Phenotypic and molecular diversity in a collection of ‘Pomodoro di Sorrento’ Italian tomato landrace

(1)

Contents lists available atScienceDirect

Scientia

Horticulturae

j o u r n a l h o m e p a g e :w w w . e l s e v i e r . c o m / l o c a t e / s c i h o r t i

Phenotypic

and

molecular

diversity

in

a

collection

of

‘Pomodoro

di

Sorrento’

Italian

tomato

landrace

Mario

Parisi

a,∗

_,

_Riccardo

_Aversano

b

_,

_Giulia

_Graziani

b,1

_,

_Valentino

_Ruggieri

b

_,

Veronica

Senape

c

_,

_Loredana

_Sigillo

c

_,

_Amalia

_Barone

b

a_Consiglio_per_la_ricerca_in_agricoltura_e_l’analisi_{dell’economia}_agraria,_Centro_di_ricerca_per_{l’Orticoltura,}_Via_{Cavalleggeri,}_25-84098,_Pontecagnano_Italy b_Department_of_Agricultural_Sciences,_University_of_Naples_‘Federico_II’,_Via_Università,_100-80055,_Portici,_Italy

c_Consiglio_per_la_ricerca_in_agricoltura_e_l’analisi_{dell’economia}_agraria,_Centro_di_{sperimentazione}_e_{certiﬁcazione}_delle_sementi,_S.S.₁₈_km_77.700,_84091, Battipaglia,Italy

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received23October2015

Receivedinrevisedform17February2016 Accepted29February2016 Keywords: SolanumlycopersicumL. Fruitmorphology SNPgenotyping Verticilliumwilt

a

b

s

t

r

a

c

t

The‘PomodorodiSorrento’,originatingfromSouthernItaly,representsoneofthemostimportantand remunerativeItaliantomatolandracesfor“freshmarket”.Despiteitsoutstandingorganolepticand nutri-tionalqualitiesscantyinformationstillexistsonthestructuralvariationofthistraditionaltomato.Here wereporttheevaluationoftenaccessionsofthe‘PomodorodiSorrento’tomatoatphenotypicand molec-ularlevelsinordertoprovideabasistounequivocallydistinguishthemfromthosebelongingtotheother mostfamoussimilarlandracesgrowninItaly.Morphologicalcharacterizationbasedon13descriptors highlightedthepresenceofasignificantlevelofvariationforpuffiness,greencolorandgreenshoulder ofunripefruits,allowingdistinguishingtwo‘Sorrento’morphotypes.Genome-wideSNPgenotypingwas carriedoutusingtheILLUMINAInfiniumtomatoarrayandrevealedageneticallywelldefinedstructure ofthe‘Sorrento’landracewithrespecttotheothermultilocularlandracesmostwidespreadinItaly.In particular,1450SNPsresultedpolymorphicamong20testedgenotypes,andoneclearlydistinguishesthe ‘Sorrento’fromothersimilarlandraces.Finally,resistanceassaytoVerticilliumwiltallowedthe identifi-cationofpromising‘Sorrento’accessions.InterestinglytwomarkerslocalizedintotheVe2geneshowed astrongassociationwiththeresistance/susceptibilitytrait.Takenaswhole,ourresultsprovidea descrip-tionof‘PomodorodiSorrento’landracediversity,whichprovideusefulinformationforitsutilizationin breedingprogramsaswellasforitsdirectuseinqualitymarkets.

1. Introduction

Thecultivatedtomato(SolanumlycopersicumL.)wasintroduced inItalyintheearly16thcentury,foundinginthiscountryan impor-tantsecondarycenterfordiversiﬁcation,whichresultedinawide arrayofvariationsespeciallyforthetraitsdeterminingfruitshapes (e.g.,ﬂat,round,long,heart)andqualitytraits(Mazzucatoetal., 2010).Suchphenotypicvariationhasgivenrisetoarangeof lan-draces,manyofthemarestillcommonlycultivatedandarefound at local market for freshconsumption (“salad tomatoes”). Sev-eraltraditionaltomatoesarecurrentlycultivatedinItaly,suchas ‘PeraLigureodiAlbenga’,‘Perad’Abruzzo’,‘CanestrinodiLucca’, ‘PomodorodiBelmonte’,‘Scatolone diBolsena’,‘Costolutorosso

∗ Correspondingauthor.

E-mailaddress:[email protected](M.Parisi).

1 _Present_address:_Department_of_Pharmacy,_University_of_Naples_‘Federico_II’,_Via DomenicoMontesano,49-80131Napoli,Italy.

di Rotonda’, ‘Costoluto ﬁorentino’, ‘Pantano romanesco’, ‘Cuore di Bue’,‘Pomodoro diSorrento’, ‘Pomodoropiatto diCambiano’ (Mazzucatoetal.,1998,2008,2010;Acciarrietal.,2007;Parisietal., 2008).

The‘PomodorodiSorrento’tomato(alsocalled‘Sorrentino’or ‘RosadiSorrento’)comesfromtheSouthernItaly.Itscultivation beganinearly1900snearSorrento,wheresomelocalship-owners introduced theseed brought fromSouth America (Parisiet al., 2008).Sincethen,thislandracespreadthroughoutthesurrounding region,especiallyasaprotectedcultivation.Similarto‘CuorediBue’ and‘Belmonte’landraces,the‘Pomodoro diSorrento’ripefruits arepink,duetothecolorlessepidermids(y)mutation(Mazzucato etal.,2010).Furthermore,‘PomodorodiSorrento’fruitsarefairly homogeneousintheirsizeandusuallygroupedinclustersof3–5 pertruss;incross sectiontheyshowa differentgradeof pufﬁ-nessandveryﬂeshyloculeswitha lownumberofseeds(Parisi etal.,2008).Moreover,the‘PomodorodiSorrento’landraceshows outstandingorganolepticandnutritionalqualities(Sinesioetal., 2007;Ercolanoetal.,2008;Lisantietal.,2008)thataregreatly http://dx.doi.org/10.1016/j.scienta.2016.02.038

(2)

appreciatedbyconsumers,whooftenarewillingtopayapremium thancommercialmodernvarieties.Thepromotionoftraditional tomatoesthatcombinesgoodorganolepticqualitiesandfunctional compoundswouldreinforcetheirqualitynichesandconsolidating theirpremiumprices(Cortés-Olmosetal.,2014).

