Protein Purification Tags: Which One Fits Your Model?

Protein purification tags are usually chosen in the first ten minutes of a cloning project and then lived with for years. That one decision sets your chromatography chemistry, your buffer compatibility, your solubility ceiling and — if the tag stays on — whether your binding data means anything. This guide compares the tags researchers actually use, and is explicit about what each one costs. Every product named below was checked against the live BioHippo antibody catalog on 31 August 2026.

Protein Purification Tags Are Part of the Experiment, Not an Accessory

Most tag comparison tables rank tags as though one is simply better. In practice none dominates, and the reviews that survey the field say so directly: no single affinity tag is optimal with respect to yield, solubility and folding simultaneously, which is why combinatorial tagging exists at all.¹

The trade-offs are concrete. A 0.8 kDa His tag that is invisible in a crystal structure is also the tag most likely to co-purify a dozen native histidine-rich contaminants. A 42.5 kDa MBP fusion that rescues a hopelessly insoluble target is also a fusion partner large enough to mask the very activity you are trying to measure. The useful question is never “which tag is best” in the abstract. It is: given this expression host, this protein’s behaviour, and this downstream assay, which trade-off can I afford? The literature frames the same decision as a question of purpose rather than of ranking.²

Three Questions to Answer Before You Choose a Protein Purification Tag

Tag selection gets much easier once these three are answered, because everything downstream follows from them.

1. What host, and is the protein soluble there? A well-behaved cytoplasmic protein in E. coli needs an affinity handle and nothing more. A protein that goes straight to inclusion bodies needs a solubility partner, which is a different class of tag with a different size penalty. Solubility enhancement is a property of large, well-folded fusion partners — MBP, SUMO, thioredoxin, NusA — and insoluble expression in E. coli remains one of the most commonly encountered obstacles to producing functional protein.³ A hexahistidine tag has no meaningful effect on folding either way, so “add a His tag and hope” is not a solubility strategy.
2. Does the tag have to come off? Structural work, enzyme kinetics, oligomerization studies and anything destined for a functional assay usually require cleavage. Antibody production, pulldown baits and routine Western blot standards usually do not. If cleavage is required, the protease site and the residual scar it leaves become design constraints from day one — retrofitting a site after the construct is built means recloning.
3. What buffer will the purification run in? Chelators, reducing agents, divalent cations and detergents are not universally compatible across tag chemistries, and this catches people out more often than any other factor. Immobilised metal affinity chromatography (IMAC) strips its own metal in the presence of EDTA, and high concentrations of DTT can reduce the resin. Strep-Tactin tolerates both comfortably.⁴ If your protein needs 5 mM EDTA and 10 mM DTT to stay intact, that rules out His before you have run a single column.

The Five Workhorse Affinity Tags for Protein Purification

These five account for the overwhelming majority of published recombinant protein preparations.⁵ Each is described below by what it binds, what it costs you, and the situation where it is the right call.

Polyhistidine (His6 / His8 / His10) — the default affinity handle

~0.8–1.4 kDa · IMAC on Ni²⁺ or Co²⁺ · imidazole elution · any host

Consecutive histidines chelate immobilised divalent metal ions, most commonly Ni²⁺ on nitrilotriacetic acid (NTA) or iminodiacetic acid resin. Elution is competitive, using an imidazole gradient, or by pH drop. The tag is small enough that it rarely perturbs structure or function, which is why His tag protein purification dominates structural biology pipelines.

Its most underrated advantage is that binding is not conformation-dependent. His tags work under fully denaturing conditions — 8 M urea or 6 M guanidine hydrochloride — so inclusion body protein can be captured before refolding. No other tag in this list does this.

Use when you need a small, host-agnostic handle, you may need to purify under denaturing conditions, or the tag will remain on the final protein. The cost: single-step purity is the weakest of the five. Native proteins with surface histidine clusters co-elute, and E. coli contributes several notorious ones, so expect to follow IMAC with size exclusion or ion exchange. Chelators and high reducing agent concentrations are restricted; use TCEP over DTT where possible. Co²⁺ resin gives cleaner but lower-yield binding than Ni²⁺.

