Physiological breeding

Physiological

breeding

Matthew

Reynolds

and

Peter

Langridge

Physiologicalbreedingcrossesparentswithdifferentcomplex

butcomplementarytraitstoachievecumulativegeneactionfor

yield,whileselectingprogenyusingremotesensing,possiblyin

combinationwithgenomicselection.Physiologicalapproaches

havealreadydemonstratedsignificantgeneticgainsinAustralia

andseveraldevelopingcountriesoftheInternationalWheat

ImprovementNetwork.Thetechniquesinvolved(seeGraphical

Abstract)alsoprovideplatformsforresearchandrefinementof

breedingmethodologies.Recentexamplesoftheseinclude

screeninggeneticresourcesfornovelexpressionofCalvincycle

enzymes,identificationofcommongeneticbasesforheatand

droughtadaptation,andgeneticdissectionoftrade-offsamong

yieldcomponents.Suchinformation,combinedwithresultsfrom

physiologicalcrossesdesignedtotestnoveltraitcombinations,

leadtomoreprecisebreedingstrategies,andfeedmodelsof

genotype-by-environmentinteractiontohelpbuildnewplant

typesandexperimentalenvironmentsforfutureclimates.

Addresses

1_{GlobalWheatProgram,InternationalMaizeandWheatImprovement} Center(CIMMYT),Mexico,D.F.,Mexico

2_{SchoolofAgriculture,FoodandWine,UniversityofAdelaide,Adelaide,} Australia

Correspondingauthor:Reynolds,Matthew([email protected])

Current Opinion in Plant Biology 2016,31:162–171 ThisreviewcomesfromathemedissueonPhysiology and metabolism

EditedbyRobert T Furbank andRowan F Sage ForacompleteoverviewseetheIssue andtheEditorial Availableonline7thMay2016

http://dx.doi.org/10.1016/j.pbi.2016.04.005

1369-5266/#2016TheAuthors.PublishedbyElsevierLtd.Thisisan openaccessarticleundertheCCBY-NC-NDlicense( http://creative-commons.org/licenses/by-nc-nd/4.0/).

Introduction

Global cereal demand is predicted to outstrip genetic gains by 2050 [2], while climate change threatens to reduce their impact [3]. To accelerate yield improve-ment, physiological traits at all levels of integration

(Figure 1) need to be considered in breeding [4–6].

Annual genetic yield gains in cereals are currently in theregionof0.5–1%[7],duealmostentirelyto conven-tionalapproaches. Thesehavecomeaboutasaresultof twomainfactors:unspecifiedrecombination of genesof minoreffectamongelitegermplasm,andtheintroduction of new genetic diversity often associated with disease resistanceandgrainquality[1,8].Physiological breeding complementsthisapproachbyaddingtwomainelements: knowledge of well-characterized genetic resources to design crossing strategies, and the ability to enrich for favorableallelesthroughphenomic(andgenomic) screen-ingofprogeny.Thisincreasestheprobabilityofachieving cumulative gene action for yield compared to crossing physiologicallyuncharacterizedlines.Inpracticeitdiffers fromconventionalbreedingbyconsideringalargerrange of traits—including genetically complex physiological characteristics [9–12]—and differs from molecular breedingbyencompassing bothphenomicand genomic information.The keystepsarepresentedbelow: Designingaplantwiththeoreticallyimproved

adapta-tion;

Identifyinggeneticresourceslikelytoencompassnew and/orcomplementaryallelicvariation(for crossing); Developingandimplementing phenotyping protocols

andexperimentaltreatmentstomaximizeresolutionof physiologicaltraitexpression(toselectparents); Geneticdissectionoftraits,anddevelopmentof

gene-basedselectionapproaches;

Strategic hybridization to achieve cumulative gene action for yield, combined with application of high throughput phenotyping and genotyping to select progeny;

Analysis of the trait/allele combinations that achieve environmentallyrobustgenetic gains based on multi-locationtrialdata(to designnewcrosses);

Informaticsservicesunderpinning theiterative refine-mentofbreeding strategiesacross allsteps.

Sinceacomprehensive geneticbasisexplainingcultivar leveldifferencesinperformancedoesnotyetexistforany crop, physiological breeding currently relies heavily on phenomics.However,itcanalsomakeuseofmarkersfor alleles associated with genes of major effect—such as Ppd, Vrn, and Rht in wheat [8]—and will increasingly make use of both genomic selection (GS) [13,14] and markerassistedselection(MAS)associatedwithgenesof minoreffect[15].Inthatsense,phenomicsandgenomics gohandinhandin thephysiological breedingapproach. Otherexamplesinclude: selectingparents with comple-mentary traitsand alleles[8];progenyselection usinga

§_{ThispaperispartofaVirtualSpecialIssuebasedontheCurrent}

OpinionConference‘AgricultureandClimateChange–adaptingcrops toincreaseduncertainty’,chairedbyDavidEdwardsandGilesOldroyd in2015.

combinationofremotesensingofintegrativetraits,MAS [16],andGS[14];andgeneticdissectiontodemonstrate successfulcombinationsoftraitsandthetrade-offsamong them,forwhichQTLanalysisaddsconsiderableweight [16].Themainobjectiveofthisreviewistopresentthe rationale for the main activities of the physiological breedingpipeline,asoutlinedintheGraphicalAbstract.

Crop

design

Designing improved plant types is a unique aspect of physiological breeding,inthesensethatthrough imple-menting new strategic crosses, noveltrait combinations canberigorouslytestedacrossarangeof target environ-mentsintermsoftheirimpactonyield.Withthe excep-tionofyieldperse,selectionforgeneticallysimpletraits has traditionally dominated plant breeding. However, developmentsinphenomicsandgenomicsareincreasing accesstomorecomplextraits[17,18,19,20],resultingin renewed interest in designing improved plant types

[9,12,18,21–23,24,25,26], including under climate

change[4,10,27].Designingnewplanttypesisnottrivial since genotype–environment–crop management interac-tions (GEM)are unpredictable,being determined largely by future weather (E), farmer choices (M), and specificlocation(EM).Anotherbarriertoeffectivecrop designissignificantgapsinknowledgeofthephysiological

andgeneticbasesofadaptation(G),furtherexacerbatedby interactionswithE[28].