Toclearlydistinguishthis“high-valuable”‘Sorrento’fromother similartomatoes,aprotectionsystembasedontheuseofmolecular markers would be beneficial. The current availability of cost-effectiveandfastgenotypingassayshasmadethegeneticanalysis morefeasibleandinformative.Inparticular,thesinglenucleotide polymorphisms(SNPs)arethemarkersofchoiceforgenome-wide geneticanalyses.Fortomato,ahigh-densitySNParraywasrecently built(Simetal.,2012)intheframework oftheTheSolanaceae Coordinated AgriculturalProject (SolCAP). Thearray consistsof 7720SNPs,identifiedbythere-sequencingofsixtomato acces-sions(Hamiltonetal.,2012).Itwasrecentlyusedforgenome-wide associationanalyses(Ruggierietal.,2014;Sauvageetal.,2014),for phylogeneticsurveyofdifferentS.lycopersicumpopulations(Blanca etal.,2012;Simetal.,2012)andforfunctionalstudies(Hirakawa etal.,2013).Giventhat,theSolCAPrepresentsanefficienttoolfor exploringnaturalgeneticdiversitywithinthe‘Sorrento’landrace.

Inthepresentworkweaimedtoprovidetheﬁrstwide descrip-tion of ‘Sorrento’ landrace diversity. An exhaustive survey at morphologicalandmolecularlevelwascarriedouttoprovidea structureofvariationof‘Sorrento’accessionsusefulto unequivo-callydistinguishthemfromtomatoesbelongingtotheothermost famoussimilarlandracesgrowninItaly.Inaddition,the‘Sorrento’ accessionswerealsocharacterizedforresistancetoonesoil-borne pathogencausingextensivedamagetotomatocultivation.Thedata obtainedwouldbeofgreatinterestforthevalorizationofthis lan-drace,andforitsutilizationinbreedingprogrammersordirectuse inmarketsofhighqualityproducts.

2. Materialandmethods 2.1. Plantmaterials

Ten tomato accessions representative of the ‘Sorrento’ lan-drace(PS01,PS04,PS05,PS06,PS07,PS08,PS09,PS10,PS11and PS13)were compared toten Italian landracesshowing similar

fruitmorphology.Thesegenotypeswere‘Belmonte’,‘Canestrinodi Lucca’,‘CostolutoFiorentino’,‘CostolutoRossodiRotonda’,‘Cuore diBue’,‘Pantanoromanesco’,‘Parmitanella’,‘Perad’Abruzzo’,‘Pera d’Albenga’,‘ScatolonediBolsena’andhereaftercodedas:BELM, CANL,FIOR,ROTN,CBUE,PANT,PARM,PABR,PALB,BOLS, respec-tively(Table1).

Allgeneticmaterialswereprovidedbylocalfarmersobyseed companiesandhadundergonetwocyclesofself-pollinationbefore beingused.Atrialwassetupundergreenhousewiththe above-describedseedstocksattheexperimentalfarmof‘CREA-Centro diricercaperl’Orticoltura’,Battipaglia(40◦3700N;14◦5800E; 65ma.s.l.;SouthItaly).Theaccessionsweregrownatadensityof 3.7plantsm−2,accordingtoarandomizedblockdesignwiththree replicatesand10plantsperelementaryexperimentalunit.

Beforetransplanting50,100and100kgha−1 ofN(as ammo-niumsulfate),P(asmineralperphosphate),andK(aspotassium sulfate)respectively,weredistributed.Startingfromfull-blossom, 75kgha−1 ofNwerefurtherappliedbyﬁvefertigation applica-tions.CropprotectionagainstattacksofTutaabsoluta(Meyrick), Trialeurodes vaporariorum (Westwood), Frankliniella occidentalis (Pergande) and Tetranychus urticae (Koch) were ensured by applications of commercial pesticides containing spinosad and spiromesifen as active ingredients. Moreover to control some phytopatogenicfungi,such asAlternaria alternata [(Fr.) Keissl.], Phytophthorainfestans [(Mont.) de Bary], Botrytiscinerea (Pers.) and Leveillula taurica [(Lév.) G. Arnaud], treatments based on boscalid+pyraclostrobin,mandipropamid,copperoxychlorideand wettablesulfurwereadopted.Thecultivationtechniquesincluded stakes as supportand galvanized wires and weed control was obtainedusingblackplasticmulchﬁlm.

2.2. Morphologicalanalysis

Toassesspeculiarfeaturesof‘Sorrento’tomatoinrespectto sim-ilarItalianlandracesandtoestimatemorphologicalvariabilityat intra-landracelevel,13traits(oneforinﬂorescenceand12forfruit) werescoredormeasured,onsingleplantbasis.Inﬂorescencetype (IT)wasevaluatedonthe2ndtruss(1,simple;2,double;3, com-pound).Meanfruitweight(FW)(grams)andexternal/internalfruit featureswereassessedon10 representativefruits.Theexternal

Table1

Listof‘Sorrento’tomatoaccessionsandsimilarlandracesassessedforphenotypicandmoleculartraits.

AccessionCode Landrace Origin

Source Site

PS01 ’Sorrento’ CREA-ORTa _Pontecagnano_(SA)

PS04 ’Sorrento’ CREA-ORTa _Piano_di_Sorrento_(NA)

PS05 ’Sorrento’ LaSemiortoc _Piano_di_Sorrento_(NA)

PS07 ’Sorrento’ CREA-ORTa _Massa_Lubrense_(NA)

PS08 ’Sorrento’ CREA-ORTa _Massa_Lubrense_(NA)

PS09 ’Sorrento’ CREA-ORTa _Pagani_(SA)

PS10 ’Sorrento’ CREA-ORTa _Vico_Equense_(NA)

BELM ’Belmonte’ CREA-ORTa _Amantea_(CS)

CANL ’CanestrinodiLucca’ CREA-ORTa _Lucca_(LU)

FIOR ’CostolutoFiorentino’ CREA-ORAb _Firenze_(FI)

ROTN ’CostolutorossodiRotonda’ CREA-ORTb _Rotonda_(PZ)

CBUE ’CuorediBue’ Sativac _Cesena_(FC)

PANT ’Pantanoromanesco’ CREA-ORTa _Albano_Laziale_(RM)

PARM ’Parmitanella’ CREA-ORTa _Pontecagnano_(SA)

PABR ’Perad’Abruzzo’ Semencoopc _Cesena_(FC)

PALB ’PeradiAlbenga’ Semencoopc _Cesena_(FC)

BOLS ’ScatolonediBolsena’ CREA-ORAb _Bolsena_(VT)

a_CREA-ORT_[(Consiglio_per_la_ricerca_in_agricoltura_e_l’analisi_{dell’economica}_agraria,_Centro_di_ricerca_per_{l’Orticoltura}_{(Pontecagnano}₋_Italy)]. b _CREA-ORA_[(Consiglio_per_la_ricerca_in_agricoltura_e_l’analisi_{dell’economica}_agraria,_Unità_di_ricerca_per_{l’Orticoltura}_(Monsampolo_del_Tronto₋_Italy)]. c _Seed_Company.

