Also available as PDF (figures and tables appear in the PDF).
Cotton (Gossypium spp.) is one of the world’s most important fibre crops, and fibre quality is a major determinant of its economic value and suitability for textile processing. Important fibre quality character- istics include fibre length, fibre strength, micronaire, fibre elongation, fibre uniformity and fibre maturity. These traits are quantitative in nature and are controlled by multiple genes interacting with environmental factors. Quantitative trait locus (QTL) mapping has therefore become an important approach for dissecting the genetic basis of fibre quality and identifying genomic regions associated with desirable phenotypes. Advances in cotton genome sequencing, molecular marker development and high-density genetic mapping have substantially increased the resolution and reliability of QTL identification. Numerous QTLs have been detected for fibre length, strength, fineness, elongation and uniformity, with several genomic regions repeat- edly detected across populations and environments. Recent studies have further combined QTL mapping with genome-wide association studies (GWAS), transcriptomics, meta-analysis and candidate-gene analysis to identify stable loci and genes controlling fibre development. Marker-assisted selection (MAS), genomic selection and genome editing provide opportunities for translating these discoveries into breeding pro- grammes. This review discusses the genetic basis of cotton fibre quality, QTL mapping strategies, important fibre-quality QTLs, molecular markers, QTL × environment interactions, candidate genes and applications in molecular breeding. Integration of QTL information with high-throughput genomics and precision breeding is expected to accelerate the development of cotton cultivars possessing superior fibre quality together with high yield and environmental adaptability.
Keywords: Cotton; Gossypium; fibre quality; QTL mapping; molecular markers; fibre length; fibre strength; micronaire; fibrequality; QTLmapping; molecularmarkers; fibrelength; fibrestrength
1.Introduction strength. Longer fibres can generally produce finer and stronger yarns when other characteristics are Cotton(Gossypiumspp.)isagloballyimportantfibre favourable [7].Fibre strength represents the force crop whose economic value is strongly associated requiredtobreakafibrebundleandisanimportant with the quality and processing performance of its determinant of yarn durability. Strong fibres are lint.Amongcultivatedcottonspecies,uplandcotton particularlydesirableformodernhigh-speedspinning (GossypiumhirsutumL.)iswidelygrownbecauseofits technologies. Fibreelongation describesthe ability adaptabilityandproductivity,whereasG.barbadense of fibres to extend before breaking and contributes is recognized for superior fibre characteristics, par- to processing behaviour.Micronaire is an indirect ticularlylong,strongandfinefibres. Fibrequalityis measureassociatedwithfibrefinenessandmaturity determined by several interrelated characteristics, [8]. Extremely low or high micronaire values can includingfibrelength,strength,fineness,elongation negatively affect processing efficiency and textile and uniformity [1-3]. Fibre length and strength are quality. Fibre uniformity reflects the consistency of particularly important for yarn production, while fibre length within a sample, while the short fibre micronaire, which reflects a combination of fibre index represents the proportion of relatively short fineness and maturity, affects processing and dye- fibres.Together,thesetraitsdeterminethesuitability ing performance.Improving fibre quality through ofcottonforginning,spinning,weavinganddyeing. conventionalbreedingischallengingbecausethese traitsarequantitativelyinheritedandinfluencedby numerousgenesandenvironmentalconditions[4-5]. 3.GeneticBasisofCottonFibreQuality In addition, favourable fibre characteristics may be Cottonfibrequalitytraitsexhibitcomplexinheritance genetically associated with undesirable agronomic becausetheyarecontrolledbymultiplelociwithdiffer- traits,makingsimultaneousimprovementofyieldand enteffects.Somelocicontributerelativelylargeeffects, qualitydifficult.Understandingthegenomicregions whereasnumerousminor-effectlocicollectivelyinflu- responsible for fibre-quality variation is therefore encephenotypicvariation.Consequently,fibre-quality essentialforefficientbreeding.