Thesehurdlesledtoabeliefthatstochasticapproaches, supportedbymoderntools,wouldachievebreakthroughs more readily; for example‘-omics’tools combinedwith bioinformatics to cut through thebiological complexity anddeliver empiricalsolutions[29].Ontheotherhand, detailed studiesincontrolledconditionswithrapid life-cycle model species (of relatively small genome size) wereexpected toextrapolateto crop speciesand envir-onments[30].Unfortunately,neitheroftheseapproaches hasyetrevolutionizedpracticalplantbreeding[15].This is partly because they are still confounded by the GEM paradigm (e.g. [31]), in addition to being retrospective, in the sense that they focus on extant germplasmratherthanextrapolatingtotherequirements ofimprovedgenotypes.Forexample,genomicselection modelsarepredictivewithinapre-determinedgenepool, buttheybreakdownwhennewgeneticdiversityis intro-ducedintothebreedinggenepool[32].

However, physiological and genetic dissection of com-plex traits [11–13,15,16,19,20,33,34,35,36,37–43], and the development of high throughput phenotyping approaches translated to the field environment [17,44,45], have provided valuable insights for the Figure1 SELECTION APPROACH Regulatory factors • Hormones • Transcription factors Genes

TRAIT GENETIC COMPLEXITY Biochemical

markers

Metabolite

Simple traits few major genes

Complex traits multiple gene Marker assisted selection Genomic selection

Physiological characterization/remote sensing

Epig

enetics

PR

ODUCTION (Yield,

Biomass)

Current Opinion in Plant Biology

Selectionapproachesfortraitsatdifferentlevelsofintegrationandgeneticcomplexity.Thefigurerepresentsplantselectionapproachesthatmay beusedatdifferentinterventionpointsacrossthespectrumoftraits,startingwithsimplemetabolitesandculminatinginpolygenicproductivity traitssuchasyieldandbiomass.Regulatoryfactorsthatmayinteractwiththeexpressionoftraitsatanylevelareindicated.Environmentally mediatedepigeneticfactorsmaydirectlyinfluencegeneexpressionandthereforeincreaseexpressionofgenotypebyenvironmentinteraction.

designofnewplanttypes;forexamplebyestablishingthe genetic bases for trait synergies and tradeoffs (e.g.

[16,35,36]), and thereby improved crossing strategies

[11–13,16]. Analysis of the performance of new lines

developed by physiological breeding helps inform crop design by demonstrating which combinations of traits/ alleles improve yield and in which environments [16,40].Usingsuchoutputs,andwiththehelpof simula-tionmodels,hypothesesregardingthevalueofnewlevels oftraitexpressionandtraitcombinationscanalsobetested theoreticallyacrossarangeofenvironments[24,28,33]. The designand testing ofimprovedplant typescanbe founded on two main hybridization tactics: either by synergisticre-combinationoftraits/allelesalreadypresent inextantgenepools,orviatheintroductionofnewlevels oftraitexpression/allelesfromexoticsources[46,47].One ofthefunctionsofcropdesignistoestimatelikelycost– benefitsof usingdifferentclasses ofgeneticresources.

Genetic

resources

Physiological profiling among genetic resources can broaden the crop genepool in a highly targeted way. However, most breeders show a pragmatic skepticism towardscrossingwithexotic germplasmandprecedents generally relateto imperativessuch as avoidingdisease epidemics (e.g. [47]). Three main classes of genetic resources can be utilized: crop wild relatives, isolated genepools of the same genome (e.g. landraces), and modernbreedinglines.

Interspecifichybridizationisthemostdifficulttoachieve asitnormallyresultsinsterility.However,ithasresulted insomeimpressiveyieldgains[47]and—asacenturyold technology accelerated by marker technology—is rela-tivelyuncontroversialintermsofmovinggenesbetween relatedspecies[48].Nonetheless,lessthan10%ofthewild relativescollectedhavebeenusedininter-specific cross-ing,andfewerstillhavebeensurveyedforgeneticdiversity oftraits withpotentialtoboostyield oradaptation.One recentexceptionwasastudyofdiversityinCalvincycle enzymes and Rubisco in the Triticeae. Rubisco from speciesrelatedtowheatshowedpromisingcatalytic prop-ertiesandmodelingofphotosynthesisat258Cand358C demonstratedthepotentialbenefitofreplacingRubiscoof T.aestivumwithRubiscofromAe.cylindricaorH.vulgare,in termsofhigherassimilationrates[49].

Polyploidcrops like wheatcanwithstand the introgres-sionofalienchromatindueto thebufferingpresenceof homoeologous genomes. The D genome of hexaploid wheat(ABD)existsinabundanceasawildgrass(Triticum tauschii)andwillcrosswithdurumwheat(AB)togenerate asynthetichexaploidwithrelativelylittlelinkagedragof detrimental alleles [50]. While typically difficult to thresh,suchprimarysyntheticscanexpressyieldsequal tomoderncultivarsandwithsubstantiallyhigherbiomass

(Figure2).Oneormorebackcrossestocultivatedwheat

canresultinlineswithsignificantlyhigheryield, includ-ingunder heatanddroughtstress[37],althoughprecise geneticbasesstillneedelaboration.ScreeningofbothAB and D genomes for stress adaptive traits presents the opportunityto combineboth sources into asingle syn-thetichexaploid genomewith uniquealleles not repre-sentedintheconventional genepool.

Compared to crop wild relatives, landraces are much easierto cross with whilestill representingnovel pools of allelic diversity; most countries have extensive and overlapping collections (e.g. [51]). Techniques such as theFocused GermplasmIdentificationStrategy (FIGS) helpidentifyaccessionsoriginatinginconditionsof rele-vancetobreedingtargets(http://www.figs.icarda.net/).A good illustration of their value came from recent field phenotypingof FIGSwheatpanelsselectedunder heat and drought stress, revealing dozens of lines with final biomass significantly larger than adapted checks [52] (finalbiomassbeinganindicatorofagronomicpotential, especiallyunderstress).

However,themostaccessiblesourceofgeneticvariation— in terms of its use in strategic crossing—is that within currentbreedingmaterial.Interestingly,detailed physiolog-ical [53] and genetic dissection [20,38,39] is not yet a routine procedure for selecting parents among advanced breedinglines,butdevelopmentsinfieldphenotyping— in combination with high throughput genotyping—will identifymorecandidatelinesforuseintrait-basedcrossing.