(3)

traits analyzed were: green shoulder (GS) (1, absent; 4, very strong);intensityofgreencolorexcludingshoulder(GCF)(1,very light;5,dark);fruitshapeinlongitudinalsection(FSL)(1,flattened; 2,oblate;3,circular;4,cordate;5,obcordate);fruitshapecross sec-tion(FSC)(1,circular;2,angular;3,irregular);ribbingatpeduncle end(RIB)(3,weak;5,medium;7,strong;9,verystrong);stem-end shape(SES)(3,weak;5,medium;7,strong);blossom-endshape (BES)(1,indented;2,indentedtoflat;3,flat;4,elongated);fruit color(FC)(3,orange;4,pink;5,red).Internalfruittraitsdetected were:puffiness(PUF)(1,nopuffy;4,verypuffy);fruitfleshcolor (FLC)(3,orange;4,pink;5,red);pericarpthickness(PT)(mm).FW, FSL,FSC,RIB,SES,BES,FC,FLC,PUFandPTwereassessedonfull-ripe fruits,whileGSandGCFonunripefruits.

2.3. Molecularanalysis

Genome-wideSNPgenotypingwasperformedusingthe ILLU-MINAInfinium arraybuilt in theframework ofthe Solanaceae Coordinated Agricultural Project(SolCAP) fromNIFA/USDA. The SolCAPtomatopanelincludes7720markersconstructedon EST-SNPs(eSNPs)derivingfromsixtomato genomesequences(Sim etal.,2012).TheprobesequencesandSNPinformationareavailable atTheSolanaceaeCoordinatedAgriculturalProjectwebsite(http:// solcap.msu.edu)andthemapoftheSNPsonthecurrenttomato genomerelease(SL2.50)isavailableontheSGNdatabase(http:// solgenomics.net/).Foreachaccession,genomicDNAwasextracted fromleaftissueusingtheDNeasyPlantMinikit(QIAGEN,Valencia, CA)accordingtothemanufacturer’srecommendationsandthen evaluatedforqualityandconcentrationbytheNanodrop instru-ment(ThermoFisherScientific,Wilmington,USA).Genotypingwas conducted at the Genomix4Life S.r.l (http://www.genomix4life. com)andthedataobtainedwereanalyzedusingtheGenotyping ModuleoftheGenomeStudiosoftwareversion1.7.4(IlluminaInc., SanDiego,CA,USA).Inordertoperformmolecularanalyses,the setofSNPswasfilteredconsideringathresholdofMAF>10%by usingTASSELprogram(Bradburyetal.,2007).Aneighbor-joining treewasgeneratedusingPASTprogram(Hammeretal.,2001)with abootstrapof1000.ABayesianpopulationclassificationwasalso carriedoutusingSTRUCTURE2.3.3(Pritchardetal.,2000). STRUC-TURErunswerecarriedoutwithalengthofburn-inandMCMC (MarkovchainMonteCarlo)of500,000each.Twelveindependent runswereconductedallowingK(numberofpopulations)varying from2to14.OptimalKwasinferredbyusingStructureHarvester program(EarlandvonHoldt,2012)basedontheEvannomethod (Evannoetal.,2005).

2.4. Resistanceassay

Resistance to Verticillium dahliae (Cooke) Wint. race 0 was evaluated by performing in vivo tests (CPVO protocol TP/44/3) (European Union - Community Plant Variety Ofﬁce, 2007) on 20 plants for each accession. Roots of plantlets withexpanded cotyledonswerecutandsoakedinaconidiasuspensionata concen-trationof1×106_conidia/ml._Typical_symptoms_of_Verticillium_wilt

wererecordedduringfourweeksafterinoculation.Resultswere expressedasnumberofdiseasedplantsonthenumberof inocu-latedplants.Asamplewasconsideredsusceptiblewhenitshowed morethanthe20%ofdiseasedplants.Resistanttestswerecarried outinclimaticchamber,attemperatureof23±1◦C,12hlight,by usingthepathogenicstrain(codeVert5)ofV.dahliaerace0from ‘CREA-Centrodi sperimentazioneecertiﬁcazione dellesementi’ collection.Thevariety‘Gianna’wasusedasreferencesusceptible control.

2.5. Statisticalanalysis

Totestforcorrelationsbetweengeneticandmorphologicaldata, aManteltestwasperformed(Mantel,1967)usingthePearson’s r-value.Hierarchicalclusteringanalysis(HCA)wasperformedusing PairwiseManhattanDistancetosystematicallyanalyzeand visual-izethefullsetofmorphologicaldata.XLStatv.2013(Addinsoft,New York,NY)wasusedforstatisticalanalyses,exceptforcorrelation analysis,whichwascarriedoutusingRversion3.2.1(2015-06-18).

3. Results

3.1. Morphologicalassessment

Thefruitmorphologicaltraitsaresummarizedin Supplemen-taryTable1.Withinthe‘Sorrento’collectionﬁveaccessions(PS01, PS04,PS08,PS11 andPS13)had fruitsweightingfrom179.9to 241.7g,fouraccessions(PS06,PS07,PS09andPS10)from241.8to 303.8g,PS05hadthebiggestberries(333.1g).FWofPARM,FIOR, CBUElandraceswassmallerthan thoseof ‘Sorrento’ accessions (Fig.1A).RegardingthePT,theten‘Sorrento’accessionsstudied showedvaluesrangingfrom5.71mmto7.62mm.CBUEandPARM differedfromtheoverall‘Sorrento’accessionsforthepresenceofa thinpericarp,onthecontraryCANLandPALBtomatohadthicker pericarp(Fig.1B).

Within‘Sorrento’collection,GSwasstrong(PS01,PS08,PS09, PS11,PS13) orvery strong(PS04, PS05, PS06,PS07,PS10). The CANL,FIOR,PANT,PARM,BOLSlandracesdifferedfromthe ‘Sor-rento’accessionsfortheirweak-coloredgreenshoulder,whilethis traitwasabsentforPALBlandrace(Fig.1C).AsforFSL,all‘Sorrento’ accessionsshowedroundedfruits,whichdiscriminatedthemfrom thetensimilarlandracesincludedinourwork(Fig.1D).Unlike ‘Sorrento’genotypes,BOLS,PANT,CANL,PARM,PALB,PABRand ROTNfruitsdisplayedastrongorverystrongdegreeofribbingat peduncleend(Fig.1E).Finally,pufﬁness(PUF)of‘Sorrento’ acces-sionsdifferedfromFIOR,ROTN,CBUE,PANT,PARM(nopuffy)and fromBOLS(verypuffy)(Fig.1F).SESshowsmoderatevariability amongthe20genotypesstudiedandmaybeusefultodistinguish allthe‘Sorrento’accessionsfromsevensimilarlandraces(CBUE, BOLS,FIOR,PABR,PANT,PARMandROTN)(Fig.1G).FinallyFCwas pinkforBELM,CBUEandforall‘Sorrento’tomatoes,whilethefruits ofotherlandraceswereredororange(BOLS).RegardingIT,GCF,FLC andBES,alowvariabilitywasobservedforthesetraits.Therefore theyfailedtodescribethediversitywithinthe‘Sorrento’collection andtowelldistinguishthemfromsimilarlandraces.

