential, 120 and 300 s time intervals at the +3Â°C temperature differential, and 30 and 60 s time intervals at the +5Â°C. The highest average internalized Salmonella population in segment B was 0.200 Â± 0.260 log MPN/segment, recovered at the 120 s time interval at the 5Â°C tempe rature differential. In segment C, internalized Salmonella populations were not significantly different ( P > 0.05) based on submersion time; increased submersion time did not lead to increased Salmonella internalization. A distribution of average interna lized Salmonella populations recovered from segment C for each temperature differential and time interval can be found in Figure 4 11 and the values are reported in Table 4 8. Average internalized Salmonella populations in segment C were significantly lowe r ( P < 0.05) than in segment A at the positive control 1800 s time interval at the 5Â°C temperature differential. No Salmonella populations (< 0.079 log MPN/segment) were recovered from any tomatoes at the 30 and 1800 s time interval at the 5Â°C temperature differential, at the 30, 60, and 120 s time interval at the 3Â°C temperature differential, at the 120 and 300 s time intervals at the 0Â°C tem perature differential, at the 30, 60, 120, and 300 s time intervals at the +3Â°C temperature differential, and at the 30, 60, and 120 s time intervals at the +5Â°C temperature differentials. For all core segments, internalized Salmonella populations varied g reatly on a tomato to tomato basis. Individual internalized Salmonella populations in each core segment at each temperature differential are displayed in Figures 4 12 through 4 16.
65 Internalization in 4Â°C Red Tomatoes Average internalized Salmonella popul ations in red tomatoes at 4Â°C were higher in segment A than in segments B and C. Average (n=6 12) internalized Salmonella populations are displayed in Table 4 6. The distribution of average internalized Salmonella populations in segment A across temperatur e differentials and submersion times is depicted in Figure 4 17. None of the tomatoes in the negative control trials (0 s) were positive for Salmonella internalization. At the 1800 s time interval, internalized populations were significantly higher ( P < 0. 05) than the 30, 60, 120, and 300 s time intervals at the 5Â°C temperature differential. Excluding the 1800 s time interval, the highest internalized Salmonella population, 0.898 Â± 1.086 log MPN/segment, was recovered at the 300 s time interval at the 5Â°C temperature differential trial. Average internalized Salmonella populations in segment B are displayed in Table 4 7 and Figure 4 18. Although not significantly different ( P > 0.05), internalization in segment B at the 300 s at the 3Â°C temperature diffe rential was 0.172 log MPN/segment larger than the highest internalized population in segment A. For all temperature differentials examined, internalized Salmonella populations did not increase with time and populations were not significantly different ( P > 0.05) across time intervals. No Salmonella populations (<0.079 log MPN/segment) were recovered from any tomatoes at the 30 s time interval at the 5Â°C temperature differential, 60, 120, and 300 s time intervals at the 0Â°C temperature differential, and 30 s time interval at the +5Â°C temperature differential. Average Salmonella internalization in segment C was significantly lower ( P < 0.05) than in segment A at the 60 s time interval at the 5Â°C temperature differential. Average internalized Salmonella pop ulations recovered from segment C can be viewed
66 in Table 4 8 and Figure 4 19. Salmonella internalization did not increase with submersion time; however, average internalized Salmonella populations at the1800 s time interval were significantly higher ( P < 0 .05) than the 60 and 300 s time intervals at the 0Â°C temperature differential. No Salmonella populations were recovered from any tomatoes at the 30 and 60 s time interval at the 5Â°C temperature differential, at the 60 s time interval and the 3Â°C temperat ure differential, at the 60 and 300 s time interval at the 0Â°C temperature differential, at the 300 s time interval at the +3Â°C temperature differential, and at the 30 s time interval at the +5Â°C temperature differential. For all core segments, internaliz ed Salmonella populations varied greatly on a tomato to tomato basis. Individual internalized Salmonella populations in each core segment at each temperature differential are displayed in Figures 4 20 through 4 24. Comparison of Internalization Among all Maturities and Pulp Temperatures Internalized Salmonella populations were lower in red tomatoes at 21Â°C than red tomatoes at 4Â°C and green tomatoes at 21Â°C. Internalization is significantly higher ( P < 0.05) in segment A in red tomatoes at 4Â°C than in re d tomatoes at 21Â°C at the 120 s time interval at the +5Â°C temperature differential. Internalization in green tomatoes at 21Â°C is significantly higher ( P < 0.05) than in red tomatoes at 4Â°C at the 300 and 1800 s time intervals at the 0Â°C temperature differe ntial. Salmonella populations in segment A based on maturity and pulp temperature were not significantly different ( P > 0.05) at any other time intervals and/or temperature differentials. For segment B, the middle core segment, internalized Salmone lla populations were significantly higher ( P < 0.05) in red tomatoes at 4Â°C than red tomatoes at 21Â°C at the 300 s time interval at the 3Â°C temperature differential. Salmonella internalization was not significantly different ( P > 0.05) in segment C when compared across maturities
67 and pulp temperatures. Despite a lack of correlation between Salmonella internalization and submersion time, internalization in all segments was highest for most maturities and pulp temperatures at the 300 s time interval .
68 Table 4 1. pH of water in submersion media. a Values represent average of triplicate pH readings (n=3) of water used for submersion media Â± standard deviation. b Two replications were completed for 5, 0, and +5 temperature differentials (n=6). c One replication was completed for 3 and +3 temperature differe ntials (n=3). Maturity and Pulp Temperature Temperature Differential (Â°C) Time Interval (s) 30 60 120 300 pH Green (21Â°C) 5 7.6Â±0.1 a,b 7.6Â±0.0 7.6Â±0.0 7.5Â±0.0 3 7.5Â±0.0 c 7.5Â±0.0 7.6Â±0.0 7.6Â±0.0 0 7.7Â±0.0 7.6Â±0.0 7.5Â±0.0 7.6Â±0.0 +3 7.5Â±0.0 7.6Â±0.0 7.7Â±0.0 7.5Â±0.0 +5 7.7Â±0.0 7.5Â±0.0 7.6Â±0.1 7.6Â±0.0 Red (21Â°C) 5 7.7Â±0.1 7.5Â±0.0 7.6Â±0.0 7.5Â±0.0 3 7.7Â±0.0 7.6Â±0.0 7.5Â±0.0 7.6Â±0.0 0 7.7Â±0.0 7.5Â±0.0 7.5Â±0.0 7.6Â±0.0 +3 7.5Â±0.0 7.5Â±0.0 7.6Â±0.0 7.6Â±0.0 +5 7.7Â±0.0 7.6Â±0.0 7.6Â±0.0 7.6Â±0.0 Red (4Â°C) 5 7.7Â±0.1 7.6Â±0.0 7.5Â±0.0 7.5Â±0.0 3 7.6Â±0.0 7.6Â±0.0 7.5Â±0.0 7.6Â±0.0 0 7.6Â±0.1 7.6Â±0.0 7.6Â±0.0 7.5Â±0.0 +3 7.5Â±0.0 7.6Â±0.0 7.5Â±0.0 7.6Â±0.0 +5 7.6Â±0.0 7.6Â±0.0 7.5Â±0.0 7.5Â±0.0
69 Table 4 2. Oxidation Reduction Potent ial (ORP) measurements of water in submersion media. a Values represent average of triplicate ORP readings (n=3) of water used for submersion media Â± standard deviation. b Two replications were completed for 5, 0, and +5 temperature differentials (n=6). c One replication was completed for 3 and +3 temperature differ entials (n=3). Maturity and Pulp Temperature Temperature Differential (Â°C) Time Interval (s) 30 60 120 300 ORP (mV) Green (21Â°C) 5 163Â±5 a,b 147Â±4 160Â±3 149Â±3 3 166Â±2 c 158Â±0 150Â±1 161Â±1 0 162Â±7 159Â±2 147Â±5 155Â±3 +3 148Â±2 153Â±0 159Â±1 155Â±0 +5 152Â±5 147Â±5 158Â±3 163Â±6 Red (21Â°C) 5 163Â±6 149Â±2 164Â±2 158Â±6 3 170Â±0 157Â±1 153Â±1 166Â±2 0 152Â±3 155Â±2 156Â±4 156Â±3 +3 167Â±2 155Â±2 152Â±1 160Â±3 +5 150Â±3 155Â±4 163Â±3 162Â±6 Red (4Â°C) 5 151Â±4 169Â±4 160Â±2 152Â±4 3 156Â±0 165Â±1 152Â±2 149Â±1 0 170Â±2 156Â±3 166Â±4 154Â±3 +3 154Â±2 144Â±1 157Â±1 162Â±0 +5 160Â±5 171Â±5 153Â±1 152Â±4