Phenotyping

Recentadvances in high throughput field phenotyping have boosted the power of physiological breeding Figure 2 y = 0.21x + 28 R² = 0.16 100 200 300 400 500 600 700 2300 2100 1900 1700 1500 1300 1100 900 700

Above ground biomass (g m–2₎

Y

ield (g m

–2

)

Current Opinion in Plant Biology

Yieldandtotalabovegroundbiomass(AGB)ofapanelof250primary synthetichexaploidwheatlines(greendots)comparedtothebest elitecheck(Tacupetobluedot)underyieldpotentialconditions(as describedin[82]),Obregon,2014;LSDforyieldandAGB=135and 390gm 2_{,respectively.}

[17,45].Almostbydefinition,highthroughputimplies use of non-invasive approaches like proximal/remote sensing of spectral reflectance from plant tissue. The traits measured relateeither to thermal/hydration prop-ertiesofplanttissueassessedintheinfraredregionofthe electromagneticspectrum,orpigmentprofilesestimated in visible bands [45]. These spectral indices cover an important range of traits including plant temperature, waterrelations,photosynthesis,nutrientstatus,and agro-nomictraits.Waterindexinparticularhasbeenshownto bepredictiveofcultivarleveldifferencesinleafandsoil water potential, as well as yield and biomass in both stressed andfavorableenvironments [54].

Remotely sensedtraitsexpressgoodresolutionforlarge scale screening, and in some cases remote sensing is sufficientfordetailedgeneticorphysiologicaldissection, asrequiredformappingquantitativetraitloci(QTL),for example. Canopy temperature (CT) has been used to identify QTL for drought and heat stress tolerance

[35,40],andQTLcommontobothstresseswerelinked

to adaptive rootresponse[36].Dedicatedspectral sen-sors,suchastheGreenSeekerthatmeasuresthe normal-ized differencevegetationindex (NDVI), arealso used for simplegrowthanalysis[17,45].

Aerial imaging (either by manned or unmanned low-flying vehicles) is revolutionizing field phenotyping [44] by offering two main advantages that increase throughputandprecisioncomparedtoground-based sys-tems.Firstly,theabilitytoincludehundredsoffieldplots in asingleimageavoidsconfoundingeffectsof environ-mentaldriftassociatedwithlengthyplot-to-plot measure-ment.Secondly,imageanalysispermitsdatacurationby removing outlying pixels in each plot. Both of these factors improve associations between remote-sensed traitsand yieldincomparison to groundbased readings [55].Assensorandimageanalysistechnologiesdevelop, thescopeandprecisionofremotesensingisexpectedto increase. Work isunderway (atthe InternationalMaize and Wheat Improvement Center,CIMMYT,for exam-ple)to developalgorithmsassociatedwithspike charac-teristics—including size, density, and phenological stages—basedonacombinationofvisibleandinfrared bands. Elsewhere,indiceshavebeen reportedthatmay eventually substitute for gas exchange measurements (e.g.[56]),perhapsincombinationwithchlorophyll fluo-rescence [57]. However, there are still many important phenotypic traitsthatdonotlendthemselves toremote sensing, such as detailed growthanalysis where the re-quiredlevelofprecisionnecessitatesdestructiveharvests, or measurements of traits that are partially or fully ob-scured fromviewsuchasstemorrootcharacteristics. Attheotherendofthephenotypingspectrum(Figure1), somemetaboliteshavebeenassociatedwithperformance traits; for example, fructans with stress tolerance and

quality, and major loci controlling fructan biosynthesis, havebeenmapped[58].Itisnowpracticaltoinvestigate interactions between metabolome and phenotype at a large scale [59], and new loci controlling metabolites havebeenassociatedwithagronomictraits[60].Astudy indrought-stressedwheatdetectedQTLfor238 metab-olites with significant genetic associationsranging from oneQTLfor125metabolitestonineQTLjustformalate [60]. Consequently, the translation of metabolite infor-mationtopracticalbreedingremainselusive.An alterna-tive application has been the direct modification of metabolite levelsthroughgeneticengineering,focusing mainly on compounds associated with osmotic adjust-ment and amino acid metabolism including trehalose, proline,mannitol,and ornithine(reviewedin[61]). Whilephenotypingprotocolsexistforsimplethroughto highlyintegrativetraits(Figure1),thegenetic improve-ment of more integrative traits (canopy temperature, biomass, fruiting efficiency, harvest index, etc.) are among the‘lowesthangingfruits’in termsof increasing geneticgains,sincetheyshowconsiderablegenetic vari-ationwithinmoderncultivarsandhavenotbeen system-aticallyconsideredinconventionalbreeding.However,as these more integrativetraitsare optimized andfixed in elitelines(acceleratedbyhighthroughputphenotyping), subsequentgeneticgainswillcome fromunderstanding theircomponenttraits,physiologicalmechanisms, meta-bolic pathways, etc. (Figure 1), in combination with geneticanalysis.

Genetic

analysis

Geneticsandphysiologyareinextricableinthecontextof cropimprovement.Geneticdissectionofcomplextraits, boosted by the tools of modern biotechnology, permit models of improved plant functions to be rigorously tested [16,20,24]. Genome-wide association studies readily identify parents with contrasting inheritance of key traits [39] for genetic dissection using bi-parental crosses/nestedassociationmapping(aswellasfor design-ingnewcrosses).Whenphenotypinggeneticallycomplex traitsinexperimentalpopulations,itiscrucialto control genesofmajoreffecttoavoidmaskingdetectionofnovel

QTL[11,39,40].Theidealgeneticmarkerisadiagnostic

onethatallowsdirectidentificationofspecificalleles,as seenintheselectionforphenologyandplantheightgenes inwheat[8].DetailsonhowDNAsequenceinformation canbeusedtodevelopmolecularmarkersforscreeninga range of agronomictraits is provided elsewhere (http://

maswheat.ucdavis.edu/).

Nonetheless,evenincropswherethecompletegenome sequenceisavailable(reviewedin[62]),trait-based mar-kers are still not in mainstream use for complex trait selectioninmajorbreedingprograms[15].Thisgoesback to the GEM paradigm (or QTLEM), since mostQTLarenotrobustacrossenvironments[41].The

challengewill be to focus on markersclosely linked to functional genes (exome capture for example in barley [63]), and to define the environmental factors causing GxE.Inthiscontext,research intoepigeneticssuggests thatit may benecessary to consider the epi-genomeif GxEistobecomemorepredictable[64].

Marker technology is used stochastically in genomic selectionin animalbreeding and isbeing evaluated for crops [65]. Someresults are promising, for example, in wheat [34],and maize[66], while incorporating pheno-typicdatafromtrainingpopulationsintomodels signifi-cantlyimprovesprediction[13,14].

