Department of Molecular and Cell Biology, University of California Berkeley, Berkeley, California, United States of America.
PLoS Genet. 2012;8(3):e1002614. doi: 10.1371/journal.pgen.1002614. Epub 2012 Mar 29.
Transcription factors have two functional constraints on their evolution: (1) their binding sites must have enough information to be distinguishable from all other sequences in the genome, and (2) they must bind these sites with an affinity that appropriately modulates the rate of transcription. Since both are determined by the biophysical properties of the DNA-binding domain, selection on one will ultimately affect the other. We were interested in understanding how plastic the informational and regulatory properties of a transcription factor are and how transcription factors evolve to balance these constraints. To study this, we developed an in vivo selection system in Escherichia coli to identify variants of the helix-turn-helix transcription factor MarA that bind different sets of binding sites with varying degrees of degeneracy. Unlike previous in vitro methods used to identify novel DNA binders and to probe the plasticity of the binding domain, our selections were done within the context of the initiation complex, selecting for both specific binding within the genome and for a physiologically significant strength of interaction to maintain function of the factor. Using MITOMI, quantitative PCR, and a binding site fitness assay, we characterized the binding, function, and fitness of some of these variants. We observed that a large range of binding preferences, information contents, and activities could be accessed with a few mutations, suggesting that transcriptional regulatory networks are highly adaptable and expandable.
（1）它们的结合位点必须具有足够的信息，以便与基因组中的所有其他序列区分开来；（2）它们必须以适当的亲和力结合这些位点，从而调节转录的速度。由于这两者都取决于 DNA 结合域的物理特性，因此对其中一个的选择最终会影响另一个。我们有兴趣了解转录因子的信息和调节性质的可塑性以及转录因子如何进化以平衡这些限制。为了研究这一点，我们在大肠杆菌中开发了一种体内选择系统，以鉴定 MarA 螺旋-转角-螺旋转录因子的变体，这些变体与不同程度简并的结合位点结合。与以前用于鉴定新的 DNA 结合物并探测结合域可塑性的体外方法不同，我们的选择是在起始复合物的背景下进行的，既要在基因组中进行特异性结合，又要保持因子功能的生理上显著的相互作用强度。我们使用 MITOMI、定量 PCR 和结合位点适合度测定法，对其中一些变体的结合、功能和适合度进行了表征。我们观察到，只需几个突变就可以获得广泛的结合偏好、信息量和活性，这表明转录调控网络具有高度的适应性和可扩展性。