70 Table 4 3. Difference in tomato weight (g) of green tomatoes (21Â°C) before and after submersion . a Values represent average (n=6) change in weight (after submersion before submersion) Â± standard deviation. Table 4 4. Difference in tomato weight (g) of red tomatoes (21Â°C) before and after s ubmersion . a Values represent average (n=6) change in weight (after submersion before submersion) Â± standard deviation. Table 4 5. Difference in tomato weight (g) of red tomatoes (4Â°C) before and after s ubmersion . a Values represent average (n=6) change in weight (after submersion before submersion) Â± standard deviation. Temperature Differential (Â°C) Time Interval (s) 30 60 120 300 1800 5 0.07Â±0.03 a 0.08Â±0.02 0.08Â±0.04 0.08Â±0.02 0.03Â±0.10 0 0.09Â±0.01 0.09Â±0.01 0.09Â±0.02 0.09Â±0.02 0.02Â±0.13 +5 0.01Â±0.01 0.01Â±0.02 0.02Â±0.01 0.00Â±0.01 0.04Â±0.02 Temperature Differential (Â°C) Time Interval (s) 30 60 120 300 1800 5 0.00Â±0.01 a 0.01Â±0.01 0.01Â±0.01 0.00Â±0.01 0.07Â±0.02 0 0.01Â±0.01 0.01Â±0.00 0.01Â±0.01 0.00Â±0.01 0.01Â±0.02 +5 0.00Â±0.01 0.00Â±0.01 0.01Â±0.00 0.01Â±0.01 0.05Â±0.03 Temperature Differential (Â°C) Time Interval (s) 30 60 120 300 1800 5 0.00Â±0.06 a 0.05Â±0.08 0.02Â±0.12 0.07Â±0.08 0.05Â±0.10 0 0.03Â±0.05 0.02Â±0.04 0.03Â±0.05 0.05Â±0.08 0.05Â±0.06 5 0.03Â±0.05 0.02Â±0.04 0.02Â±0.04 0.05Â±0.05 0.05Â±0.10
71 Table 4 6. Salmonella populations (log MPN/segment) recovered from segment A, just below the stem scar . a Values represent average internalized Salmonella populations recovered Â± standard deviation for either one or two replications (n=6 12). b Mean values (log CFU /segment) between: (i) temperature differential in columns 1 ; (ii) time intervals in rows (Capital letters), and (iii) maturity treatments (Green 21Â°C, Red 21Â°C, Red 4Â°C); (small letters), were analyzed for significant differences ( P<0.05 ). c All samples (n=6 12) were below th e minimum level of detection (<0.079 log MPN/segment). 1 No significant differences ( P>0.05) exist between temperature differentials (in columns). Maturity and Pulp Temperature Temperature Differential (Â°C) Time Intervals (s) 0 30 60 120 300 1800 Green (21Â°C) 5 0Â±0 a 0.133Â±0.134Ba b 0.554Â±0.743ABa 0.363Â±0.515Ba 0.396Â±0.711Ba 2.090Â±0.975Aa 3 0Â±0 0.457Â±0.809Aa 0.106Â±0.037Aa 0.572Â±0.937Aa 0.169Â±0.135Aa 2.022Â±1.076Aa 0 0Â±0 0.086Â±0.022Ba 0.210Â±0.411BCa 0.092Â±0.030BCa 0.948Â±0.960ACa 1.695Â±0.711Aa +3 0Â±0 0.261Â±0.272Aa 0.272Â±0.431Aa <0.079Â±0.000Aa c 0.885Â±0.861Aa 2.002Â±1.145Aa +5 0Â±0 0.282Â±0.380Aa 0.255Â±0.216Aa 0.182Â±0.318Aab 0.299Â±0.707Aa 0.710Â±0.834Aa Red (21Â°C) 5 0Â±0 <0.079Â±0.000Ba c 0.347Â±0.705ABa 0.293Â±0.709Ba 0.588Â±0.717ABa 1.347Â±1.123Aa 3 0Â±0 <0.079Â±0.000Ba c <0.079Â±0.000Ba c 0.092Â±0.030ABa 0.520Â±0.950ABa 1.401Â±1.243Aa 0 0Â±0 0.160Â±0.181Aa 0.285Â±0.427Aa 0.490Â±0.744Aa 0.215Â±0.382Aab 0.794Â±0.865Aab +3 0Â±0 0.160Â±0.181Aa 0.080Â±0.000Aa 0.396Â±0.709Aa 0.092Â±0.030Aa 1.286Â±0.712Aa +5 0Â±0 0.092Â±0.030Ba 0.086Â±0.022Ba 0.120Â±0.135Bb 0.163Â±0.132ABa 0.737Â±0.608Aa Red (4Â°C) 5 0Â±0 0.257Â±0.425Ba 0.489Â±0.576ABa 0.596Â±0.763ABa 0.898Â±1.086 ABa 1.482Â±0.933Aa 3 0Â±0 0.272Â±0.425Aa 0.373Â±0.330Aa 0.468Â±0.387Aa 0.768Â±0.996Aa 1.010Â±1.055Aa 0 0Â±0 0.497Â±0.830Aa 0.133Â±0.134Aa 0.174Â±0.180Aa 0.244Â±0.524Ab 0.476Â±0.410Ab +3 0Â±0 0.187Â±0.173Aa 0.515Â±0.436Aa 0.209Â±0.256Aa 0.092Â±0.030Aa 1.466Â±0.913Aa +5 0Â±0 0.375Â±0.474Aa 0.707Â±0.958Aa 0.532Â±0.601Aa 0.527Â±0.510Aa 0.738Â±0.861Aa
72 Table 4 7. Salmonella populations (log MPN/segment) recovered from segment B, middle core segment. a Values represent average internalized Salmonella populations recovered Â± standard deviation for one replication (n=6). b Mean values (log CFU /segment) between: (i) temperature differential in columns 1 ; (ii) time intervals in rows 2 , and (iii) maturity treatments (Green 21Â°C, Red 21Â°C, Red 4Â°C); (Capital letters), were analyzed for significant differences ( P<0.05 ). c All samples (n=6) were below the minimum level of detection (<0.079 log MPN/segment). 1 No significant differences ( P>0 .05) exist between temperature differentials (in columns). 2 No significant differences ( P>0.05) exist between time intervals (in rows). Maturity and Pulp Temperature Temperature Differential (Â°C) Time Intervals (seconds) 0 30 60 120 300 1800 Green (21Â°C) 5 0Â±0 a <0.079Â±0.000A b 0.604Â±0.808A 0.158Â±0.143A 0.132Â±0.117A 0.064 Â±0.386A 3 0Â±0 0.092Â±0.030A <0.079Â±0.000A c <0.079Â±0 .000 A c 0.263Â±0.411AB 1.36 5Â±1.106 A 0 0Â±0 <0.079Â±0.000A c 0.106Â±0.037A 0.160Â±0.181A 0.158Â±0.146A 1.188 Â±1.00 3 A +3 0Â±0 0.092Â±0.030A 0.278Â±0.330A 0.092Â±0.030A 0.321Â±0.335A 1.25 2 Â±0.916A +5 0Â±0 0.272Â±0.431A <0.079Â±0.000A c <0.079Â±0 .000 A c 0.839Â±0.181A 0.540Â±0.211A Red (21Â°C) 5 0Â±0 <0.079Â±0.000A c <0.079Â±0.000A c 0.200Â±0.260A 0.162Â±0.185A 0.099Â±0.034A 3 0Â±0 <0.079Â±0.000A c <0.079Â±0.000A c <0.079Â±0 .000 A c <0.079Â±0 .000 B 1.217Â±1.156 A 0 0Â±0 0.132Â±0.117A 0.157Â±0.109A <0.079Â±0 .000 A c <0.079Â±0 .000 A c 0.201Â±0.211A +3 0Â±0 0.131Â±0.115 A 0.160Â±0.115A <0.079Â±0 .000 A c <0.079Â±0 .000 A c 1.199 Â±0.940A +5 0Â±0 <0.079Â±0.000A c <0.079Â±0.000A c 0.132Â±0.117A 0.131Â±0.115A 0.818Â±0.706A Red (4Â°C) 5 0Â±0 <0.079Â±0.000A c 0.092Â±0.030A 0.311Â±0.316A 0.588Â±0.936A 0.442Â±0.353A 3 0Â±0 0.145Â±0.146A 0.160Â±0.181A 0.253Â±0.320A 1.069 Â±0.923A 0.682Â±0.754A 0 0Â±0 0.680Â±0.620A <0.079Â±0 .000 A c <0.079Â±0 .000 A c <0.079Â±0 .000 A c 1.217 Â±0.921A +3 0Â±0 0.145Â±0.146A 0.171Â±0.139A 0.145Â±0.146A 0.092Â±0.030A 1.359 Â±0.911A +5 0Â±0 <0.079Â±0.000A c 0.661Â±0.712A 0.291Â±0.352A 0.298Â±0.421A 0.720Â±0.911 A
73 Table 4 8. Salmonella populations (log MPN/segment) recovered from segment C, just above the blossom end. a Values represent average internalized Salmonella populations recovered Â± standard deviation for one replication (n=6). b Mean values (log CFU /segment) between: (i) temper ature differential in columns 1 , (ii) time intervals in rows (Capital letters), and (iii) maturity treatments (Green 21Â°C, Red 21Â°C, Red 4Â°C) 2 were analyzed for significant differences ( P<0.05 ). c All samples (n=6) were below the minimum level of detection ( <0.079 log MPN/segment). 1 No significant differences ( P>0.05) exist between temperature differentials (in columns). 2 No significant differences ( P>0.05) exist between maturity treatments (Green 21Â°C, Red 21Â°C, Red 4Â°C). Maturity and Pulp Temperature Temperature Differential (Â°C) Time Intervals (seconds) 0 30 60 120 300 1800 Green (21Â°C) 5 0Â±0 a <0.079Â±0.000A b,c 0.583Â±0.484A <0.079Â±0.000A c 0.145Â±0.146A 0.665Â±0.511A 3 0Â±0 0.079Â±0.030A <0.0790Â±0.000A c <0.079Â±0.000A c 0.240Â±0.325A 1.267Â±1.195A 0 0Â±0 <0.079Â±0.000A c 0.173Â±0.178A 0.223Â±0.209A 0.119Â±0.037A 1.009Â±1.055A +3 0Â±0 0.079Â±0.000A c 0.227Â±0.330A <0.079Â±0.000A c 0.433Â±0.479A 1.502Â±1.158A +5 0Â±0 <0.079Â±0.000A c <0.079Â±0.000A c <0.079Â±0.000A c 0.945Â±0.956A 0.650Â±0.034A Red (21Â°C) 5 0Â±0 <0.079Â±0.000A c 0.092Â±0.030A 0.143Â±0.142A 0.160Â±0.181A <0.079Â±0.000A c 3 0Â±0 < 0.079Â±0.000A c <0.079Â±0.000A c <0.079Â±0.000A c 0.131Â±0.115A 0.904Â±0.844A 0 0Â±0 0.092Â±0.030A 0.197Â±0.168A <0.079Â±0.000A c <0.079Â±0.000A c 0.099Â±0.034A +3 0Â±0 <0.079Â±0.000A c <0.079Â±0.000A c <0.079Â±0.000A c <0.079Â±0.000A c 1.305Â±0.766A +5 0Â±0 <0.079Â±0.000A c < 0.079Â±0.000A c <0.079Â±0.000A c 0.132Â±0.117A 0.338Â±0.229A Red (4Â°C) 5 0Â±0 <0.079Â±0.000A c <0.079Â±0.000A c 0.310Â±0.185A 0.585Â±0.926A 0.221Â±0.202A 3 0Â±0 0.092Â±0.030A <0.079Â±0.000A c 0.092Â±0.030A 0.726Â±0.950A 0.285Â±0.570A 0 0Â±0 0.362Â±0.633AB < 0.079Â±0.000B c 0.301Â±0.256AB <0.079Â±0.000B c 0.702Â±0.725A +3 0Â±0 0.092Â±0.030A 0.092Â±0.030A 0.092Â±0.030A <0.079Â±0.000A c 1.001Â±0.610A +5 0Â±0 <0.079Â±0.000A c 0.661Â±0.712A 0.291Â±0.352A 0.298Â±0.421A 0.720Â±1.110A
74 Fi gure 4 1. Average Salmonella populations plus or minus standard deviation recovered from Segment A (below the stem scar; n=6 12) in green tomatoes at ambient temperatures (21Â°C). Submersion times are as follows: 30 s (white bar), 60 s (light grey bar), 120 s (grey bar), 300 s (dark grey bar), and 1800 s (black bar). The solid line on the graph indicates the limit of detection (0.079 log MPN/segment). 0 0.5 1 1.5 2 2.5 3 3.5 -5 -3 0 3 5 Salmonella population (log MPN/segment) Temperature Differential ( C)