Crossing

and

selection

With amodest investmentin some phenotyping equip-ment,3anybreedingprogramcanimplementphysiological trait(PT)basedcrossingstrategies(seeGraphAbst).Based ontraitsidentifiedinconceptualmodels(e.g.[9,10,12])and theuseofsimplequantitativemodelstoestimate poten-tiallycomplementarytraits[11,52],systematicscreeningof genetic resources has been employed at CIMMYT to identify complementary parental sources for adaptive

traits.Forexample,parentswithcoolcanopiesassociated withmore extensiverootsystems are crossedwithlines expressing ability to store and remobilize stem water solublecarbohydratesunderheatanddroughtstress(see [52]andreferencestherein).Toimproveyieldpotential, parentswithgoodspikefertilitycharacteristics,for exam-ple, [26] arecrossed with sourcesof high radiation use efficiency[67].Progenyselectionisfacilitatedbyremote sensing.

Suchapproacheshaveachievedimpactsoverconventional breedingin waterstressedAustralian environments[12]. CIMMYTandcollaboratorshaveachievedconsistent ge-neticgainsinheatanddroughtstressedenvironments,as wellasraisingyieldpotential[11,41,42,52,68,69](Table1), resultinginuptakeofnewlinesandphenotypingmethods by national programs [68]. The best new PT lines expressedthehighestaverageyieldacrossallsites, indi-catingyieldstability(Table1).

Experimental PT crosses between judiciously selected parentshavealsodemonstratedsubstantialgeneticgains. Double haploid progeny of the wheat cross Bacanora/ Weebil—where both parents express high yield via contrastingexpressionofyieldcomponents—generated progenywithexceptionalyieldpotential[42](Table 1). Similarly, random inbred lines of the Seri/Babax wheat Table1

Summary of genetic progress achieved through physiological trait (PT) based breeding/crossing at CIMMYT for different target environmentsasindicatedbymulti-locationyieldtrials(inAsia,NorthAfrica,andMexico),conductedwithintheInternationalWheat ImprovementNetwork(IWIN)

Breedingtarget environment

Trialnameand#PT linesincluded

#International sites

Yearof harvest

Indicationofgeneticprogress

Waterdeficit 17thSAWYT PTadvancedalines

64 2010 BestPTlinehigheryielding(avg16%)thanlocalcheckc_at 50/64sites;bestPTlinesexpressedcoolercanopiesand largerbiomassthanchecks(http://apps.cimmyt.org/ wpgd/index.htm;[68]).

Yieldpotential 2ndWYCYT

35rapid-cycleb_PTlines

26 2014 BestPTlinehigheryielding(avg10%)thanbestcheckd_at 23/26sites;bestPTlinesexpressedlargerbiomassthan checksatallsiteswheremeasured[67].

Heatstress 4thSATYN

25rapid-cyclePTlines

24 2015 BestthreePTlineshigheryielding(avg8%)thanbest checkat23/24sites;bestPTlinesexpressedlarger biomassthanchecksatallsiteswheremeasured[69]. Yieldpotential Bacanora/Weebilexperimental

populationwith105double haploidPTlines

9 2007–2010 BestPTlinesexpressedhigheryieldthanbestparentinall 9environments,mostspectacularlyinS.Chilewhere34% ofPTlineswerehigheryielding,andthebestPTline showed22%higheryieldthanthebestparent(avg 2seasons)[16_,42].

Heat/drought Seri/Babaxexperimental populationwith 167recombinant-inbred PTlines

12&9 2002–2013 Consideringyieldaveragedacrossall12international sites,15PTlineswerehigheryieldingthanthebestparent byasmuchas13%;Considering9environmentsin Mexico,thebest3PTlineswerehigheryielding(avg24%) thanthebestparentatall9sites[40].

Abbreviations:SAWYT,semi-aridwheatyieldtrial;WYCYT,wheatyieldcollaborationyieldtrial;SATYN,stressadaptivetraityieldnursery;DH, doubledhaploid(lines);RIL,recombinantinbredline.

a

Advancedlineshaving<1%ofgenelociheterozygous.

b_{Rapidcyclelineshaving<10%ofgenelociheterozygous(i.e.productsofpre-breeding).} c_{Localcheck=bestadaptedlocalcultivar.}

d_{Bestcheck=bestperformingconventionalCIMMYTeliteline(wheresuperiortolocalcheck).}

3_{Infra-red thermometer for canopy temperature, Greenseeker for} NDVI,labfacilitytoestimateabovegrounddrybiomassandassimilate partitioningatkeygrowthstages[68].

cross—chosentocombineawidelyadaptedparent(Seri) withasourceofdroughtadaptivetraits(Babax)—expressed superior yield to either parent across multiple sites [40] (Table1).Resultssuchastheselendthemselvesto dissec-tionofgeneticgainstorefinePT-basedcrossingstrategies.

Evaluation

of

genetic

gains

Strategic crossing for PTs, in addition to incorporating genetic diversity for complex traits into the genepool,

provides an ‘acid test’ for validating hypotheses about trait interactions, genetic background effects, and the valueofplantideotypesacrossenvironments.For exam-ple, among the material developed for drought adapta-tion,expressionof thekeyphysiologicaltraitsidentified inparents—suchasdeepwaterextraction,cooler cano-pies, andstorageofwatersolublecarbohydrates—were also expressed in the best progeny [36,52]. Figure 3 shows an example of progeny from parents crossed for Figure3 349 275 807 707 246 257 618 703 0 100 200 300 400 500 600 700 800 900 (a) (b) BM-Drip BM-Grav YLD-Drip YLD-Grav New PT-SA PT+Land PT+Syn Drt Adap Check Yield / Biomass (g/m 2)

Gravity irrigation simulates post monsoon drought

Drip irrigation simulates Mediterranean drought

20.3 21.5 13.0 23.7 21.4 21.7 21.3 24.6 10.0 12.0 14.0 16.0 18.0 20.0 22.0 24.0 26.0 28.0 Res H2O 60-90 cm (Grav) (mm) CT-G (Grav) CT-V (Grav) CT (Drip) New PT-SA PT+Land PT+Syn Drt Adap Check

Canopy Temp (C) & Resid H

2

O (mm

)

