Whole genome co-expression analysis of soybean cytochrome P450 genes identifies nodulation-specific P450 monooxygenases

Background: Cytochrome P450 monooxygenases (P450s) catalyze oxidation of various substrates using oxygen and NAD(P)H. Plant P450s are involved in the biosynthesis of primary and secondary metabolites performing diverse biological functions. The recent availability of the soybean genome sequence allows us to identify and analyze soybean putative P450s at a genome scale. Co-expression analysis using an available soybean microarray and Illumina sequencing data provides clues for functional annotation of these enzymes. This approach is based on the assumption that genes that have similar expression patterns across a set of conditions may have a functional relationship. Results: We have identified a total number of 332 full-length P450 genes and 378 pseudogenes from the soybean genome. From the full-length sequences, 195 genes belong to A-type, which could be further divided into 20 families. The remaining 137 genes belong to non-A type P450s and are classified into 28 families. A total of 178 probe sets were found to correspond to P450 genes on the Affymetrix soybean array. Out of these probe sets, 108 represented single genes. Using the 28 publicly available microarray libraries that contain organ-specific information, some tissue-specific P450s were identified. Similarly, stress responsive soybean P450s were retrieved from 99 microarray soybean libraries. We also utilized Illumina transcriptome sequencing technology to analyze the expressions of all 332 soybean P450 genes. This dataset contains total RNAs isolated from nodules, roots, root tips, leaves, flowers, green pods, apical meristem, mock-inoculated and Bradyrhizobium japonicum-infected root hair cells. The tissue-specific expression patterns of these P450 genes were analyzed and the expression of a representative set of genes were confirmed by qRT-PCR. We performed the co-expression analysis on many of the 108 P450 genes on the Affymetrix arrays. First we confirmed that CYP93C5 (an isoflavone synthase gene) is co-expressed with several genes encoding isoflavonoid-related metabolic enzymes. We then focused on nodulation-induced P450s and found that CYP728H1 was co-expressed with the genes involved in phenylpropanoid metabolism. Similarly, CYP736A34 was highly co-expressed with lipoxygenase, lectin and CYP83D1, all of which are involved in root and nodule development. Conclusions: The genome scale analysis of P450s in soybean reveals many unique features of these important enzymes in this crop although the functions of most of them are largely unknown. Gene co-expression analysis proves to be a useful tool to infer the function of uncharacterized genes. Our work presented here could provide important leads toward functional genomics studies of soybean P450s and their regulatory network through the integration of reverse genetics, biochemistry, and metabolic profiling tools. The identification of nodule-specific P450s and their further exploitation may help us to better understand the intriguing process of soybean and rhizobium interaction.