Groupingofgenotypesbasedonthe13qualitative morpholog-icaltraitsallowedtodividethemintotwomainclusters(M1and M2)(Fig.2,SupplementaryTable1).

Thefirstgroup(M1)comprisedmostlytheSorrento-like Ital-ianlandraces,exceptforBELMandCBUE,whichclusteredinthe M2alongwithalltrue‘Sorrento’accessions.Overall,thelandraces belongingtotheM1groupshowedflattened,oblateorobcordate fruitshape inlongitudinalsection,irregularFSCand orange/red externalfruitcolor.Twolandraces(PALBandCANL)withobcordate fruitsexhibitinglowpuffinessandhighestthicknessofpericarp slightlydivergedfromtheotherM1genotypes.Ontheotherhand, thesecondmajorgroupincludedlandraceswithcircularorcordate (heart-shaped)fruitsshowingcircularorangular shapeincross section,externallypink-coloredaswellasthefruitflesh.Within theM2group,anintra-landracevariabilitywasobservedforthe ‘Sorrento’tomatoeswhichallowedtofurtherdividethemintotwo sub-clusters:thefirstgroupingaccessionscharacterizedbyvery stronggreenshoulder(PS04,PS05,PS06,PS07andPS10),highest valuesofGCF,weakdegreeofpuffinessandribbingatpeduncleend, whiletheothersub-groupclusteredaccessionswithpuffyfruits

(4)

BELM CANL FIOR ROTN CBUE PANT PARM PABR PALB BOLS Sorrento 0 2 4 6 8 10 12 14

1

2

3

4 Green shoulder

0 2 4 6 8 10 12 14

1

2

3

4

5 Fruit shape in longitudinal section

0 2 4 6 8 10 12 14

3

5

7

9 Ribbing at peduncle end

0 2 4 6 8 10 12 14

1

2

3

4 Puffiness

0 2 4 6 8 10 12 14

3

5

7 Stem-end shape

0 2 4 6 8 10 12 14

3

4

5 Fruit color

0 2 4 6 8 10 12 14 very small

small medium large very

large

Fruit weight

0 2 4 6 8 10 12 14

very thin thin medium thick very thick

Per

icarp thickn

ess

very thin

Fig.1.Morphologicalassessment.(Forinterpretationofthereferencestocolourinthisﬁgurelegend,thereaderisreferredtothewebversionofthisarticle.)

Numberofaccessionsofthe‘Sorrento’tomato(n=10,orangebar)andothersimilarlandraces(n=barofeachofthem)fallingintodifferentclassesforeight morpho-logicaldescriptors.X-axesreportclasslimitsasdimension.Ordinateaxesreportabsolutefrequencies.Fruitweight:verysmall(55.5–117.5g),small(117.6–179.6g), medium(179.7–241.7g),large(241.8–303.8g),verylarge(303.9–365.9g);Pericarpthickness:verythin(4.75–5.70mm),thin(5.71–6.66mm),medium(6.67–7.62mm), thick(7.63–8.58mm),verythick(8.59–9.54mm);Greenshoulder:from1(absent)to4(verystrong);Fruitshapeinlongitudinalsection;1(ﬂattened),2(oblate),3(circular), 4(cordate),5,(obcordate);Ribbingatpeduncleend:from3(absent)to9(verystrong);Pufﬁness:from1(nopuffy)to4(verypuffy);Stem-endshape:3(weak),5(medium), 7(strong);Fruitcolor:3(orange),4(pink),5(red).

showingmediumvaluesofGSandGCF(PS01,PS08,PS09,PS11and PS13)(Fig.3).

TheSpearmanrankcorrelationcoefficientscalculatedbetween pairsofvariables revealedhow somewereratherindependent, whereasagroupoftraitsclusteredtogetherbecauseofareciprocal tightcorrelation(Fig.4).Forexamplethehighpositivecorrelation foundbetweenRIB-SES,RIB-FC,FSC-FC,FSC-RIB,FLC-FSC,FLC-RIB, whereasthenegativecorrelationbetweenPUF-FC,PUF-SES and FSL-SESreflectedthetendencyforflattened/oblatefruitstohave

astrongribbinganddepressionatpeduncleend,irregularshape incrosssection,externalandinternalredcolorandlowestvalues ofpuffiness.Moreoverthefruitsshowingalittlegreenshoulder seemstohavehighdegreeofribbing,irregularshapeincross sec-tionand red-coloredflesh(negativecorrelationGS-RIB, GS-FSC, GS-FLC).Thesefeaturesclearlyidentifiedthefollowinglandraces: ROTN,PARM, PANT, PABR, FIORand BOLS. However, thelatter landracestandsout alittlefromthepreviousgroupdue tothe remarkablepuffinessandtheorangecoloroftheripefruitsasabove

(5)

CBUE BELM PS08 PS11 PS09 PS01 PS13 PS06 PS05 PS04 PS07 PS10 CANL PALB BOLS ROTN PARM PABR FIOR PANT 0 5 10 15 20

Dissimilarity

M1

M2

Fig.2. Clusteranalysis.

Thedendrogramwasbuiltusing13morphologicaltraitsevaluatedonten‘Sorrento’ accessionsandtensimilarlandraces.

mentioned.Mostofgenotypesshowingcompoundinﬂorescences producedfruitswithindentedtoﬂatblossom-endshape(negative correlationbetweenIT-BES).

3.2. Genotyping

Outof7720scorableSNPsdeﬁnedbytheInﬁniumassay,7639 (98.7%) were successfully genotyped and 1450 (19%) resulted polymorphicamongthe20testedlineswithaMAF>10% (Supple-mentaryTable2).Onaverage,around120SNPsforchromosome weresurveyed.Thehighestnumberwasscoredforchromosome 11(n=258)andthelowestforchromosome7(n=62).

AccordingtoLDdecayvaluesreportedinRuggierietal.(2014), weselectedasubsetof364potentiallyunlinkedSNPsforinferring populationstructure.ThemodelusedindicatedK=2asthebest numberof sub-populations,providingsupportfortheexistence oftwodistinctclustersinourpopulation.STRUCTUREresultsand DeltaKplotaregraphedinFig.5andSupplementaryFig.1, respec-tively.Thegeneticdistancesbetweenallthelineswerecalculated basedonthegenotypingdataoftheinformativeSNPlociwitha MAF>10%,andadendrogramwasthenconstructed(Fig.5).

