A holistic approach to rapid protein engineering: the synergistic combination of optimised protein library generation and screening.

这项建议旨在开发和结合‘MAX’和‘ProxiMAX’随机化技术与Isgenica的专利CIS筛选技术，以产生一个无缝平台技术，同时解决蛋白质工程的所有关键问题。这两种MAX随机化技术都是在BBSRC的支持下(分别获得B14245和BB/D525756/1)和随后的内部资金支持下由阿斯顿大学开发的。由Isgenica开发的顺式筛选代表了筛选工程蛋白质文库方面的最新技术。目前高通量蛋白质工程的方法要么优化突变，要么优化筛选。很少有人同时优化两者(而那些做到这一点的人，则依赖于复杂、昂贵的化学合成库)。该项目将代表简单、优化的突变和筛选的第一个组合，其中最大限度地减少文库大小，同时最大限度地提高筛选能力，从而使基于体外的简单技术能够实现比以前更大的多样性。此外，该项目将不使用模型蛋白质，而是使用多肽靶标来开发当前学术/商业相关性。从概念上讲，蛋白质/肽是一串折叠成3D结构的氨基酸，定义了活性。如果一种或多种氨基酸发生改变，蛋白质的活性可能不会改变、取消或改变。后者是蛋白质工程的目标：产生具有新的或增强活性的合成蛋白质。在商业上，蛋白质工程是以高通量进行的，要么通过饱和突变(在有结构数据的情况下)，要么通过基因改组(在缺乏信息的情况下)，或者通过这两种方法的组合。无论采用哪种方法，结果都是由一个基因的变体编码的蛋白质文库。在饱和突变中，这种变异仅限于特定密码子(S)，代之以随机密码子(S)，如NNN或NNG/T(N为任意核苷酸)。一旦表达出来，就会对文库进行筛选，以找到具有新的、所需活性的“最佳”蛋白质。不幸的是，饱和突变有缺点(与遗传密码退化有关)，包括高损失率和不相等的蛋白质浓度，导致蛋白质文库可能危及筛选过程。“MAX”和“ProxiMAX”随机化是两种技术，它们都解决了饱和突变的陷阱，使组合蛋白质工程成为可能，而这在以前是不可能的，除非使用高度专业化的化学。通过消除文库中的遗传密码退化，他们提供了小型文库，这些文库每次都应该产生“最好的”蛋白质，从而极大地降低了筛选成本。“MAX”被开发用来设计蛋白质，其中关键残基被顺序分离(例如，酶活性部位内的残基)，而“ProxiMAX”被用来随机化连续的氨基酸。通过消除克隆的需要，CIS筛选消除了蛋白质文库的毒性效应和克隆偏见，并将可实现的最大文库大小增加了4-5个数量级。基因在体外表达，作为REPA蛋白的融合。DNA表达盒还含有CIS和ORI位点，它们共同导致每个新表达的REPA融合与其编码的DNA结合。这种DNA(成功的蛋白质)是通过聚合酶链式反应扩增出来的。只有在经过几轮“淘洗”之后，才能克隆出为数不多的盒式磁带，并对它们的蛋白质进行鉴定/过度表达。在整合这两项广泛兼容的技术时，我们的目标是同时设计神经生长因子的多肽调节剂(神经生长因子)和降钙素基因相关多肽的多肽变体(阿斯顿大学特别感兴趣的降钙素基因相关多肽)。主要的研究挑战将包括在“MAX”/“ProxiMAX”随机化中引入程序化偏见、技术兼容性的开发以及特定多肽产品的工程设计。

This proposal aims to develop and combine 'MAX' and 'ProxiMAX' randomisation technologies with Isogenica's proprietary CIS screening technology, to generate a seamless platform technology that addresses all key issues of protein engineering simultaneously. Both 'MAX' randomisation technologies were developed at Aston University with BBSRC support (grants B14245 and BB/D525756/1 respectively) and subsequent in-house funding. CIS screening, developed by Isogenica, represents the state-of-the-art in terms of screening engineered protein libraries. Current approaches to high-throughput protein engineering optimise either mutagenesis, or screening. Few optimise both (and those that do, rely on complex, expensive, chemical synthesis of libraries). This project will represent the first combination of simple, optimised mutagenesis AND screening, where library size is minimised, whilst screening capacity is maximised, thus enabling diversity greater than previously achievable, from simple, in vitro-based technology. Moreover, rather than utilising model proteins, the project will employ peptide targets for development of current academic/commercial relevance. Conceptually, a protein/peptide is a string of amino acids folded into a 3D structure that defines activity. If one or more amino acids are changed, protein activity may be unchanged, abolished or altered. The latter is protein engineering's goal: to generate synthetic proteins with novel or enhanced activities. Commercially, protein engineering is performed in high-throughput, either by saturation mutagenesis (where structural data is available); gene shuffling (where information is lacking) or by a combination of the two approaches. Whichever is employed, a protein library results that is encoded by variants of just one gene. Within saturation mutagenesis, that variation is limited to specific codon(s), replaced with randomised codon(s) such as NNN or NNG/T (N is any nucleotide). Once expressed, the library is screened to find the 'best' protein with the new, required activity. Unfortunately, saturation mutagenesis has drawbacks (associated with genetic code degeneracy) including high wastage ratios and unequal protein concentrations resulting in protein libraries that are likely to compromise screening processes. 'MAX' and 'ProxiMAX' randomisation are technologies that each address the pitfalls of saturation mutagenesis, enabling combinatorial protein engineering that was previously impossible, except by using highly-specialised chemistry. By eliminating genetic code degeneracy from the libraries, they deliver small libraries that should produce the 'best' protein every time, for vastly-reduced screening costs. 'MAX' was developed to engineer proteins where the key residues are sequentially separated (e.g. the residues within the active site of an enzyme), whereas 'ProxiMAX' is used to randomise contiguous amino acids. By removing the need for cloning, CIS screening eliminates toxicity effects and cloning bias from protein libraries and increases the maximum achievable library size by 4-5 orders of magnitude. Genes are expressed in vitro, as fusions to RepA protein. The DNA expression cassettes also contain CIS and ori sites which together, cause each newly-expressed RepA fusion to bind to its encoding DNA. This DNA (successful proteins) is amplified by PCR. Only after several cycles of 'panning' are the few resulting cassettes cloned and their proteins identified / over-expressed. In integrating these two broadly-compatible technologies, we aim to engineer both peptide modulators of Nerve Growth Factor (NGF - of particular interest to Isogenica); and peptide variants of Calcitonin Gene Related Peptide (CGRP - of particular interest to Aston University). Key research challenges will include the introduction of programmed bias into 'MAX'/'ProxiMAX' randomisation, development of technology compatibility and engineering of specific peptide products.