Babu M Madan, Balaji S, Aravind L
National Center for Biotechnology Information, National Institutes of Health, Bethesda, Md., USA.
Genome Dyn. 2007;3:66-80. doi: 10.1159/000107604.
Gene expression in organisms is controlled by regulatory proteins termed transcription factors, which recognize and bind to specific nucleotide sequences. Over the years, considerable information has accumulated on the regulatory interactions between transcription factors and their target genes in various model prokaryotes, such as Escherichia coli and Bacillus subtilis. This has allowed the representation of this information in the form of a directed graph, which is commonly referred to as the transcriptional regulatory network. The network representation provides us with an excellent conceptual framework to understand the structure of the transcriptional regulation, both at local and global levels of organization. Several studies suggest that the transcriptional network inferred from model organisms may be approximated by a scale-free topology, which in turn implies the presence of a relatively small group of highly connected regulators (hubs or global regulators). While the graph theoretical principles have been applied to infer various properties of such networks, there have been few studies that have actually investigated the evolution of the transcriptional regulatory networks across diverse organisms. Using recently developed computational methods that exploit various evolutionary principles, we have attempted to reconstruct and compare these networks across a wide-range of prokaryotes. This has provided several insights on the modification and diversification of network structures of various organisms in course of evolution. Firstly, we observed that target genes show a much higher level of conservation than their transcriptional regulators. This in turn suggested that the same set of functions could be differently controlled across diverse organisms, contributing significantly to their adaptive radiations. In particular, at the local level of network structure, organism-specific optimization of the transcription network has evolved primarily via tinkering of individual regulatory interactions rather than whole scale reuse or deletion of network motifs (local structure). In turn, as phylogenetic diversification proceeds, this process appears to have favored repeated convergence to scale-free-like structures, albeit with different regulatory hubs.
生物体中的基因表达由称为转录因子的调节蛋白控制，这些转录因子识别并结合特定的核苷酸序列。多年来，在各种模式原核生物（如大肠杆菌和枯草芽孢杆菌）中，转录因子与其靶基因之间的调节相互作用积累了大量信息。这使得这些信息能够以有向图的形式呈现，通常称为转录调控网络。网络表示为我们提供了一个极好的概念框架，用于在局部和全局组织水平上理解转录调控的结构。几项研究表明，从模式生物推断出的转录网络可能近似于无标度拓扑结构，这反过来意味着存在相对较少的一组高度连接的调节因子（枢纽或全局调节因子）。虽然图论原理已被应用于推断此类网络的各种属性，但实际上很少有研究调查转录调控网络在不同生物体中的进化。利用最近开发的利用各种进化原理的计算方法，我们试图在广泛的原核生物中重建和比较这些网络。这为各种生物体在进化过程中网络结构的修改和多样化提供了一些见解。首先，我们观察到靶基因的保守程度远高于其转录调节因子。这反过来表明，同一组功能在不同生物体中可能受到不同的控制，这对它们的适应性辐射有很大贡献。特别是，在网络结构的局部水平上，转录网络的生物体特异性优化主要是通过调整个体调节相互作用而不是通过整个网络基序（局部结构）的重复使用或删除来实现的。反过来，随着系统发育多样化的进行，这个过程似乎有利于反复收敛到类似无标度的结构，尽管具有不同的调节枢纽。
