Combining in vitro peptide display and in vivo directed evolution to search for entirely de novo folds of DNA-binding domains

本项目将利用蛋白质进化方法的最新进展，开发全新的DNA结合域，为基因调控中新的合成转录因子奠定基础。即使考虑到最近在蛋白质结构确定方面的计算进展，开发新的自然蛋白质功能仍然是一个重大挑战。为了解决这个问题，我们将采用一种实验方法，结合CIS肽的体外表达和大量文库的选择（Patel等人）。蛋白质工程，设计与选择26,307-315,2013)，与噬菌体为基础的体内选择(Brodel等。自然通讯7,13858,2016)，它给出了基因选择特性的指数扩增。在66个氨基酸中，lambda cro可能是已知的在大肠杆菌中起作用的最小的转录因子，它可以转化为激活剂（Brodel等人）。科学进展6:eaba2728, 2020)。因此，提供一个编码70 - 100 aa的非结构化基因序列，原则上可以为转录因子的从头进化提供一个合理的起点。~1091 - 10130个氨基酸组合看起来令人望而生畏，但这也是为什么问进化的复合优势（我们将采用）是否会导致指数选择轨迹是一个有趣的问题。该项目将采取分阶段的方法，以确保项目期间技术的可行进展和发展。最初，我们将基于熟悉的DNA调控元件（例如helix-turn-helix）筛选大型结构化但非DNA结合序列文库，使用体外CIS显示。随后，这将扩展到创建大量的非结构化蛋白质文库，以筛选新的DNA结合支架。从这些研究中获得的DNA结合域将被开发为转录因子，使用进一步的蛋白质工程，使用基于噬菌体的微生物基因驱动在体内加速进化，以及通过基因定向DNA损伤加速蛋白质进化的新技术。结果将通过生化、计算和结构方法进一步分析。

This project will take recent advances in protein evolution methods to develop entirely novel DNA binding domains as a basis for new synthetic transcription factors in genetic regulation. The development of new-to-nature protein functionality remains a significant challenge even given recent computational advances in protein structure determination. To address this we will take an experimental approach that combines CIS peptide display for the in vitro expression and selection of massive libraries (Patel et al. Protein Engineering, Design & Selection 26, 307-315, 2013), with phage-based in vivo selection (Brodel et al. Nature Communications 7, 13858, 2016), which gives exponential amplification of genetically selectable traits.At 66 amino acids, lambda cro is perhaps the smallest known transcription factor that functions in E. coli, and it can be converted into an activator (Brodel et al. Science Advances 6: eaba2728, 2020). Therefore, providing an unstructured gene sequence coding for 70 - 100 aa, could in principle provide a reasonable starting point for de novo evolution of a transcription factor. The ~1091 - 10130 amino acid combinations seem daunting, but that is also why it is a fascinating question to ask whether the compounding advantages of evolution, which we will employ, will lead to an exponential selection trajectory. The project will take a staged approached to ensure viable progression and development of the techniques during the project. Initially, we will screen large libraries of structured but non-DNA binding sequences based on familiar DNA regulatory elements (e.g. helix-turn-helix), using in vitro CIS display. Subsequently, this will be expanded to create massive libraries of unstructured proteins to screen for novel DNA binding scaffolds. DNA binding domains from these studies will then be developed as transcription factors using further protein engineering and in vivo accelerated evolution using a phage-based microbial gene drive as well as a novel technique for accelerating protein evolution through gene-directed DNA damage. The outcomes will be further analysed through biochemical, computational and structural approaches.