Glutathione S-transferase (GST) — affinity handle plus mild solubility help

~26 kDa · glutathione resin · reduced glutathione elution · bacterial expression

GST binds immobilised glutathione and elutes under genuinely gentle conditions — reduced glutathione at neutral pH, no imidazole, no low pH. It improves solubility modestly in E. coli and gives cleaner single-step purity than IMAC because few native proteins bind glutathione.

The complication is that GST is an obligate homodimer. A GST fusion is therefore dimeric by construction, which will produce artefactual dimerization in any interaction or oligomerization assay run before cleavage. For pulldown baits this is often tolerable; for biophysical characterisation it is not.

Use when you want gentle elution and good single-step purity from a bacterial host and the tag will be cleaved before functional work — or when building a GST pulldown bait. The cost: forced dimerization, a 26 kDa mass penalty, poor performance for secreted or mammalian-expressed protein, and no compatibility with denaturing conditions, since GST must stay folded to bind.

Maltose-binding protein (MBP) — the strongest solubility partner

~42.5 kDa · amylose resin · maltose elution · cytoplasm or periplasm

MBP is the most reliable solubility enhancer in routine use. In a direct comparison against GST and thioredoxin across six aggregation-prone proteins, MBP was a far more effective solubilising agent than either, and in some cases promoted folding into the biologically active conformation — behaviour consistent with an intramolecular chaperone acting during co-translational folding.⁶

It doubles as an affinity handle on amylose resin with maltose elution, though amylose has lower binding capacity and shorter lifetime than IMAC or glutathione media. Many groups therefore use MBP purely for solubility and put a His tag on the far terminus to do the actual capture.

Use when the target is insoluble or aggregation-prone in E. coli and you have exhausted temperature, induction and strain optimisation. The cost: a fusion partner larger than many targets. Solubility conferred by MBP does not always survive cleavage — a protein can precipitate the moment the fusion is cut, which means the tag was masking aggregation rather than preventing it. Always test cleavage on a small scale first.

Strep-tag II — highest single-step purity, gentlest conditions

8 residues (WSHPQFEK) · Strep-Tactin resin · desthiobiotin elution · any host

Strep-tag II binds an engineered streptavidin variant and elutes competitively with desthiobiotin under fully physiological conditions. Purification of the fusion — including its complexes with interacting partners — can be completed within about an hour from bacterial or eukaryotic lysate, and the tag is short, biologically inert and proteolytically stable.⁴ Single-step purity is typically the best of any tag in this list, because almost nothing in a cell lysate binds Strep-Tactin.

Critically, it tolerates EDTA, high concentrations of reducing agents and most detergents. That makes it the tag of choice when the protein needs a chelator to stay stable, or when you are purifying an intact multi-subunit complex that would not survive imidazole or low-pH elution.

Use when you are isolating a native complex, your buffer requires EDTA or strong reductant, or you need near-homogeneous protein in one step. The cost: lower binding capacity per millilitre of resin than IMAC, higher resin cost per preparation, and no use under denaturing conditions.

FLAG (DYKDDDDK) — small epitope with true purification capability

8 residues · anti-FLAG affinity gel · peptide or low-pH elution · highly hydrophilic

FLAG is unusual among short epitopes in being a genuine purification tag, captured on immobilised anti-FLAG monoclonals. Its hydrophilic, charged sequence tends to sit on the protein surface, and it doubles as a high-quality Western blot and immunoprecipitation epitope with mature commercial antibody support.

The sequence also embeds an enterokinase recognition site (DDDDK), so a C-terminally positioned enterokinase cut removes the tag without a scar. Note the practical distinction between clone families: M1-type antibodies bind calcium-dependently and require the tag at a free N-terminus, while M2-type clones bind at either terminus and are the general-purpose choice.

Use when you want one tag that serves purification, Western blot and immunoprecipitation, particularly in mammalian expression where a small hydrophilic epitope is preferred. The cost: antibody-based affinity resin is the most expensive medium in routine use and has a limited lifetime. Low-pH elution can denature sensitive targets; peptide elution avoids this but adds cost.

