Identification of genes involved in spikelet fertility in rice using RNA sequencing

The identification of genes controlling spikelet fertility is directly important for the yield and quality of rice. This study developed two groups of genetically stable Recombinant Inbred Lines (RILs) with significantly different seed set rates. One group had a high seed set rate (≥ 70%) and the other had a low seed set rate (≤ 25%). These populations were created by crossing the Japanese rice variety Nipponbare with the Thai rice variety Khao Dawk Mali 105. To identify genes, the study examined the stages and tissues affecting the different seed set rates in the two groups. In the F5 generation, the populations with different seed set rates still had complete pollen but varied significantly in seed set rates. The study further observed differences in pollen adhesion and pollen tube growth abnormalities in the low seed set population 4-5 hours after pollination. Samples were collected, and RNA sequencing was conducted at this stage in the F6 generation. Analysis of the differentially expressed genes (DEGs) between the high and low seed set populations identified 249 DEGs in the high seed set group, with 118 up-regulated and 131 down-regulated genes. The low seed set group had 473 DEGs, with 214 up-regulated and 259 down-regulated genes. Validation of 10 DEGs using qRT-PCR showed consistent results with the RNA sequencing data. GO ontology analysis of DEGs in the high seed set group identified 18 GO terms, while the low seed set group had 16 GO terms. KEGG pathway analysis revealed that 11 DEGs in the high seed set group were involved in metabolic pathways, whereas 6 DEGs in the low seed set group were related to pentose and glucuronate interconversions and flavonoid biosynthesis pathways. Further study of DEGs interaction with other genes involved in pollen development, using rice databases, found several interactions, including some previously reported genes related to spikelet fertility. This study discovered new genes and gene functions associated with spikelet fertility, such as Os02g0467600 (cinnamate 4-hydroxylase, CYP73), which is involved in flavonoid synthesis and is expressed more in the high seed set group post-pollination. Additionally, OsPME35 and OsPLL9, which control pectin methylesterase (PME) and pectate lyase (PLL) respectively, showed higher expression in the ovary before pollination in the high seed set group compared to the low seed set group. This indicates their necessity for cell wall formation of the pistil during the interaction between pistil and pollen. The DEGs identified were also located near QTLs related to pollen tube growth and yield components, as previously reported. These findings are valuable as they provide new knowledge about the genetic control of spikelet fertility and can be utilized for developing high-yield rice varieties.