Structure and evolution of gene regulatory networks in microbial genomes

Abstract

With the availability of genome sequences for hundreds of microbial genomes, it has become possible to address several questions from a comparative perspective to understand the structure and function of regulatory systems, at least in model organisms. Recent studies have focused on topological properties and the evolution of regulatory networks and their components. Our understanding of natural networks is paving the way to embedding synthetic regulatory systems into organisms, allowing us to expand the natural diversity of living systems to an extent we had never before anticipated.

Fig. 1

Genomic view of a gene regulatory network. Transcriptional regulation from a genomic perspective can be viewed as a complex interplay between cis-regulatory elements on the DNA and trans-acting factors like TFs and sigma factors. Depending on the set of input signals (extracellular or intracellular in nature), upon a cascade of events, transcription factors positively or negatively control transcription of the target gene. Often, transcription factors work in a combinatorial fashion to produce the final output. Protein products formed after transcription and translation of the gene regions are responsible for various cellular functions and ultimately feedback transcription factors at several levels to control their own transcription.

Fig. 2

Network view of transcription regulation. Transcriptional regulatory interactions at a genomic level can be visualized as a network between TFs (shown in red) and target genes (shown in green). a) The transcriptional regulatory network is a multi-layer hierarchical modular structure without feedback regulation at the transcription level [38; 68], with the global regulators at the top of this layout and local TFs at the bottom, regulating a few genes. b) Modules are interconnected clusters which divide the network of transcriptional interactions into subnetworks. Modules have been identified using a variety of approaches [17; 38; 57] and have been found to be semi-independent in nature. Modules are in turn formed by one or more different types of network motifs. c) Motifs are patterns of interconnections which are overrepresented in transcriptional networks. Known transcriptional regulatory networks were found to have feed forward loops (FFLs), multiple input modules (MIMs) and SIMs, with each kind of motif playing a different role [1].

Fig. 3

Different subtypes of feed forward loop motifs. Directed arrow represents positive regulation and is shown in green, while negative transcriptional control is shown in red.