The two main clusters in the dendrogram corroboratedthe structuresubdivisions.Inparticulartheﬁrstclusterincludednine out10‘Sorrento’accessions,besidesthreesimilarlandraces(CANL,

Fig.4.Correlogramcalculatedusing13morphologicalvariables.(Forinterpretation ofthereferencestocolourinthisﬁgurelegend,thereaderisreferredtotheweb versionofthisarticle.)

Positivecorrelationsaredisplayedinblueandnegativecorrelationsinredcolor. Colorintensityandthesizeofthecircleareproportionaltothecorrelation coeffi-cients.Intherightsideofthecorrelogram,thelegendcolorshowsthecorrelation coefficientsandthecorrespondingcolors.IT(Inflorescencetype),GS(Green shoul-der),GCF(Intensityofgreencolorexcludingshoulder),FLS(Shapeinlongitudinal section),FSC(Shapeincrosssection),RIB(Ribbingatpeduncleend),SES(Stem-end shape),BES(Blossom-endshape),FC(Fruitcolor),PUF(Puffiness),FLC(Fruitflesh color),FW(Fruitweight),PT(Thicknessofpericarp).

PALBandPABR).Bycontrast,thesecondclusterincluded8 geno-types, and only one (PS09) of them belonged tothe ‘Sorrento’ group.Notably,PS10andPS11,aswellasgenotypesPS04andPS06, resultedhighlysimilaratmolecularlevel(99.5%and99.4%, respec-tively).The‘Sorrento’accessionsshowedprevalentlyoneancestor (redbar),eventhoughinthreecasesareducedinfluenceofthe secondancestor(bluebar)wasrevealed,ora 50%admixture,as incaseofPS09.AmongtheotherItalianlandraces,fourofthem (BOLS,FIOR,PART,PARM) showedonlyone ancestor,while the remainingsixexhibitedanadmixtureoftwoancestors. Interest-ingly,theclassificationofthetomatolandracesobtainedwithSNP datawascomplementarytothatonedrawnusingmorphological data,asrevealedbythehighlysignificantcorrelationbetweenthe two independentcluster analyses(r=0.611,P=0.0001,datanot shown).

Fig.3. Representativeof‘Sorrento’morphotypescomparedtootherpink-coloredgenotypes.(Forinterpretationofthereferencestocolourinthisﬁgurelegend,thereader isreferredtothewebversionofthisarticle.)

Puffy/round-shapedfruitswithmediumshoulderandlightgreencoloratunripestage(PS01);round-shapedfruitsshowingweakpufﬁnessanddarkshoulder,mediumgreen color(PS04).‘CuorediBue’(CBUE)and‘Belmonte’(BELM)fruitsarereportedforcomparison.Foreachaccession,thecorrespondingcodeisreportedatbottom.Baris1cm.

(6)

Fig.5.Molecularscreeningof20genotypesby1450polymorphicSNPs.

Neighbourjoininganalysiswasusedtobuildthetree.Numbersatthenodesarebootstrapvaluesfor1000re-samplings.Structureforeachgenotypewasalsoshowedon therightside.

Inordertoenabletheclearidentiﬁcationof‘Sorrento’tomato,a surveywascarriedoutontheentireSNPcollectiontoselect mark-ersthatcoulddistinguishthislandraceunequivocally. Aunique SNP(solcapsnpsl4963)wasfoundtobespeciﬁcforall‘Sorrento’ accessions,exceptPS09.Inparticular,thehomozygouslocusGG (SupplementaryTable2)wasdetectedinnineaccessions represen-tativeof‘Sorrento’landrace,whiletheothergenotypesshowedthe homozygouslocusAA.Thesolcapsnpsl4963fallsinthe3UTRof thegeneSolyc01g068200.2annotatedas“Abscisicacidinsensitive 8homologue”mappingonchromosome1.

3.3. ResistanceassaytoVerticilliumdahliae

The resistanceassay showed that four ‘Sorrento’ accessions (PS04, PS05, PS06 and PS07) were resistant to V. dahliae race 0,whiletheothersix onesweresusceptible(PS01, PS08,PS09, PS10,PS11andPS13)(SupplementaryFig.2).Inordertoidentify putativemarkers associated withresistanceto V. dahliae, SNPs withinthegenesVe1(Solyc09g005090)andVe2(Solyc09g005080) conferringresistanceagainstthispathogenintomato(Fradinetal.,

2009)weresought.Survey ontheSolCAParray highlightedsix markersinthegeneVe2andnoneinVe1(Fig.6).

OnlyfourmarkersintheVe2gene(solcapsnpsl17547; sol-capsnpsl17546; solcapsnpsl17545; solcapsnpsl17544) resulted polymorphic. Solcapsnpsl17545 and sol-capsnpsl17544showedaheterozygousconditionfortheminor and themajorallele,respectively. Since thesetwo markersare probablystillsegregatinginthecollection,albeitco-segregating withtheothers,wedecidedtofocusonlyonthetwohomozygous markerssolcapsnpsl17547andsolcapsnpsl17546.

Fig. 6 also shows the haplotype combinations of the markers aforementioned, and the resistance or susceptibil-ity of the genotypes tested. Being the markers physically linked, their alleles co-segregated. Indeed only two haplo-type combinations were detected. The ﬁrst combination, sol-capsnpsl17547(A)−solcapsnpsl17546(G),includedallthesix susceptiblegenotypeswhilethesecond,solcapsnpsl17547(C)− solcapsnpsl17546(A),includedallthefourresistantgenotypes. Thisresultsuggestedastrongassociationofthesetwomarkerswith theresistancetoV.dahliaerace0insomeofthe‘Sorrento’

(7)

acces-Fig.6. SolCapMarkersintheVe1andVe2genesonchromosome9.

Foreachmarker,thepositioninbp(onleftsideofthechromosome)andSolCapID(onrightside)arereported.HaplotypecombinationsofmarkerssurveyedintheVe2gene andindicationrelatedtoresistance/susceptibilitytoVerticilliumdahliaerace0(Ve0)for‘Sorrento’genotypesisalsoreported.LettersofthehaplotypesarereferredtoIUPAC nucleotidecode.

sions,whichcouldbeconsideredefﬁcientmarkersforscreening genotypesforresistancetothissoil-bornepathogen.

