Mining of co-expression genes in response to cold stress at maize (Zea mays L.) germination and sprouting stages by weighted gene co-expression networks analysis.

Abstract

Background: Maize (Zea mays L.) is one of the main agricultural crops with the largest yield and acreage worldwide. Maize at the germination and sprouting stages are highly sensitive to low-temperatures, especially in high-latitude and high-altitude regions. Low-temperature damage in early spring presents a major meteorological disaster in maize, severely affecting plant growth and maize yield. Therefore, mining genes tolerant to low temperatures is crucial. We aimed to analyze differential gene expression and construct co-expression networks in maize under low temperatures.

Methods: Inbred lines, Zhongxi 091/O2 and Chang 7-2, are tolerant and sensitive to low temperatures at the germination and sprouting stages, respectively. We grew these lines at 10 °C and 2 °C at the germination and sprouting stages, respectively. Samples were taken at five time points (0, 6, 12, 24, and 36 h) during the two stages, and transcriptome sequencing was performed. The analyses were conducted using weighted gene co-expression networks analysis (WGCNA), Gene Ontology (GO), the Kyoto Encyclopedia of Genes and Genomes (KEGG), and gene co-expression networks.

Results: WGCNA was used to construct co-expression networks at two stages, resulting in six and nine co-expression modules, respectively. Two modules at the germination stage (blue and yellow) and two modules at the sprouting stage (turquoise and magenta) were identified. These were significantly associated (p < 0.01) with tolerance at low temperature. The differentially expressed genes (DEGs) in the four modules revealed entries related to hormone and oxygen-containing compound responses by GO functional enrichment. Among the four modules, DEGs from three modules were all significantly enriched in the MAPK signaling pathway. Based on the connectivity, the top 50 genes for each module were selected to construct a protein interaction network. Seven genes have been proven to be involved in the response to low-temperature stress.

Conclusion: WGCNA revealed the differences in the response patterns of genes to low-temperature stress between tolerant and sensitive lines at different time points. Seven genes involved in low-temperature stress were functionally annotated. This finding suggests that WGCNA is a viable approach for gene mining. The current findings provide experimental support for further investigation of the molecular mechanisms underlying tolerance to low temperatures in maize.