Combining the polymerase incomplete primer extension method for cloning and mutagenesis with microscreening to accelerate structural genomics efforts

Successful protein expression, purification, and crystallization for challenging targets typically requires evaluation of a multitude of expression constructs. Often many iterations of truncations and point mutations are required to identify a suitable derivative for recombinant expression. Making and characterizing these variants is a significant barrier to success. We have developed a rapid and efficient cloning process and combined it with a protein microscreening approach to characterize protein suitability for structural studies. The Polymerase Incomplete Primer Extension (PIPE) cloning method was used to rapidly clone 448 protein targets and then to generate 2143 truncations from 96 targets with minimal effort. Proteins were expressed, purified, and characterized via a microscreening protocol, which incorporates protein quantification, liquid chromatography mass spectrometry and analytical size exclusion chromatography (AnSEC) to evaluate suitability of the protein products for X-ray crystallography. The results suggest that selecting expression constructs for crystal trials based primarily on expression solubility is insufficient. Instead, AnSEC scoring as a measure of protein polydispersity was found to be predictive of ultimate structure determination success and essential for identifying appropriate boundaries for truncation series. Overall structure determination success was increased by at least 38% by applying this combined PIPE cloning and microscreening approach to recalcitrant targets.     

Parallel screening and optimization of protein constructs for structural studies

A major challenge in structural biology remains the identification of protein constructs amenable to structural characterization. Here, we present a simple method for parallel expression, labeling, and purification of protein constructs (up to 80 kDa) combined with rapid evaluation by NMR spectroscopy. Our approach, which is equally applicable for manual or automated implementation, offers an efficient way to identify and optimize protein constructs for NMR or X‐ray crystallographic investigations.     

High-throughput Limited Proteolysis/Mass Spectrometry for Protein Domain Elucidation

High-resolution structural information is important for improving our understanding of protein function in vitro and in vivo and providing information to enable drug discovery. The process leading to X-ray structure determination is often time consuming and labor intensive. It requires informed decisions in expression construct design, expression host selection, and strategies for protein purification, crystallization and structure determination. Previously published studies have demonstrated that compact globular domains defined by limited proteolysis represent good candidates for production of diffraction quality crystals [1–7]. Integration of mass spectrometry and proteolysis experiments can provide accurate definition of domain boundaries at unprecedented rates. We have conducted a critical evaluation of this approach with 400 target proteins produced by SGX (Structural GenomiX, Inc.) for the New York Structural GenomiX Research Consortium (NYSGXRC; ) under the auspices of the National Institute of General Medical Sciences Protein Structure Initiative (). The objectives of this study were to develop parallel/automated protocols for proteolytic digestion and data acquisition for multiple proteins, and to carry out a systematic study to correlate domain definition via proteolysis with outcomes of crystallization and structure determination attempts. Initial results from this work demonstrate that proteins yielding diffraction quality crystals are typically resistant to proteolysis. Large-scale sub cloning and subsequent testing of expression, solubility, and crystallizability of proteolytically defined truncations is currently underway.     

Parallel N- and C-Terminal Truncations Facilitate Purification and Analysis of a 155-kDa Cold-Adapted Type-I Pullulanase

The cold-adapted pullulanase Pul13A is an industrial useful amylolytic enzyme, but its low solubility is the major bottleneck to produce the protein in recombinant form. In a previous approach, a complex and time-consuming purification strategy including a step-wise dialysis procedure using decreasing concentrations of urea to renature the insoluble protein from inclusion bodies had been established. In this study, a truncation strategy was developed to facilitate the purification and handling of the type-I pullulanase. Pul13A has a size of 155-kDa with a multidomain architecture that is composed of the following predicted modules: CBM41/E-set/Amy-Pul/DUF3372/E-set/E-set/E-set, with CBM and E-set domains being putative carbohydrate-binding modules, Amy-Pul is the catalytic region and DUF is a domain of unknown function. Consecutive N- and C-terminal deletions of domains were applied to construct minimized enzyme variants retaining pullulanase activity and exhibiting improved renaturation efficiencies. A total of seven truncation constructs were generated and tested, which still led to the production of inclusion bodies. However, the parallel deletion of the exterior CBM41 and E-set domain enabled the direct refolding of active enzymes during one-step dialysis in urea-free buffer. Catalytic properties of truncation construct Pul13A-N1/C1 were not impaired indicating that this enzyme variant may be superior for industrial applications over the full-length pullulanase.     

