Protein Expression and Purification in the Fast Lane

我们之前证明了大肠杆菌麦芽糖结合蛋白（MBP）具有显著的增强溶解度和促进融合伙伴正确折叠的能力。由于这个原因，并且因为MBP融合蛋白在大肠杆菌中通常积累到非常高的水平，我们已经将MBP作为我们高通量蛋白表达和纯化方法的基石。然而，MBP融合蛋白并不总是有效地与直链淀粉树脂结合，即使他们这样做，融合蛋白在直链淀粉亲和层析后也很少是纯的。因此，为了弥补MBP作为亲和力标签相对较差的性能，我们试图在MBP融合蛋白的总体框架内加入补充标签。我们在MBP融合蛋白的框架内确定了几个位置，在这些位置上可以添加辅助标签，而不会影响MBP促进其融合伙伴溶解度的能力。然后，我们设计并成功测试了一种利用双His6-MBP亲和标签在大肠杆菌中生产蛋白质的通用方案。MBP片段提高了产率并增强了乘客蛋白的溶解度，而his标签则有助于其纯化。在2009财年，我们一直在酵母乳酸克鲁维菌（Kluyveromyces lactis）中测试这种方法，以生产重组蛋白（包括分泌型和胞内型）。我们发现细胞内HisMBP融合蛋白的产率通常极低，分泌的HisMBP融合蛋白在MBP结构域后经常被截断。未来的实验将探索不同的融合伙伴对K. lactis重组蛋白产量的影响。最近（2010年），利用绿色荧光蛋白作为融合体，在乳酸菌中同时表达细胞内和细胞外（分泌）蛋白，获得了可喜的结果。开发和测试在狼蛛利什曼原虫（Leishmania tarantolae）中生产重组蛋白的工具的实验正在进行中，评估宿主短芽孢杆菌（Brevibacillus choshinensis）分泌重组蛋白的能力也在进行中。因为大多数亲和标签都有可能干扰结构研究，所以去除它们的可靠方法是绝对必要的。因此，我们投入了大量的努力，试图利用高特异性烟草蚀刻病毒（TEV）蛋白酶来实现这一目的。为了提高TEV蛋白酶在大肠杆菌中的溶解度，我们设计了一种表达载体，以MBP融合蛋白的形式产生酶，该蛋白在体内自我裂解，产生一个n端his标记的TEV蛋白酶催化结构域，该结构域不含MBP。通过使用tRNA辅助质粒来补偿在大肠杆菌中很少使用的精氨酸密码子的存在，实现了TEV蛋白酶产量的显着增加。我们还设计了一种简单的方法，通过TEV蛋白酶在细胞内处理融合蛋白，用于确定当融合到MBP时，乘客蛋白是否可能被正确折叠。我们已经证明，许多不同的氨基酸侧链可以被容纳在TEV蛋白酶识别位点的P1'位点上，而对加工效率几乎没有影响。因此，在许多情况下，可以使用TEV蛋白酶来生产重组蛋白，其n端没有附着非天然残基。野生型TEV蛋白酶在一个特定的位点上自我切割，产生一个截断的多肽，酶活性大大降低。通过构建一种突变酶（S219V），我们成功地克服了自溶问题，该酶几乎不受自失活的影响，催化活性几乎是野生型酶的两倍。我们已经向全球数百个研究实验室分发了S219V TEV蛋白酶表达载体。我们还确定了TEV蛋白酶与肽底物和抑制剂络合的晶体结构，揭示了其严格序列特异性的结构基础。我们目前正专注于表征其他高度特异性的蛋白酶，例如由烟草静脉斑驳病毒（TVMV）编码的蛋白酶，我们最近将其与肽底物结合结晶。该共晶结构提示TVMV蛋白酶在S1'口袋中应该具有更严格的序列特异性，我们已经能够通过实验证实这一点。最近，我们鉴定了由Sindbis病毒、Semliki森林病毒和委内瑞拉马脑炎病毒编码的三种甲病毒蛋白酶，我们希望它们能成为TEV蛋白酶的有用替代品。尽管这些蛋白酶被证明具有足够的特异性，可以作为去除亲和标签的试剂，但它们的催化效率远远低于多病毒TEV和TVMV蛋白酶。最后，我们一直在研究重组形式的真菌羧肽酶（MeCPA）的效用，用于从重组蛋白的c端去除短亲和力标签（例如，多组氨酸）。我们对酶的底物特异性进行了彻底的分析，并表明它能够在晶体学的制备水平上用于去除c端his标签。然而，从杆状病毒表达系统中获得的MeCPA的产量相当低（每升250微克），因此正在努力寻找一种更有效的方法来生产重组酶。替代酶，如牛羧肽酶A和B也在各种系统中表达。我们最近（2010年）开发了一种在大肠杆菌中生产MeCPA的方法，其产量是杆状病毒系统的两倍，并且对当前程序进行轻微修改，似乎可以实现更大的产量。关键是在携带突变的大肠杆菌细胞中以MBP融合蛋白的形式表达这种酶，这种突变将细胞质转化为更氧化的环境，同时过量产生通常存在于细胞质外周的蛋白二硫异构酶DsbC。

We previously demonstrated that E. coli maltose binding protein (MBP) has a remarkable ability to enhance the solubility and promote the proper folding of its fusion partners. For this reason, and because MBP fusion proteins routinely accumulate to very high levels in E. coli, we have made MBP the cornerstone of our approach for high-throughput protein expression and purification. However, MBP fusion proteins do not always bind efficiently to amylose resin, and even when they do the fusion proteins are rarely pure after amylose affinity chromatography. Therefore, to compensate for the relatively poor performance of MBP as an affinity tag, we attempted to incorporate supplementary tags within the general framework of an MBP fusion protein. We identified several locations within the framework of an MBP fusion protein where accessory tags could be added without compromising the ability of MBP to promote the solubility of its fusion partners. We then designed and successfully tested a generic protocol for protein production in E. coli that utilizes a dual His6-MBP affinity tag. The MBP moiety improves the yield and enhances the solubility of the passenger protein while the His-tag facilitates its purification. During FY2009, we have been testing this method for the production of recombinant proteins (both secreted and intracellular formats) in the yeast Kluyveromyces lactis. We have found that the yield of intracellular HisMBP fusion proteins is extremely poor in general, and secreted HisMBP fusion proteins are frequently truncated after the MBP domain. Future experiments will explore the impact of alternative fusion partners on the yield of recombinant proteins in K. lactis. Promising results have recently been obtained (2010) using green fluorescent protein as a fusion parter for both intracellular and extracellular (secretion) protein expression in K. lactis. Experiments to develop and test tools for the production of recombinant proteins in the protozoan Leishmania tarantolae are just getting underway, as is an effort to evaluate the host Brevibacillus choshinensis for the secretion of recombinant