Expanding the biochemical toolbox for protein modification at cysteine

几十年来，半胱氨酸硫醇优越的亲核性，相对于存在于蛋白质原氨基酸中的其他功能，已被认识并利用来实现选择性蛋白质修饰。然而，由于难以区分具有相似可用性的游离半胱氨酸残基，特别是在表达和/或未折叠的多肽链中，半胱氨酸定向生物学研究仍有一些未开发的机会。为了充分利用半胱氨酸可以提供的靶向潜力，必须能够区分肽序列内的各种半胱氨酸残基。由于n- epsilon-乙酰赖氨酸和2-乙酰氨基甲基半胱氨酸（Cys(Acm)）的结构相似，我们旨在从Methanosarcina barkeri （Mb）中开发正交t-RNA/RNA合成酶对，用于将n- epsilon-乙酰赖氨酸结合到响应琥珀色（TAG）停止密码子的表达蛋白中，以引入Cys（Acm）。Cys（Acm）与大多数常规的肽转化兼容，重要的是与金属游离去硫化（MFD）兼容，但在温和的条件下被切割，能够区分可用的半胱氨酸残基，而不管它们的序列背景如何。这种保护策略目前只适用于合成肽化学，但在这里，我们的目标是使用完全重组的前体，可以从可再生资源中经济有效地产生。最初，我们将随机选取靠近底物的Mb tRNA合成酶残基，从而产生突变酶库。那些能够成功地用Cys（Acm）给Mb tRNA充电的细胞将通过既定的遗传选择进行扩增和分离。在这一进展的同时，我们还致力于合成我们的蛋白质目标，尽管是以简化的形式，以优化“后期”修饰和肽/蛋白质重折叠。而不是作为一个不必要的重复工作，合成工作的目的是为蛋白质连接领域的创新创造新的机会。首先，保护/靶向策略将首先应用于模型治疗蛋白。用2 × Cys（Acm）表达该蛋白，并用半胱氨酸取代单个糖基化位点，可以用糖溴乙酰胺明确靶向糖基化位点。在糖基化和去保护Acm基团之后，蛋白质将被氧化重新折叠。接下来，该过程将应用于设计的四肽重复序列（TPR’s）。这些设计的重复蛋白的从头到尾寡聚化已被证明可以产生多种蛋白质相互作用支架。然而，现有的策略受到不断生长的多肽与蛋白酶的反复暴露以及每个天然化学连接（NCL）步骤引入一个新的半胱氨酸残基的限制，这导致不必要的二硫键形成。在引入单个n端Acm保护的半胱氨酸残基之后，迭代的NCL/MFD反应消除了重复暴露于蛋白酶的需要，并防止了在寡聚化过程中游离半胱氨酸残基的干扰。最后，受保护的Cys残基将被引入单链胰岛素类似物。无论胰岛素是化学还是生物产生的，影响效率的主要因素是形成二硫键的氧化步骤。因此，为了形成一个关键的折叠中间体，一对半胱氨酸残基（对应于A7-B7）将以Acm保护形式表达。折叠将使用包括核磁共振和CD光谱学在内的标准生物物理技术进行检查，并与合成对口物进行比较。每个模型研究的成功完成都是重要的原理证明，并突出了这种蛋白质工具箱的新成员的多功能性和丰富的潜在应用。

For decades the superior nucleophilicity of the cysteine thiol, relative to other functionality present in proteinogenic amino acids, has been recognised and exploited to enable selective protein modification. However, there are still several untapped opportunities for cysteine-directed biological investigations as a consequence of the difficulty in distinguishing between free cysteine residues of similar availability, particularly in an expressed and/or unfolded polypeptide chain. In order to take full advantage of the targeting potential that cysteine can provide, it is essential that the various cysteine residues within a peptide sequence can be distinguished from each other. Due to the structural similarity between N-epsilon-acetyllysine and 2-acetamidomethyl cysteine (Cys(Acm)) we aim to develop the orthogonal t-RNA/RNA synthetase pair from Methanosarcina barkeri (Mb), employed to incorporate N-epsilon-acetyllysine into expressed proteins in response to an amber (TAG) stop codon, to introduce Cys(Acm). Cys(Acm) is compatible with most routine peptide transformations, importantly with metal free dethiylation (MFD), yet is cleaved under mild conditions, enabling discrimination between available cysteine residues regardless of their sequence context. This protection strategy is currently only available in synthetic peptide chemistry, yet here we aim to use fully recombinant precursors that can be generated cost-effectively from renewable resources.Initially we will randomise Mb tRNA synthetase residues that are proximal to the substrate giving rise to a library of mutant enzymes. Those which can successfully charge Mb tRNA with Cys(Acm) will be amplified and isolated using an established genetic selection. While this progresses we additionally aim to synthesise our protein targets, albeit in simplified form, in order to optimise "late-stage" modification and peptide/protein refolding. Rather than serving as a needless duplication of effort, the synthetic work is designed to create new opportunities for innovation in the protein ligation arena. First, the protection/targeting strategy will first be applied to a model therapeutic protein. Expression of this protein with 2 x Cys(Acm) and a single glycosylation site replaced with cysteine allows unambiguous targeting of the glycosylation site with sugar bromoacetamides. Following glycosylation and deprotection of the Acm groups the protein will be oxidatively refolded. Next the process will be applied to designed tetratricopeptide repeats (TPR's). Head-to-tail oligomerisation of these designed repeat proteins has been shown produce a diverse range of protein-protein interaction scaffolds. However the existing strategy is limited by repeated exposure of the growing polypeptide to proteases, and the introduction of one new cysteine residue per native chemical ligation (NCL) step which results in unwanted disulphide bond formation. Following introduction a single N-terminal Acm protected cysteine residue, iterative NCL/MFD reactions obviates the need for repeated exposure to proteases and prevents interference from free cysteine residues as the oligomerisation progresses. Finally, protected Cys residues will be introduced to a single chain insulin analogue. Whether Insulin is produced chemically or biologically the major factor compromising efficiency is the oxidation step to form the disulfide bonds. Consequently, in order to bias formation of a crucial folding intermediate one pair of cysteine residues (corresponding to A7-B7) will be expressed in Acm protected form. Folding will be examined using standard biophysical techniques including NMR and CD spectroscopy, and compared with a synthetic counterpart. Successful completion of each model study serves as an important proof of principle and highlights the versatility, and wealth of potential applications for this new addition to the protein toolkit.