4. Discussion

Theincreasingdemandforhigh-qualityandhigh-status prod-uctshasboostedinEuropethemarketforvalue-addedproducts that carry a strong identiﬁcation with a particular geographic region(McCluskeyandLoureiro,2003).Unfortunately,inItaly a greatnumberoflandraceshavebeendroppedfromtheNational Register,duetolackofcommercialinterestfrommajorseed com-panies(Mazzucatoetal.,2010).Theonly twotomatolandraces thathavebeenacceptedintoalternativevarietyprotection sys-temsare‘SanMarzano’and‘Vesuviano’labeledas‘PomodoroS. Marzanodell’AgroSarnese-Nocerino’and ‘Pomodorinodel Pien-nolo del Vesuvio’ (Protected Designation of Origin – PDO and ProtectedGeographicalIndication–PGI,respectively).Thesetwo landraceshavetheirownpeculiarorganolepticandquality charac-teristics,aswellasthe‘Sorrento’tomato(Sinesioetal.,2007;Lisanti etal.,2008).Therefore,thesamevarietyprotectionisdesirablefor thelatterthat,atthispurpose,shouldbeclearlydistinguishable fromotherlandracescharacterizedbymultilocularfruits.

By combining a wide morphological and a high-throughput genotypiccharacterization,ourstudyevidencedthatispossibleto clearlydistinguishthe‘Sorrento’landraceformother“salad”local varietiesmostlycommercializedonournationalmarketoffresh tomatoes.Indeed,themorphologicaltraitsanalyzedinourstudy mightbeusedtodiscriminate‘Sorrento’tomatofromother sim-ilarItalianlandraces.Forexample,theround-shaped‘Abruzzese’ or‘Perad’Abruzzo’tomatodiffersfromall‘Sorrento’accessionsfor

itsinternal/externalred-coloredfruitsshowinghighdegreeof rib-bingatpeduncleend.Furthermore‘Sorrento’landracewasclearly distinguishedfrom otherobcordate (CANLand PALB), ﬂattened (BOLS,PARM,PANTandFIOR)andoblate(ROTN)tomatoes,while groupedwithBELMandCBUElandraces,whichshowheart-pinked fruitsandaresometimeswronglyreferredto‘Sorrento’.In partic-ular,BELMdiffersfrom‘Sorrento’foritselongatedblossom-end shapeandverybigfruitweightingupto1000g.Circularshapein crosssection,nopufﬁnessandverythinpericarpdiscriminateCBUE tomatofromall‘Sorrento’accessions.

Inourstudyweexploitedthehigh-throughputSolCAPplatform to integrate morphological analysesin the clear differentiation between ‘Sorrento’ tomato and other multilocular Italian lan-draces.Uptillnow,highlyinformativegenotypingsystemshave beeninvestigatedonlargetomatogermplasmcollectionsaimedat studyingpopulationstructure,tomatoevolutionaryhistoryandthe geneticarchitectureofagronomictraits(Rancetal.,2008;Blanca etal.,2012;Shirasawaetal.,2013;Linetal.,2014),butfewregarded themolecularvariability atintra-speciﬁclevel(Andreakisetal., 2004;Raoetal.,2006;Mazzucatoetal.,2010;Cebolla-Cornejoetal., 2013).Eventhoughthemorphologicaldendrogramclusteredall the‘Sorrento’accessionsinthesamegrouptogetherwithCBUEand BELM,thehigh-throughputSNPanalysisevidencedastrong clus-teringof‘Sorrento’accessionsbutwithanoutgrouppositionofPS09 accession.ThisisinlinewiththeuniqueSNP(solcapsnpsl4963) thatdistinguishedall‘Sorrento’typesfromtheothergenotypes,but notPS09,evenwhencomparedwithawidercollectionofaround 120landraces(Saccoetal.,2015).ThedifferentoriginsiteofPS09, comingfromPagani(SA),mightexplainadiversiﬁedevolutionof thisaccessionfromtheothersoriginatingintheSorrento

(8)

penin-sula.Moreover,onlysix‘Sorrento’accessionsshowedtosharethe sameancestor,whileotherfourderivedfromtwoancestors,asalso happenedforCANL,PALB,PABR,CBUE,BELM,ROTNandthe out-standingPS09.Theaccessionsshowingoneancestor(PS10,PS11, PS05,PS07,PS04and PS06)wereprobablygrownin conditions ofconsiderablesexualisolation.Thishypothesismightbe reason-ableconsideringthatthecollectingareacorrespondstoamountain geographicalregion(upto400mta.s.l.), withmarginal farming unsuitableforcultivationoflandracesrequiringhigh-fertilitysoils typicalofAgroSarnese-Nocerinovalley.

Whenonlythe‘Sorrento’groupwasconsidered,theresultsof thepresentstudyhighlightedanappreciablevariabilityforfruit weight,puffiness,greencolorandgreenshoulderofunripefruits,as itwasalsoobservedbyCortés-Olmosetal.(2015),whenanalyzed thevariabilityofvariousSpanishlandraces.Inourwork,according tofruitmorphologyitwaspossibletodistinguishtwo‘Sorrento’ morphotypes:puffyfruitswithmediumshoulderandlightgreen coloratunripestage(PS01,PS08,PS09;PS11andPS13)andfruits showingweakpuffinessanddarkshoulder,mediumgreencolor excludingtheshoulder (PS04, PS05,PS06,PS07and PS10). Dif-ferent morphotypes have been already reported by Mazzucato et al. (2010) and Andreakis et al. (2004) for two other Italian tomatolandraces(‘AperaAbruzzese’ and‘Corbarino’). Previous studiesevaluatedthegeneticdiversityandstructureincollections oftomatolandracesthatalsoincluded‘Sorrento’accessions. García-Martínezetal.(2013)ina collectionof26cultivatedaccessions belongingtofourtomatotypes,‘Muchamiel’and‘DelaPera’from Spain,‘San Marzano’and ‘Sorrento’fromItaly, showedthatthe five‘Sorrento’accessionsmostlygroupedtogether,albeitthesame genotypeswerecharacterizedbysomedegreeofpossible admix-turewithonefrom‘DelaPera’orthe‘SanMarzano’group.Corrado etal.(2014)characterizedthepopulationstructureof75Italian landraces(mainlycollectedintheSouth)usingacustom-made Illu-minaSNP-panel.Theresultshighlightedthatthelandraceshada geneticstructuremainlyrelatedtothefruittype,andthatseven outofeight‘Sorrento’accessionswereincludedinacluster group-ingwithfewotheroxheart-shapedgenotypes.Unfortunately,these studiesonlyanalyzedthelandracesattheDNAlevelwithout com-biningmoleculardatawithmorphologicalones.

Tobetterselectthemostsuitable‘Sorrento’accessionstobe sub-mittedtovarietyprotectionsystemswecarriedoutaresistancetest toVerticilliumwilt.Inapreviousstudy,Acciarrietal.(2010)testeda collectionof169tomatolandracesofdifferentfruittypeand prove-nienceforreactiontoinfectionwithV.dahliae:someindividualsof the‘Sorrentino’landracewereasresistantas‘FloridaVe’control cultivar.Similarlyinourstudysome‘Sorrento’accessions exhib-itedresistancetothispathogen,thusconﬁrmingthatitispossible toselectaccessionsresistanttoV.dahliaerace0thatarepromising forbreedingpurposes.WealsodetectedtwopolymorphicSNPsin thegeneVe2,mappingonchromosome9,thatmightbesuitable inthefuturetocarryoutamarker-assistedselectionofresistant genotypes.

