Identification of a key locus, qNL3.1, associated with seed germination under salt stress via a genome-wide association study in rice.

KEY MESSAGE: Two causal OsTTL and OsSAPK1 genes of the key locus qNL3.1 significantly associated with seed germination under salt stress were identified via a genome-wide association study, which could improve rice seed germination under salt stress. Rice is a salt-sensitive crop, and its seed germination determines subsequent seedling establishment and yields. In this study, 168 accessions were investigated for the genetic control of seed germination under salt stress based on the germination rate (GR), germination index (GI), time at which 50% germination was achieved (T50) and mean level (ML). Extensive natural variation in seed germination was observed among accessions under salt stress. Correlation analysis showed significantly positive correlations among GR, GI and ML and a negative correlation with T50 during seed germination under salt stress. Forty-nine loci significantly associated with seed germination under salt stress were identified, and seven of these were identified in both years. By comparison, 16 loci were colocated with the previous QTLs, and the remaining 33 loci might be novel. qNL3.1, colocated with qLTG-3, was simultaneously identified with the four indices in two years and might be a key locus for seed germination under salt stress. Analysis of candidate genes showed that two genes, the similar to transthyretin-like protein OsTTL and the serine/threonine protein kinase OsSAPK1, were the causal genes of qNL3.1. Germination tests indicated that both Osttl and Ossapk1 mutants significantly reduced seed germination under salt stress compared to the wild type. Haplotype analysis showed that Hap.1 of OsTTL and Hap.1 of OsSAPK1 genes were excellent alleles, and their combination resulted in high seed germination under salt stress. Eight accessions with elite performance of seed germination under salt stress were identified, which could improve rice seed germination under salt stress.

Identification and analysis of the genetic integrity of different types of rice resources through SSR markers.

Seed aging is the key factor leading to the loss of genetic integrity. In this study, the seeds of Dongxiang wild rice, Xianggu, 9194 and Nipponbare were kept in a plant incubator with constant temperature and humidity for artificial aging treatment. The genetic integrity of germplasm resources with different germination gradients were analyzed using 44 SSR markers. The results suggested that different accessions could be ranked in order of aging resistance from highest to lowest as common wild rice > Xianggu > 9194 > Nipponbare. In order to maintain the genetic diversity of rice, the population size for reproduction and regeneration should be between 60 and 140. After aging, the number of polymorphic alleles, the number of specific single plant, the ratio of polymorphic bands, the number of alleles, the number of effective alleles, gene diversity index and Shannon index of different accessions all decreased with the decrease of germination rate. The germination rate of 60% was the critical value to maintain genetic integrity. Besides, the genetic integrity of eighteen SSR markers was rapidly lost or significantly increased. The regions of these markers were closely related to seed viability or genetic integrity. This study provides a theoretical basis for determining the population size for reproduction and regeneration and the critical value of germination rate of rice resources.

Genome-wide association analysis of panicle exsertion and uppermost internode in rice (Oryza sativa L.).

BACKGROUND: Rice (Oryza sativa L.) yield is seriously influenced by panicle exsertion (PE) and the uppermost internode (UI) through panicle enclosure or energy transport during grain-filling stages. We evaluated the traits of PE and UI of 205 rice accessions in two independent environments and performed genome-wide association (GWAS) to explore the key genes controlling PE and UI, which could be used to improve panicle enclosure in rice breeding. RESULTS: In this study, extensive genetic variation was found in both PE and UI among the 205 rice accessions, and 10.7% of accessions had panicle enclosure (PE/UI ≤ 0). Correlation analysis revealed that PE was significantly positively correlated with 1000-grain weight (1000-GW) but negatively correlated with heading date (HD), and UI was significantly positively correlated with HD but no significantly correlated with 1000-GW. A total of 22 and 24 quantitative trait loci (QTLs) were identified for PE and UI using GWAS, respectively. Eight loci for PE and nine loci for UI were simultaneously detected both in 2015 and in 2016, seven loci had adjacent physical positions between PE and UI, and ten loci for PE and seven loci for UI were located in previously reported QTLs. Further, we identified the CYP734A4 gene, encoding a cytochrome P450 monooxygenase, and the OsLIS-L1 gene, encoding a lissencephaly type-1-like protein, as causal genes for qPE14 and qUI14, and for qPE19, respectively. PE and UI were both significantly shorter in these two genes' mutants than in WT. Allelic Hap.1/2/4 of CYP734A4 and Hap.1/2/4 of OsLIS-L1 increased PE, UI, PE/UI, and 1000-GW, but Hap.3 of CYP734A4 and Hap.3 of OsLIS-L1 reduced them. In addition, six candidate genes were also detected for four key novel loci, qPE16, qPE21, qUI1, and qUI18, that seemed to be related to PE and UI. CONCLUSIONS: Our results provide new information on the genetic architecture of PE and UI in rice, confirming that the CYP734A4 and OsLIS-L1 genes participate in PE and UI regulation, which could improve our understanding of the regulatory mechanism of PE and UI for rice breeding in the future.