Transcription factor modularity in a gene-centered C. elegans core neuronal protein–DNA interaction network

Abstract

Transcription regulatory networks play a pivotal role in the development, function, and pathology of metazoan organisms. Such networks are comprised of protein-DNA interactions between transcription factors (TFs) and their target genes. An important question pertains to how the architecture of such networks relates to network functionality. Here, we show that a Caenorhabditis elegans core neuronal protein-DNA interaction network is organized into two TF modules. These modules contain TFs that bind to a relatively small number of target genes and are more systems specific than the TF hubs that connect the modules. Each module relates to different functional aspects of the network. One module contains TFs involved in reproduction and target genes that are expressed in neurons as well as in other tissues. The second module is enriched for paired homeodomain TFs and connects to target genes that are often exclusively neuronal. We find that paired homeodomain TFs are specifically expressed in C. elegans and mouse neurons, indicating that the neuronal function of paired homeodomains is evolutionarily conserved. Taken together, we show that a core neuronal C. elegans protein-DNA interaction network possesses TF modules that relate to different functional aspects of the complete network.

Computational and Systems Biology

“…These were mostly worm transcription factors (855), 26 but also included proteins involved in signal transduction (187) and chromatin modification (6) and proteins that have been shown to bind DNA in yeast one-hybrid assays (15). 15,17 Each of the 248 coexpression modules received on average four predicted regulators with a weight of 100 or higher; only one module had none and the maximum number of regulators per module was eight. In total, 426 regulators were uniquely assigned to coexpression modules, 49% of which were only assigned to one module and less than one percent was assigned to 10 modules or more.…”

“…These data consisted of 598 yeast one-hybrid protein-DNA interactions, 15,17,26 and 525 Wormbase WS195 regulatory interactions. The first are direct regulator-target binding interactions, while for the latter this is not necessarily the case.…”

“…37 From the partition of experiments, we observed major differences in gene expression between the conditions of experiment 6 and Two genes in this module, bbs-5 and dyf-11, are reported to be regulated by DAF-19 (WormBase), whose promoter is bound by DAF-3 in yeast one-hybrid assays, which in turn is a direct target for UNC-42 in yeast one-hybrid assays. 17 UNC-42 is bound by ODR-7, 17 which is regulated by LIN-11 (WormBase). CLR predicts ODR-3 as regulator for several genes in module 108 (dashed-dotted line).…”

“…While a genome-wide collection of transcription regulatory interactions has been reported for E. coli and S. cerevisiae, 11,12 transcription regulatory networks for metazoan organisms, including C. elegans, have been mapped experimentally only to a limited extent. In C. elegans, direct transcription factor regulatory interaction networks have been identified through chromatin immunoprecipitation, 13,14 and yeast one-hybrid experiments, [15][16][17] while indirect regulatory interaction networks have mainly been identified through combinations of gene knockout or knockdown and gene reporter, RT-PCR or microarray experiments. [18][19][20][21] In the latter case, cis-regulatory motif finding algorithms have been applied in order to pinpoint the transcription factors directly responsible for the change in gene expression.…”

“…24 WormNet is a probabilistic functional gene network of protein-encoding genes of C. elegans, constructed using a modified Bayesian integration of many different data types from several different organisms, with each data type weighted according to how well it links genes that are known to function together in C. elegans. Next, we compared the predicted transcription regulatory interactions to a set of known regulatory interactions in C. elegans, 15,17,25,26 and to the predicted interactions obtained by another reverseengineering method, the context likelihood of relatedness (CLR) algorithm. 2 CLR is a direct reverse-engineering method that scores all possible pairwise regulator-target gene interactions based on the mutual information of their expression profiles as compared to an interaction specific background distribution.…”