Identification and mapping of QTL/genes for salinity tolerance in barley (Hordeum vulgare L.)

Salinity stress is one of the major abiotic stresses which affect grain yield and quality in barley. The response of plants to salinity stress presents complex quantitative traits that are affected by multiple environmental factors, involving complex physiological and molecular mechanisms. The key to salt tolerance molecular breeding is to identify and explore genes and elite alleles that control the tolerance. Salt tolerance is a developmentally regulated, stage-specific phenomenon that varies during the ontogeny of plant development. The tolerance at one stage of plant development may be poorly correlated with tolerance at other developmental stages. The mechanisms of salt tolerance at different stages of the whole ontogeny of the plant were reviewed. Barley (Hordeum vulgare L.) is the fourth largest cereal crop grown worldwide, which is extensively utilized in livestock feed, food, brewing or even medicine. As the most salinity tolerant cereal crop, barley is one of the pioneer crops grown popularly for improving saline or alkali farming land. However, many salt-tolerant QTL in barley were difficult to be used in salt-tolerant breeding due to low phenotypic contributions. In addition, until now, there are no salinity tolerance genes have been positional cloned by QTL analysis or GWAS analysis. A DH population constructed from a wild barley Tam407227 and Australian cultivar Franklin was used to map salinity tolerance QTL by scoring the plant damage under either single salinity stress or combined stress of salinity and waterlogging. Na\\(^+\\) content was measured under salinity only stress. A significant correlation was found between the damage scores under salinity stress and those under combined stress (R\\(^2\\) = 0.475). The results showed that the performance of the population under combined stress was mainly determined by salinity tolerance. Damage scores under salinity conditions showed a weak positive correlation with Na\\(^+\\) content (R\\(^2\\) = 0.146), suggesting that Na\\(^+\\) content in leaves was not the major determinant of salinity tolerance in this population. Seven QTL for salinity tolerance were identified with three major ones (qS5.1/qSW5.1, qS5.2/qSW5.2 and qS7.1/qSW7.1) being detected under both well-drained and waterlogged condition. Two QTL derived from different parents were identified for Na\\(^+\\) content in leaves. Two pairs of NILs containing qS7.1/qSW7.1 or HvNax3 only were constructed. qS7.1/qSW7.1 is a gene closely linked to HvNax3 while qNa1.1 is more likely the HvNax4. Covariate QTL analysis and correlation analysis showed that Na\\(^+\\) exclusion is not the major factor contributing to salinity tolerance in this wild barley. Different mechanisms regulate salinity tolerance under well-drained and waterlogged conditions from the wild barley. QTL mapping for salinity tolerance was also conducted in another two DH populations and genome wide association study was performed with three models GLM, MLM and FarmCPU using a natural population consisting of 445 barley accessions. A major salinity tolerant QTL named QSl.TxNn.2H, was co-localized in the two DH populations and the natural population by linkage mapping and association analysis, determining 82.7% (CG population), 26.1% (SG population) and 40% (natural population) of the phenotypic variation. Another QTL on 3H (qSSG3.1) detected by a DH population was mapped to 118.18-122.78 cM genetic interval, determining 14.40 % phenotypic variation. A significant association marker on 4H was identified using FarmCPU and GLM model, with the P value of 1.33E-16 and 3.84E-08 which is close to a salinity tolerance gene HvHKT1;5. Two pairs of NILs (N33 and N53 are salinity sensitive lines, T46 and T66 are salinity tolerance lines) differing in QSl.TxNn.2H were used to analyze the differences of ion homeostasis, proline content and scavenging capacity of reactive oxygen species (ROS) in roots and leaves after 0, 2, 4 and 6 days salinity application. The roots and leaves of one pair of NILs (N33 and T46) were used for proteomic analysis under normal conditions and after exposure to 300mM NaCl for 96 hours. Tolerant NILs (T46 and T66) showed salinity tolerance with little symptoms of leaf chlorosis or wilting, while sensitive ones, N33 and N53, showed salinity sensitivity with severe chlorosis and a low survival rate. Both tolerant NILs showed a significantly lower Na\\(^+\\)/K\\(^+\\) ratio, H\\(_2\\)O\\(_2\\) MDA, and proline but a greater enhancement in antioxidant enzymatic activities than both sensitive NILs. A total of 53 and 51 differentially expressed protein spots were identified through tandem mass spectrometry analysis in the leaves and roots, respectively. Proteins that are associated with photosynthesis, ROS scavenging, and ATP synthase were found to be specifically upregulated in the tolerant NILs. It is suggested that the tolerance allele of the QTL QSl.TxNn.2H improves salinity tolerance by controlling Na\\(^+\\) loading into xylem through Ca\\(^{2+}\\) signal. This signaling network involves reductions of Na\\(^+\\) toxicity in leaves, upregulations of proteins related to photosynthesis, ROS scavenging, and ATP synthase to protect the photosynthetic apparatus, thus alleviating oxidative damage, providing additional energy which contributes to the salinity tolerance. To assist in the understanding of salinity tolerance mechanism and identification of candidate genes, the roots of a pair of near-isogenic lines (NILs) containing QSl.TxNn.2H were analyzed by RNA-seq after exposing the plants to 300 mM NaCl for 48h. In total, 1256 and 809 differentially expressed genes were identified in the salt-tolerant and sensitive lines, T46 and N33, respectively. Of these genes, 572 were specifically up-regulated and 326 down-regulated in the salt-tolerant line T46, while 170 genes were specifically up-regulated and 281 down-regulated in the sensitive line N33. There are 146 genes in the QSl.TxNn.2H mapping region. The candidate gene for QSl.TxNn.2H may confer barley salinity tolerance by participating in Ca\\(^{2+}\\) signaling and hormone metabolism, maintaining the integrity of cell wall, regulating ion homeostasis, participating in lipid metabolism and regulating nitrogen and sugar transportation. The F\\(_2\\) population developed from N33 and T46 was used for fine mapping of QSl.TxNn.2H. This QTL has been mapped to a 215kb physical distance, containing a total of 21 predicted genes using 3000 F\\(_2\\) and F\\(_3\\) population lines. In the fine mapped region, the most significant association SNPs SNP-9538933 were located within a gene, which encodes a lectin receptor kinase. Among the 21 genes in the fine-mapping region, a gene encoding heat shock protein 90 was significantly up-regulated in both N33 and T46, but the tolerant line T46 showed a higher expression than N33. The genes encoding lectin receptor kinase and heat shock protein 90 were two possible candidate genes for QSl.TxNn.2H.