traitsgenerallyexhibitcontinuousvariationratherthan Quantitativetraitlocusmappingprovidesapowerful discretephenotypicclasses.Theallotetraploidgenome method for identifying chromosome regions asso- ofcultivatedcottonfurthercomplicatesgeneticanal- ciated with quantitative phenotypes. Earlier cotton ysis. Upland cotton contains A and D subgenomes, studiesgeneratednumerousQTLsforfibretraits,while andinteractionsbetweenhomoeologouslocicanin- advancesinmolecularmarkersandgenomesequenc- fluencefibredevelopment.Comparativeanalysesin- inghaveenabledincreasinglydensegeneticmapsand dicatethatQTLsforfibretraitsaredistributedacross improvedQTLresolution[6].Amajorreviewreported bothsubgenomes,althoughsomechromosomescon- more than 1,000 QTLs associated with cotton fibre tainrelativelygreaterconcentrationsofparticulartrait- qualityfromnumerousmappingstudies,illustrating associatedloci.Forexample,studiesoftetraploidcot- thecomplexgeneticarchitectureofthesetraits.Recent tonhavereportedsubstantialconcentrationsoffibre- developments have extended QTL analysis beyond qualityQTLsonseveralAandDsubgenomechromo- conventional linkage mapping. GWAS can identify somes[9].Fibredevelopmentitselfoccursthroughdis- marker–trait associations using diverse germplasm, tinctdevelopmentalstagesinvolvinginitiation,elon- while QTL meta-analysis can integrate results from gation,secondarycell-wallsynthesisandmaturation. independent experiments and identify more stable Differentgenesandregulatorypathwaysbecomeac- genomicregions.A2025meta-analysisintegrated379 tive during these stages. Hormonal signalling, tran- original cotton fibre-quality QTLs from 21 mapping scriptionfactors,cell-wallbiosynthesis,carbohydrate studiesandidentified74meta-QTLsdistributedacross metabolismandcytoskeletalprocessesallcontribute all26cottonchromosomes,demonstratingthevalue tofibredevelopment.Thisdevelopmentalcomplexity ofcombiningindependentQTLdatasets.
explainswhydifferentQTLsmaycontrolfibrecharac- teristicsatdifferentstages.
2.FibreQualityTraitsinCotton 4.QTLMappingApproachesinCotton Cottonfibrequalityisamultidimensionalcharacter- istic comprising several physical and technological QTLmappingidentifiesgenomicregionsstatistically properties.Fibrelengthisgenerallyexpressedthrough associatedwithphenotypicvariationwithinasegre- measurementssuchasupper-halfmeanlengthand gating population. In cotton, commonly used map- is associated with spinning performance and yarn pingpopulationsincludeFpopulations,recombinant inbredlines,backcrosspopulations,chromosomesub- polymorphism [12-14]. The development of cotton stitutionlines,introgressionlinesandotheradvanced reference genomes and high-throughput sequenc- populations. Thesepopulationsaregenotypedwith ing has greatly expanded the availability of single molecularmarkersandphenotypedforfibre-quality nucleotide polymorphism markers. SNP arrays and characteristics.Linkagemapsareconstructedbydeter- genotyping-by-sequencing technologies permit miningtheorderandgeneticdistancesbetweenmolec- high-density genome-wide genotyping, improving ularmarkers.Statisticalanalysisisthenusedtoiden- mapping resolution and facilitating the detection tifygenomicintervalsassociatedwithvariationinfibre of smaller-effect loci.High-density markers also traits[10].Dependingonthemappingpopulationand facilitatefinemapping,inwhichabroadQTLinterval experimentaldesign,intervalmapping,compositein- is progressively narrowed using additional markers tervalmappingandinclusivecompositeintervalmap- and larger or specially designed populations. Fine pingcanbeused.TheaccuracyofQTLidentification mappingisimportantbecauseabroadQTLinterval dependsonpopulationsize,markerdensity,pheno- may contain hundreds of genes, whereas a smaller typingaccuracyandenvironmentalreplication.QTLs interval can facilitate candidate-gene identification detectedrepeatedlyacrossgenerations,locationsand andfunctionalvalidation.
yearsareparticularlyvaluablebecausetheirstability increases the likelihood of successful application in 7.QTLMeta-AnalysisandStableQTLIdentifi- breeding.Recentstudieshavedemonstratedthatsta- cation bleQTLscanbeidentifiedthroughmulti-environment Individual QTL mapping experiments frequently evaluationandsubsequentvalidation.