5. Conclusions

Ourdatashowedthatthe‘Sorrento’tomatoiswelldistinctfrom othermultilocularlandracesmostwidespreadinItaly.Eventhetwo pink-coloredtraditionaltomatoes‘CuorediBue’and‘Belmonte’, oftenwronglyreferredto‘Sorrento’,canbeclearlydistinguished fromthelatterlandrace.

Thankstothelargegenotypingperformed,weidentiﬁedone SNPmarkerthatclearly distinguishesthe‘Sorrento’ fromother similarlandraces.ThepossibilityofusingasingleSNPmarkerto discriminate‘Sorrento’accessionsfromsimilar/confounding geno-typeswouldbeanefﬁcientandcost-effectivewaytocharacterize

thislandrace.Thiswillbeacrucialsteptodefendthequality mar-ketsassociatedwiththislandraceandtocontrolthecommercial fraudassociatedwithit.

Resistance assay for Verticillium wilt allowed for the iden-tiﬁcation of promising ‘Sorrento’ accessions and also for the identiﬁcationoftwoSNPmarkerssuitableforgenotypingforthe susceptibilitytothissoil-bornepathogen.

Acknowledgements

WethankseedcompaniesandNazzarenoAcciarri(CREA-ORA) forprovidingaccessionseeds.Thisworkhasbeencarriedoutwithin theprojects“Recuperoevalorizzazionedelgermoplasmaorticolo campano”fundedbyCampaniaRegionandgrant“GenoPOM-pro −Integratingpost-genomicplatformstoenhancethetomato pro-ductionchain”project(PON02003953082360)fundedbyItalian MinistryofUniversityandResearch(MIUR).Thisarticleisdedicated tothememoryofAnielloParisiinrecognitionofhisextraordinary tenacity,expertiseandwisdomin‘PomodorodiSorrento’ exploita-tionandvalorization.Awonderfullegacyremainsinthepeoplehe trainedandinspired.

AppendixA. Supplementarydata

Supplementarydataassociatedwiththisarticlecanbefound, intheonlineversion,athttp://dx.doi.org/10.1016/j.scienta.2016. 02.038.

References

Acciarri,N.,Rotino,G.L.,Voltattorni,S.,Mennella,G.,Nigro,C.,Cerbino,D.,Baudino, M.,Maestrelli,A.,Costa,G.,Noferini,M.,Fiori,G.,Valentino,D.,Tamietti,G., Sabatini,E.,2007.Recupero,caratterizzazioneevalorizzazionedivarietàlocali italianedipomodorodamensa.ItalusHortus14(2),74–84.

Acciarri,N.,Sabatini,E.,Ciriaci,T.,Rotino,G.L.,Valentino,D.,Tamietti,G.,2010.The presenceofgenesforresistanceagainstVerticilliumdahliaeinItaliantomato landraces.Eur.J.Hortic.Sci.75(1),8–14.

Andreakis,N.,Giordano,I.,Pentangelo,A.,Fogliano,V.,Graziani,G.,Monti,L.M., Rao,R.,2004.DNAﬁngerprintingandqualitytraitsofcorbarinocherry-like tomatolandraces.J.Agric.FoodChem.52(11),3366–3371.

Blanca,J.,Canizares,J.,Cordero,L.,Pascual,L.,Diez,M.J.,Nuez,F.,2012.Variation revealedbySNPgenotypingandmorphologyprovidesinsightintotheorigin ofthetomato.PLoSOne7(10),e48198.

Bradbury,P.J.,Zhang,Z.,Kroon,D.E.,Casstevens,T.M.,Ramdoss,Y.,Buckler,E.S., 2007.TASSEL:softwareforassociationmappingofcomplextraitsindiverse samples.Bioinformatics23(19),2633–2635.

Cebolla-Cornejo,J.,Roselló,S.,Nuez,F.,2013.Phenotypicandgeneticdiversityof Spanishtomatolandraces.Sci.Hortic.162(10),150–164.

Corrado,G.,Caramante,M.,Piffanelli,P.,Rao,T.,2014.GeneticdiversityinItalian tomatolandraces:implicationsforthedevelopmentofacorecollection.Sci. Hortic.168,138–144.

Cortés-Olmos,C.,Leiva-Brondo,M.,Roselló,J.,Raigón,M.D.,Cebolla-Cornejo,J., 2014.Theroleoftraditionalvarietiesoftomatoassourcesoffunctional compounds.J.Sci.FoodAgric.11,2888–2904,http://dx.doi.org/10.1002/jsfa. 6629.

Cortés-Olmos,C.,Vilanova,S.,Pascual,L.,Roselló,J.,Cebolla-Cornejo,J.,2015.SNP markersappliedtothecharacterizationofSpanishtomato(Solanum lycopersicumL.)landraces.Sci.Hortic.194,100–110,http://dx.doi.org/10.1016/ j.scienta.2015.07.036.

Earl,D.A.,vonHoldt,B.M.,2012.STRUCTUREHARVESTER:awebsiteandprogram forvisualizingSTRUCTUREoutputandimplementingtheEvannomethod. Conserv.Genet.Resour.4(2),359–361.

Ercolano,M.R.,Carli,P.,Soria,A.,Cascone,A.,Fogliano,V.,Frusciante,L.,Barone,A., 2008.Biochemical,sensorialandgenomicproﬁlingoftraditionalItalian tomatovarieties.Euphytica164(2),571–582.

EuropeanUnion–CommunityPlantVarietyOfﬁce(2007)–Protocolfor Distinctness,UniformityandStabilityTests.LycopersiconlycopersicumL. KarstenexFarw.Adoptedon21/03/2007.CPVO-TP/44/3Final.English.website www.cpvo.europa.eu.

Evanno,G.,Regnaut,S.,Goudet,J.,2005.Detectingthenumberofclustersof individualsusingthesoftwareSTRUCTURE:asimulationstudy.Mol.Ecol.14, 2611–2620.

Fradin,E.F.,Zhang,Z.,JuarezAyala,J.C.,Castroverde,C.D.,Nazar,R.N.,Robb,J.,Liu, C.M.,Thomma,B.P.,2009.GeneticdissectionofVerticilliumwiltresistance mediatedbytomatoVe1.PlantPhysiol.150(1),320–332,http://dx.doi.org/10. 1104/pp.109.136762.

(9)

García-Martínez,S.,Corrado,G.,Ruiz,J.J.,Rao,R.,2013.Diversityandstructureofa sampleoftraditionalItalianandSpanishtomatoaccessions.Genet.Resour. CropEvol.60(2),789–798.