Solubility Partners Versus Affinity Handles: MBP, SUMO, Thioredoxin and NusA

Conflating these two categories is the single most common error in tag selection. An affinity handle exists to bind a resin. A solubility partner exists to keep your protein folded and in solution during expression. Some tags do both; most do one well.

The SUMO case — a scarless native N-terminus

The SUMO tag deserves separate mention because it solves a problem the others cannot. The SUMO protease Ulp1 recognises the tertiary structure of the SUMO domain rather than a linear sequence, and cleaves precisely at the C-terminal Gly-Gly. The result is a target protein with its authentic native N-terminus and no residual scar residues, and attaching SUMO N-terminally also markedly enhances expression of under-expressed proteins in E. coli.⁷

That matters whenever the N-terminal residue is functionally significant — mature cytokines, proteins where an added glycine alters activity, or any preparation intended to match a natural sequence exactly. The documented limitation is specific: purified SUMO fusions were efficiently cleaved when any amino acid except proline occupied the +1 position of the cleavage site, so check your target’s first residue before committing.⁷

A caution on solubility rescue. Solubility conferred by a large fusion partner is not the same as a stable, correctly folded target. A protein can be perfectly soluble as an MBP or SUMO fusion and precipitate within minutes of cleavage. Before scaling up, always run a small-scale cleavage and a post-cleavage spin, and confirm the released target stays in the supernatant.

Epitope Tags for Detection and Immunoprecipitation

HA, Myc and GFP tags are short or fluorescent handles optimised for antibody recognition. They are excellent for Western blot, immunofluorescence and immunoprecipitation, and generally poor for preparative purification — antibody-based media are expensive and low-capacity relative to metal chelate or glutathione resin.⁸

A common and sensible design is dual tagging: a purification handle at one terminus and a detection epitope at the other. This lets you run IMAC or Strep-Tactin for the preparation while retaining a clean, well-supported antibody epitope for every downstream blot — and, if the tags flank the target, a two-step tandem purification will select only full-length product.

N-Terminal or C-Terminal? Where to Put the Tag

Position is not a coin flip. Four considerations decide it.

• Signal peptides and processing. If the protein carries a signal sequence for secretion or periplasmic export, an N-terminal tag will be cleaved off with it. Use a C-terminal tag.
• Full-length verification. A C-terminal tag is only present on protein that was translated to completion, so C-terminal affinity capture automatically selects against truncation products. This is valuable for large or difficult targets.
• Solubility partners belong at the N-terminus. The chaperone-like effect of MBP, SUMO and thioredoxin operates during co-translational folding, so it requires the partner to emerge from the ribosome first.
• Functional termini. Check whether either terminus is buried, part of the active site, or required for interaction. If the C-terminus carries a PDZ-binding motif or an ER retention signal, do not put a tag on it.

Where the choice is genuinely open, the pragmatic answer is to build both. Parallel N- and C-terminal constructs cost one extra cloning reaction and routinely differ several-fold in soluble yield for reasons that are not predictable in advance.

Linkers matter more than people assume. Fusing a tag directly against a folded domain can sterically occlude it from the resin, producing poor binding that looks like a failed expression. A short flexible linker — glycine- and serine-rich, typically 5 to 15 residues — restores accessibility. Where a protease site is included, the linker also gives the protease room to engage, which is a frequent cause of stalled or incomplete cleavage.

Tag Removal: Comparing TEV, SUMO/Ulp1, HRV 3C and Thrombin

Cleavage specificity determines how much of the tag stays behind. “Scarless” means the target begins at its authentic first residue; most proteases leave one or more extra amino acids.⁹

The subtractive purification step

The cleanest way to separate cleaved target from cut tag is to make the protease itself removable. His-tagged TEV protease acting on a His-tagged fusion allows a second IMAC pass in which the free tag, the protease and any uncleaved fusion all bind the column while the tag-free target flows through. That single trick recovers most of the purity lost to IMAC’s poor selectivity in the first step.

