Transcriptional multiomics reveals the mechanism of seed deterioration in Nicotiana tabacum L. and Oryza sativa L.

Introduction: Mature seeds deteriorate gradually and die eventually during long-term storage. Controlled deterioration is often used to accelerate the seed deterioration rate to assess the seed vigor and physiological quality of seed lots.

Objectives: Although it is well known that the process of seed deterioration produced by controlled deterioration is distinct from that caused by long-term storage, the differences in transcriptional levels have not been reported. Clarifying the mechanism of seed deterioration is critical for identifying, conserving and utilizing germplasm resources.

Methods: Tobacco (Nicotiana tabacum L.) seeds were studied thoroughly using transcriptome, small RNA, and degradome sequencing after long-term storage (LS) and controlled deterioration (CD). Co-expression trend analysis identified transcripts involved in tobacco seed deterioration, while phylogenetic analysis helped to uncover comparable targets in rice (Oryza sativa L.) for further verification and utilization.

Results: In LS and CD, a total of 2,112 genes and 164 miRNAs were differentially expressed, including 20 interaction miRNA-mRNA pairs with contrasting expression. Transcriptional multiomics found that the main causes of LS were plant hormone signal transduction and protein processing in the endoplasmic reticulum, whereas the primary cause of CD was nucleotide excision repair dysfunction. The homeostatic balance of RNA degradation and the spliceosome occurred in both modes of seed deterioration. Additionally, co-expression trend analysis identified two coherent pairs, nta-miR160b-NtARF18 and nta-miR396c-NtMBD10, as being significant in LS and CD, respectively. For utilization, rice homologous targets OsARF18 and OsMBD707 were verified to play similar roles in LS and CD, respectively.

Conclusion: This study demonstrated the transcriptional mechanism of tobacco and key genes in seed deterioration. And the application of key genes in rice also verified the feasibility of the multiomics method, guiding the identification of candidate genes to precisely delay seed deterioration in other species of seed research.