growthisblockedbeforepollinationandthatauxinisa keyregulatorofovarygrowthderepressionatfruitset Goetzetal2007Pandolfinietal2007Auxinrespon siveproteinIAA9andauxinresponsefactorARF8repress ovar ID: 487336
Download Pdf The PPT/PDF document "Thecurrentmodelofregularfruitsetimpliest..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Thecurrentmodelofregularfruitsetimpliesthatovary growthisblockedbeforepollinationandthatauxinisa keyregulatorofovarygrowthde-repressionatfruitset (Goetzetal.2007;Pandolfinietal.2007).Auxinrespon- siveproteinIAA9andauxinresponsefactorARF8repress ovarygrowthbeforefertilisation.Followingpollinationin raspberrythereisaperiodofrapidgrowthduetocell division.Thisisfollowedbyaperiodofslowgrowth duringwhichtheembryodevelopsandtheendocarp becomeshardened,untilfinallycellenlargementresults inaperiodofrapidgrowth.Otherphytohormones (giberellin,cytokinin,brassinosteroids,ethyleneand abscisicacid)playaroleinfru itinitiationanddevelop- ment(SchwabeandMills1981;Vriezenetal.2008). Delayeddifferentiationoftheembryosachasbeenassoci- atedwithlowdrupeletsetinsomeclonesofthediploid cultivarSumner.ThecultivarLathamcanalsoshowa crumblyphenotypeandthisisthoughttobeduetomuta- tionofthedominantalleleataheterozygousgenelocus causingplantstobecomehomozygousforadeleterious recessivegene(Jennings1967b).InSumnerithasbeen suggestedthattheeffectsontheembryosacandalsothe reducedproductionoffertilepollenarecausedbya mutationgivinghomozygosityfortworecessivegene pairs(Daubenyetal.1967). Jennings(1967a)suggestedthatincultivarNorfolk Giantembryosacdevelopmentceasedatanearlystage. Jennings(1971)alsosuggestedthatanoptimumstatus foramaternalgrowthsubstancewasrequiredforgood fruitsetandseeddevelopment.Howevertherewasevi- dencethatthestrengthofmaternaleffectswasconsider- ablyinfluencedbyenvironmentalfactors. Fromacv.Lathamself,Jennings(1967b)demonstrated thatseedlingsobtainedcouldbeclassifiedintothree groups:normal,crumblyandsterile.Heproposedamodel oftwogeneticloci,designated St foronewhoserecessive formgivescompletesterilityand Cr whoserecessiveform givescrumblyfruit,toexplainthe9:3:4segregationratio obtainedofnormal(StCr):crumbly(Stcr):sterile(stCror stcr)as st isepistaticto Cr.Cr werepostulatedtobe linkedto GeneH (pubescentcanes)andalso geneT (fruit colour)(Jennings,1988). GeneH hasbeenidentifiedin raspberryonlinkagegroup2(Grahametal.2006)anda QTLforfruitcolourwasalsoidentifiedcloseto GeneH (McCallumetal.2010).Thismaysuggest Cr isasJennings suggestedalsoonthischromosome. Thisworksetouttoexaminewhetherthecrumbly fruitsyndromesegregatedinapopulationusingLatham asoneparentandtoexaminetheseasonalandenviron- mentalimpactonexpressionofthecrumblytraitand identifyQTLassociatedwiththecrumblyphenotype.It alsosetouttoidentifyanyassociationwiththe GeneH regionandimpactofgenesforfertilityonthetrait. Resultsanddiscussion Phenotypicscoring Overaperiodof7fruitingseasonswehaveinvestigated thesegregationofcrumblyfruitsyndromeinaLathamx GlenMoycross.Theresultsshowacomplexpatternof two crumbly phenotypesbasicallydifferinginseverity, oneofwhichmaybethesterilephenotypeproposedby Jennings(1967b).Table1showstheproportionofthe offspringwithcrumblyfruitforeachyearandenviron- ment.Thehighestproportionoccurredinthefieldin 2011,where73%oftheoffspringhadcrumblyfruit. Thelowestproportionofcrumblyfruitinthefieldwas in2008,whereonly4%werescoredascrumbly.The proportionsofcrumblyfruitwerealwayslowerinthe polytunnelthaninthefield,rangingfrom22%in2010 to1%in2008.In2010thecrumblyandsterilepheno- typeswerescoredseparately,forasinglereplicateofthe 188linesofthemappingpopulation.Inthefield,115 werescoredasnormal,50ascrumblyand15assterile, with8missingscores.Inthepolytunnel,117were scoredasnormal,29ascrumblyand5assterile,with37 missingscores.Comparingthetwoenvironments,87 werescoredasnormalinbothfieldandpolytunnelin 2010,15werescoredascrumblyinbothenvironments and3werescoredassterileinbothenvironments.In viewofthesmallnumberofplantsclassedas sterile plants andthelackofagreementinscoringthisbetween environmentswehavecombinedthesterileandcrumbly classesinfurtheranalyses. Table2showsgammastatisticsmeasuringassociations betweenthefieldcrumblyscoresforthedifferentyears. Thepolytunnelscoreswereexcludedhereduetothe Table1 Proportionoftheoffspringwithcrumblyfruit YearEnvNo.scoredProportioncrumblyfruits.e. 2004FieldA940.540.051 2004FieldB940.310.048 2007FieldA1880.270.032 2007Poly1880.090.021 2008FieldA1880.040.015 2008Poly1880.010.007 2009FieldA1880.320.036 2009Poly1880.040.007 2010 a FieldA1800.360.036 2010 a Poly1530.220.034 2011FieldA1680.730.035 2012FieldA1420.640.040 Severity(0 4)means.e. 2011FieldA1681.160.077 2012FieldA1421.160.093 2009valuesarebasedon3replicates,othersonone.s.e.