75 Figure 4 2. Average Salmonella populations plus or minus standard deviation recovered fro m Segment B (middle segment ; n=6 ) in green tomatoes at ambient temperatures (21Â°C). Submersion times are as follows: 30 s (white bar), 60 s (light grey bar), 120 s (grey bar), 300 s (dark grey bar), and 1800 s (black bar). The solid line on the graph indicates the limit of detection (0.0 79 log MPN/segment). 0 0.5 1 1.5 2 2.5 3 3.5 -5 -3 0 3 5 Salmonella population (log MPN/segment) Temperature Differential ( C)
76 Figure 4 3. Average Salmonella populations plus or minus standard deviation recovered fro m Segment C (above the blossom end ; n=6) in green tomatoes at ambient temperatures (21Â°C). Submersion times are as follows: 30 s (white bar), 60 s (light grey bar), 120 s (grey bar), 300 s (dark grey bar), and 1800 s (black bar). The solid line on the graph indicates the limit of detection (0.079 log MPN/segment). 0 0.5 1 1.5 2 2.5 3 3.5 -5 -3 0 3 5 Salmonella population (log MPN/segment) Temperature Differential ( C)
77 Figure 4 4. Internalized Salmonella populations recovered from (i) segment A (open tomatoes (21Â°C) in submersion media at the 5Â°C temperature differential. Segment A values are reported from two replications (n=1 2), segment B and C values are reported from one replication (n=6). 0 0.5 1 1.5 2 2.5 3 0 50 100 150 200 250 300 350 Salmonella population (log MPN/segment) Time Interval (s)
78 Figure 4 5. Internalized Salmonella populations recovered from (i) segment A (open tomatoes (21Â°C) in sub mersion media at the 3Â°C tempera ture differential. Segment A, B and C values are reported from one replication (n=6). 0 0.5 1 1.5 2 2.5 3 0 50 100 150 200 250 300 350 Salmonella population (log MPN/segment) Time Interval (s)
79 Figure 4 6. Internalized Salmonella populations recovered from (i) segment A (open (iii) segment C (Ã—) of green tomatoes (21Â°C) in submersion media at the 0Â°C temperature differential. Segment A values are reported from two replications (n=12), segment B and C values are reported from one replication (n=6). 0 0.5 1 1.5 2 2.5 3 0 50 100 150 200 250 300 350 Salmonella population (log MPN/segment) Time Interval (s)
80 Figure 4 7. Internalized Salmonella populations recovered from (i) segment A (open tomatoes (21Â°C) in submersion media at the 3Â°C temperat ure differential. Segment A, B and C values are reported from on e replication (n=6). 0 0.5 1 1.5 2 2.5 3 0 50 100 150 200 250 300 350 Salmonella population (log MPN/segment) Time Interval (s)
81 Figure 4 8. Internalized Salmonella populations recovered from (i) segment A (open tomatoes (21Â°C) in submersion media at the +5Â°C temperature different ial. Segment A values are reported from two replications (n=12), segment B and C values are reported from one replication (n=6). 0 0.5 1 1.5 2 2.5 3 0 50 100 150 200 250 300 350 Salmonella population (log MPN/segment) Time Interval (s)
82 Figure 4 9. Average Salmonella populations plus or minus standard deviation recovered from Segment A (below the stem scar ; n=6 12) in red tomatoes at ambient temperatures (21Â°C). Submersion times are as follows: 30 s (white bar), 60 s (light grey bar), 120 s (grey bar), 300 s (dark grey bar), and 1800 s (black bar). The solid line on the graph indicates the limit of detectio n (0.079 log MPN/segment). 0 0.5 1 1.5 2 2.5 3 3.5 -5 -3 0 3 5 Salmonella population (log MPN/segment) Temperature Differentials ( C)
83 Figure 4 10. Average Salmonella populations plus or minus standard deviation r ecovered from Segment B (middle segment ; n=6) in red tomatoes at ambient temperatures (21Â°C). Submersion times are as follows: 30 s (white bar), 60 s (light grey bar), 120 s (grey bar), 300 s (dark grey bar), and 1800 s (black bar). The solid line on the graph indicates the limit of detection (0.079 log MPN/segment). 0 0.5 1 1.5 2 2.5 3 3.5 -5 -3 0 3 5 Salmonella population (log MPN/segment) Temperature Differentials ( C)
84 Figure 4 11. Average Salmonella populations plus or minus standard deviation recovered fro m Segment C (above the blossom end ; n=6) in red tomatoes at ambient temperatures (21Â°C). Submersion times are as follows: 30 s (white bar), 60 s (light grey bar), 120 s (grey bar), 300 s (dark grey bar), and 1800 s (black bar). The solid line on the graph indicates the limit of detection (0.079 log MPN/segment). 0 0.5 1 1.5 2 2.5 3 3.5 -5 -3 0 3 5 Salmonella population (log MPN/segment) Temperature Differentials ( C)
85 Figure 4 12. Internalized Salmonella populations recovered from (i) segment A (open iii) segment C (Ã—) of red tomatoes (21Â°C ) in submersion media at the 5Â°C temperature differential. Segment A values are reported from two replications (n=12), segment B and C values are reported from one replication (n=6). 0 0.5 1 1.5 2 2.5 3 0 50 100 150 200 250 300 350 Salmonella population (log MPN/segment) Time Interval (s)
86 Figure 4 13. Internalized Salmonella populations recovered from (i) segment A (open tomatoes (21Â°C ) in submersion media at the 3Â°C temperatu re differential. Segment A, B and C values are reported from one replication (n=6). 0 0.5 1 1.5 2 2.5 3 0 50 100 150 200 250 300 350 Salmonella population (log MPN/segment) Time Interval (s)
87 Figure 4 14. Internalized Salmonella populations recovered from (i) segment A (open tomatoes (21Â°C ) in submersion media at the 0Â°C temperature differential. Segment A values are reported from two replications (n=12), segment B and C values are reported from one replication (n=6). 0 0.5 1 1.5 2 2.5 3 0 50 100 150 200 250 300 350 Salmonella population (log MPN/segment) Time Interval (s)
88 Figure 4 15. Internalized Salmonella populations recovered from (i) segment A (open ), and (iii) segment C (Ã—) of red tomatoes (21Â°C ) in submersion media at the +3Â°C temperature differen tial. Segment A, B and C values are reported from one replication (n=6). 0 0.5 1 1.5 2 2.5 3 0 50 100 150 200 250 300 350 Salmonella population (log MPN/segment) Time Interval (s)
89 Figure 4 16. Internalized Salmonella populations recovered from (i) segment A (open tomatoes (21Â°C ) in submersion media at the +5Â°C temperature differential. Segment A values are reported from two replications (n=12), segment B and C values are reported fr om one replication (n=6). 0 0.5 1 1.5 2 2.5 3 0 50 100 150 200 250 300 350 Salmonella population (log MPN/segment) Time Interval (s)
90 Figure 4 1 7. Average Salmonella populations plus or minus standard deviation recovered from Segment A (below the stem scar; n=6 12) in red tomatoes at refrigeration temperatures (4Â°C). Submersion times are as follows: 30 s ( white bar), 60 s (light grey bar), 120 s (grey bar), 300 s (dark grey bar), and 1800 s (black bar). The solid line on the graph indicates the limit of detection (0.079 log MPN/segment). 0 0.5 1 1.5 2 2.5 3 3.5 -5 -3 0 3 5 Salmonella population (log MPN/segment) Temperature Differentials ( C)
91 Figure 4 18. Average Salmonella populations plus or minus standard deviation recovered fro m Segment B (middle segment ; n=6) in red tomatoes at refrigeration temperatures (4Â°C). Submersion times are as follows: 30 s (white bar), 60 s (light grey bar), 120 s (grey bar), 300 s (dark grey b ar), and 1800 s (black bar). The solid line on the graph indicates the limit of detection (0.079 log MPN/segment). 0 0.5 1 1.5 2 2.5 3 3.5 -5 -3 0 3 5 Salmonella population (log MPN/segment) Temperature Differentials ( C)
92 Figure 4 19. Average Salmonella populations plus or minus standard deviation recovered fro m Segment C (above the blossom end ; n=6) in re d tomatoes at refrigeration temperatures (4Â°C). Submersion times are as follows: 30 s (white bar), 60 s (light grey bar), 120 s (grey bar), 300 s (dark grey bar), and 1800 s (black bar). The solid line on the graph indicates the limit of detection (0.079 l og MPN/segment). 0 0.5 1 1.5 2 2.5 3 3.5 -5 -3 0 3 5 Salmonella population (log MPN/segment) Temperature Differentials ( C)
93 Figure 4 20. Internalized Salmonella populations recovered from (i) segment A (open tomatoes (4Â°C) in submersion media at the 5Â°C temperature differential. Se gment A values are reported from two replications (n=12), segment B and C values are reported from one replication (n=6). 0 0.5 1 1.5 2 2.5 3 0 50 100 150 200 250 300 350 Salmonella population (log MPN/segment) Time Interval (s)