Fast plasmid based protein expression analysis in insect cells using an automated SplitGFP screen

Recombinant protein expression often presents a bottleneck for the production of proteins for use in many areas of animal‐cell biotechnology. Difficult‐to‐express proteins require the generation of numerous expression constructs, where popular prokaryotic screening systems often fail to identify expression of multi domain or full‐length protein constructs. Post‐translational modified mammalian proteins require an alternative host system such as insect cells using the Baculovirus Expression Vector System (BEVS). Unfortunately this is time‐, labor‐, and cost‐intensive. It is clearly desirable to find an automated and miniaturized fast multi‐sample screening method for protein expression in such systems. With this in mind, in this paper a high‐throughput initial expression screening method is described using an automated Microcultivation system in conjunction with fast plasmid based transient transfection in insect cells for the efficient generation of protein constructs. The applicability of the system is demonstrated for the difficult to express Nucleotide‐binding Oligomerization Domain‐containing protein 2 (NOD2). To enable detection of proper protein expression the rather weak plasmid based expression has been improved by a sensitive inline detection system. Here we present the functionality and application of the sensitive SplitGFP (split green fluorescent protein) detection system in insect cells. The successful expression of constructs is monitored by direct measurement of the fluorescence in the BioLector Microcultivation system. Additionally, we show that the results obtained with our plasmid‐based SplitGFP protein expression screen correlate directly to the level of soluble protein produced in BEVS. In conclusion our automated SplitGFP screen outlines a sensitive, fast and reliable method reducing the time and costs required for identifying the optimal expression construct prior to large scale protein production in baculovirus infected insect cells. Biotechnol. Bioeng. 2016;113: 1975–1983. © 2016 The Authors. Biotechnology and Bioengineering Published by Wiley Periodicals, Inc.     

Protein Production and Crystallization at the Joint Center for Structural Genomics

By definition, structural genomics centers must be able to address a large number of diverse protein targets. The methods developed should permit parallel and cost-effective processing while allowing for the diverse nature of proteins. Our approach to this problem is a multi-tiered effort where targets are characterized and categorized by behavior and processed in parallel by appropriate methods. The Joint Center for Structural Genomics (JCSG) has applied this tactic to create a fully integrated and scaleable structure determination pipeline. Highlights of the development of the current pipeline for protein production and crystallization are presented here.     

Addition of a polypeptide stretch at the N-terminus improves the expression, stability and solubility of recombinant protein tyrosine phosphatases from Drosophila melanogaster

The production of recombinant proteins in Escherichia coli involves substantial optimization in the size of the protein and over-expression strategies to avoid inclusion-body formation. Here we report our observations on this so-called construct dependence using the catalytic domains of five Drosophila melanogaster receptor protein tyrosine phosphatases as a model system. Five strains of E. coli as well as three variations in purification tags viz., poly-histidine peptide attachments at the N- and C-termini and a construct with Glutathione-S-transferase at the N-terminus were examined. In this study we observe that inclusion of a 45 residue stretch at the N-terminus was crucial for over-expression of the enzymes, influencing both the solubility and the stability of these recombinant proteins. While the addition of negatively charged residues in the N-terminal extension could partially rationalize the improvement in the solubility of these constructs, conventional parameters like the proportion of order promoting residues or aliphatic index did not correlate with the improved biochemical characteristics. These findings thus suggest the inclusion of additional parameters apart from rigid domain predictions to obtain domain constructs that are most likely to yield soluble protein upon expression in E. coli.     