=standarderror a 2010showstheproportionofeithercrumblyorsterilefruit Graham etal.SpringerPlus (2015) 4:223 Page2of9 lowincidenceofcrumbliness.Thegammastatistics showedthattherewerenosignificantassociations( p � 0.05)ofotheryearswiththe2008scores,whichhadthe lowestlevelofcrumbliness.Theassociationsofother yearswiththe2004scores,whichwerebasedonMP1 only,werealsogenerallysmall.Theassociationsamong thescoresonMP1andMP2in2007,2009,2010,2011 and2012wereallhighlysignificant(gamma 0.67, p 0.001),withaparticularlyhighgammastatisticof0.99 betweentheseverityscoresin2011and2012. Noprogenywerealwaysscoredascrumbly,though somewereassessedasbeingcrumbly75%ormoreof thetimesscored.Someindividualsneverexhibitedthe crumblyphenotype.Thecrumblyphenotypewasalways moresevereundertheopenfieldconditionsthaninthe polytunnelandvariedsignificantlyfromseasontoseason, accordingtotheoveryearsandsitesanalysis.Aswellas theenvironmentalandseasonaleffect,theagreementover yearsmeasuredbythegammascoresindicatedastrong geneticeffectforthecrumblytrait.MetOfficemonthly weatherdatawasexaminedonmaximumandminimum temperature,frost,rainandsunshinebutnoassociations couldbeidentifiedbetweenweatherconditionsandextent ofthecrumblyconditionacrossseasons.Forexamplein thetwosevereseasons2004and2012theweathercondi- tionswereverydifferent,with2004beingawarmdrysea- sonand2012coolerandwetter.Therewasalatespring frostin2012whichdidnotoccurin2004.In2008where littlecrumblyfruitoccurred,themajordifferencewasin theamountofrainfallattheopenflowerstage. LinkagemappingandQTLanalysis MappingandQTLanalysis Basedonpermutationtests,athresholdof13.8forthe Kruskal-Wallis(KW)statisticwithonedegreeoffree- domwasused,correspondingtoagenome-widesignifi- canceof p =0.05.TheKruskal-Wallisanalysisindicated associationsabovethisthresholdofthecrumblyphenotypes withmarkersonLG1forthefieldscoresin5ofthe7 seasonsanalysed(2007,2009,2010,2011and2012).No significantassociationswiththisregionweredetected inthefieldtrialsin2004(thefirstfullfruitingyear)or in2008,whentheincidenceofcrumblywasverylow (4%),orwithanyofthepolytunneltrials,whichhada muchlowerincidenceofthecondition.Themostsig- nificantregionincludedmarkerRUB256e,anSSRwith fouralleles(abinLatham,cdinGlenMoy)at101cM, althoughtypicallymarkersbetween90cMand110cM weresignificant.Figure1showsthelinkagemapofLG 1,withthemostsignificantmarkerindicatedalongwith one-LODsupportintervalsfortheseverityscores. Table3showstheresultsofmodellingtherelationship betweeneachofthecrumblytraitsandthismarker, usingageneralisedlinearmodelwithbinomialerrors andalogitlinkfunctionforthebinaryscoresanda normalmodelfortheseverityscores,andexpressing themarkereffectsasadditiveeffectsofeachparentto- getherwithadominanceeffect,asdefinedinequation(1). Forthefieldscoresfrom2007,2009,2010,2011and2012 (iewheretheKruskal-Wallistestwasabovethegenome- widepermutationthreshold),theadditiveeffectofthe Lathamparentwassignificant( p 0.001)intheGLM,but theadditiveeffectoftheMoyparentandthedominance effectwerenotsignificant( p �0.05).Foreachofthese traitsthedirectionoftheeffectwasconsistent,withthe meanproportionofcrumblyfruitbeingsignificantly higherinthegenotypescarryingtheLatham b allelethan inthosewiththeLatham a allele.Forthefieldscores from2004andthepolytunnelscoresfrom2007and2009, theadditiveeffectoftheLathamalleleatRUB256ewas significant(0.005 p 0.05)butagaintheadditiveeffect oftheMoyparentandthedominanceeffectwerenotsig- nificant( p �0.05).Again,themeanproportionofcrumbly fruitwashigherinthegenotypescarryingtheLatham b allelethaninthosewiththeLatham a allele.Thelasttwo columnsofTable3showthepredictedproportionof crumblyfruitinthetwogenotypeclasses. TheKruskal-Wallisanalysisalsoindicatedassociations abovethegenome-widepermutationthresholdbetween thecrumblyscoresfromthepolytunnelin2007,2009 and2010andthefieldseverityscoresfrom2011and 2012withmarkerssegregatingonLG3,althoughthe bestmarkervariedslightly(region107-133cM).Again thesignificantmarkerssegregatedintheLathamparent. Figure2showsthelinkagemapofLG3,withthemost significantmarkerforeachtraitindicatedalongwithone- LODsupportintervalsfortheseverityscores.Themarker ERubLR_SQ05.3_D11AOCat121cMwaschosenas