Hamilton,J.P.,Sim,S.C.,Stoffel,K.,VanDeynze,A.,Buell,C.R.,Francis,D.M.,2012. Singlenucleotidepolymorphismdiscoveryincultivatedtomatovia sequencingbysynthesis.PlantGenome5(1),17–29.

Hammer,Ø.,Harper,D.A.T.,Ryan,P.D.,2001.PAST:PaleontologicalStatistic softwarepackageforeducationanddataanalysis.Paleontol.Eletron.4(1),1–9 http://palaeoelectronica.org/20011/past/issue101.htm.

Hirakawa,H.,Shirasawa,K.,Ohyama,A.,Fukuoka,H.,Aoki,K.,Rothan,C.,Sato,S., Isobe,S.,Tabata,S.,2013.Genome-wideSNPgenotypingtoinfertheeffectson genefunctionsintomato.DNARes.20(3),221–233,http://dx.doi.org/10.1093/ dnares/dst005.

Lin,T.,Zhu,G.,Zhang,J.,Xu,X.,Yu,Q.,Zheng,Z.,Zhang,Z.,Lun,Y.,Li,S.,Wang,X., Huang,Z.,Li,J.,Zhang,C.,Wang,T.,Zhang,Y.,Wang,A.,Zhang,Y.,Lin,K.,Li,C., Xiong,G.,Xue,Y.,Mazzucato,A.,Causse,M.,Fei,Z.,Giovannoni,J.J.,Chetelat, R.T.,Zamir,D.,Stadler,T.,Li,J.,Ye,Z.,Du,Y.,Huang,S.,2014.Genomicanalyses provideinsightsintothehistoryoftomatobreeding.Nat.Genet.46, 1220–1226.

Lisanti,M.T.,Piombino,P.,Genovese,A.,Pessina,R.,Moio,L.,2008.Traditional Italiantomato(LycopersiconesculentumMill.)cultivarsandtheircommercial homologues:differencesinthevolatilecomposition.Ital.J.FoodSci.3(20), 333–350.

Mantel,N.,1967.Thedetectionofdiseaseclusteringandageneralizedregression approach.CancerRes.27,209–220.

Mazzucato,A.,Pepponi,F.,Chioccia,G.,Soressi,G.P.,1998.Ricostituzionee valorizzazionedellavarietàlocaledipomodoroScatolonediBolsena.Italus Hortus5(3),30–35.

Mazzucato,A.,Papa,R.,Bitocchi,E.,Mosconi,P.,Nanni,L.,Negri,V.,Picarella,M.E., Siligato,F.,Soressi,G.P.,Tiranti,B.,Veronesi,F.,2008.Geneticdiversity, structureandmarker-traitassociationsinacollectionofItaliantomato (SolanumlycopersicumL.)landraces.Theor.Appl.Genet.116(5),657–669. Mazzucato,A.,Ficcadenti,N.,Caioni,M.,Mosconi,P.,Piccinini,E.,Sanampudi,

V.R.R.,Sestili,S.,Ferrari,V.,2010.Geneticdiversityanddistinctivenessin tomato(SolanumlycopersicumL.)landraces:theItaliancasestudyof‘Apera Abruzzese’.Sci.Hortic.125(1),55–62.

McCluskey,J.J.,Loureiro,M.L.,2003.Consumerpreferencesandwillingnesstopay forfoodlabeling:adiscussionofempiricalstudies.J.FoodDistrib.Res.34(3), 95–102.

Parisi,M.,DOnofrio,B.,Pentangelo,A.,Villari,G.,Giordano,I.,2008.Morphology, productivityandqualitativecharacterizationofthetraditionaltomatoecotype PomodorodiSorrentooriginatingfromCampaniaRegion,SouthernItaly.Acta Hortic.789,205–210.

Pritchard,J.K.,Stephens,M.,Donnelly,P.,2000.Inferenceofpopulationstructure usingmultilocusgenotypedata.Genetics155(2),945–959.

Ranc,N.,Mu ˜nos,S.,Santoni,S.,Causse,M.A.,2008.ClariﬁedpositionforSolanum lycopersicumvar.cerasiformeintheevolutionaryhistoryoftomatoes (Solanaceae).BMCPlantBiol.8,130, http://dx.doi.org/10.1186/1471-2229-8-130.

Rao,R.,Corrado,G.,Bianchi,M.,DiMauro,A.,2006.(GATA)4DNAﬁngerprinting

identiﬁesmorphologicallycharacterized‘SanMarzano’tomatoplants.Plant Breed.125(2),173–176.

Ruggieri,V.,Francese,G.,Sacco,A.,D’Alessandro,A.,Rigano,M.M.,Parisi,M., Milone,M.,Cardi,T.,Mennella,G.,Barone,A.,2014.Anassociationmapping approachtoidentifyfavorableallelesfortomatofruitqualitybreeding.BMC PlantBiol.14,337,http://dx.doi.org/10.1186/s12870-014-0337-9.

Sacco,A.,Ruggieri,V.,Parisi,M.,Festa,G.,Rigano,M.M.,Picarella,M.E.,Mazzucato, A.,Barone,A.,2015.Exploringatomatolandracescollectionforfruit-related traitsbytheaidofahigh-throughputgenomicplatform.PLoSOne10(9), e0137139,http://dx.doi.org/10.1371/journal.pone.0137139.

Sauvage,C.,Segura,V.,Bauchet,G.,Stevens,R.,TheyDo,P.,Nikoloski,Z.,Fernie, A.R.,Causse,M.,2014.Genomewideassociationintomatoreveals44 candidatelociforfruitmetabolictraits.PlantPhysiol.165(3),1120–1132. Shirasawa,K.,Fukuoka,H.,Matsunaga,H.,Kobayashi,Y.,Kobayashi,I.,Hirakawa,

H.,Isobe,S.,Tabata,S.,2013.Genome-wideassociationstudiesusingsingle nucleotidepolymorphismmarkersdevelopedbyre-sequencingofthe genomesofcultivatedtomato.DNARes.20,593–603.

Sim,S.C.,Durstewitz,G.,Plieske,J.,Wieseke,R.,Ganal,M.W.,VanDeynze,A., Hamilton,J.P.,Buell,C.R.,Causse,M.,Wijeratne,S.,Francis,D.M.,2012. DevelopmentofalargeSNPgenotypingarrayandgenerationofhigh-density geneticmapsintomato.PLoSOne7(7),e40563,http://dx.doi.org/10.1371/ journal.pone.0040563.

Sinesio,F.,Paparario,M.,Moneta,E.,2007.Sensorycharacteristicsoftraditional ﬁeldgrowntomatogenotypesinsouthernItaly.J.FoodQual.30(6),878–895.