C-Terminal truncation of rabbit flavin-containing monooxygenase isoform 2 enhances solubility

Flavin-containing monooxygenases (FMO) are membrane-associated enzymes contributing to oxidative metabolism of drugs and other chemicals. There are no known structures similar enough to FMO to provide accurate insights into the structural basis for differences in metabolism observed among FMOs. To develop an FMO amenable to crystallization, we introduced mutations into rabbit FMO2 (rF2) to increase solubility, decrease aggregation, and simplify isolation. Alterations included removal of 26 AA (Delta26) from the carboxyl-terminus, His(6)-fusion to the amino-terminus and a double Ser substitution designed to reduce local hydrophobicity. Only Delta26 FMO variants retained normal activity, increased the yield of cytosolic rF2 and decreased protein aggregation. Delta26 constructs increased rF2 in cytosol in low (from 2 to 13%), and high salt (from 24 to 62%) conditions. His-fusion proteins, while active and useful for purification, did not affect solubility. Delta26 variants should prove useful for identifying conditions suitable for production of an FMO crystal.     

A solubility-enhancement tag (SET) for NMR studies of poorly behaving proteins

Protein-fusion constructs have been used with great success for enhancing expression of soluble recombinant protein and as tags for affinity purification. Unfortunately the most popular tags, such as GST and MBP, are large, which hinders direct NMR studies of the fusion proteins. Cleavage of the fusion proteins often re-introduces problems with solubility and stability. Here we describe the use of N-terminally fused protein G (B1 domain) as a non-cleavable solubility-enhancement tag (SET) for structure determination of a dimeric protein complex. The SET enhances the solubility and stability of the fusion product dramatically while not interacting directly with the protein of interest. This approach can be used for structural characterization of poorly behaving protein systems, and would be especially useful for structural genomics studies.     

High-throughput protein purification under denaturating conditions by the use of cation exchange chromatography

A high-throughput protein purification strategy using the polycationic Z(basic) tag has been developed. In order for the strategy to be useful both for soluble and less soluble proteins, a denaturating agent, urea, was used in all purification steps. First, four target proteins were genetically fused to the purification tag, Z(basic). These protein constructs were purified by cation exchange chromatography and eluted using a salt gradient. From the data achieved, a purification strategy was planned including stepwise elution to enable parallel protein purification using a laboratory robot. A protocol that includes all steps, equilibration of the chromatography resin, load of sample, wash, and elution, all without any manual handling steps, was handled by the laboratory robot. The program allows automated purification giving milligram amounts of pure recombinant protein of up to 60 cell lysates. In this study 22 different protein constructs, with different characteristics regarding pI and solubility, were successfully purified by the laboratory robot. The data show that Z(basic) can be used as a general purification tag also under denaturating conditions. Moreover, the strategy enables purification of proteins with different pI and solubility using ion exchange chromatography (IEXC). The procedure is highly reproducible and allows for high protein yield and purity and is therefore a good complement to the commonly used His(6)-tag.     

Structural genomics for Caenorhabditis elegans: high throughput protein expression analysis

The structural genomics initiatives have begun with the aim to create a so-called "basic set library" of protein folds that will be used to improve protein prediction methods. Such a library is thought to require the determination of up to 10,000 new structures, including representative structures of several sequence variants from each protein fold. To meet this goal in a reasonable time frame and cost, automated systems must be utilized to clone and to identify the soluble recombinant proteins contained in multiple genomes. This paper presents such a system, developed using the genome of Caenorhabditis elegans (19,099 genes) as a model eukaryotic organism for structural genomics. This system successfully automates nearly all aspects of recombinant protein expression analysis including subcloning, bacterial growth, recombinant protein expression, protein purification, and scoring protein solubility.     

His tag effect on solubility of human proteins produced in Escherichia coli: a comparison between four expression vectors

We have compared four different vectors for expression of proteins with N- or C-terminal hexahistidine (His6) tags in Escherichia coli by testing these on 20 human proteins. We looked at a total recombinant protein production levels per gram dry cell weight, solubility of the target proteins, and yield of soluble and total protein when purified by immobilized metal ion affinity purification. It was found that, in general, both N- and C-terminal His6 tags have a noticeable negative affect on protein solubility, but the effect is target protein specific. A solubilizing fusion tag was able to partly counteract this negative effect. Most target proteins could be purified under denaturing conditions and about half of the proteins could be purified under physiological conditions. The highest protein production levels and yield of purified protein were obtained from a construct with C-terminal His tag. We also observe a large variation in cell growth rate, which we determined to be partly caused by the expression vectors and partly by the targets. This variation was found to be independent of the production level, solubility and tertiary structure content of the target proteins.     