representativeofthisregion,onthegroundsofalow numberofmissingscores,andwasincludedtogether withRUB256efromLG1inafurtherGLMtotesttheir jointsignificance.Thesignificanceoftheadditiveeffectof Table2 Gammastatisticsforassociationsamongthecrumbly scoresfromthefieldsites 2004, FieldA 2004, FieldB 20072008200920102011 2004, FieldB 0.59* 20070.400.71** 20080.260.65 0.04 20090.62**0.54*0.97***0.34 20100.62*0.81***0.81***0.420.79*** 20110.48*0.57*0.78***0.050.70***0.78*** 20120.72***0.61*0.75***0.080.67***0.80***0.99*** 2009valuesarebasedon3replicates,othersonone.Theseverityscoresare usedin2011and2012 *** p 0.001;** p 0.01;* p 0.05 Graham etal.SpringerPlus (2015) 4:223 Page3of9 theLathamparentatmarkerERubLR_SQ05.3_D11AOC forthetraitsabovewasconfirmed,with p =0.003forthe polytunnelin2007and p 0.001fortheothertraits. ERubLR_SQ05.3_D11AOCalsoshowedalowersignifi- canceintheGLMfortraits2004FieldB( p =0.003),2010 Field( p =0.004),2011Fieldincidence( p =0.017)and 2012Fieldincidence(Field=0.011).Forallofthesetraits thedirectionoftheeffectwasconsistent,andtherewas nosignificantinteraction( p �0.05)betweenthetwo markersERubLR_SQ05.3_D11AOCandRUB256e.Table4 showsthepredictedmeanseverityscoresin2011and 2012fromtheGLMforthetwomarkerstogether,ranging from0.45to2.0. Acombinedanalysisoveryearsandsites(fieldorpoly- tunnel)wasconductedonthebinaryscores,omittingthe datafrom2004asthiswasonlyscoredontheMP1lines. Theanalysisofdeviancetablefortheincidenceofcrum- blinessisshowninAdditionalfile1:TableS1.Thisshows significanteffectsofyear,siteandtheirinteractionand significanteffectsoftheLathamparentatthemarkers RUB256eonLG1andERubLR_SQ05.3_D11AOConLG3. Therewerenosignificantinteractionsinvolvingyear, buttherewasasignificantinteractionbetweensiteand eachofthemarkers.Table5showsthemeancrumbly scoresfromtheseinteractions,withtheeffectof RUB256eonLG1beinggreateratthefieldsitesand thatofERubLR_SQ05.3_D11AOConLG3being greaterinthepolytunnelsites.Asimilarcombinedana- lysiswasconductedonthe2011and2012fieldseverity scorestogether,butnosignif icantinteractionsbetween theyearandthemarkerweredetected. Relationshipwithripening Thecrumblyscoresin2007,2009,2010,2011and2012 showedsignificantcorrelations( p 0.05)withsomeof thetimetoripeningscoresrecordedforthispopulation in2006byGrahametal.(2009),asshowninTable6. Thelargestcorrelationswerewiththetimetoreachthe fruitsetstageandthetimetoreachgreenfruit.Some crumblyscoresalsohadasignificantcorrelationwith thetimetoreachthegreen/redstage,butthetimeto reachtheopenflowerstagewasnotcorrelatedwiththe crumblyscores.Thecorrelationswerepositivei.e.the proportionofcrumblyfruitincreaseswiththetime takentoreachfruitsetandgreenfruit.Ripeningisalso associatedwithmanymarkersonLG3includingthere- gionidentifiedabove(Grahametal.2009).AGLMwith markersERubLR_SQ05.3_D11AOCandRUB256eand timetofruitsetwasinvestigatedusingall-subsetregression toidentifythemostsignificantexplanatoryvariablesfor eachcrumblyscore,buttherewasnoconsistencyinthe choiceamongtimetofruitset,ERubLR_SQ05.3_D11AOC orbothofthese.Wecannotthereforedrawanyconclu- sionsatpresentastowhetherripeningtimeaffects Ri25D10SSR04 0 Ri4CL3SNP 3 RUB137a 9 454c1717Embdef 19 RUB124a 24 ERubParaSQ008D04 27 ERubLR_SQ9.2_C12PE 29 JHI_25076_3_deh 36 E40M60-106 44 P13M60-117 47 ERUBLR_SQ5.2_H12Cell 53 E40M60-125 55 P13M58-86 59 P13M40-131 61 Ripgl 63 E40M43-116 65 454c2261HS 66 RiCell1 68 bes_Ri36G22F 70 454c1149_CCDr 71 JHIRi34405_158_RAD 73 454c7333bChHS 76 454C6398PGD_snp212 80 454C6398PGD_snp155 454C6398PGD_snp91 81 P12M31-171 83 nudix256e JHI_Ri_29680 JHI_Ri_5828 JHI_Ri_37529 89 E41M39-125 ara4-2_256esnp222 90 RUB262b 92 ara4-2_256esnp303 93 454C2375_COP1b ara4-2_256esnp234 95 signalo_256esnp115 96 454C1037_ACCsynth ara4_1_snp139 97 E41M60-315 98 RUB256e*** 101 ERubLR_SQ10.2_H07Aq 103 RUB243a RibHLH 104 bes_Ri31G22R 105 RUB119a 106 P12M95-95 110 Rub232b 112 RibHLH2 123 Severity 2011 Severity 2012 LG1 Fig.1 LinkagemapforLG1.Themostsignificantmarkeraccording totheKruskal-Wallistestisthesameforallbinarytraits,andis shownby***.One-lodsupportintervalsfortheseveritytraitsare alsoshown Graham etal.SpringerPlus (2015) 4:223 Page4of9 crumblinessdirectlyorwhetherbothtraitsarecontrolled byoneormoregenesonLG3.Theassociationwithripen- ingisinteresting,withthelongerthefruittakestogetto thefruitsetandgreenfruitstage,themorelikelyitistobe crumbly.Thisisparticularlyapparentwhenconsidering thedifferencebetweenpolytunnelgrownfruitandfield grownfruitwhereunderfieldconditionsthefruitalways takelongertogettothesestagesandbeyond.Graham etal.