Two practical checks before you scale: run a pilot cleavage across temperature, time and protease-to-substrate ratio, since sites embedded near folded domains cleave far more slowly than the same site in a flexible linker; and confirm your target survives the buffer the protease requires.

Decision Matrix: Matching a Protein Purification Tag to Your Situation

Tag Antibodies, Proteases and Pre-Tagged Proteins in the BioHippo Catalog

Once a tag is chosen, three reagent classes carry the work: antibodies to detect and capture the tag, proteases to remove it, and — where you would rather not express anything at all — ready-made tagged proteins.

Tag antibodies for detection and capture

The Abbkine tag antibody family is available in four formats per clone: unconjugated for Western blot, HRP-conjugated for direct detection, and agarose- or magnetic-bead-conjugated for immunoprecipitation and small-scale pulldown. The conjugated formats are the practical capture route for epitope tags that have no dedicated chromatography medium.

HRP-conjugated versions of the His, HA, Myc, GST and GFP clones are also stocked at $112.27 for direct blot detection without a secondary antibody. Browse the full range in the BioHippo antibody catalog.

Proteases for tag removal

More cleavage and modifying enzymes are listed in the enzymes collection.

Pre-tagged recombinant proteins

If the goal is a positive control, an assay standard or a binding partner rather than a protein you must express yourself, the Proteins & Peptides catalog carries a large library already supplied in defined tag configurations — C-His, N-His, N-His-SUMO, N-GST & C-His, and C-Strep among them. The configuration is stated in each product title, so you can match the tag to your detection reagent before ordering. A Recombinant myc tag Control Protein (C-terminal) is also available as an antibody validation control.

What BioHippo does not stock — and where to source it

Being straightforward about this: BioHippo does not currently carry preparative affinity chromatography media. If your workflow needs any of the following, source them elsewhere:

• Ni-NTA, IDA or Co²⁺ IMAC resins and cartridges
• Glutathione agarose or Sepharose for GST capture
• Amylose resin for MBP capture
• Strep-Tactin resin and desthiobiotin elution buffer
• Anti-FLAG M1/M2 affinity gel for preparative FLAG purification
• Thrombin, Factor Xa, HRV 3C and enterokinase proteases
• Halo- and SNAP-tag ligands and capture resins

What the catalog covers well is the verification and analytical-capture half of the workflow — tag antibodies in blot, HRP, agarose and magnetic-bead formats — plus TEV and SUMO proteases for cleavage, and a deep library of ready-tagged proteins. If you are unsure which combination your construct needs, request a quote and describe the host, the target and the assay.

Frequently Asked Questions About Protein Purification Tags

Do I actually need to remove the tag?

Often not. For antibody production, assay standards, pulldown baits and most Western blot work, a small tag such as His6 or FLAG can stay on with no measurable consequence. Remove it when the protein is going into crystallography or cryo-EM, enzyme kinetics, binding affinity measurement, oligomerization analysis, or any application where an added 26 to 42 kDa partner would dominate the behaviour. Cleavage always costs yield and a purification step, so it should be justified rather than automatic.

Why is my His-tagged protein not binding the column?

The usual causes, roughly in order of frequency: the tag is sterically buried and needs a linker or the opposite terminus; the lysis buffer contains EDTA or excess DTT that has stripped or reduced the resin; imidazole in the binding buffer is too high; or the protein is in the insoluble fraction and was never in the load to begin with. Run an SDS-PAGE on the pellet before troubleshooting the column — a surprising proportion of “binding failures” are solubility failures.

Which protein purification tag gives the highest purity in one step?

Strep-tag II, typically. Very little in a cell lysate binds Strep-Tactin, whereas IMAC co-purifies native histidine-rich proteins and is realistically a capture step rather than a polishing step. If you need near-homogeneous protein without a second column, Strep-tag II is the strongest option — at the cost of lower resin capacity and higher per-preparation expense.⁴

Can I use two tags at once?

Yes, and it is often the better design. Tandem affinity purification — two different tags at opposite termini, captured sequentially — enriches only species carrying both, which excludes truncation products and most contaminants. The common pairing of a solubility partner at the N-terminus with a small affinity handle at the C-terminus is the same idea: each tag does one job well instead of one tag doing both jobs adequately.¹

Will the tag interfere with my activity assay?