identify different genomic regions because of differ- encesingeneticbackgrounds,mappingpopulations, 5. MajorQTLsAssociatedwithFibreQuality markersystemsandenvironments.QTLmeta-analysis Traits provides an approach for integrating results from NumerousQTLshavebeenidentifiedforfibrelength, multiplestudiesandidentifyingconsensusgenomic strength, micronaire, elongation and uniformity. Fi- regions.Recentmeta-analysisofcottonfibre-quality brelengthisparticularlywellstudied,andQTLshave QTLscombined379originalQTLsfrom21independent beenidentifiedacrossmanycottonchromosomes.Are- studies and identified 74 meta-QTLs across the 26 centoverviewreportedthatapproximately1,850QTLs chromosomes [15-16]. These meta-QTLs contained havebeenmappedforfibrelength,fibrestrengthand thousands of genes, and integration with RNA se- finenessoverthepasttwodecades,althoughonlya quencing, functional annotation and expression subsethavebeensufficientlyvalidatedfordirectbreed- analysishelpednarrowthelistofpotentialcandidate ingorgene-editingapplications.FibrestrengthQTLs genes.Meta-QTL analysis is therefore particularly havealsobeendetectedacrossthecottongenome.For valuable for distinguishing stable genomic regions example, high-resolution mapping on chromosome fromQTLsthatarespecifictoindividualpopulations 25 identified a stable fibre-strength QTL, qFS-chr25- or environments. Stable meta-QTLs can serve as 4,thatwasdetectedacrosssevenenvironmentsand prioritiesforfinemapping,functionalcharacterization explainedapproximately6.53–11.83%ofphenotypic andmarker-assistedselection.
variation.Suchstablelociareespeciallyvaluablefor marker-assistedbreeding[11].Micronaireisinfluenced 8.Genome-WideAssociationStudiesforFibre bynumerousgenomicregionsandisoftenassociated Quality withfibrematurityandfineness.Becausemicronaire Genome-wideassociationstudieshavecomplemented canshowcomplexrelationshipswithotherfibretraits, conventionallinkagemappingbyexploitinghistorical breedingstrategiesmustcarefullyevaluatecorrelated recombinationindiversegermplasm.Insteadofrely- responseswhenselectingfavourablealleles. ingonasinglesegregatingpopulation,GWASevalu- atesmarker–traitassociationsacrossacollectionof 6.MolecularMarkersforQTLDetection geneticallydiversecottonaccessions.GWAScanpro- Molecular markers are fundamental tools for con- videhighermappingresolutionthanconventionalQTL structing genetic maps and detecting QTLs. Earlier mappingbecausehistoricalrecombinationcannarrow cotton mapping studies extensively used restriction associationintervals.However,populationstructure fragmentlengthpolymorphisms,amplifiedfragment andgeneticrelatednessmustbecarefullycontrolledto lengthpolymorphismsandsimplesequencerepeats. reducefalse-positiveassociations[17-18].Recentcot- SSR markers were particularly useful because of ton research has increasingly combined GWAS with their codominant inheritance and relatively high genomicselectionandselective-sweepanalysisand Fibrequalitytrait Majorcharacteristic Geneticnature Breedingimportance Fibrelength Lengthofindividualfibres Quantitative,polygenic Importantforspinningandyarnqual- ity Fibrestrength Resistancetofibrebreakage Quantitative Determinesyarnstrengthanddura- bility Micronaire Indicatoroffibrefinenessandmatu- Quantitative Influencesprocessinganddyeing rity Fibreelongation Abilitytoextendbeforebreaking Quantitative Affectsprocessingperformance Fibreuniformity Consistencyoffibrelength Quantitative Importantforuniformyarnforma- tion Shortfibreindex Proportionofshortfibres Quantitative Higherlevelscanreduceprocessing efficiency Fibrematurity Degreeofsecondarywalldevelop- Quantitative Influencesdyeuptakeandprocess- ment ingquality identificationofsignificantassociationsforfibreelon- underfieldconditions.Forfibrequality,markerslinked gation,lengthanduniformityandreportedstableQTL to stable QTLs can be incorporated into breeding regionsacrossenvironments,illustratingthegrowing populations. Individuals carrying favourable alleles integrationofmultiplegenomicapproachesinfibre- canbeselectedatearlydevelopmentalstages,thereby qualityimprovement.