(2009)identifiedmarkersonLG3asassociatedwith timetoripening.Atthisstagehoweverconclusionscannot bedrawnastowhetherripeningtimeaffectscrumbliness directlyorwhetherbothtraitsarecontrolledbyoneor moregenesonLG3.Contrarytothesuggestionby Jennings(1967b)thatcrumblyfruitwasrelatedtothe GeneHregion,nogeneticassociationwiththisregion onLG2couldbeidentifiedwiththecrumblyfruitsyn- drome.TheGeneHregionisaninterestingregionassoci- atedwithanumberofdiversetraits(KnightandKeep 1958;JenningsandBrydon1989;Jennings1962;Keep 1968,1976;JenningsandMcGregor1988;Anthonyetal. 1986;Jennings1967a).Interestingly,the Hh genotypeof GeneHwasassociatedwithaslowingdownofripening acrossallstagesfromopenflowerstothegreen/redstage comparedtothe hh genotype(Grahametal.2009).The correlationwithGeneHandcrumblyfruitidentifiedby Jenningsmayactuallybeduetothisassociationwithrip- eningtimeratherthantotheregionitself.Interestingly allelesassociatedwithlongertimetoripeningintheGene HregionandalsoonLG3andLG5areassociatedwith smallerrootdensityanddiametermeasuresandmaybe regardedasgeneralvigorgenes(Grahametal.2011).This mayalsobeafactorincrumblyfruitandwillneedfurther investigation. GenecontentinRub256eregion Theraspberrygenomepseudomolecules(providedby JoshuaUdallBYU,GeneticsandBiotechFaculty (pws.byu.edu))weresearchedusingBLAST(Altschul etal.1990)foranyregionsthatmatchedRUB256e.Six differentgeneswerepredictedintheregionasfollows: Methyltransferase(XP004133879.1),signalosomecomplex (XP002511799.1),cysteineprotease(XP002306369.1),Ara4 interactingprotein(XP002511798.1),Nudixhydrolase. (XP002266987.1)andMethyltransferase(AER13155.1) containingtheRub256emarkerandthosecontaininga polymorphismbetweentheparentsweremappedusing primersinTable7toconfirmlocationandallowfuture geneexpressionstudiestobecarriedout. Conclusion Thisstudyhashighlightedthatenvironmental,seasonal andgeneticfactorsallplayaroleinthedevelopmentof crumblyfruitinredraspberry.AregiononLG1atthe Rub256emarker,andanassociationwithripeningtime andripeningassociatedmarkersonLG3wereidentified forfurtheranalysis.Noassociationwithcrumblyfruit andGeneHwasdetermined.Thisworkhasallowedus toidentifyageneticcomponenttotheconditionwhich Table3 EffectoftheRUB256emarkeronLG1onthecrumblyscores YearEnv.KWstatistic (3df) Sig.ofLathamalleleMeancrumblyscorefora-offspring(s.e.)Meancrumblyscoreforb-offspring(s.e.) Incidence(0 1) 2004FieldA4.70.0460.44(0.071)0.65(0.071) 2004FieldB7.40.0070.18(0.056)0.44(0.074) 2007FieldA26.9***0.0010.12(0.033)0.43(0.053) 2007Poly5.80.0360.05(0.022)0.14(0.037) 2008FieldA0.50.8490.04(0.019)0.05(0.022) 2008Poly5.70.0780.00(0.001)0.02(0.016) 2009FieldA20.8***0.0010.16(0.037)0.49(0.054) 2009Poly5.60.0050.02(0.007)0.06(0.012) 2010FieldA25.1***0.0010.22(0.042)0.53(0.055) 2010Poly4.40.0940.17(0.041)0.28(0.053) 2011FieldA19.7***0.0010.58(0.052)0.89(0.036) 2012FieldA24.4***0.0010.46(0.059)0.83(0.044) Severity(0 4) 2011FieldA27.8***0.0010.77(0.097)1.59(0.104) 2012FieldA26.3***0.0010.69(0.121)1.62(0.121) KW=theKruskal-Wallisstatisticforthismarker;df=degreesoffreedom.ThelastthreecolumnsshowthesignificanceoftheadditiveeffectoftheLa thamallele inageneralisedlinearmodel,andthepredictedmeancrumblyscorefortheoffspringinheritingeitherthe a alleleorthe b allelefromLatham *** p 0.001 Graham etal.SpringerPlus (2015) 4:223 Page5of9 canbeassessedforbreedinglineslesspronetocrumbly fruit.Controlledenvironmentstudieswillbecarriedout inanattempttodefinetriggersoftheconditioninthose sampleswherethephenotypevariesbetweenseasons andenvironments. Materialsandmethods Fieldandpolytunneltrials Thepopulation,asdescribedpreviously(Grahametal. 2004,2006,2009,2011,2014;Woodheadetal.2013), consistsofafullsibfamilygeneratedfromacrossbe- tweentheEuropeanredraspberrycv.GlenMoyandthe NorthAmericanredraspberrycv.Latham.Trialswere arrangedinarandomisedblockdesignwiththreerepli- cateseachcontainingtworeplicatedplantsof330geno- typesattwofieldsites,andthreesingle-plantreplicates of188genotypes(randomlyselectedfromtheoriginal 330fullsibfamilyformappingpurposes)underpoly- thenetunnelprotection(McCallumetal.2010). Phenotypicdatacollectionforcrumblyfruit Mappingofthispopulationhasfocusedontwosubsets, aninitialpopulationmappingpopulation1(MP1)of94 seedlings,andafurthermappingpopulationofanadd- itional94seedlings(MP2).Phenotypicdataoncrumbly fruit,scoredascrumblyornotaccordingtoJennings 1967b,wascollectedonMP1onlyin2004onasingle