It depends on size and position more than on identity. A His6 tag on a flexible terminus rarely perturbs anything. A GST fusion will dimerize your protein whether you want it to or not, and a 42.5 kDa MBP partner can occlude an interaction surface. The rigorous control is to compare tagged and cleaved material side by side in the same assay — if the two agree, the tag is inert for that readout and you can keep it.

My protein precipitated right after cleavage. What happened?

The fusion partner was maintaining solubility rather than promoting correct folding — a real and well-documented limitation of MBP and similar partners.⁶ Options are to screen buffer conditions at the cleavage step, cleave at higher dilution and lower temperature, add a stabilising ligand or cofactor before cutting, or accept that the target requires the partner and design the downstream assay around a tagged protein.

At a Glance: The Full Protein Tag Lineup

References

All literature identifiers below were verified in PubMed. Where a statement rests on established methodological consensus rather than a single study, that is stated in the text.

1. Waugh DS. Making the most of affinity tags. Trends Biotechnol. 2005;23(6):316–320. doi:10.1016/j.tibtech.2005.03.012 · PMID 15922084
2. Bell MR, Engleka MJ, Malik A, Strickler JE. To fuse or not to fuse: what is your purpose? Protein Sci. 2013;22(11):1466–1477. doi:10.1002/pro.2356 · PMID 24038604
3. Costa S, Almeida A, Castro A, Domingues L. Fusion tags for protein solubility, purification and immunogenicity in Escherichia coli: the novel Fh8 system. Front Microbiol. 2014;5:63. doi:10.3389/fmicb.2014.00063 · PMID 24600443
4. Schmidt TGM, Skerra A. The Strep-tag system for one-step purification and high-affinity detection or capturing of proteins. Nat Protoc. 2007;2(6):1528–1535. doi:10.1038/nprot.2007.209 · PMID 17571060
5. Terpe K. Overview of tag protein fusions: from molecular and biochemical fundamentals to commercial systems. Appl Microbiol Biotechnol. 2003;60(5):523–533. doi:10.1007/s00253-002-1158-6 · PMID 12536251
6. Kapust RB, Waugh DS. Escherichia coli maltose-binding protein is uncommonly effective at promoting the solubility of polypeptides to which it is fused. Protein Sci. 1999;8(8):1668–1674. doi:10.1110/ps.8.8.1668 · PMID 10452611
7. Malakhov MP, Mattern MR, Malakhova OA, Drinker M, Weeks SD, Butt TR. SUMO fusions and SUMO-specific protease for efficient expression and purification of proteins. J Struct Funct Genomics. 2004;5(1–2):75–86. doi:10.1023/B:JSFG.0000029237.70316.52 · PMID 15263846
8. Kimple ME, Brill AL, Pasker RL. Overview of affinity tags for protein purification. Curr Protoc Protein Sci. 2013;73:9.9.1–9.9.23. doi:10.1002/0471140864.ps0909s73 · PMID 24510596
9. Young CL, Britton ZT, Robinson AS. Recombinant protein expression and purification: a comprehensive review of affinity tags and microbial applications. Biotechnol J. 2012;7(5):620–634. doi:10.1002/biot.201100155 · PMID 22442034
10. Kapust RB, Tözsér J, Fox JD, Anderson DE, Cherry S, Copeland TD, Waugh DS. Tobacco etch virus protease: mechanism of autolysis and rational design of stable mutants with wild-type catalytic proficiency. Protein Eng. 2001;14(12):993–1000. doi:10.1093/protein/14.12.993 · PMID 11809930

Molecular weights and recognition sequences are nominal values and vary with construct design. Product availability, pricing, species reactivity and validated applications were verified against the live BioHippo catalog on 31 August 2026 and are subject to change; confirm the specification table and intended-use statement on each product page before purchase. Methodological guidance summarises published literature and does not replace validation in your own system. Figures 1–4 are illustrative schematics, not experimental data.