Protein engineering of the colony-stimulating factor-1 receptor kinase domain for structural studies

A parallel approach to designing crystallization constructs for the c-FMS kinase domain was implemented, resulting in proteins suitable for structural studies. Sequence alignment and limited proteolysis were used to identify and eliminate unstructured and surface-exposed domains. A small library of chimeras was prepared in which the kinase insert domain of FMS was replaced with the kinase insert domain of previously crystallized receptor-tyrosine kinases. Characterization of the newly generated FMS constructs by enzymology and thermoshift assays demonstrated similar activities and compound binding to the FMS full-length cytoplasmic domain. Two chimeras were evaluated for crystallization in the presence and absence of a variety of ligands resulting in crystal structures, and leading to a successful structure-based drug design project for this important inflammation target.     

Extra terminal residues have a profound effect on the folding and solubility of a Plasmodium falciparum sexual stage-specific protein over-expressed in Escherichia coli.

The presence of extra N- and C- terminal residues can play a major role in the stability, solubility and yield of recombinant proteins. Pfg27 is a 27K soluble protein that is essential for sexual development in Plasmodium falciparum. It was over-expressed using the pMAL-p2 vector as a fusion protein with the maltose binding protein. Six different constructs were made and each of the fusion proteins were expressed and purified. Our results show that the fusion proteins were labile and only partially soluble in five of the constructs resulting in very poor yields. Intriguingly, in the sixth construct, the yield of soluble fusion protein with an extended carboxyl terminus of 17 residues was several fold higher. Various constructs with either N-terminal or smaller C-terminal extensions failed to produce any soluble fusion protein. Furthermore, all five constructs produced Pfg27 that precipitated after protease cleavage from its fusion partner. The sixth construct, which produced soluble protein in high yields, also gave highly stable and soluble Pfg27 after cleavage of the fusion. These results indicate that extra amino acid residues at the termini of over-expressed proteins can have a significant effect on the folding of proteins expressed in E. coli. Our data suggest the potential for development of a novel methodology, which will entail construction of fusion proteins with maltose binding protein as a chaperone on the N-terminus and a C-terminal 'solubilization tag'. This system may allow large-scale production of those proteins that have a tendency to misfold during expression.     

Activation of the Abl tyrosine kinase in vivo by Src homology 3 domains from the Src homology 2/Src homology 3 adaptor Nck.

The nonreceptor tyrosine kinase c-Abl is tightly regulated in vivo, but the mechanisms that normally repress its activity are not well understood. We find that a construct encoding the first two Src homology 3 (SH3) domains of the Src homology 2/SH3 adaptor protein Nck can activate c-Abl in human 293T cells. A myristoylated Nck SH3 domain construct, which is expected to localize to membranes, potently activated Abl when expressed at low levels. An unmyristoylated Nck SH3 domain construct, which localizes to the cytosol and nucleus, also activated Abl but only at high levels of expression. Activation by both myristoylated and unmyristoylated Nck constructs required the C terminus of Abl; a C-terminally truncated form of Abl was not activated, although this construct could still be activated by deletion of its SH3 domain. Activation did not require the major binding sites in the Abl C terminus for Nck SH3 domains, however, suggesting that the mechanism of activation does not require direct binding to the C terminus. Activation of c-Abl by Nck SH3 domains provides a robust experimental system for analyzing the mechanisms that normally repress Abl activity and how that normal regulation can be perturbed.     

A rapid and universal tandem-purification strategy for recombinant proteins

A major goal in the production of therapeutic proteins, subunit vaccines, as well as recombinant proteins needed for structure determination and structural proteomics is their recovery in a pure and functional state using the simplest purification procedures. Here, we report the design and use of a novel tandem (His)(6)-calmodulin (HiCaM) fusion tag that combines two distinct purification strategies, namely, immobilized metal affinity (IMAC) and hydrophobic interaction chromatography (HIC), in a simple two-step procedure. Two model constructs were generated by fusing the HiCaM purification tag to the N terminus of either the enhanced green fluorescent protein (eGFP) or the human tumor suppressor protein p53. These fusion constructs were abundantly expressed in Escherichia coli and rapidly purified from cleared lysates by tandem IMAC/HIC to near homogeneity under native conditions. Cleavage at a thrombin recognition site between the HiCaM-tag and the constructs readily produced untagged, functional versions of eGFP and human p53 that were >97% pure. The HiCaM purification strategy is rapid, makes use of widely available, high-capacity, and inexpensive matrices, and therefore represents an excellent approach for large-scale purification of recombinant proteins as well as small-scale protein array designs.     