replicateattwofieldlocations(A&B).In2007and 2008,crumblyfruitwasassessedsimilarlyonasingle replicateofthelinesinMP1andMP2atonefieldsite (A)andonplantsgrownunderapolytunnel.In2009, threereplicateswereassessedforthelinesinMP1and LEAF86 0 E41M60-184 10 P13M39-177R 16 P14M61-121 RUB22a 20 FRUITE8OMT 21 ERUB271PR 23 ERubLR_SQ01_P18 26S_prot 25 Risnf4 26 ERubLR_SQ1_FG23Pgl 27 ERubLR_SQ071_E10TF JHIRi681_152_ATN 28 ERubLR_SQ13.2C12IPPI RubnebH15 30 ERubLR_SQ12.4A04DMQ 31 ERubLR_SQ4.2_A08LTP 32 P13M58-112 36 JHIRi681_131_ATN 40 P12M121-186 48 JHIRi_10729_AlkR 52 ERubLR_SQ13.2E09Exp 54 454C6570_ISPH 58 RUB160a Rub17a 60 Rub242a 63 JHIRi_54644GR_snp97 66 454C1803_PGIP2 67 RiM015 RiCTR1 68 454CL8848C1_ZFP8 70 Ri4CL1SNP JHI_20114_3_NIF water_c0035_GTIP 71 JHI_33046_HS 75 454C0016_Maldehy 77 ERubLRcont74PME-I 81 Pip_La_E02V0_Aq1 84 RißGal1 85 JHI_42376_unk 86 JHI_44413_226MyB 88 Ri9022SSR01 90 RiMYB 92 454C3991_PME 95 ERubLath2_C21 105 454C2985_PSY***F12 108 JHIRi_197775PIF***P09 JHIRi_5456PIF***P10 110 JHIRi20954_snp187 113 water_c0020_PIP2 114 ERubLR_SQ12.2C05Acon 116 P14M60-129***F11 118 ERubLR_SQ05.3_D11AOC 121 ERubLRSQ10.2E02SAMDC 122 SAMDC_LIZ1200_1024 125 JHIRi41071_RGP2 127 JHIRI34798b_snp522 132 JHIRi37996_snp298***P07 134 Rub120a 136 Rub238h 143 Rub259b 149 RUB259f 157 P13M40-203 164 Severity 2011 Severity 2012 LG3 Fig.2 LinkagemapforLG3.Themostsignificantmarkersaccordingto theKruskal-Wallistestareshownby***foreachbinarytrait,togetherwith yearandsite.One-lodsupportinterval sfortheseveritytraitsarealsoshown Table4 Predictedmeans(se)atbothlocifortheseverityscores usingatwo-markermodel YearLG1=a-; LG3=a- LG1=a-; LG3=b- LG1=b-; LG3=a- LG1=b-; LG3=b- 20111.23(0.128)0.53(0.102)2.00(0.125)1.29(0.113) 20121.16(0.158)0.45(0.128)2.00(0.145)1.29(0.138) LG1isrepresentedbyRUB256eandLG3byERubLR_SQ05.3_D11AOC Table5 Predictedmeans(se)forthecrumblyscoresfromthe generalisedlinearmodelcombiningdataoveryearsandsites (a)Site.LG1interaction SiteLG1=a-LG1=b- Field0.17(0.013)0.51(0.020) Poly0.05(0.009)0.10(0.013) (b)Site.LG3interaction SiteLG3=a-LG3=b- Field0.35(0.018)0.30(0.015) Poly0.13(0.016)0.04(0.008) LG1isrepresentedbyRUB256eandLG3byERubLR_SQ05.3_D11AOC Graham etal.SpringerPlus (2015) 4:223 Page6of9 MP2atbothfieldsites(A&B)andinthepolytunnel.In 2010,asinglereplicateofMP1andMP2wasassessedat onefieldsite(A)andinthepolytunnel,butthescoring wasmodifiedtodistinguishbetweencrumblyandamore severeformwherenodrupedevelopmentoccurred,which wereferredtoas sterilefruit .Thiswasexaminedtoseeif itshowedanyassociationwithsterilityasdescribedprevi- ously(Jennings1967b)orwasanextensionofthecrumbly fruitphenotype.In2011and2012asinglereplicateof MP1andMP2wasscoredatthefieldsite(A)only;for theseyearscrumblinesswasscoredasbothcrumblyor notandona0 4scalewhere0wasnocrumblyfruitand 4wasthesevere sterile condition. RaspberryBushyDwarfVirus(RBDV)testingwas carriedoutasstandardtoensureplantswerefreeofthe virus(http://www.fruithealth.co.uk). Linkagemapping,summarystatisticsandQTLanalysis Previousversionsofthelinkagemapforthispopulation havebeendescribedbyGrahametal.(2004,2006,2009, 2011,2014),McCallumetal.(2010)andWoodhead etal.(2010,2013).Furthermarkershavebeenaddedto themapusedhere,usingJoinMap4.1(VanOoijen 2006)Table7. Asthecrumblyscoresarebinaryorordinaltraits, associationsbetweenthem werecalculatedusingthe gammastatistic(SiegelandCastellan1988),whichvar- iesbetween 1and+1. TheQTLmappinganalyseswerechosentobesuitable forbinaryandordinaltraits.Anon-parametricmapping basedontheKruskal-Wallis(KW)testwasusedinitially totesteachmarkeronthemapforassociationswiththe crumblyfruitscoresforeachyearandenvironment, usingtheMapQTL5software(VanOoijen2004).The KWteststatistichasanapproximatechi-squaredistri- butionwithdegreesoffreedomequaltothenumberof genotypeclassesminusoneunderthehypothesisofno segregatingQTL.ThethresholdfortheKruskal-Wallis mappingacrossthegenomewasestablishedusinga smallpermutationtestof400permutations(carriedout inGenStat16forWindows(Payneetal.2013)).Fora normallydistributedtrait,thesubsequentanalysiswould betocombineinformationacrossgeneticmarkersalong thechromosometoestimatetheprobabilitiesofeach possibleQTLgenotypeforeachoffspringateachpos- ition(the geneticpredictors )andtomodeleachtrait asafunctionoftheseusingastandardlinearmodel.Ina crosssuchasthiswithoutbreedingparents,theparental genotypesataQTLareusuallyrepresentedasabxcd, withoffspringgenotypesac,ad,bcandbdandtheprob- abilitiesforthesegenotypescanbeusedinthelinear