94 Figure 4 21. Internalized Salmonella populations recovered from (i) segment A (open and (iii) segment C (Ã—) of red tomatoes (4Â°C) in submersion media at the 3Â°C temperature differential. Seg ment A, B and C values are reported from one replication (n=6). 0 0.5 1 1.5 2 2.5 3 0 50 100 150 200 250 300 350 Salmonella population (log MPN/segment Time Intervals (s)
95 Figure 4 22. Internalized Salmonella populations recovered from (i) segment A ( open tomatoes (4Â°C) in submersion media at the 0Â°C temperature differential. Segment A values are reported from two replications (n=12), segment B and C values are reported from on e replication (n=6). 0 0.5 1 1.5 2 2.5 3 0 50 100 150 200 250 300 350 Salmonella population (log MPN/segment) Time Interval (s)
96 Figure 4 23. Internalized Salmonella populations recovered from (i) segment A (open tomatoes (4Â°C) in submersion media at the +3Â°C temperature different ial. Segment A, B and C values are reported from one replication (n=6). 0 0.5 1 1.5 2 2.5 3 0 50 100 150 200 250 300 350 Salmonella population (log MPN/segment) Time Interval (s)
97 Figure 4 24 . Internalized Salmonella populations recovered from (i) segment A (open tomatoes (4Â°C) in submersion media at the +5Â°C temperature differential. Segment A values are reported from two replications (n=12), segment B and C values are reported from one replication (n=6). 0 0.5 1 1.5 2 2.5 3 0 50 100 150 200 250 300 350 Salmonella population (log MPN/segment) Time Interval (s)
98 CHAPTER 5 DISCUSSION In 1990, tomatoes were first implicated as a vehicle for foodborne illness outbreaks (Hedberg et al., 1999). From 1990 to 2009, a total of 1,095 illnesses have been caused by tomatoes contaminated with Salmonella enterica (Valadez et al., 2012). Tomatoes can be contaminated with human pathogens can occur at all steps within the tomato supply chain ; guidance documents for safe practices to minimize bacterial contamination of tomatoes have been put in place by the Food and Drug Administration (FDA, 2006, 2009) and United Fresh (FDA, 2008,). The Florida Department of Agriculture and Consumer Services (FDACS) has mandated that tomato Good Agricultural Practices and Best Management Practices be implemented throughout the tomato supply chain (FDACS, 2008). Previous research conducted on mature green tomatoes and plant pathogens, such as Pectobacterium car o tovorum , determined that submersion of tomato fruits into water that is cooler than the internal fruit pulp temperature results in infiltration of water and internalization of plant pathogens (Showalter, 1979; Bartz and Showalter, 1981; Bartz, 1982). Inte rnalization into other commodities during immersion of warm apples, oranges, and mangoes into cold water has also been reported (Buchanan et al., 1999; Burnett et al., 2000; Eblen et al., 2004; Penteado et al., 2004). In 2009, the United Fresh Produce Asso ciation commented on put tomatoes in ice water for firming immediately before slicing. Un ited Fresh commented that this and other immersion practices present a potential for infiltration of
99 any present pathogens and that the FDA should expressly note that this practice is not recommended (United Fresh, 2009). In order to minimize the influx o f microorganisms into tomatoes, postharvest wash water at all steps within the supply chain must be maintained at a temperature that is 5Â°C/10Â°F warmer than the internal fruit temperature (FDA, 2006; FDA, 2008; FDACS, 2008). This recommendation is based on the internalization of plant pathogens and the subsequent incidence of bacterial soft rot; it was intended for the primary packinghouse where internal tomato temperatures are high due to field heat. The prescribed positive temperature differential between tomatoes and water has been put in place for subsequent washing steps beyond the initial washing and packing of tomatoes at the packinghous e despite a lack of research linking bacterial internalization to a negative temperature differential between water and tomatoes with cooler pulp temperatures. The aim of this research was to determine if temperature differential between tomato fruits and water plays a significant role in the internalization of Salmonella into tomatoes that have previously been washed a nd waxed and are maintained at temperatures typical of re packing and fresh cut operations. In order to determine the influence of temperature differential on Salmonella internalization into tomatoes with cooler temperatures, mature green and red round to matoes at 21Â°C and red tomatoes at 4Â°C were submerged to a depth of 6 cm in inoculated water to create a 0, Â±3, and Â±5Â°C temperature differential. Temperature differentials of up to 5Â°C between tomato pulp and immersion water did not significantly impact Salmonella internalization under any conditions examined. This finding does not support the current metrics in the TGAPs regulations, based upon previous finding
100 where a negative temperature differential led to increased internalization. Here, negative tem perature differentials of 35Â°C between tomatoes and water, with submersion times ranging from 10 to 30 min, led to increased water and postharvest phtyopathogens bacterial infiltration into tomatoes, resulting in extensive decay (Bartz and Showalter, 198 1). Positive, and small negative temperature differentials ( 2Â°C) t emperature differentials led to less disease . This research on plant pathogens suggests that interna lization is increased following imm ersion of fruits into cold water; it is difficult to c ompare these results determined by visual inspections of fruits to decay, to the cu rrent study where MPN analysis allowed us to quantify internalization . Other studies also support increased internalization when warm fruit are submerged into cool water. Z huang et al. (1995) observed more Salmonella uptake by core tissues of tomatoes when exposed to bacterial suspensions 15Â°C cooler than pulp temperatures than when exposed to suspensions 12Â°C warmer than pulp temperatures. Here, however, exterior core tissu es were not removed from samples so they are unable to ensure that Salmonella present on the outer stem scar surface rather than internalized did not influence this finding. Salmonella internalized into ripened and un ripened mangoes following immersion of fruit with substantial heat following a quarantine heat treatment, into water 26Â°C cooler than the pulp temperature and stored at different temperatures for up to one week following immersion (Penteado et al., 2004). Interestingly, the literature also reports occurrences when internalization is higher when fruits were submerged into warmer water. For example, when mature green tomatoes at 21Â°C were submerged in water with temperature differentials of 5.4 or
101 +11.2Â°C fo r 5 min, incidence of soft rot was lower in tomatoes submerged in water at a temperature differential of 5.4Â°C. The tomatoes at the negative temperature differential also had fewer skin cracks and detrimental effects during storage (Segall et al., 1977). There are also a number of published studies that support the observation that small temperature differentials and short contact times do not impact Salmonella internalization. Internalization of Escherichia coli O157:H7 into apples at either 2Â°C or 22Â°C following submersion of fruit for 20 min into water at 2Â°C, was less when cold apples were submerged into cold water (0Â°C temperature differential) than when warm apples were submerged into cold water ( 20Â°C temperature differential); however, E. coli O157 :H7 internalized primarily into the outer core of apples at both temperatures (Buchanan et al., 1999). Burnett et al. (2000) also evaluated influence of 23, 0, and +23Â°C temperature differentials on E. coli O157:H7 internalization into apples and reports infiltration of E. coli O157:H7 into core tissues of all apples, regardless of temperature differential (Burnett et al., 2010). Temperature differentials of 5.6, 0, and +5.6Â°C were not a significant factor in Salmonella internalization incidence into matu re green tomatoes at 32.2Â°C under conditions typically in a packinghouse; no more than 2 min submersion at a depth of two layers of tomatoes; temperature differential also had a limited effect on the populations of internalized cells (Xia et al., 2011). In order to quantify water ingress into tomatoes in this study, tomato weight before and after submersion was recorded at the 5, 0, +5Â°C temperature differential. No weight gain, equating to no water uptake, was observed at any temperature differential eval uated. This suggests that smaller temperature differentials, those within 5Â°C of the fruit temperature, may not be a factor in the infiltration of water and internalization of bacteria