CT= canopy temperature V=vegetative stage; G=grain-filling

Current Opinion in Plant Biology

NewPTlineSOKOLL/3/PASTOR//HXL7573/2*BAU/4/WBLL4//OAX93.24.35/WBLL1undertwodistinctdroughtenvironments:(a)yieldand biomassundergravityanddripirrigationsimulatingpost-monsoonstoredsoilmoisture,andMediterraneandroughtenvironments,respectively.(b) ImprovedwaterrelationsinnewPTlineshowingcanopytemperatureinseason,andresidualsoilmoistureatharvest.Traitsarecomparedwith thedroughtadaptedcheckVorobey(from[52]).LSDvalues(P<0.05)ofgenotypicdifferencesforyield,biomass,CT,andresidualsoilmoisture

goodexpression(underdrought)ofdeepwaterextraction inonecase(purplebar),andcoolercanopies(bluebar)in theother.Theprogeny(greenbar)showssuperioryield andbiomasstoeitherparentundertwodifferentdrought environments, suggesting that combining both traits resulted in cumulative gene action. Another PT line, resultingfrom a cross betweena genetic resource with good expression of stem water soluble carbohydrate (WSC), and a line adapted in terms of yield to the Mediterraneantypeofdrought(whereWSCisexpected tobeofmostvalue),showedsuperiorexpressionofboth traitscomparedtoeitherparentinthetargetenvironment [52].Itisimportanttokeepinmindthattheexpression and value of anyPT will be a function of thegrowing environment[33]as wellas itsgeneticbackground. Determiningthegenetic basesofsuccessfultrait com-binationshelpsvalidatecumulativegeneaction.Genetic dissectionofthewheatcrossBacanora/Weebildescribed above revealed two loci influencing grain yield on

chromosomes1Band7B,increasing grainnumberand grainweight, respectively[16]. Thesetwoyield com-ponentstypicallyshowanegativecorrelation,however, thelackofatrade-offbetweentheminsomeprogenyof thiscrossledtoextremelyhighyieldsintheselines[42], makingtheseloci goodtargets forMAS[16].Genetic dissection of the Seri/Babax wheat population [40] showedthat yield QTL underheatand droughtstress werecollocatedwithQTLforthefollowingphysiological traits:CT(threeQTL),NDVI(threeQTL),and chloro-phyll(one QTL), demonstrating thegenetic contribu-tionofeachtraittoyield(Figure4).Insummary,while phenotyping can indicatethe potential for cumulative geneactionincontrastinggenotypes,geneticdissection ofphysiologicaltraitscanindicatepotentiallyfavorable alleliccombinationsbetweengenotypesexpressing sim-ilarphenotypes.Furthermore,genetic dissectionisthe onlydefinitivewaytoshow,forexample,thattwo appar-entlydifferenttraitsmayshareacommongeneticbasis, andviceversa,orthattwoapparentlyvaluabletraitsmay be mutually exclusive when combined in a common background.

Conclusions

An increasingly challenging crop environment and the rapidadvancesingenetictechnologiesbothcallforbetter understandingofthephysiologicalprocessesinvolvedin achieving crop productivity, and their interaction with environment. Three new factors canhelp achieve this, while at the same time contributing directly to crop improvement: first,new models of improvedplant pro-cessesandcropideotypescapitalizingonmorethanhalfa centuryofphysiological research;second,high through-put phenotyping technologiesthatpermit evaluation of complextraitexpressiononabreeding scalein realistic field environments; third, renewedfocus on preserving andutilizingplantgeneticresourcesandagrowing aware-nessthatclimatechangewillmakeitincreasinglydifficult toachieveneededgeneticgainsunlessnewallelic diver-sity is brought into existing genepools. Along with hy-pothesisdrivenphysiologicalbreedingandmulti-location testing,thesefactorscontributetobettergenetic under-standing,which itselfdrivesthedesign andselection of improvedcultivars. Physiological breeding as described hereinisacentralpillarofthenewlyformedInternational Wheat Yield Partnership (http://iwyp.org/) that aims to raise wheatyield potentialcloser to itsbiological limit, andwillbeimportantinsimilarinitiativesoftheCGIAR (e.g.theHeatandDroughtWheatImprovement Consor-tium) that aim to adapt crops to climate change and underpintheneed forglobal foodsecurity.

Acknowledgements

AuthorsacknowledgetheGrainsResearch&DevelopmentCorporation (GRDC),U.S.AgencyforInternationalDevelopment(USAID),German FederalMinistryforEconomyCooperationandDevelopment(BMZ), SecretariatofAgriculture,Livestock,RuralDevelopment,Fisheriesand Food(SAGARPA),andtheInternationalWheatYieldPartnership(IWYP) Figure4 YIELD

CT

Current Opinion in Plant Biology

Geneticdissectionofyieldbyphysiologicaltraits.TheVenndiagram showshowQTLforyieldaresharedbyanumberofdifferent physiologicaltraitscontributingtoitsexpression,includingcanopy temperature(CT);NDVI,andchlorophyll(CHL);weakerQTLeffectsfor watersolublecarbohydrateswerecommontoallfour[40].Boldfonts denoteQTLforyieldthatarestableacrossenvironments,while non-boldfontsdenoteenvironmentdependantQTLforyield(adaptedfrom [40]).

forsupporttoprojectsthathavepermittedthedevelopmentoftheideas presentedinthisreview.EditorialassistancefromEmmaQuilliganisalso gratefullyacknowledged.

References

and

recommended

reading

Papersofparticularinterest,publishedwithintheperiodofreview, havebeenhighlightedas:

ofspecialinterest

ofoutstandinginterest

1. BraunH,AtlinG,PayneT:Multi-locationtestingasatoolto identifyplantresponsetoglobalclimatechange.In Climate ChangeandCropProduction.EditedbyReynoldsMP. Wallingford,UK:CABIPublishers;2010:115-138. 2. RayDK,MuellerND,WestPC,FoleyJA:Yieldtrendsare

insufficienttodoubleglobalcropproductionby2050.PLOS ONE2013,8.

3.

ChallinorAJ,WatsonJ,LobellDB,HowdenSM,SmithDR, ChhetriN:Ameta-analysisofcropyieldunderclimatechange andadaptation[Internet].Nat.Clim.Chang.2014,27:1-5. This meta-analysis examines uncertainty and timing of temperature increases using 1700 published simulations to evaluate impacts of climatechangeonyield.Yieldlossesaregreaterinmagnitudeforthe second half ofthecentury than for thefirstwhile increasesin yield variabilityarelikely.

4. ChapmanSC,ChakrabortyS,DreccerMF,HowdenSM:Plant adaptationtoclimatechange—opportunitiesandprioritiesin breeding.CropPastureSci2012,63:251-268.

5. MirRR,Zaman-AllahM,SreenivasuluN,TrethowanR, VarshneyRK:Integratedgenomics,physiologyandbreeding approachesforimprovingdroughttoleranceincrops.Theor ApplGenet2012,125:625-645.

6. LongSP,Marshall-ColonA,ZhuX-G:Meetingtheglobalfood demandofthefuturebyengineeringcropphotosynthesisand yieldpotential.Cell2015,161:56-66.

7. FischerRA,ByerleeD,EdmeadesGO:CropYieldsandFood Security:WillYieldIncreasesContinuetoFeedtheWorld? Canberra:ACIARMonogr.158.AustralianCentreforInt.Agric. Res.;2014.