model.Alternativelygeneticpredictorsforthematernal additiveeffect( P 1 ) ,thepaternaladditiveeffect( P 2 )and thedominanceeffect( D )canbederivedforeachoffspring ateachpositionas: P 1 ¼ prbc ðÞþ prbd ðÞ prac ðÞ prad ðÞ P 2 ¼ prbd ðÞþ prad ðÞ prbc ðÞ prac ðÞ D ¼ prbd ðÞ prbc ðÞ prad ðÞþ prac ðÞ ð 1 Þ where pr(ac) istheprobabilitythattheoffspringhas genotypeacatthatposition,andthesecanbeusedin thelinearmodel.Forthebinarytraitshere,thelinear modelwasreplacedbyageneralisedlinearmodel (GLM)withbinomialerrorsandalogitlinkfunction torelatethecrumblytraittotheadditiveanddominance effectsinthegeneticregionsidentifiedbytheKWana- lysis.Thegeneticpredictorsoftheadditiveeffectsofeach parentandthedominanceeffectwerecalculatedateach markerpositionusingtheQIBDPROBABILITIESproced- ureofGenStat16forWindows(Payneetal. 2013 )and Table6 Correlationbetweencrumblyscoresandthetimeto eachoftheripeningstagesfrom2006 YearEnvOpenFruitsetGreenGreen/RedRipe 2004FieldA 0.180.190.170.140.08 2004FieldB 0.030.110.11 0.090.11 2007FieldA 0.050.26***0.29***0.140.08 2007Poly 0.150.16*0.100.21**0.07 2008FieldA 0.09 0.02 0.11 0.120.00 2008Poly 0.060.04 0.010.040.04 2009FieldA 0.050.30***0.32***0.17*0.07 2009Poly0.010.22**0.140.00 0.17* 2010 a FieldA 0.010.31***0.27***0.17*0.10 2010 a Poly 0.050.30***0.27***0.18* 0.06 2011FieldA 0.070.18*0.17*0.140.02 2012FieldA 0.130.21*0.22**0.23**0.08 Severity(0 4) 2011FieldA 0.030.29***0.29***0.20*0.01 2012FieldA 0.020.29***0.29***0.24**0.01 a 2010showstheproportionofeithercrumblyorsterilefruit *** p 0.001**p0.01;*p0.05 Table7 PrimerstoconfirmlocationofgenesinRub256eregion GeneinRub256eregionPrimersequence Ara41256eGgcaagtttacccagctgaa catatgagtgcgcagatacag Ara42256eCattccctgcgttgaaatct Ttctgagtcgtctggtgtgc Nudix256eGaaggttttcggtaccacca tcctgcttctggatgtcaaa Signalo256eTgcatcctggatatggattt ccaagttgcccatgagaataa Graham etal.SpringerPlus (2015) 4:223 Page7of9 thisprogramwasalsousedtofittheGLM.FinallyaGLM analysiswascarriedoutovertheyearsandsitestogether, fittingyear,site,thegeneticeffectsandallinteractions. Linearmodelswithnormallydistributederrorswasused forthefieldseverityscoresfrom2011and2012. IdentificationofgenecontentinRub256eregionand mappingtoconfirmlocation Littleinformationintermsoffunctionalmarkerswas availablefortheRub256elinkagemapregiononlinkage group(LG)1,thereforetheraspberrygenomepseudo- molecules(providedbyJoshuaUdallBYU,Geneticsand BiotechFaculty(pws.byu.edu))weresearchedusing BLAST(Altschuletal.1990)foranyregionsthatmatched theRUB256esequence.Primersweredesignedtosome genesintheregion(Table6)andaddedtothelinkage mapaspreviouslydescribed(Grahametal.2011)to confirmlocationofsequence. Associationwithripening Grahametal.(2009)studiedthedevelopmentoffruitin thesamepopulationandidentifiedQTLassociatedwith theripeningprocess.Heretheassociationbetweenrip- eningratesandcrumblinesswasinvestigatedusingthe ripeningfieldscoresfrom2006,asthesewereavailable forbothMP1andMP2.Theripeningdatawasoriginally scoredasdevelopmentalstagesusinga1 7scale(1= budbreak,2=openflowers,3=fruitset,4=greenfruit, 5=green/redfruit,6=ripeand7=over-ripe),withthe firstscoringon19thMay2006whenallscoreswere equaltoone.FromthesescoresGrahametal.(2009) estimatedthenumberofdaystoreacheachofthede- velopmentalstagesforeachgenotypeandmappedQTL forthese.Theassociationwithcrumblinesswasinvesti- gatedhereusingcorrelationcoefficients.Thecrumbly traitswerealsomodelledasafunctionofboththe geneticmarkersandthetimetoripening,usingagen- eralisedlinearmodelwithb inomialerrorsandalogit linkfunction. Additionalfile Additionalfile1:TableS1. Analysisofdeviancetablefortheincidence ofcrumblyfruit,modelledasafunctionofsite,year,geneticeffects andtheirinteractions.LG1isrepresentedbyRub256eandLG3by ERubLR_SQ05.3_D11AOC,usingtheadditiveeffectoftheLatham alleleineachcase. Abbreviations QTL: Quantitativetraitloci;MP:Mappingpopulation;LG:Linkagegroup; GLM:Generalisedlinearmodel. Competinginterest Theauthorsdeclaretheyhavenocompetinginterests. Authors contributions JGsupervisedthestudy,JG&CHdraftedthemanuscript,CHandJMcN analysedthedata,SM,KS,DC,PHparticipatedindatacollection.KScarried outthemarkeranalysis.LMprovidedthegenomicsequenceforRub256e.All authorsreadandapprovedthefinalmanuscript. Acknowledgements TheauthorswouldliketoacknowledgeRESASforfundingsupport. Authordetails 1 JamesHuttonInstitute,Dundee,ScotlandDD25DA,UK. 2 Biomathematics andStatisticsScotland,Dundee,ScotlandDD25DA,UK. Received:10February2015Accepted:30April2015 