102 into tomatoes which have previously been washed and waxed and with subs tantial postharvest stem removal time. Infiltration of water and bacteria through the stem scar can be explained by the general gas law, as tomato fruits cool, decreases in the internal gas pressure leads to a combination of partial vacuums inside the frui t and an influx from the external environment (Bartz and Showalter, 1981). Increased hydrostatic pressure, from fruit being submerged into water well below the surface may also impact infiltration (Bartz, 1982). Tomatoes have extensive intercellular air sp aces interconnected among the loosely bound ceils (Showalter, 1979). The relatively thick external wall of tomatoes and the heavily cutinized epidermis with no stomatal openings, mean that the exchange of gases in tomatoes occurs almost entirely at the ste m scar (Showalter, 1979). Bartz and Showalter (1981) confirmed that water can be drawn into the stem scar along with any present microorganisms based on the general gas law, when successful isolations of the inoculum bacteria were recovered from tissues ju st beneath the stem scar. The extent of the water and bacteria influx may be regulated by factors such as length of exposure, amount of fruit cooling, depth of immersion, viscosity of external environment, presence of sanitizers in the water, and size an d number of pores leading to internal airspaces (Bartz and Showalter, 1981).Internalization of Salmonella into tomatoes was highest in segment A, the segment just below the stem scar, suggesting the stem scar is the major point of entry for Salmonella inte rnalization. Populations in this segment were significantly greater than populations in segment C following an 1800 s (30 min) submersion (21Â°C red tomatoes) and a 60 s submersion (4Â°C red tomatoes) at the 5Â°C temperature differential. Xia et al. (2011) r emoved several internal tissue
103 samples in order to determine the distribution of internalized Salmonella within tomatoes. They also observed the highest internalization into the segment immediately below the stem scar and that internalized Salmonella popul ations declined as distance from the stem scar increased. Similarly in mangoes, internalized populations were significantly higher (P > 0.05) in the stem end segment than in middle and blossom end segments following cold water immersion (Penteado et al., 2 004). Infiltration of water and bacterial infiltration due to cooling requires fruits be in contact with cold water long enough for the pulp temperature to cool (Bartz, 1982); submersion time, in addition to temperature differential, plays an important ro le in infiltration. Here , positive control tomatoes submerged for 1800 s (30 min) have significantly higher internalized Salmonella populations than, shorter submersion times. At a temperature differential of 17Â°C, immersion for 10 min in P. car o tovorum resulted in significant decay, while those immersed for 2 min did not increase in weight and had a decay incidence similar to the control treatment (Bartz, 1982). Since both immersion time and temperature differential are central to the infiltration of bac teria into tomatoes, small submersion times and small temperature differentials may not result in bacterial internalization. Infiltration into mature green and red tomatoes at 21Â°C treatment was evaluated as both maturity stages may be repacked in Florida . While not significantly different, Salmonella internalization into green tomatoes at 21Â°C was higher than into red tomatoes at 21Â°C for all temperature differentials analyzed ; no published studies have evaluated the internalization of Salmonella into red tomatoes. Bartz and Showalter (1981) studied water infiltration and subsequent soft rot development in tomatoes of
104 varying maturities. Red, pink, and mature green tomatoes were exposed to bacterial suspensions at negative, zero, and positive temperature di fferentials. In red tomatoes exposed to a 14Â°C temperature differential, weight gain was not significantly different from tomatoes exposed to positive temperature differential treatments. Similar to what is report ed here for Salmonella internalization, wa ter uptake in red fruits was significantly lower than pink and green fruits. The tomatoes evaluated in this study had been previously washed and waxed as the goal was to replicate conditions typical of repacking and fresh cut operations where Florida toma toes would have been previously washed and waxed. The amount of time between when a tomato has been harvested and when it enters a water system, also microorganisms. Fresh stem sc ars may be congested with water, particularly in the vascular tissue, and can easily absorb water. Old stem scars will absorb less water because they become congested with air that does not allow water to move into the tissues easily (Bartz and Showalter, 1981; Smith et al., 2007). Smith et al. (2007) studied water infiltration into tomatoes with post stem removal times of 2, 4, 6, 8, 14, and 26 h; tomatoes 2 h after harvest absorbed significantly more water and water absorption declined as post stem remov al times increased. This trend is also true for Salmonella internalization into green tomatoes; internalization was significantly higher into tomatoes whose stems had immediately been removed than into tomatoes whose stems were removed 2 or 16 h before su bmersion (Xia et al.,2011). Using washed and waxed tomatoes ensured the stem scars had dried and sealed, effectively eliminating this variable. The average internalized Salmonella populations recovered was generally
105 low (<1.0 log MPN/segment) or below the limit of detection (<0.079 log MPN/segment). Most samples had fewer than 100 MPN/g (2 log MPN/g) recoverable Salmonella cells in Xia et al. (2011) and populations varied greatly depending on tissue location. Average cell populations in tissues directly bel ow the stem scar were 527 MPN/g (2.7 log MPN/g) (Xia et al., 2011). Internalized Salmonella populations varied greatly across temperature differentials, submersion times, and maturity treatment, suggesting significant tomato to tomato variability. This va riability is also reported in previous studies, and may be due to differences in stem scar porosity even being seen amongst tomatoes from one plant (Bartz, 1982). Stem scar congestion with water or air may also differ depending on tomato location in gondol as prior to entering the packinghouse or in packed boxes leaving the packinghouse. Tomatoes at the top of gondolas or boxes may dry out faster than those in the middle or at the bottom, which may still be congested with water and prone to water uptake duri ng immersion. Currently, a 5Â°C/10Â°F temperature differential between internal tomato fruit temperature and water is required for washing systems at all steps in the tomato supply chain, including repacking and fresh cut operations in Florida. The results from this research suggest that for prevention of Salmonella internalization, temperature differential, within 5Â°C/10Â°F of tomato pulp temperature is not a significant factor. Submersion time, especially those exceeding 5 min may contribute to pathogenic bacteria internalization, regardless of temperature In order to conserve energy in repacking and fresh cut operations, water could be maintained at the same temperature as the incoming fruit provided that submersion time does not exceed two minutes.
