Drought Tolerance Candidate Gene Mining Based On High-throughput Photosynthetic Traits In Wheat(Triticum Aestivum L.)

Wheat is one of the key crops affecting food security and global economic development,and photosynthetic traits of the leaves are critical for the formation of carbohydrates in the wheat grain,which affect the yield of wheat.Unmanned remote sensing is allowed for effective detection of chlorophyll content（SPAD）in crops,which is an important tool for high-throughput phenotype acquisition.Breeding drought-tolerant varieties of wheat is a fundamental way to improve the water utilisation of the crop.SPAD,leaf area index（LAI）,crown air temperature difference（CTD）and normalised vegetation index（NDVI）,have become important phenotypic traits in recent years for the study of high light efficiency and selection of drought tolerant genotypes in wheat.Therefore,62 winter wheats from Xinjiang,300 winter wheats from around the world and 309 lines from the Berkut/Worrakatta F_2:6RIL population were used as test materials for this study.Multispectral images of two water treatments（normal irrigation,drought stress）at three key wheat fertility stages（tassel,flowering and filling）and corresponding phenotypic data of photosynthetic traits were obtained by DJI P4M UAV（Unmanned Aerial Vehicle）.Model inversion studies were carried out for different genotypes of canopy SPAD,followed by association analysis of natural populations using 90K SNP chips and genetic linkage mapping and Quantitative Trait Locus（QTL）localization using 50K SNP chips,in order to find the optimal model for SPAD inversion and drought-related loci for photosynthetic traits in wheat.Finally,the drought resistance genes in wheat were predicted in conjunction with the database of the Chinese spring genome to provide a reference for drought resistance response in wheat.The main results were as follows:1.Under both water treatments at different fertility stages,SPAD values were higher under normal irrigation than under drought stress;several vegetation indices and SPAD showed highly significant correlations,with the highest correlation reaching 0.51.In the construction of prediction models for SPAD,different models had high estimation accuracy under both normal irrigation and drought stress,with the highest model prediction accuracy under normal irrigation being the Ridge CV model（r=0.63,RMSE=3.28,NRMSE=16.20%）,and the highest prediction accuracy of the model under drought stress was the SVM model（r=0.63,RMSE=3.47,NRMSE=19.20%）at the tasseling stage,and the SVM model was more stable overall.2.Drought resistance was evaluated by drought index（DI）and drought tolerance coefficient（DC）for300 winter wheat and 309 RIL populations plus lines,respectively.SPAD and LAI data were obtained for300 winter wheat in the field,and a comprehensive evaluation of drought resistance of wheat was carried out by DI and DC,with normal irrigation as the control treatment.Eight drought-resistant varieties were screened using the SPAD trait and four using the LAI trait.The DI and DC of the 12 varieties screened exceeded the top 10%of the participating material,with the identified Yumai 34 and Yumai 18 being consistent with the results of the group’s earlier identification of drought-tolerant varieties at the sprouting stage.In addition,309 RIL populations were evaluated for drought tolerance by SPAD,LAI and NDVI traits,of which the number of materials with DC and DI greater than 1 in SPAD reached 10,accounting for 3.00%;the number of materials with DC and DI greater than 1 in NDVI reached 19,accounting for 6.00%;the number of materials with DC and DI greater than 1 in LAI reached 34,accounting for 11.00%.The number of materials with DC and DI greater than 1 for LAI was 34,accounting for more than 11.00%.Among them,B057,B088and B103 were more consistent with the drought evaluation results previously identified by the group and could be used as candidates for later drought-resistant varieties of parents.3.Through quality screening of 90K SNP microarray loci,16710 SNP loci covering 21 chromosomes of the whole wheat genome were obtained for genome-wide association analysis using a hybrid linear model of Q+K.76 candidate genes related to leaf physiological indicators were obtained,mainly involved in hormone protein synthesis and transduction,zinc finger protein regulation and cytochrome synthesis.4.The construction of genetic linkage map was performed by 50K SNP microarray using 309 RIL populations from Berkut/Worrakatta.QTL localization of leaf physiological indicators was performed to mine candidate genes and 41 genes were identified.Among them,Traes CS1D01G295400 encodes cytochrome b559 protein;Traes CS5B01G400600 and Traes CS1D01G389300 encode color class proteins;Traes CS7D01G137200 encodes chlorophyll oxygenase;Traes CS6D01G103100 encodes zinc finger protein VAR3 chloroplast protein,etc.The chlorophyll content of cells had a huge impact on photosynthesis,and these genes regulating cytochromes may be closely related to photosynthesis in crops.5.By comparing the physical locations of SNP markers mined by genome-wide association analysis and linkage analysis in wheat,the SNP markers corresponding to 28 significantly associated loci for SPAD,LAI,CTD and NDVI for light and traits were compared to those corresponding to QTL with the same or similar physical locations.This region was also compared with previous studies,except for the presence of loci on chromosomes 3A and 2B that are closer to those previously mined,and others that may be new loci for photosynthetic traits in crops.