reducing the number of plants requiring extensive fieldevaluation.However,MASismosteffectivewhen 9.CandidateGenesUnderlyingFibreQuality QTL effects are stable and markers are sufficiently QTLs closetothecausalgene.Therefore,validationacross multiple genetic backgrounds and environments is IdentificationofaQTLisanimportantfirststep,but essentialbeforewidespreaddeployment[20-21].The determiningthegeneresponsibleforthephenotype integrationofQTLmapping,high-densitymarkersand provides greater biological and breeding value. MAS has already been recognized as an important Candidate-gene analysis integrates QTL intervals strategyforcottonfibre-qualityimprovement. with genome annotation, gene-expression profiles, functionalgenomicsandsequencevariation.Several 11.IntegrationofQTLMappingwithGenomic classes of genes have been associated with cotton Selection fibre development, including transcription factors, hormone-related genes, genes involved in cell-wall GenomicselectiondiffersfromconventionalMASbe- biosynthesis and genes controlling carbohydrate causeitusesgenome-widemarkerinformationtopre- metabolism. Transcription-factor families such dictthebreedingvalueofindividuals.Insteadofselect- as MYB, HD-ZIP IV, bHLH and ERF have received ingonlyforalimitednumberofmajorQTLs,genomic considerableattentionbecauseoftheirrolesinfibre selectioncapturescontributionsfromnumerousloci, initiationandelongation.Recentresearchhasfurther includingsmall-effectgenes.Thisapproachispartic- integratedQTLandGWASsignalswithtranscriptomic ularlyattractiveforcottonfibrequalitybecausethe information to identify candidate genes controlling traitsarehighlyquantitativeandcontrolledbymany fibre quality.Functional validation is increasingly loci. Atrainingpopulationwithbothgenotypicand important for confirming candidate genes [19-20]. phenotypic information is developed, and genomic Gene-editingapproachessuchasCRISPR/Cascanbe predictionmodelsaresubsequentlyusedtoestimate usedtodeterminewhetherdisruptionormodification breedingvaluesforuntestedindividuals[23].Combin- ofacandidategenechangesfibrecharacteristics.For ingstableQTLinformationwithgenomicprediction example, recent work has demonstrated the use of mayimproveselectionaccuracy.QTLswithlargeand CRISPRvalidationforcandidategenesassociatedwith reproducibleeffectscanbeexplicitlyincorporatedinto fibrelength.
predictionmodels,whilegenome-widemarkerscap- tureadditionalsmall-effectvariation.
10.Marker-AssistedSelectionforFibreQuality 12.QTL×EnvironmentInteractions Marker-assistedselectionenablesbreederstoselect plants carrying favourable alleles using molecular Environmentalconditionscansubstantiallyinfluence markersratherthanrelyingexclusivelyonphenotypic theexpressionoffibre-qualitytraits.Temperature,wa- measurements.Thisapproachisparticularlyusefulfor teravailability,soilconditionsandcrop-management traitsthatareexpensiveortime-consumingtoevaluate practicescanaffectfibredevelopment.Consequently, andforQTLswhoseeffectsaredifficulttodistinguish a QTL identified in one environment may have a smaller effect or fail to be detected in another samepredictivevalueinanothergeneticbackground. environment.Multi-environment trials are therefore Therefore,validationacrosspopulationsandenviron- essential for identifying stable QTLs. Stable QTLs mentsshouldprecedelarge-scaledeployment. are particularly valuable for breeding because they are less dependent on environmental conditions. 15.FuturePerspectives Conversely, environment-specific QTLs may still The future of cotton fibre-quality improvement will be useful when breeding for specific production depend increasingly on the integration of genetic regions [24-25].Understanding QTL × environment mapping, genomics, phenomics and precision interactions can also contribute to climate-resilient breeding.High-densitySNPgenotyping,pan-genome cotton breeding. Future studies should combine resources and long-read sequencing will improve genomic information with detailed environmental the ability to detect structural variants and rare measurements to identify alleles that maintain favourable alleles.Combining QTL mapping with fibre quality under variable temperature and water GWASandmeta-analysiscanidentifyrobustgenomic conditions.
regions, while transcriptomics and proteomics can helprevealbiologicalmechanismsunderlyingthese 13. RelationshipBetweenFibreQualityand loci. Recentstudieshavealreadydemonstratedthe Yield value of integrating QTL data with RNA sequencing One of the major challenges in cotton breeding is andfunctionalannotationtoidentifycandidategenes. achievingsimultaneousimprovementoffibrequality Genomeeditingprovidesanotherimportantopportu- andyield. Selectionforhigheryieldmaysometimes nity.Oncecandidategenesarevalidated,CRISPR/Cas resultinundesirablechangesinfibrecharacteristics, technologiescanpotentiallymodifyfavourablealleles while selection for superior fibre quality may nega- orremoveundesirablegenefunctions.Recentreviews tively affect productivity.QTL mapping provides an indicate that several genes involved in cotton fibre opportunity to identify genomic regions controlling quality are now being investigated or edited for fibrequalityindependentlyofyield[21]. Itcanalso improvementoffibrecharacteristics.Futurebreeding reveal genomic regions affecting multiple traits, programmesshouldmovetowardintegratedgenomic allowingbreederstodeterminewhetherassociations strategies in which stable QTLs, GWAS signals, can- arecausedbylinkageorpleiotropy.Modernbreeding didategenesandgenomicpredictionarecombined programmesshouldthereforeconsiderfibrequality, withhigh-throughputphenotyping.Suchapproaches yield, maturity, plant architecture and stress adap- couldacceleratethedevelopmentofcottoncultivars tation as interconnected traits. Multi-trait genomic that combine superior fibre quality, high yield and selection and strategic QTL pyramiding may help resistancetoenvironmentalstresses.