106 CHAP TER SIX CONCLUSIONS AND FUTURE WORK For all treatments examined, Salmonella internalization was highest in segment A, just below the stem scar suggesting that this is the major entry point for water and bacteria; this is also supported by previous research (Xia et al., 2011). Temperature differential, up to 5Â°C did not play a sig nificant role in Salmonella internalization for all maturities and pulp temperatures. A logical next step for this research would be to include larger temperature differentials between tomatoes and water. Plant pathogens are known to internalize into tomat oes when exposed to large negative temperature differentials to a greater extent than when exposed to small temperature differentials. No studies have evaluated the effect of large (Â± 20Â°C) temperature differentials on internalization of human pathogens in to tomatoes. While internalization rarely increased with increased submersion time up to the 500 s experimental time point, average internalized Salmonella populations were significantly higher following 1800 s submersion in many cases. During experimental time intervals, internalization was highest following immersion for 300 s in most treatments. Previous research has shown that infiltration of water and bacteria into the stem scar involves both a time and temperature component as the fruit must have suff icient time to cool and cause a decrease in pressure of gases within the fruit (Bartz, 1982). Internalization was higher in green tomatoes than in red tomatoes at both pulp temperatures. Previous research suggests that red fruits will imbibe less water th an green fruits (Bartz and Showalter, 1981); however no research has been conducted on Salmonella internalization into red fruits. Further research could be performed to
107 determine if large temperature differentials will effect internalization of human path ogens into red tomatoes despite the fact that the stem scar has had sufficient time to seal. No Salmonella populations (< 0.079 log MPN/segment) were recovered from some tomatoes at time intervals and temperature differentials in all maturities and pulp te mperatures tested and average internalized populations were low (< 2.0 log MPN/segment) in all treatments. This suggests that when the stem scar has had sufficient time to dry and become congested with air, the risk of pathogen internalization is lowered. Pathogen internalization was still noted in most cases at low levels. Tomato variety has been shown to affect both phyto and foodborne pathogen uptake into tomatoes, although the reason is unknown (Batrz et al., 1975; Smith et al., 2007, Xia et al., 201 1). These studies were primarily performed on mature green tomatoes; further research is needed to determine if tomato variety plays a significant role in internalization into washed and waxed tomatoes at ambient and refrigeration temperatures. Further re search is needed to determine if temperature differentials currently used in the tomato industry are necessary to prevent the internalization of human pathogens in the packinghouse. Tomatoes vary greatly in size, stem scar size, and stem scar exposure to a ir. Due to great variability between tomatoes in each replication, additional replications should be performed. Also, this experiment was conducted on a small scale to create a bench top replica of industry practice. A larger version of this experiment, us ing larger quantities of tomatoes could provide more insight into an
108 industry operation. Following further replications, advisement on water temperature used in re packing and fresh cut operations could be made to the industry.
109 LIST OF REFERENCES Allen, R.L., Warren, B.R., Archer, D.L., Sargent, S.A., Schneider, K.R., 2005. Survival of Salmonella spp. on the surfaces of fresh tomatoes and selected packing line materials. Hort. Tech. 15, 831 836. Baloda, S.B., Christensen, L.,Trajcevska, S., 2001. P ersistence of a Salmonella enterica serovar Typhimurium DT12 clone in a piggery and in agricultural soil amended with Salmonella contaminated slurry. Appl. Environ. Microbiol. 67, 2859 2862. Bartz, J.A., 1982. Infiltration of tomatoes immersed at different temperatures to different depths in suspensions of Erwinia Carotovora Subsp Carotovora. Plant Dis. 66, 302 306. Bartz, J.A., 1988. Potential for postharvest disease in tomato fruit infiltrated with chlorinated water. Plant Dis. 72, 9 13. Bartz, J.A., 1991 . Relation between resistance of tomato fruit to infiltration by Erwinia Carotovora and bacterial soft rot. Plant Dis. 75, 153 155. Bartz, J.A., 2006. Internalization and infiltration. In: Sapers, G.M., Gorny, J.R., Yousef, A.E. (Eds.), Microbiology of fru its and vegetables. CRC Taylor and Francis, Boca Raton, FL., pp. 75 95. Bartz, J. A., 2009. Raw tomatoes and Salmonella. In: Sapers, G.M., Soloman, E.B., Matthews, K.R. (Eds.), The produce contamination problem: causes and solutions. Academic Press of Elsevier, Inc. Burlington, MA, pp. 223 243. Bartz, J.A., Crill, J.P., John, C.A., 1975. Inheritance of tolerance of Erwinia Carotovora in Florida MH 1 tomato. Phytopathol. 65, 1146 1150 . Bartz, J.A., Showalter, R.K., 1981. Infiltration of tomatoes by aqueous bacterial suspensions. Phytopathol. 71, 515 518. Behravesh, C.B., Mody, R.K., Jungk, J., Gaul, L., Redd, J.T., Chen, S., Cosgrove, S., Hedican, E., Sweat, D., Chavez Hauser, L., Snow, S.L., Hanson, H., Nguyen, T.A., Sodha, S.V., Boore, A.L., Russo, E., Mikoleit, M., Theobald, L., Gerner Smidt, P., Hoekstra, R.M., Angulo, F.J., Swerdlow, D.L., Tauxe, R.V., Griffin, P.M., Williams, I.T., 2011. 2008 outbreak of Salmonella Saintpaul infect ions associated with raw produce. N. Engl. J. Med. 364, 918 927. Beuchat, L.R., Ryu, J.H., 1997. Produce handling and processing practices. Emerg. Infect. Dis. 3, 459 465. Brackett, R.E., 1999. Incidence, contributing factors, and control of bacterial pa thogens in produce. Post. Biol. Tech. 15, 305 311.