8. EaglesHA,CaneK,TrevaskisB,VallanceN,EastwoodRF, GororoNN,KuchelH,MartinPJ:Ppd1,Vrn1,ALMT1andRht genesandtheireffectsongrainyieldinlowerrainfall environmentsinsouthernAustralia.CropPastureSci2014,

65:159.

9. ReynoldsM,FoulkesJ,FurbankR,GriffithsS,KingJ,MurchieE, ParryM,SlaferG:Achievingyieldgainsinwheat.PlantCell Environ2012,35:1799-1823.

10. CossaniCM,ReynoldsMP:Physiologicaltraitsforimproving heattoleranceinwheat.PlantPhysiol2012,160:1710-1718. 11. ReynoldsM,ManesY,IzanlooA,LangridgeP:Phenotyping

approachesforphysiologicalbreedingandgenediscoveryin wheat.AnnApplBiol2009,155:309-320.

12. RichardsRA,RebetzkeGJ,WattM,CondonAG,SpielmeyerW, DolferusR:Breedingforimprovedwaterproductivityin temperatecereals:phenotyping,quantitativetraitloci, markersandtheselectionenvironment.FunctPlantBiol2010,

37:85-97.

13. CooperM,MessinaCD,PodlichD,TotirLR,BaumgartenA, HausmannNJ,WrightD,GrahamG:Predictingthefutureof plantbreeding:complementingempiricalevaluationwith geneticprediction.CropPastureSci2014,65:311-336. 14. RutkoskiJE,PolandJ,MondalS,AutriqueE,Gonza´lezPe´rezL,

CrossaJ,ReynoldsM,SinghRP:Predictinggrainyieldbasedon predictortraitsfromhigh-throughputphenotyping, genome-widemarkers,andpedigreetoincreaseratesofgeneticgain perunittimeandcost.PlantandAnimalGenomeConference XXIV;SanDiego,California:January8–13,2016.

15. LangridgeP,ReynoldsMP:Genomictoolstoassistbreedingfor droughttolerance.CurrOpinBiotechnol2015,32:130-135.

16.

GriffithsS,WingenL,PietragallaJ,GarciaG,HasanA,MirallesD, CalderiniDF,AnkleshwariaJB,WaiteML,SimmondsJetal.:

Geneticdissectionofgrainsizeandgrainnumbertrade-offsin CIMMYTwheatgermplasm.PLOSONE2015,10:e0118847. IncreasedgrainweightassociatedwithaQTLonwheatchromosome7B wasnotaccompaniedbyasignificantreductioningrainnumber,making thelocusanattractivetargetformarkerassistedselectiontodevelophigh yieldingboldgrainvarieties.

17.

ArausJL,CairnsJE:Fieldhigh-throughputphenotyping:the newcropbreedingfrontier.TrendsPlantSci2014,19:52-61. Excellentoverviewofhighthroughputfieldphenotypingapproachesand technologies.

18. GhanemME,MarrouH,SinclairTR:Physiological

phenotypingofplantsforcropimprovement.TrendsPlant Sci2014,20:139-144.

19. ZhaoK,TungC-W,EizengaGC,WrightMH,AliML,PriceAH, NortonGJ,IslamMR,ReynoldsA,MezeyJetal.:Genome-wide associationmappingrevealsarichgeneticarchitectureof complextraitsinOryzasativa.NatCommun2011,2:467. 20.

LiX,YanW,AgramaH,JiaL,JacksonA,MoldenhauerK,YeaterK, McClungA,WuD:Unravelingthecomplextraitofharvestindex withassociationmappinginrice(OryzasativaL.).PLoSOne 2012,7:1-10.

Red-shifted chlorophylls could increase the ability of photosynthetic organisms touselightinan additionalregionofthesolar spectrum; theoreticallygiving Chla,b, orc-containing oxygenicphotosynthetic organisms access to approximately an additional 19% photon flux comparedtostandardPARabsorbingantenna.

21. HallAJ,RichardsRA:Prognosisforgeneticimprovementof yieldpotentialandwater-limitedyieldofmajorgraincrops. FieldCropsRes2013,143:18-33.

22. WilkinsonS,KudoyarovaGR,VeselovDS,ArkhipovaTN, DaviesWJ:Planthormoneinteractions:innovativetargetsfor cropbreedingandmanagement.JExpBot2012,63:3499-3509. 23. LynchJP:Steep,cheapanddeep:anideotypetooptimize

waterandNacquisitionbymaizerootsystems.AnnBot2013,

112:347-357. 24.

TardieuF,ParentB,CaldeiraCF,WelckerC:Geneticand physiologicalcontrolsofgrowthunderwaterdeficit.Plant Physiol2014,164:1628-1635.

Theauthorssuggestthattheabilitytomaintaingrowthunderwaterdeficit providesameasureofsinkstrengthandislinkedtohydraulicprocesses. Thistraitshowshighlevelsofgeneticvariationandrepresentsagood targetforQTLmappingandselection.

25.

BlumA:TowardsaconceptualABAideotypeinplantbreeding forwaterlimitedenvironments.FunctPlantBiol2015,42 :502-513.

Ausefuloverviewandsynthesisof50yearsofresearchonABAandits manyeffectsonplantgrowthandinteractionswithotherplantgrowth regulators, concludingthatABA’sbenefit dependsverymuchonthe droughtstressprofile.

26.

SlaferGA,EliaM,SavinR,Garcı´aGA,TerrileII,FerranteA, MirallesDJ,Gonza´lezFG:Fruitingefficiency:analternative traittofurtherrisewheatyield.FoodEnergySecur2015,

4:92-109.

Firstreviewofthetrait‘fruitingefficiency’—theratioofgrainstospike weight—anintegrative‘sink’traitthatshowssignificantgeneticdiversity inelitematerialaswellasgoodassociationwithyield.

27. AssengS,EwertF,MartreP,Ro¨tterRP,LobellDB,CammaranoD, KimballBA,OttmanMJ,WallGW,WhiteJWetal.:Rising temperaturesreduceglobalwheatproduction.NatClim Change2014,5:143-147.

28. ChenuK,DeihimfardR,ChapmanSC:Large-scale characterizationofdroughtpattern:acontinent-wide modellingapproachappliedtotheAustralianwheatbelt– spatialandtemporaltrends.NewPhytol2013,198:801-820. 29. LangridgeP,FleuryD:Makingthemostof‘‘omics’’forcrop

breeding.TrendsBiotechnol2011,29:33-40. 30. GehanMA,GreenhamK,MocklerTC,McClungCR:

Transcriptionalnetworks–crops,clocks,andabioticstress. CurrOpinPlantBiol2015,24C:39-46.