proteins. Because most affinity tags have the potential to interfere with structural studies, reliable ways to remove them are absolutely necessary. Accordingly, we have invested a substantial effort in trying to exploit the highly specific tobacco etch virus (TEV) protease for this purpose. To improve the solubility of TEV protease in E. coli, we designed an expression vector that produces the enzyme in the form of an MBP fusion protein that cleaves itself in vivo to generate an N-terminally His-tagged TEV protease catalytic domain that is free of MBP. A dramatic increase in the yield of TEV protease was realized by using a tRNA accessory plasmid to compensate for the presence of arginine codons that are rarely used in E. coli. We also devised a simple method for intracellular processing of fusion proteins by TEV protease, which is used to determine whether or not a passenger protein is likely to be properly folded when it is fused to MBP. We have shown that many different amino acid side chains can be accommodated in the P1' site of a TEV protease recognition site with little or no impact on the efficiency of processing. Consequently, in many cases it is possible to use TEV protease to produce recombinant proteins with no non-native residues attached to their N-termini. Wild-type TEV protease cleaves itself at a specific site to generate a truncated polypeptide with greatly reduced enzymatic activity. We managed to overcome the autolysis problem by constructing a mutant enzyme (S219V) that is nearly impervious to autoinactivation and almost twice as catalytically active as the wild-type enzyme. We have distributed S219V TEV protease expression vectors to hundreds of research laboratories around the world. We have also determined crystal structures of TEV protease complexed with a peptide substrate and an inhibitor, which revealed the structural basis of its stringent sequence specificity. We are currently focusing on the characterization of other highly specific proteases, such as that encoded by the tobacco vein mottling virus (TVMV), which we have recently crystallized in complex with a peptide substrate. The co-crystal structure suggested that TVMV protease should have more stringent sequence specificity in the S1' pocket, and we have been able to confirm this experimentally. More recently, we characterized a trio of alphavirus proteases encoded by Sindbis Virus, Semliki Forest Virus and Venezuelan Equine Encephalitis Virus that we hoped would be useful alternatives to TEV protease. Although these proteases were shown to have adequate specificity for use as reagents to remove affinity tags, their catalytic efficiency is far less than that of the potyviral TEV and TVMV proteases. Finally, we have been investigating the utility of a recombinant form of a fungal carboxypeptidase (MeCPA) for removing short affinity tags (e.g., polyhistidine) from the C-termini of recombinant proteins. We have carried out a thorough analysis of the enzyme's substrate specificity and shown that it is capable of being used to remove C-terminal His-tags on a preparative level for crystallography. However, the yield of MeCPA obtained from the baculovirus expression system is rather low (250 micrograms per liter), and so efforts are underway to find a more efficient way to produce the recombinant enzyme. Alternative enzymes, such as bovine carboxypeptidases A and B are also being expressed in a variety of systems. We recently (2010) developed a method for the production of MeCPA in E. coli that produces twice the yield that can be obtained from the baculovirus system, and it seems likely that substantially greater yields will can be achieved with minor modifications of the current procedure. The key was to express the enzyme as an MBP fusion protein in E. coli cells that carry mutations that transform the cytosol into a more oxidative environment, while simultaneously overproducing the protein disulfide isomerase DsbC, which normally resides in the periplasm, in the cytosol.