110 Brandt, S., Pek, Z., Barna, E., Lugasi, A., Helyes, L., 2006. Lycopene content and colour of ripening tomatoes as affected by environmental conditions. J. Sci. Food Agr. 86, 568 572. Buchanan, R. L., Ede lson, S.G., Miller, R.L., Sapers, G.M., 1999. Contamination of intact apples after immersion in an aqueous environment containing Escherichia coli O157:H7. J. Food Prot. 62, 444 450. Burnett, S.L., Chen, J., Beuchat, L.R., 2000. Attachment of Escherchia co li O157:H7 to the surfaces and internal structures of apples as detected by confocal scanning laser microscopy. Appl. Environ. Microbiol. 66, 4679 4687. Cabral, J.P.S., 2010. Water microbiology, bacterial pathogens, and water. Int. J. Environ. Res. Pub. H ealth. 7, 3657 3703. Callaway, T.R., Edrington, T.S., Anderson, R.C., Byrd, J.A., Nisbet, D.J., 2008. Gastrointestinal microbial ecology and the safety of our food supply as related to Salmonella . J. Anim. Sci. 8, E163 E172. Centers for Disease Control a nd Prevention (CDC). 2005. Outbreaks of Salmonella infections associated with eating Roma tomatoes United States and Canada, 2004. MMWR. 54, 325 328. Accessed July 2013 from: http://www. cdc.gov/mmwr/preview/mmwrhtml/mm5413a1.htm Center for Disease Control and Prevention (CDC). 2007. Multistate outbreaks of Salmonella infections associated with raw tomatoes eaten in restaurants United States, 2005 2006. MMWR. 56, 909 911. Accessed July 201 3 from: http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5635a3.htm Center for Disease Control and Prevention (CDC). 2008. Annual listing of foodborne disease outbreaks. Accessed October 2012 from: http://wwwn.cdc.gov/foodborneoutbreaks/ Center for Disease Control and Prevention (CDC). 2010. Preliminary FoodNet data on the incidence of infection with pathogens transmitted commonly through food 10 States, 2009. MMWR. 59, 418 422. Accessed July 2013 from: http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5914a2.htm Center for Disease Control and Prevention (CDC). 2011a. Estimates of foodborne illness in the United States. Accessed October 2012 from: http://www.cdc.gov/foodborneburden/201 1 foodborne estimates.html Center for Disease Control and Prevention (CDC). 2011b. Salmonellosis Outbreak investigation, October 2006. Accessed November 2012 from: http://www.cdc.gov/ncidod/dbmd/diseaseinfo/salmonellosis_2006/110306_outbre ak_notice.htm
111 Center for Disease Control and Prevention (CDC). 2013. Surveillance for foodborne disease outbreaks United States, 2009 2010. MMWR. 62, 41 47. Accessed Ju ly 2013 from: http://www.cdc.gov/mmwr/preview/mmwrhtml/mm6203a1.htm?s_cid=mm6203a1 _w Center for Disease Control and Prevention (CDC). 2014. Foodborne disease outbreak, 2 011 case definition. Accessed May 2014 from: http://wwwn.cdc.gov/nndss/script/casedef.aspx?CondYrID=677&DatePub=1/1/20 11%2012:00:00%20AM Cummings, K., Barrett, E., Mohle Boetani,J., Brooks, J., Farrar, J., Hunt, T., Fiore, A., Komatsu, K., Werner, S.B., Slutsker, L., 2001. A multistate outbreak of Salmonella Enterica serotype Balidon associated with domestic raw tomaotes. Emerg. Infec. Dis. 7, 1046 1048. Davies, J.N., Hobson, G.E., McGlasson, W.B., 1981. The constituents of tomato fruit the influence of environment, nutrition, and genotype. CRC Crit. Rev. Food Sci. Nutri. 15, 205 280. Davies, R.H., Wray, C., 1996. Seasonal variations in the isolation of Salmonella Typhimurium , Salmonella Enteritidis , Bacillus cereus and Clostridium perfringens from environmental samples. J. Vet. Med. Ser. 43, 119 127. DiRita, V.J., 2001. Molecular basis of Vibrio cholera pathogenesis. In: Groisman , E.A. (Ed.), Prinicples of bacterial pathogenesis. Academic Press, California, pp. 457 508. Eblen, B.S., Walderhaug, M.O., Edelson Mammel, S., Chirtel, S.J., De Jesus, A., Merker, R.I., Buchanan, R.L., Miller, A.J., 2004. Potential for internalization, gr owth, and survival of Salmonella and Escherichia coli O157:H7 in oranges. J. Food Prot. 67, 1578 1584. Ellermeir, C., Slauch, J., 2006. The genus Salmonella. The prokaryotes: gamma subclass. Springer, New York, pp. 124 158. Erickson, M.C., 2012. Internaliz ation of fresh produce by foodborne pathogens. In: Doyle, M.P., Klaenhammer, T.R. (Eds.), Annual review of food science and technology. pp. 283 210. Fatica, M.K., Schneider, K.R., 2011. Salmonella and produce survival in the plant environment and implicati ons in food safety. Virulence. 2, 573 579. Florida Department of Agriculture and Consumer Services (FADCS). 2008. Tomato best practices manual (TGAPs) Accessed October 2012 from: h ttp://www.freshfromflorida.com/fs/TomatoBestPractices.pdf
112 Foster, J.W., 1991. Salmonella acid shock proteins are required for the adaptive acid tolerance response. J. Bacteriol. 173, 6896 6802. Garcia, R., Baelum, J., Fredslund , L., Santorum, P., Jacobsen, C.S., 2010. Influence of temperature and predation on survival of Salmonella enterica serovar Typhimurium and expression of invA in soil and manure amended soil. Appl. Environ. Microbiol. 76, 5025 5031. Gillaspy, G., Ben David , H., Gruissem, W., 1993. Fruits: A developmental perspective. Plant Cell. 5, 1439 1451. Greene, S.K., Daly, E.R., Talbot, E.A., Demma, L.J., Holzhauer, S., Patel, N.J., Hill, T.A., Walderhaug, M.O., Hoekstra, R.M., Lynch, M.F., Painter, J.A., 2008. Recur rent multistate outbreak of Salmonella Newport associated with tomatoes from contaminated fields. 2005 . Epidemiol. Infect. 136, 157 165. Gu, G., Hu, J., Cevallos Cevallos, J.M., Richardson, S.M., Bartz, J.A., van Bruggen, A.. 2011. Internal colonization o f Salmonella enterica serovar Typhimurium in tomato plants. Plos One. 6, 1 11. Guo, X., Chen, J., Brackett, R.E., Beuchat, L.R., 2001. Survival of Salmonelae on and in tomato plants from the time of inoculation at flowering and early stage of fruit develop ment through fruit ripening. Appl. Environ. Microbiol. 67, 4760 4764. Guo, X., Chen, J., Brackett, R.E., Beuchat, L.R., 2002. Survival of Salmonella on tomatoes stored at high relative humidity, in soil, and on tomatoes in contact with soil. J. Food Prot. 65, 274 279. Gupta, S.K., Nalluswami, K., Snider, C., Perch, M., Balasegaram, M., Burmeister, D., Lockett, J., Sandt, C., Hoekstra, R.M., Montgomery, S., 2007. Outbreak of Salmonella Braenderup infections associated with Roma tomatoes, Northeastern United States, 2004: A useful method for subtyping exposures in field investigations. Epidemiol. Infect. 135, 1165 1173. Hedberg, C.W., Angulo, F.J., White, K.E., Langkop, C.W., Schell, W.L., Stobierski, M.G., Schuchat, A., Besser, J.M., Dietrich, S., Helsel, L. , Griffin, P.M., McFarland,J.W., Osterholm, M.T., 1999. Outbreaks of Salmonellosis associated with eating uncooked tomatoes: implications for public health. Epidemiol. Infect. 122, 385 393. Hobson, G.E., 1987. Low temperature injury and the storage of ripe ning tomatoes. J. Hort. Sci. 62, 55 62. Hong, J.H., Gross, K.C., 2001. Maintaining quality of fresh cut tomato slices through modified atmosphere packaging and low temperature storage. J. Food Sci. 66, 960 965.
113 Hong, J.H., Gross, K.C., 2000. Involvement o f ethylene in development of chilling injury in fresh cut tomato slices during cold storage. J. Amer. Hort. Sci. 125, 736 741. Islam, M., Morgan, J., Doyle, M.P., Phatak, S.C., Millner, P., Jiang, X., 2004. Persistence of Salmonella enterica serovar Typhi murium on lettuce and parsley and in soils on which they were grown in fields treated with contaminated manure composts or irrigation water. Food Path. Dis. 1, 27 35. Jacobsen, A., Hendriksen, R.S., Aaresturp, F.M., Ussery, D.W., Friis, C., 2011. The Salmo nella Enterica pan genome. Microb. Ecol. 62, 487 504. Kader, A.A., 2002. Postharvest technology of horticultural crops, third ed. Regents of the University of California, Oakland. Kader, A.A., L.L. Morris., 1976. Correlating subjective and objective measur ements of maturation and ripeness of tomatoes. In: Proc. 2nd Tomato quality workshop, vegetable crops series 178. University of California, Davis. Klee, H.J., Giovannoni, J.J., 2011. Genetics and control of tomato fruit ripening and quality attributes. Ann u. Rev. Genet. 45, 41 59. Knapp, S., 2002. Tobacco to tomatoes: A phylogenetic perspective on fruit diversity in the Solanaceae . J. Exp. Bot. 53, 2001 2022. Lemiare Chamley, M., Petit, J., Garcia, V., Just, D., Baldet, P., Germain, V., Fagard, M., Mouassite, M., Cheniclet, C., Rothan, C., 2005. Changes in transcriptional profiles are associated with early fruit tissue specialization in tomato. Plant Physiol. 139, 750 769. LeStrange, M., Schrader, W.L., Hartz, T.K., 2000. Fresh market tomato product ion in California. division of agriculture and natural resources. 8017. University of California, Oakland. Accessed July 2013 from: http://anrcatalog.ucdavis.edu/pdf/8017.pdf Lin, C.M., Wei, C.I., 1997. Transfer of Salmonella Montevideo onto the interior of tomatoes by cutting. J. Food Prot. 60, 858 863. Pan, W., Schaffner, D.W., 2010. Modeling the growth of Salmonella in cut red round tomatoes as a function of temperature. J. Food Prot. 73, 1502 1 505. Penteado, A.L., Eblen, B.S., Miller, A.J., 2004. Evidence of Salmonella internalization into fresh mangoes during simulated postharvest insect disinfestation procedures. J. Food Prot. 67, 181 184. Reller, M.E., Tauxe, R., Kalish, L., Molbak, K., 2008. Excess Salmonellosis in women in the United States: 1968 2000. Epidemiol. Infect. 136, 1109 1117.
