Characterisation of selected bread wheat (Triticum aestivum L.) genotypes for drought tolerance based on SSR markers, morpho-physiological traits and drought indices.

Abstract

Bread wheat (Triticum aestivum L.) and durum wheat (T. turgidum L. var. durum) are staple cereal food crops worldwide. In South Africa, bread wheat is the second most economically important cereal after maize. Drought stress associated with climate change is a major cause of the yield gap in wheat production in South Africa. Drought tolerant wheat cultivars are yet to be developed and released in the country. Wheat improvement for drought tolerance is one of the major breeding goals in South Africa. Integrative pre-breeding techniques involving genotypic and phenotypic characterisation ensure an accurate selection of potential drought tolerant parents for breeding. Therefore, the specific objectives of the current study were: 1) to determine the genetic diversity and population structure of forty-seven diverse bread wheat genotypes introduced from the International Maize and Wheat Improvement Center (CIMMYT) using ten selected polymorphic Simple Sequence Repeat (SSR) markers, 2) to characterise fifteen bread wheat genotypes introduced from CIMMYT using physiological and morphological traits, and 3) to assess drought tolerance amongst fifteen selected bread wheat genotypes using nine drought tolerance indices. Genetic diversity and population structure of 47 CIMMYT derived bread wheat genotypes were examined using 10 SSR molecular markers. All the SSR markers used in the study were highly polymorphic. The highest PIC values were recorded for XGWM 132, WMS 179 and WMS 30 with 0.93, 0.89 and 0.89, respectively. Cluster analysis detected 3 distinct clusters with Clusters A and C consisting of most diverse genotypes. Two distinct heterotic patterns were identified to select unique parents for crosses. Analysis of molecular variance (AMOVA) detected significant genetic diversity among populations, among individuals and within individuals with explained percentage variance of 3%, 37% and 60%, respectively. Genetic diversity and population stratification was mainly due to private alleles detected. Based on detected genetic variability, a total of 15 genotypes were selected and subjected for phenotypic characterisation. The selected genotypes included SYM2016-037, SYM2016-038, SYM2016-029, SYM2016-010 and SYM2016-012 from Cluster A, SYM2016-044, SYM2016-004, SYM2016-016, SYM2016-019, SYM2016-014, SYM2016-008, SYM2016-006 and SYM2016-047 from Cluster B and SYM2016-042 and SYM2016-027 from Cluster C. The above selected 15 bread wheat genotypes were evaluated under field and greenhouse conditions using a randomised complete block design with 3 replications. Drought stress was imposed as follows: 1 week before 50% heading (WBH) and 1 week after 50% heading (WAH). A fully-irrigated water regime (NS, non-stress) was used as a comparative control. Genotypes were evaluated using 2 physiological and 8 morphological traits. Significant differences (P < 0.05) were detected among genotypes and genotype x test environment interaction. Genotype effect was significant for days to flowering, days to maturity, plant height, number of productive tillers, number of spikelets per spike, grain number and 100 grain weight. Genotype x test environment interaction was significant for canopy temperature, days to flowering, days to maturity, plant height, number of spikelets per spike, grain number, 100 seed weight and the yield. Significant correlations were detected between yield and days to flowering, days to maturity, plant height, number of productive tillers, number of spikelets per spike, grain number and 100 seed weight under greenhouse condition. The number of productive tillers per plant and the number of spikelets per spike were positively associated with yield under field evaluation. Principal component analysis revealed PC1 to be consistently associated with yield, 100 seed weight and number of spikelets per spike. Days to flowering and maturing, plant height and canopy temperature were positively associated with either PC2 or PC3 under greenhouse and field conditions. A yield penalty was noted for early flowering and maturing genotypes such as SYM2016-014, SYM2016-027 and SYM2016-029 relative to late flowering and maturing genotypes SYM2016-016, SYM2016-037 and SYM2016-006. Crossing of these complementary lines and continuous selection of progenies is essential to develop early maturing genotypes with stable and high yield potential. In this study, days to flowering and maturity, plant height, canopy temperature and 100 seed weight were favourable traits to screen genotypes for drought tolerance. Screening for drought tolerance under greenhouse condition was more reliable than under field evaluation. The above 15 wheat genotypes were evaluated using 9 drought tolerance indices based on yield data. The drought indices used were drought resistance (DR), mean productivity (MP), harmonic mean of yield (HM), stress susceptibility index (SSI), stress tolerance index (STI), tolerance index (TOL), yield index (YI), yield reduction index (YR) and yield stability index (YSI). Analysis of variance detected significant differences among genotypes (P < 0.001) and genotype by water regime interaction (P < 0.01) affecting yield response. Significant differences were also recorded among genotypes (P < 0.05) for DR, HM, MP, STI, YI and YSI. Consistent mean genotype ranking was recorded for HM, MP, STI, SSI and YI enabling selection of genotypes SYM2016-006, SYM2016-016 and SYM2016-037. PC analysis detected high variation of 82.2% among genotypes, with percentage variation partitioned as follows: 42.64% for PC1, 22.37% for PC2 and 12.18% for PC3. Both PC and bi-plot analyses revealed strong associations between HM, MP, STI, YI and yield under drought stressed and non-stressed conditions. High yielding genotypes such as SYM2016-006, SYM2016-016 and SYM2016-037 scored higher values for HM, MP, STI, YI and yield under drought stressed and non-stressed conditions. DR was associated with early maturing genotypes such as SYM2016-014, SYM2016-029 and SYM2016-38. These genotypes were considered as potential parents for future wheat breeding programmes emphasizing drought tolerance.