References AltschulSF,GishW,MillerW,MyersEW,LipmanDJ(1990)Basiclocalalignment searchtool.JMolBiol215:403 410 AnthonyVM,WilliamsonB,JenningsDL,ShattockRC(1986)Inheritanceof resistancetoyellowrust( Phragmidiumrubi-idaei )inredraspberry.AnnAppl Biol109:365 374 DaubenyHA,CrandallPC,EatonGW(1967)Crumblinessintheredraspberry withspecialreferencetothe Sumner variety.ProcAmSocHorticultSci 9:224 230 DaubenyHA,FreemanJA,Stace-SmithR(1978)Theoccurrenceandsomeeffects ofraspberrybushydwarfvirusinredraspberry.JAmSocHorticultSci 103:519 522 GoetzM,HooperLC,JohnsonSD,CarlyleJ,RodriguesM,Vivian-SmithA,Koltunow A(2007)ExpressionofaberrantformsofAuxinResponseFactor8stimulates parthenocarpyinArabidopsisandTomato.PlantPhys145:351 366 GrahamJ,SmithK,MacKenzieK,JorgensonL,HackettCA,PowellW(2004)The constructionofageneticlinkagemapofredraspberry( Rubusidaeus subsp. idaeus )basedonAFLPs,genomic-SSRandEST-SSRmarkers.TheorAppl Genet109:740 749 GrahamJ,SmithK,TierneyI,MacKenzieK,HackettCA(2006)MappinggeneH controllingcanepubescenceinraspberryanditsassociationwithresistance tocane botrytis andspurblight,rustandcanespot.TheorApplGenet 112:818 831 GrahamJ,HackettC,SmithK,WoodheadM,HeinI,McCallumS(2009)Mapping QTLfordevelopmentaltraitsinraspberryfrombudbreaktoripefruit.Theor ApplGenet118:1143 1155 GrahamJ,HackettCA,SmithK,WoodheadM,MacKenzieK,TierneyI,CookeD, BayerM(2011)Towardsanunderstandingofthenatureofresistanceto Phytophthora rootrotinredraspberry:isitmainlyrootvigour?TheorAppl Genet123:585 601 GrahamJ,HackettCA,SmithK,KarleyA,MitchellC,RobertsH,O NeillT(2014) Geneticandenvironmentalregulationofplantarchitecturaltraitsand opportunitiesforpestanddiseasecontrolinraspberry.AnnApplBiol. doi:10.1111/aab.12134 JenningsDL(1962)Someevidenceontheinfluenceofthemorphologyof raspberrycanesupontheirabilitytobeattackedbycertainfungi.HortRes 1:100 111 JenningsDL(1967a)Balancedlethalsandpolymorphismin Rubusidaeus . Heredity22:465 479 JenningsDL(1967b)Observationsonsomeinstancesofpartialsterilityinred raspberrycultivars.HortRes7:116 122 JenningsDL(1971)Somegeneticfactorsaffectingfruitdevelopmentin raspberries.NewPhytol70:361 370 JenningsDL(1988)RaspberriesandBlackberries:TheirBreedingDiseasesand Growth.Academic,London,p145 JenningsDL,BrydonE(1989)Furtherstudiesonbreedingforresistanceto Leptosphaeriaconiothyrium inredraspberryandrelatedspecies.AnnAppl Biol115:499 506 JenningsDL,McGregorGR(1988)Resistancetocanespot( Elsinoeveneta )inred raspberryanditsrelationshiptoresistancetoyellowrust( Phragmidium rubi-idaei ).Euphytica37:173 180 KassimA,PoetteJ,PatersonA,ZaitD,McCallumS,WoodheadM,SmithK, HackettCA,GrahamJ(2009)Environmentalandseasonalinfluencesonred raspberryanthocyaninantioxidantcontentsandidentificationofquantitative traitsloci(QTL).MolecularNutriFoodRes53:625 634 Graham etal.SpringerPlus (2015) 4:223 Page8of9 KeepE(1968)Inheritanceofresistancetopowderymildew Sphaerotheca macularis (Fr.)Jaczewskiintheredraspberry Rubusidaeus L.Euphytica 17:417 438 KeepE(1976)Progressin Rubus breedingatEastMalling.ActaHort60:123 128 KnightRL,KeepE(1958)DevelopmentsinsoftfruitbreedingatEastMalling. ReptEastMallingResStn1957:62 67 McCallumS,WoodheadM,HackettCA,KassimA,PatersonA,GrahamJ(2010) Geneticandenvironmentaleffectsinfluencingfruitcolour.TheorApplGenet 121:611 627 MurantAF,ChambersJ,JonesAT(1974)Spreadofraspberrybushydwarfvirus bypollination,itsassociationwithcrumblyfruit,andproblemsofcontrol. AnnApplBiol77:271 281 PandolfiniT,MolensiniB,SpenaA(2007)Moleculardissectionoftheroleof auxininfruitinitiation.TrendsPlSci12:327 329 PatersonA,KassimA,McCallumS,WoodheadM,SmithK,ZaitD,GrahamJ (2013)Environmentalandseasonalinfluencesonredraspberryflavour volatilesandidentificationofquantitativetraitloci(QTL)andcandidate genes.TheorApplGenet126:33 48 PayneR,MurrayD,HardingS,BairdD,SoutarD(2013)IntroductiontoGenStat forWindows,16thedn.VSNInternational,HemelHempstead SchwabeWM,MillsJJ(1981)Hormonesandparthenocarpicfruitset.Aliterature survey(temperate,subtropicalandtropicalfruisandvegetables).Hort Abstracts51:661 698 SiegelS,CastellanNJ(1988)Nonparametricstatisticsforthebehaviouralsciences, secondthedn.McGraw-Hill,NewYork VanOoijenJW(2004)MapQTL®5,softwareforthemappingofquantitativetrait lociinexperimentalpopulations.KyazmaBV,Wageningen VanOoijenJW(2006)JoinMap®4;Softwareforthecalculationofgeneticlinkage mapsinexperimentalpopulations.KyazmaB.V,Wageningen VriezenWH,FeronR,MarettoF,KeijmanJ,MarianiC(2008)Changesintomato