overcomeundesirablegeneticcorrelations. 16.Conclusion 14.ChallengesinQTLMappingofFibreQuality QTLmappinghassignificantlyadvancedunderstand- Traits ingofthegeneticarchitectureofcottonfibrequality. Despiteconsiderableprogress,severalchallengesre- Numerous genomic regions controlling fibre length, main.First,fibrequalityiscontrolledbynumerousloci strength, micronaire, elongation and uniformity withvariableeffects,makingitdifficulttodistinguish have been identified through linkage mapping, majorgenesfromminor-effectQTLs.Second,QTLex- associationstudiesandmeta-analysis.Theincreasing pressioncanbestronglyinfluencedbyenvironmental availability of high-density molecular markers and conditions.Third,differentstudiesoftenidentifydiffer- cottongenomeresourceshasimprovedtheresolution entgenomicregionsbecauseofvariationinparental of QTL detection and facilitated the identification genotypesandexperimentalpopulations.Anotherchal- of candidate genes.Stable QTLs identified across lengeisthepolyploidnatureofcultivatedcotton[22]. multiple environments are particularly valuable for HomoeologousgenesintheAandDsubgenomesmay marker-assistedbreeding,whileGWASandmeta-QTL haveoverlappingordistinctfunctions,makingfunc- analysis provide complementary approaches for tionalvalidationmorecomplicated.Inaddition,link- discovering robust genomic regions. Integration of agebetweenfavourablefibre-qualityallelesandunde- QTLmappingwithtranscriptomics,genomicselection sirableagronomictraitscanlimitthedirectuseofsome and genome editing is creating new opportunities QTLs.ThetranslationofQTLdiscoveriesintobreeding for precise improvement of cotton fibre quality. gainsalsorequiresrigorousvalidation.Amarkerasso- Futureresearchshouldfocusonvalidatingstableloci, ciatedwithatraitinonepopulationmaynothavethe identifying causal genes and developing breeding Approach Mainapplication Majoradvantage Mainlimitation SSRmarkers LinkagemappingandQTLdetection Codominantandinformative LowerdensitythanSNPsystems SNPmarkers High-densitygeneticmapping Genome-widecoverage Genotypinginfrastructurerequired QTLmapping Identificationofgenomicregions Effectiveinsegregatingpopulations Resolutiondependsonpopulation size QTLfinemapping NarrowingQTLintervals Facilitatescandidate-geneidentifica- Requires large populations and tion densemarkers QTLmeta-analysis IntegrationofindependentQTLstud- Identifiesstableconsensusloci RequirescomparableQTLinforma- ies tion GWAS Association mapping in diverse Highmappingresolution Populationstructurecancreatefalse germplasm associations RNAsequencing Identification of differentially ex- Providesfunctionalinformation Expressiondoesnotalwaysindicate pressedgenes causality Genomicselection Predictionofbreedingvalues Capturesmanysmall-effectloci Requires suitable training popula- tions Marker-assistedselection SelectionoffavourableQTLalleles Acceleratesbreeding Lesseffectiveforhighlypolygenic traits CRISPR/Cas Functionalvalidationandtargeted Precisegeneticmanipulation Technicalandregulatorychallenges modification strategiesthatsimultaneouslyimprovefibrequality, prehensivePCR-basedmarkerlinkagemapand yieldandenvironmentaladaptability.Suchintegrated QTLmappingforfiberqualitytraitsinuplandcot- approaches will contribute to the development of ton(GossypiumhirsutumL.).MolecularBreeding, next-generationcottoncultivarscapableofmeeting 24,49–61.
thechangingrequirementsofthetextileindustryand [7] Yang,X.L.,Wang,Z.W.,Zhang,G.Y.,Pan,Y.X.,Wu, sustainablecottonproduction.
L.Q.,Li,Z.K.etal.(2009).Constructionofmolecu- largeneticmapandQTLanalysisoffiberquality