To fuse or not to fuse: what is your purpose?

Bell et al. (2013). Protein Sci DOI: 10.1002/pro.2356 Citations: 142

Earlier fusion-tag reviews listed tags without saying which suit a given application. This review compares GST, MBP, thioredoxin, His6, SUMO, inteins and Fc by use, and states that over 75% of proteins made for crystallization are fusions.

Parts used

Key findings

  • Over 75% of proteins produced for crystallization are expressed as fusion constructs, and the His6 tag is the most common fusion partner, used in 60% of crystallographic studies.

  • The His6 tag (0.84 kDa) does not increase expression or solubility of its partner, is immunologically inactive, does not dimerize, and can sit at either terminus of a target protein, where most proteins remain functional. His6-tagged proteins are purified by immobilized metal-affinity chromatography on Ni2+ and eluted with an imidazole gradient, but high imidazole concentrations can strip metal ions from some proteins and leave them inactive.

  • A His6 tag on a protein expressed in E. coli is variably gluconoylated, which produces heterogeneity that does not favor crystal formation. The Structural Genomics Center estimates that up to 50% of prokaryotic proteins are insoluble when expressed in E. coli with a His tag, and studies suggest the fraction is higher for eukaryotic proteins.

  • Glutathione-S-transferase (GST), a 26 kDa protein from Schistosoma japonicum, binds glutathione with high affinity, so a GST fusion is captured on immobilized glutathione and eluted with 10 mM reduced glutathione. GST fused to hydrophobic proteins or to proteins larger than 100 kDa tends to form insoluble aggregates and inclusion bodies that render the protein inactive.

  • Native GST is a homodimer, so GST fused to a target that also oligomerizes can form large complexes that elute poorly from glutathione resin. Elution can be improved by adjusting salt or pH, adding dithiothreitol, adding extra free glutathione, or switching to S-butylglutathione, which has a 25-fold higher affinity for GST than glutathione does, but these measures can disrupt the function of the tagged protein.

  • GST, MBP, thioredoxin and SUMO fusions enhance the solubility of difficult proteins and are used as a rescue strategy for proteins that do not express well with a small tag. In most structural projects the tag is removed by an engineered endoprotease site between tag and protein of interest, which adds the cost of the protease and the risk of failing to recover soluble, structurally intact protein.

  • Thrombin, enterokinase and Factor Xa, commonly used for tag removal, show spurious cleavage at sites other than the engineered site. The viral proteases TEV and human rhinovirus 3C protease have lower turnover and therefore cleave sub-optimal sites less often, but they are needed in larger quantities, which raises the cost of tag removal at multi-milligram scale.

  • SUMO protease recognizes the tertiary structure of the SUMO tag, which is not found elsewhere in the proteome, rather than a linear peptide sequence, and cleaves precisely at the C-terminus of the tag. SUMO protease has a kcat 25-fold higher than TEV protease, but it removes only SUMO tags.

  • Crystallization of the intact fusion protein, with the tag left in place (carrier-driven crystallization), has produced structures with MBP, GST, thioredoxin, lysozyme and SUMO fusions. The tag supplies surfaces favorable to lattice formation and can serve as a molecular-replacement search model to solve the phase problem.

  • MBP has been the most successful carrier for carrier-driven crystallization, with over 25 structures determined as MBP fusions. The C-terminus of MBP forms a solvent-accessible alpha-helix that gives a rigid attachment point, but MBP adopts different conformations with and without bound maltose, so partial maltose occupancy can cause heterogeneity that hinders crystal formation, and MBP fusions often fail to bind amylose resin, so MBP is frequently combined with a His6 tag.

  • Carrier-driven crystallization with GST appears limited to protein fragments under 100 amino acids, and GST does not improve solubility in all cases and can promote aggregation of certain targets through dimerization. Crystallization of a fusion needs a short, rigid linker between tag and protein of interest, because a linker that is too long gives excessive flexibility and one that is too short may influence the structure of the target.

  • Thioredoxin ends in a rigid, solvent-accessible alpha-helix at its C-terminus and has a polar surface that readily forms crystal contacts, yet only two crystals and one solved structure have been reported with thioredoxin as a fusion partner. In the beta2-adrenergic G-protein coupled receptor, lysozyme replaced an unstructured loop to add polar surface for lattice formation rather than serving as a solubility or affinity tag.

  • SUMO fusions have been crystallized with the tag attached, as in a Burkholderia pseudomallei peptidyl-prolyl cis-trans isomerase bound to 8-deethyl-8-[but-3-enyl]-ascomycin solved to 1.9 A resolution (PDB 3UQB), or after enzymatic removal of SUMO, as in human Parkin E3 ubiquitin ligase solved to 2.0 A resolution (PDB 4I1H). The SUMO fusion system can be expressed in eukaryotic hosts, so proteins that need eukaryotic folding or post-translational modifications can be produced for structural work.