31.

Fu¨llnerK,TempertonVM,RascherU,JahnkeS,RistR,SchurrU, KuhnAJ:Verticalgradientinsoiltemperaturestimulates developmentandincreasesbiomassaccumulationinbarley. PlantCellEnviron2012,35:884-892.

Excellentpaperindicatingtherisksofphenotypinginpotssinceshoot growthinteractsstronglywithsoiltemperaturegradientsthatexistinsoil butdonotgenerallyoccurinpotexperiments.

32. DaetwylerHD,CalusMPL,Pong-WongR,delosCamposG, HickeyJM:Genomicpredictioninanimalsandplants: simulationofdata,validation,reporting,andbenchmarking. Genetics2013,193:347-365.

33.

VadezV,KholovaJ,Zaman-AllahM,BelkoN:Water:themost important‘‘molecular’’componentofwaterstresstolerance research.FunctPlantBiol2013,40:1310-1322.

Areviewontherangeofstrategiesthatplantscanemploytoimprove performanceunderdroughtandtheimportanceofunderstandingwhich strategiestouseindifferentsituations,aswellasthevalueofcontrolled environmentsandsimulationmodeling totestalternativehypotheses, giventheunpredictabilityofrainfallinthefield.

34. LonginCF,MiX,Wu¨rschumT:Genomicselectioninwheat: optimumallocationoftestresourcesandcomparisonof breedingstrategiesforlineandhybridbreeding.TheorAppl Genet2015,128:1297-1306.

35. RebetzkeGJ,RatteyAR,FarquharGD,RichardsRA,CondonAG:

Genomicregionsforcanopytemperatureandtheirgenetic associationwithstomatalconductanceandgrainyieldin wheat.FunctPlantBiol2012,40:14-33.

36.

PintoRS,ReynoldsMP:Commongeneticbasisforcanopy temperaturedepressionunderheatanddroughtstress associatedwithoptimizedrootdistributioninbreadwheat. TheorApplGenet2015,128:575-585.

Demonstratesacommongeneticbasisforrootadaptationtoheatand droughtstress,indicatingallelesassociatedwithsensitivitytopresence ofwater,irrespectiveofthesoildepthprofile.

37. CossaniCM,ReynoldsMP:Heatstressadaptationinelitelines derivedfromsynthetichexaploidwheat.CropSci2015,

55:2719.

38. SukumaranS,DreisigackerS,LopesM,ChavezP,ReynoldsMP:

Genome-wideassociationstudyforgrainyieldandrelated traitsinanelitespringwheatpopulationgrownintemperate irrigatedenvironments.TheorApplGenet2014,128:353-363. 39. LopesMS,DreisigackerS,Pen˜aRJ,SukumaranS,ReynoldsMP:

Geneticcharacterizationofthewheatassociationmapping initiative(WAMI)panelfordissectionofcomplextraitsin springwheat.TheorApplGenet2015,128:453-464. 40. PintoRS,ReynoldsMP,MathewsKL,McIntyreCL,

Olivares-VillegasJJ,ChapmanSC:HeatanddroughtadaptiveQTLina wheatpopulationdesignedtominimizeconfounding agronomiceffects.TheorApplGenet2010,121:1001-1021. 41. BonneauJ,TaylorJ,ParentB,BennettD,ReynoldsM,FeuilletC,

LangridgeP,MatherD:Multi-environmentanalysisand improvedmappingofayield-relatedQTLonchromosome3B ofwheat.TheorApplGenet2013,126:747-761.

42. BustosDV,HasanAK,ReynoldsMP,CalderiniDF:Combining highgrainnumberandweightthroughaDH-populationto improvegrainyieldpotentialofwheatinhigh-yielding environments.FieldCropsRes2013,145:106-115. 43. BennettD,ReynoldsM,MullanD,IzanlooA,KuchelH,

LangridgeP,SchnurbuschT:Detectionoftwomajorgrainyield QTLinbreadwheat(TriticumaestivumL.)underheat,drought andhighyieldpotentialenvironments.TheorApplGenet2012,

125:1473-1485. 44.

ChapmanS,MerzT,ChanA,JackwayP,HrabarS,DreccerM, HollandE,ZhengB,LingT,Jimenez-BerniJ:Pheno-Copter:a low-altitude,autonomousremote-sensingrobotichelicopter forhigh-throughputfield-basedphenotyping.Agronomy2014,

4:279-301.

Describesacustomizedrobotichelicopterwithautonomousflight con-trol,andsoftwaretoplanflightsoverexperiments0.5–3hainarea,and extractmultiple experimental fieldplots fromimages taken by three camerasforvariationinearlyseasongroundcoverinsorghum;canopy temperatureinsugarcane,and3Dmeasuresofcroplodginginwheat.

45.

FahlgrenN,GehanMA,BaxterI:Lights,camera,action: high-throughputplantphenotypingisreadyforaclose-up.Curr OpinPlantBiol2015,24:93-99.

Reviewofthepotentialofremote-sensingoftheelectromagnetic reflec-tancefromplanttissuetorapidlyandnon-destructivelyphenotypeplants. 46. FeuilletC,LangridgeP,WaughR:Cerealbreedingtakesawalk

onthewildside.TrendsGenet2008,24:24-32.

47. OrtizR,BraunHJ,CrossaJ,CrouchJH,DavenportG,DixonJ, DreisigackerS,DuveillerE,HeZ,HuertaJetal.:Wheatgenetic resourcesenhancementbytheInternationalMaizeandWheat ImprovementCenter(CIMMYT).GenetResourCropEvol2008,

55:1095-1140.

48. HarperJ,ArmsteadI,ThomasA,JamesC,GasiorD,BisagaM, RobertsL,KingI,KingJ:Alienintrogressioninthegrasses

Loliumperenne(perennialryegrass)andFestucapratensis

(meadowfescue):thedevelopmentofsevenmonosomic substitutionlinesandtheirmolecularandcytological characterization.AnnBot2011,107:1313-1321.

49. PrinsA,DouglasJ,OrrP,AndralojcPJ,ReynoldsMP, Carmo-SilvaE,ParryMAJ:Rubiscocatalyticpropertiesofwildand domesticatedrelativesprovidescopeforimprovingwheat photosynthesis.JExpBot2016:erv574.

50. TrethowanRM,Mujeeb-KaziA:Novelgermplasmresourcesfor improvingenvironmentalstresstoleranceofhexaploidwheat. CropSci2008,48:1255-1265.