114 Rushing, J., Angulo, F., Beuchat, L., 1996. Implementation of a HACCP program in a commercial fresh market tomato packinghouse: A model for the industry. Dai r. Food Environ. Sanit. 16, 549 553. Saltveit, M.E., 1991. Determining tomato fruit maturity with nondestructive in vivo nuclear magnetic resonance imaging. Post. Biol. Tech. 1,153 159. Sargent, S.A., Brecht, J.K., Zoellner, J.J., 1992. Sensitivity of tom atoes at mature green and breaker ripeness stages to internal bruising. J. Amer. Soc. Hort. Sci. 117, 119 123. Scallan, E., Hoekstra, R., Angulo, R., Tauxe, R., Widdowson, R., Roy, S., Jones, J., Griffin, P., 2011. Foodborne illness acquired in the United States major pathogens. Emerg. Infect. Dis. 17, 7 15. Schraff, R.L., 2012. Economic burden from health losses due to foodborne illness in the United States. J. Food Prot. 75, 123 131. Segall, R.H., Henry, F.E., Dow, A.T., 1977. Effect of dump tank water t emperature on the incidence of bacterial soft rot of tomatoes. Proc. Fla. State Hort. Soc. 90, 204 205. Showalter, R.K., 1979. Postharvest water intake by tomatoes. (Abstract) Hortsci. 14, 125. Sivapalasingam, S., Friedman, C.R., Cohen, L., Tauxe, R.V., 2 004. Fresh produce: A growing cause of outbreaks of foodborne illness in the United States, 1973 through 1997. J. Food Prot. 67, 2342 2353. Smith, S.M., Scott, J.W., Bartz, J.A., Sargent, S.A., 2007. Effect of time after harvest on stem scar water absorpt ion in tomato. Hortsci. 42, 1227 1230. Steele, M., Odumeru, J., 2004. Irrigation water as source of foodborne pathogens on fruit and vegetables. J. Food Prot. 67, 2839 2849. Stine, S.W., Song, I., Choi, C.Y., Gerba, C.P., 2005. Effect of relative humidit y on preharvest survival of bacterial and viral pathogens on the surface of cantaloupe, lettuce, and bell peppers. J. Food Prot. 68, 1352 1358. Tauxe, R., 1991. Salmonella : A postmodern pathogen. J. Food Prot. 54, 563 568. Tindall, B.J., Grimont, P.A.D., Garrity, G.M., Euzeby, J.P., 2005. Nomenclature and taxonomy of the genus Salmonella . Int. J. Sys. Evol. Microbiol. 55, 521 524. United Fresh Produce Association. 2009. Re: FDA 2009 D 0346; guidance to minimize microbial food safety hazards of tomatoes; draft guidance. Accessed June 2014 from:
115 http://www.unitedfresh.org/assets/files/UFPA%20comments%20on%20FDA%20 Tomato%20Guidance.pdf US Department of Agriculture (USDA). 2013. Most Probable Number procedure and t ables. Accessed August 2013 from: http://www.fsis.usda.gov/wps/wc m/connect/8872ec11 d6a3 4fcf 86df 4d87e57780f5/MLG_Appendix_2_03.pdf?MOD=AJPERES US Department of Agriculture: Agricultural Marketing Service ( USDA AMS). 1991. United States standards for grades for fresh t omatoes. Accessed July 2013 from: http://www.ams.usda.gov/AMSv1.0/getfile?dDocName=STELPRDC5050331 US Department of Agriculture: Agricultural Marketing Service (USDA AMS). 2009. Microbiological data program: progress update and 2009 data summary. Accessed June 2014 from: ht tp://www.ams.usda.gov/AMSv1.0/getfile?dDocName=STELPRDC5088761 US Department of Agriculture: Economic Research Services (U SDA ERS). 2008. Vegetables and melons situation and outlook y earbook. Accessed May 2014 from: http://usda.mannlib.cornell.edu/usda/ers/VGS yearbook//2000s/2008/VGS yearbook 05 15 2008.pdf US Department of Agriculture: Economic Research Services (U SDA ERS). 2012. Vegetables and pu lses: t omatoes. Accessed December 2013 from: http://www.ers.usda.gov/topics/crops/vegetables pulses/tomatoes.aspx US Food and Drug Adminis tration (FDA). 2006. Commodity specific food safety guidelines for the fresh tomato supply c hain. Edition 1.0. Silver Spring, Maryland. Accessed January 2014 from: http://www.fda.gov/downloads/Food/Guidan ceRegulation/UCM168629.pdf US Food and Drug Admini stration (FDA). 2007a. Program information manual: retail food protection storage and handling of t omatoes. Silver Spring, Maryland. Accessed June 2014 from: http://www.fda.gov/Food/GuidanceRegulation/RetailFoodProtection/IndustryandR egulatoryAssistanceandTrainingResources/ucm113843.htm#attachc US Fo od and Drug Admin istration (FDA). 2007b. Tomato safety i nitiative. Silver Spring, Maryland. Accessed November 2012 from: http://www.fda.gov/Food/FoodSafety/Product SpecificInformation/FruitsVegetablesJuices/FDAProduceSafetyActivities/ucm115 334.htm US Food and Drug Adminis tration (FDA). 2008. Commodity specific food safety guidelines for the fresh tomato supply c hain. 2 nd Edition. Silver Spring, Maryland. Accessed May 2014 from: http://www.fda.gov/downloads/Food/GuidanceRegulation/UCM171708.pdf
116 US Food and Drug Administra tion (FDA). 2009 . Guidance for industry: guide to minimize microbial food safety hazards of t omatoes; Draft Guidance. 2011. Silver Spring, Maryland. Accessed December 2012 from: http://www.fda.gov/Food/GuidanceComplianceRegulatoryInformation/GuidanceD ocuments/ProduceandPlanProducts/ucm173902.htm US Food and Drug Administra tion (FDA) 2012. Bad bug book, foodborne pathogenic microorganisms an d natural t oxins. Second Edition. pp 9 14. Accessed July 2013 from: http://www.fda.gov/downloads/Food/FoodborneIllnessContaminants/UCM297627 .pdf Valadez, A.M., Sch neider, K.R., Dan yluk, M.D., 2012. Outbreaks of foodborne disease associated with t omatoes. FSHN 12 08, Food science and human nutrition department, Florida cooperative extension service, Institute of food and agricultural sciences, University of Florida. Accessed June 2014 from: http://edis.ifas.ufl.edu/fs192 Vigneault, C., Bartz, J.A., Sargent, S.A., 2000. Postharvest decay risk associated with hydrocooling tomatoes. Plant Dis. 84, 1314 1318. Xia, X.D., Luo, Y.G., Yang, Y., Vinyard, B., Schneider, K., Meng, J.H., 2011. Effects of tomato variety, temperature differential, and post stem removal time on internalization of Salmonella Enterica serovar Thompson in tomatoes. J. Food Prot. 75, 297 303. Yuk, H.G., Bartz, J.A., Schneider, K.R., 2006. The effectiveness of sanitizer treatments in inactivation of Salmonella spp. from bell pepper, cucumber, and strawberry. J. Food Sci. 71, 3. Zheng, J., Allard, S., Reynolds, S., Millner, P., Arce, G., Blodgett, R.J., Br own, E.W., 2013. Colonization and internalization of Salmonella enterica in tomato plants. Appl. Environ. Microbiol. 79, 2494 2502. Zhuang, R.Y., Beuchat, L.R., Angulo, F.J., 1995. Fate of Salmonella Montevideo on and in raw tomatoes as affected by tempera ture and treatment with chlorine. Appl. Environ. Microbiol. 61, 2127 2131.
117 BIOGRAPHICAL SKETCH Ashley Nicole Turner was born in Decatur, Georgia to Terry and Cheryl Turner . Ashley received her Bachelor of Science and Agriculture in food science and technology from the University of Georgia in May 2012. She began her Master of Science degree in August 2012. Her major advisor was Dr. Michelle Danyluk and she studied food science and human nutrition with an emphasis in food microbiology. In the futu re Ashley plans to pursue a career in the industry focusing on food safety.