  • GST and thioredoxin fusions perform inconsistently for solubilizing cytokines, while MBP and NusA increase solubility and expression of cytokines, and NusA has been used to express and purify cytokine homologs with the IL8-like fold. Removing these tags with endoproteases leaves extra amino acids on the N-terminus of the processed protein, which can alter its structure or function.

  • SUMO fusion followed by SUMO protease cleavage yields cytokines with an authentic N-terminus. IL-1 beta and IL-8 were expressed at high levels as SUMO fusions and were biologically active, IFN-gamma was obtained in active form after SUMO expression although it had previously been expressed as an inclusion body, and TNF-alpha was generated as a mature active product.

  • A His6-SUMO tag produced 15 chemokines in active form with authentic N-termini, where SUMO fusion did not improve solubility but was essential for the authentic N-terminus that chemokine activity depends on. A tandem His6-thioredoxin-SUMO fusion promoted refolding of the chemokines without a redox buffer.

  • GST gives inconsistent results as a fusion tag for antimicrobial peptides and in several cases failed to protect the precursor peptide from proteolytic cleavage, possibly because its 28 kDa size is large relative to the peptide. MBP fusions of human beta-defensin 25 and 28 needed refolding steps to recover protein from aggregates, whereas the smaller thioredoxin (11.8 kDa) and SUMO (11.2 kDa) tags have been used more consistently, and a SUMO fusion yielded 166 mg of human beta-defensin-4 per 1 L fermentation after purification.

  • Intein-based purification (the IMPACT system) couples a yeast-derived protein splicing element to a bacterial chitin-binding domain. The fusion is captured on chitin, and the intein undergoes an S-N acyl shift at its N-terminal cysteine to form a thioester with the protein of interest, which 2-mercaptoethanol, 2-mercaptoethanesulfonic acid or DTT cleaves, releasing the protein as a thioester that supports C-terminal modification such as expressed protein ligation or can be hydrolyzed to a C-terminal carboxylate.

  • Chitin-agarose binds only 1 to 2 mg of recombinant protein per mL of matrix, compared with 40 to 80 mg/mL for Ni2+-IMAC or ion-exchange sepharose columns. On-column intein cleavage is slow, usually 16 h at room temperature or up to 5 days at 4 C, and the 30 to 100 mM reducing agent used can impair recovery of disulfide-linked proteins and require refolding.

  • Placing a His6-Ub tag or His6-SUMO tag at the N-terminus of an intein with a C-terminal target protein allows Ni2+-IMAC purification followed by intein-induced release of the protein without a protease, and the Ub and SUMO tags increase expression levels compared with the intein alone.

  • Protease-based self-cleaving tags include a His6-tagged cysteine protease domain from Vibrio cholerae activated by inositol hexakisphosphate, a His6-tagged catalytic sortase domain from Staphylococcus aureus activated by Ca2+, and the Profinity system, which uses the subtilisin prodomain as affinity tag bound to an engineered subtilisin on agarose and releases the target on addition of fluoride. Profinity requires separate expression, purification and conjugation of the mutant subtilisin, whereas in the other two systems the protease is part of the fusion protein.

  • An Fc fusion links a protein of interest to the roughly 250-amino-acid Fc domain of an immunoglobulin, at either terminus, and all commercial Fc-fusion therapeutics use the Fc domain of human IgG1. The Fc domain binds Protein A for purification with elution by a pH gradient or detergents, binds the neonatal Fc receptor (FcRn) to prolong serum half-life, and folds independently of its fusion partner, which can improve solubility, and no cleavage step is needed because the Fc is a functional part of the therapeutic.

  • SUMO tags can be used in E. coli, Pichia pastoris and mammalian HEK and CHO cells, and with a His6 or other affinity tag on both the SUMO tag and the SUMO protease a purification scheme based on IMAC alone is possible. Expression of the antiviral protein cyanovirin-N as a His6-SUMO fusion in E. coli reached over 30% of total soluble protein in a 30 L bioreactor.

1 of 23

Abstract

Since the dawn of time, or at least the dawn of recombinant DNA technology (which for many of today's scientists is the same thing), investigators have been cloning and expressing heterologous proteins in a variety of different cells for a variety of different reasons. These range from cell biological studies looking at protein-protein interactions, post-translational modifications, and regulation, to laboratory-scale production in support of biochemical, biophysical, and structural studies, to large scale production of potential biotherapeutics. In parallel, fusion-tag technology has grown-up to facilitate microscale purification (pull-downs), protein visualization (epitope tags), enhanced expression and solubility (protein partners, e.g., GST, MBP, TRX, and SUMO), and generic purification (e.g., His-tags, streptag, and FLAG™-tag). Frequently, these latter two goals are combined in a single fusion partner. In this review, we examine the most commonly used fusion methodologies from the perspective of the ultimate use of the tagged protein. That is, what are the most commonly used fusion partners for pull-downs, for structural studies, for production of active proteins, or for large-scale purification? What are the advantages and limitations of each? This review is not meant to be exhaustive and the approach undoubtedly reflects the experiences and interests of the authors. For the sake of brevity, we have largely ignored epitope tags although they receive wide use in cell biology for immunopreciptation.