Analysis of gene expression data generated by high-throughput microarray transcript profiling experiments coupled with cis-regulatory elements enrichment study and cluster analysis can be used to define modular gene programs and regulatory networks. Unfortunately, the high molecular weight glutenin subunits of wheat (Triticum aestivum) are more similar than microarray data alone would allow to distinguish between the three homoeologous gene pairs. However, combining complementary DNA (cDNA) expression libraries with microarray data, a co-expressional network was built that highlighted the hidden differences between these highly similar genes. Duplex clusters of cis-regulatory elements were used to focus the co-expressional network of transcription factors to the putative regulatory network of Glu-1 genes. The focused network helped to identify several transcriptional gene programs in the endosperm. Many of these programs demonstrated a conserved temporal pattern across the studied genotypes; however, few others showed variance. Based on this network, transient gene expression assays were performed with mutated promoters to inspect the control of tissue specificity. Results indicated that the interactions of the ABRE│CBF cluster with distal promoter regions may have a dual role in regulation by both recruiting the transcription complex as well as suppressing it in non-endosperm tissue. A putative model of regulation is discussed.
Keywords: Co-expression network; Endosperm; Gene program; High molecular weight glutenin; Promoter; Regulation; Tissue specificity; Transient gene assay; Triticum aestivum; cis-regulatory module.