CoESPRIT: a library-based construct screening method for identification and expression of soluble protein complexes

Structural and biophysical studies of protein complexes require multi-milligram quantities of soluble material. Subunits are often unstable when expressed separately so co-expression strategies are commonly employed since in vivo complex formation can provide stabilising effects. Defining constructs for subunit co-expression experiments is difficult if the proteins are poorly understood. Even more problematic is when subunit polypeptide chains co-fold since individually they do not have predictable domains. We have developed CoESPRIT, a modified version of the ESPRIT random library construct screen used previously on single proteins, to express soluble protein complexes. A random library of target constructs is screened against a fixed bait protein to identify stable complexes. In a proof-of-principle study, C-terminal fragments of the influenza polymerase PB2 subunit containing folded domains were isolated using importin alpha as bait. Separately, a C-terminal fragment of the PB1 subunit was used as bait to trap N-terminal fragments of PB2 resulting in co-folded complexes. Subsequent expression of the target protein without the bait indicates whether the target is independently stable, or co-folds with its partner. This highly automated method provides an efficient strategy for obtaining recombinant protein complexes at yields compatible with structural, biophysical and functional studies.     

High‐throughput mAb expression and purification platform based on transient CHO

A high‐cell‐density transient transfection system was recently developed in our laboratory based on a CHO‐GS‐KO cell line. This method yields monoclonal antibody titers up to 350 mg/L from a simple 7‐day process, in volumes ranging from 2 mL to 2 L. By performing transfections in 24‐deep‐well plates, a large number of mAbs can be expressed simultaneously. We coupled this new high‐throughput transfection process to a semiautomated protein A purification process. Using a Biomek FX^p liquid handling robot, up to 72 unique mAbs can be simultaneously purified. Our primary goal was to obtain >0.25 mg of purified mAb at a concentration of >0.5 mg/mL, without any concentration or buffer‐exchange steps. We optimized both the batch‐binding and the batch elution steps. The length of the batch‐binding step was important to minimize mAb losses in the flowthrough fraction. The elution step proved to be challenging to simultaneously maximize protein recovery and protein concentration. We designed a variable volume elution strategy based on the average supernatant titer. Finally, we present two case studies. In the first study, we produced 56 affinity maturation mAb variants at an average yield of 0.33 ± 0.05 mg (average concentration of 0.65 ± 0.10 mg/mL). In a second study, we produced 42 unique mAbs, from an early‐stage discovery effort, at an average yield of 0.79 ± 0.31 mg (average concentration of 1.59 ± 0.63 mg/mL). The combination of parallel high‐yielding transient transfection and semiautomated high‐throughput protein A purification represents a valuable mAb drug discovery tool. © 2014 American Institute of Chemical Engineers Biotechnol. Prog., 31:239–247, 2015     

Selective high throughput protein quantification based on UV absorption spectra

The application of high throughput experimentation (HTE) in protein purification process development has created an analytical bottleneck. Recently, a new label‐free and non‐invasive methodology for analyzing multicomponent protein mixtures by means of spectral measurements was presented. Analytics based on the methodology was shown to increase analytical throughput for selective protein quantification significantly, however this was only demonstrated for one particular protein combination. In this work, the possibilities and limitations of the analytical method are investigated further. Principal component analysis (PCA) was performed on a broad range of absorption spectra to investigate their common characteristics and differences. The PCA was used both for cluster analysis and to define a measure for spectral similarity. For binary protein combinations, the calibration precision was shown to decrease exponentially with the defined spectral similarity factor. Knowledge of this correlation can be used to determine a priori whether a calibration will be successful or not. Calibration robustness was investigated by applying the analytics to liquid chromatography performed in HTE mode. Further it was shown, that a spectral difference of 0.6% was sufficient to sucessfully preform a spectral based calibration of two IgG1 monoclonals. Biotechnol. Bioeng. 2013; 110: 448–460. © 2012 Wiley Periodicals, Inc.