51. WingenLU,OrfordS,GoramR,Leverington-WaiteM,BilhamL, PatsiouTS,AmbroseM,DicksJ,GriffithsS:EstablishingtheA.E. Watkinslandracecultivarcollectionasaresourcefor systematicgenediscoveryinbreadwheat.TheorApplGenet 2014,127:1831-1842.

52. ReynoldsMP,TattarisM,CossaniCM,EllisM, Yamaguchi-ShinozakiK,SaintPierreC:Exploringgeneticresourcesto increaseadaptationofwheattoclimatechange.In Advances inWheatGenetics:FromGenometoField.EditedbyOgiharaY, TakumiS,HandaH.Japan:Springer;2015:355-368.

53.

DrieverSM,LawsonT,AndralojcPJ,RainesCA,ParryMAJ:

Naturalvariationinphotosyntheticcapacity,growth,andyield in64field-grownwheatgenotypes.JExpBot2014,65:1-15. Significant variation in photosynthetic capacity and biomass were observedinapanelof64elitewheatcultivars,mainlyrelatedtomaximum capacityandoperationalratesofCO2assimilation,suggestsignificant underutilizedphotosyntheticcapacityamongexistingwheatvarieties.

54. GutierrezM,ReynoldsMP,KlattAR:Associationofwater spectralindiceswithplantandsoilwaterrelationsin contrastingwheatgenotypes.JExpBot2010,61:3291-3303. 55. TattarisM,ReynoldsMP,PietragallaJ,MoleroG,CossaniCM,

EllisM:Airborneremotesensingforhigh-throughput phenotypingofwheat.In In Proceedingsoftheworkshopon UAV-basedremotesensingmethodsformonitoringvegetation. EditedbyBendigJ,BarethG.Proceedingsoftheworkshopon UAV-basedremotesensingmethodsformonitoringvegetation Germany:UniversityofCologne;2013:125-136.

56. SerbinSP,DillawayDN,KrugerEL,TownsendPA:Leafoptical propertiesreflectvariationinphotosyntheticmetabolismand itssensitivitytotemperature.JExpBot2012,63:489-502. 57. Zarco-TejadaPJ,CatalinaA,Gonza´lezMR,Martı´n P:

Relationshipsbetweennetphotosynthesisand steady-statechlorophyllfluorescenceretrievedfrom airbornehyperspectralimagery.RemoteSensEnviron2013,

136:247-258.

58. HuynhBL,MatherDE,SchreiberAW,ToubiaJ,BaumannU, ShoaeiZ,SteinN,AriyadasaR,StangoulisJCR,EdwardsJetal.:

Clustersofgenesencodingfructanbiosynthesizingenzymes inwheatandbarley.PlantMolBiol2012,80:299-314. 59.

Carreno-QuinteroN,BouwmeesterHJ,KeurentjesJJB:Genetic analysisofmetabolome–phenotypeinteractions:frommodel tocropspecies.TrendsGenet2013,29:41-50.

Theproblemofusingmetabolomicsdatasetstounderstandphysiological processandphenotypesisaddressedinthisreview.Linkingstudiesof metabolitediversitytophenotypeisproposedasarouteto understand-ingcomplextraits.

60. HillC,TaylorJ,EdwardsJ,MatherD:Whole-genomemappingof agronomicandmetabolictraitstoidentifynovelquantitative traitlociinbreadwheatgrowninawater-limitedenvironment. PlantPhysiol2013,162:1266-1281.

61. HuH,XiongL:Geneticengineeringandbreedingof drought-resistantcrops.AnnuRevPlantBiol2014,65:715-741. 62. MichaelTP,VanBurenR:Progress,challengesandthefutureof

cropgenomes.CurrOpinPlantBiol2015,24C:71-81. 63.

WendlerN,MascherM,No¨hC,HimmelbachA,ScholzU, Ruge-WehlingB,SteinN:Unlocking the secondary gene-pool of barley with next-generation sequencing.PlantBiotechnolJ

2014http://dx.doi.org/10.1111/pbi.12219.

Newgenomesequencingtechnologiesareshowntooffernewstrategies foranalyzingintrogressedchromosomesegmentsfromthesecondary gene-poolofbarleyandenhancingtheuseofgeneticresourcesforcrop improvement.

64. KingGJ:Cropepigeneticsandthemolecularhardwareof genotypeenvironmentinteractions.FrontPlantSci2015,

6:968.

65. HeslotN,JanninkJ-L,SorrellsME:Perspectivesforgenomic selectionapplicationsandresearchinplants.CropSci2015,

55:1-12.

66. BeyeneY,SemagnK,MugoS,TarekegneA,BabuR,MeiselB, SehabiagueP,MakumbiD,MagorokoshoC,OikehSetal.:

Geneticgainsingrainyieldthroughgenomicselectionineight bi-parentalmaizepopulationsunderdroughtstress.CropSci 2015,55:154-163.

67. ReynoldsMP,PaskA,TorresA,QuicheYN,MoleroG, SukumaranS,FigueroP,SolisE,IretaJ,BorbonAetal.: Pre-breedingforyieldpotential:summaryofinternationaldata from2ndWYCYTandperformanceofpipelinematerial.In In ProceedingsoftheInternationalTRIGO(Wheat)YieldPotential Workshop2015.CENEB,CIMMYT.EditedbyReynoldsMP, MoleroG,MollinsJ,BraunHJ.ProceedingsoftheInternational TRIGO(Wheat)YieldPotentialWorkshop2015.CENEB, CIMMYTCd.Obregon,Sonora,Mexico,24–26thMarch2015: Mexico,DF:CIMMYT;2015:17-22.

68. PaskA,JoshiAK,Mane`sY,SharmaI,ChatrathR,SinghGP, SohuVS,MaviGS,SakuruVSP,KalappanavarIKetal.:A wheatphenotypingnetworktoincorporatephysiological traitsforclimatechangeinSouthAsia.FieldCropsRes2014,

168:156-167.

69. SukumaranS,ReynoldsMP,PaskAJD,MoleroG,TorresA, QuicheYN,PayneT,Solis-MoyaRE,Camacho-CasasMA, Figueroa-LopezPetal.:Internationalperformanceof physiologicaltraitbasedlinesofthe4thStressAdaptiveTrait YieldNursery(4thSATYN).In In Proceedingsofthe2ndWheat (TRIGO)YieldPotentialWorkshop.EditedbyReynoldsMP,Molero G,QuilliganE.Proceedingsofthe2ndWheat(TRIGO)Yield PotentialWorkshopCIMMYT;2016:15-24.