Abstract
BACKGROUND: Conserved non-coding sequences (CNS) are islands of non-coding sequences conserved across species and play an important role in regulating the spatiotemporal expression of genes. Identification of CNS provides valuable information about potentially functional genomic elements, regulatory regions, and helps to gain insights into the genetic basis of crop agronomic traits. RESULTS: Here, we comprehensively analyze CNS in maize, by comparing the genomes of maize inbred line B73 (Zea mays ssp. mays), its close wild relative Zea mays spp. mexicana, and other grasses in Poaceae, including sorghum (Sorghum bicolor), foxtail millet (Setaria italica) and two adlay (Coix lacryma) cultivars. There were 289,931 CNS found in two syntenic gene pairs, while 51,701 CNS were conserved within at least three species. To explore the regulatory characteristics of the CNS identified, the flanking regions of CNS were compared with the peaks called using both transposase-accessible chromatin with high-throughput sequencing (ATAC-seq) and chromatin immunoprecipitation with high-throughput sequencing (ChIP-Seq) data of histone modifications. It was found that CNS in maize were enriched in open chromatin regions compared with randomly selected non-coding regions of similar length. A significant enrichment of transcription factor binding sites was found within CNS sequences, including different transcription factors involved in abiotic stress response, such as OBP (OBF-BINDING PROTEIN) family and Adof1 (Encodes dof zinc finger protein). To investigate the epigenetic modification patterns in CNS, ChIP-Seq data for histone modifications H3K9ac, H3K4me3, H3K36me3, H3K9me3, and H3K27ac were further analyzed to depict the changes along CNS. Our findings revealed significantly elevated levels of transcription-promoting histone modifications in the CNS regions compared to randomly selected non-coding sequences with an equal number and similar length. Notably, CNS were also identified on both Vgt1 (Vegetative to generative transition 1) and ZmCCT10. In addition, CNS with potential functions were identified based on SNPs within CNS significantly associated with various agronomic traits in maize, which holds potential utility in molecular breeding for maize. CONCLUSIONS: In summary, we identified and characterized CNS in maize through genomic comparative analysis, which provides valuable insights into their potential regulatory effects on gene expression and phenotypic variation.