ovarytranscriptomedemonstratecomplexhormonalregulationoffruitset. NewPhytol177:60 76 WoodheadM,WeirA,SmithK,McCallumS,MacKenzieK,GrahamJ(2010) Functionalmarkersforredraspberry.JAmSocHortSci135:418 427 WoodheadM,WeirA,SmithK,McCallumS,JenningsN,HackettCA,GrahamJ (2013)IdentificationofQTLsforcanesplittinginredraspberry( Rubusidaeus ). MolBreed31:111 122 Submit your manuscript to a journal and bene t from: 7 Convenient online submission 7 Rigorous peer review 7 Immediate publication on acceptance 7 Open access: articles freely available online 7 High visibility within the eld 7 Retaining the copyright to your article Submit your next manuscript at 7 springeropen.com Graham etal.SpringerPlus (2015) 4:223 Page9of9 RESEARCHOpenAccess Towardsanunderstandingofthecontrolof crumbly fruitinredraspberry J.Graham 1* ,K.Smith 1 ,S.McCallum 1 ,P.E.Hedley 1 ,D.W.Cullen 1 ,A.Dolan 1 ,L.Milne 1 ,J.W.McNicol 2 andC.A.Hackett 2 Abstract Thegeneticdisorderknownas crumbly fruitisbecomingaseriousproblemintheEuropeanraspberryindustry. Thestudysetouttoexaminethecrumblyphenotypeinaredraspberrymappingpopulationundertwoenvironments (fieldandpolytunnel)acrosssixseasonsinanefforttounderstandvariabilityofthesyndromeandtoexaminewhether geneticfactorswereimportantandifso,whetherQTLa ssociatedwiththephenotypec highlightedthatseasonal,environmental(fieldorpolytun nel)andgeneticfactorsallinfluencethecondition.Two QTLthatareimportantforthegeneticcontroloftheconditionhavebeenlocatedonlinkagegroupsoneand three,andanassociationwithripeningtimehasbeenidentified. Keywords: Raspberry;Crumblyfruit;QTLmapping Background Raspberryfruitsareformedfromanaggregationofmul- tiplefertilizedovarieseachofwhicharereferredtoas drupeletsastheybecomefleshy.Intheconditionknown ascrumblyfruit,whichhasbeenlinkedwithpollen abortionandembryosacdegeneration,drupeletsare generallyreducedinnumberbutgreatlyenlargedor,in thecaseofsmallreductions,cohereimperfectlysofruit readilycrumbleswhenpicked(Daubenyetal.1967; Jennings1988). Crumblyfruitisanindicationofapartialfailurein oneormorephysiologicalprocessesconcernedwith fruitdevelopment(Jennings1967b)andisanincreasing problemfortheEuropeanraspberryindustry,withpar- ticularproblemsoccurringinwidelygrowncommercial cultivarsTulameenandGlenAmple.Therehavebeena Itisknownthatinfectionwithcertainvirusescanincrease thelikelihoodthatplantsbecomecrumbly(Jennings 1988).RaspberryBushyDwarfVirus(RBDV)infects pollen,reducingitscapacitytoinducefruit-setandcan leadtofailureofalmosthalfofalldrupeletstoset(Murant etal.1974;Daubenyetal.1978).Ageneticcausehasbeen demonstratedwherethecrumblyphenotypearisesfrom virus-testedmotherplants(Jennings1988).Studieshave alsoshownthatextensivetissueculturingofplantsmay increasetheemergenceofthecondition(N.Jenningspers comm.).Additionally,environmentalfactorssuchaslow orhightemperaturesatparticulartimepointsindevelop- mentappeartoplayanimportantrolewithvariations intheextentofcrumblinessapparentfromyeartoyear (A.Dolanperscomm). Assessmentofthefruitofmotherplantsiscurrently theonlymethodofdetectingcrumblinessinnuclear stockmaterial,andithasbeenobservedthatasmall numberofplantswithvaryingdegreesofcrumbliness canbedetectedeachyear.Somecultivarsappeartobe morepronetotheconditionthanothers.Howeverifthe environmentalconditionsdifferfromthenormalseasonal levels,ithasbeenobservedthatrandomsymptomsof crumblinesscanbedisplayedincultivarsnotpreviously knownfortheproblem.Also,knowncrumblyaffected cultivarscanshowmoreextremesymptoms. Thismaterialisnotreleasedtoindustrybutmayre- sult,unneccessarily,inacultivarpermanentlylosingits positioninthemarketplace. Thegeneticbasisofraspberryfruitdevelopmentisnot wellunderstood,althoughsomestudieshavebeencar- riedouttolookatoverallcontroloffruitdevelopment andripening(Grahametal.2009)andalsospecificrip- eningrelatedprocessessuchasanthocyaninproduction (Kassimetal.2009)colourdevelopment(McCallum etal.2010)andvolatileproduction(Patersonetal.2013). *Correspondence: Julie.Graham@hutton.ac.uk 1 JamesHuttonInstitute,Dundee,ScotlandDD25DA,UK Fulllistofauthorinformationisavailableattheendofthearticle a SpringerOpen Journal ©2015Grahametal.;licenseeSpringer.ThisisanOpenAccessarticledistributedunderthetermsoftheCreativeCommons AttributionLicense(http://creativecommons.org/licenses/by/4.0),whichpermitsunrestricteduse,distribution,andreproduction inanymedium,providedtheoriginalworkisproperlycredited. Graham etal.SpringerPlus (2015) 4:223 DOI10.1186/s40064-015-1010-y