Unified Protein Engineering Platform™

Protein-Class Purification Protocols with Cleavable Affinity Tags

Tag-free end-product strategy | English edition | UPE-PP-2026-002 · v3.0 · verified 2026-08-12

Figure 1. End-to-end workflow for tag-free protein purification using a cleavable affinity handle.

Core principles before use

This document starts by classifying the protein, selecting an expression host and cellular location, and then choosing the identity and position of a cleavable affinity handle. Every default recombinant-tag route includes post-capture tag cleavage and subtractive affinity chromatography. This separates host proteins, the released tag, the protease, and uncleaved fusion more effectively than a single affinity step [1-5,19,20].

Affinity-tag configurations without a removal route are excluded from the default protocols. Ion exchange, heparin, hydrophobic interaction, and size-exclusion chromatography remain because they are polishing operations rather than retained tags. Protein A capture of an intact IgG is included as an exception because it uses the antibody's native Fc and introduces no foreign tag. An Fc fusion must contain a protease site whenever Fc is not part of the required final product.

Executive summary

The most broadly useful first choices are N-terminal His8-SUMO or His8-TEV for soluble bacterial proteins, a 3C-tandem SBP tag at an exposed C terminus of the mature protein for secreted eukaryotic proteins, and 3C-His8 or a 3C-tandem SBP tag on a topology-accessible terminus for membrane proteins. A C-terminal tag enriches full-length material when premature termination or terminal proteolysis is a concern. When integrity of both ends matters, use orthogonal N- and C-terminal tags sequentially and remove both [11,23,30,31].

His tags provide the best economics and capacity, but single-capture purity is usually moderate; reverse IMAC followed by SEC or IEX is therefore the default. A tandem SBP tag offers mild competitive elution and high selectivity and suits dilute supernatants and complexes. A DYKDDDDK tag is highly selective but is favoured at smaller eukaryotic scale because of resin cost and capacity. MBP, SUMO, and thioredoxin can reduce insolubility, yet the target must be checked for precipitation immediately after tag removal [2,3,8-14].

Table of contents

1. Purification design starts with protein classification

1.1 Required project inputs

1.2 Expression-host selection rules

2. Cleavable construct design and tag placement

2.1 Placement checklist

2.2 Protease selection

3. Selection of cleavable purification tags

3.1 Recommended default combinations

3.2 Tag modes excluded from default protocols

4. Platform protocol: expression through storage

4.1 Small-scale construct and condition screen

4.2 Harvest, lysis, and clarification

4.3 Capture through polishing

5. Default protocols by protein class

5.1 Soluble cytosolic enzymes and single domains

5.2 Poorly soluble proteins, IDPs, and short peptides

5.3 Secreted, disulfide-rich, and glycosylated proteins

5.4 Intact IgG, Fab, and Fc fusions

5.5 Membrane proteins

5.6 DNA/RNA-binding proteins

5.7 Multisubunit complexes and oligomers

5.8 Inclusion bodies and refolded proteins

5.9 Protease-sensitive, toxic, and cofactor-dependent proteins

6. Detailed protocols by purification tag

6.1 His6/8/10 - IMAC - TEV/3C

6.2 SBP tag (WSHPQFEK) - streptavidin mutein resin - 3C/TEV

6.3 DYKDDDDK tag - antibody resin - 3C/TEV

6.4 GST - glutathione - 3C/TEV

6.5 MBP - amylose - TEV

6.6 His-SUMO - IMAC - Ulp1

6.7 His-thioredoxin - IMAC - TEV/3C

6.8 Intein-CBD self-cleavage

7. Polishing, formulation, and quality control

7.1 Selecting a polishing step

7.2 Running SEC and concentration

7.3 Formulation screen

7.4 Research-grade release criteria and analytical panel

8. Troubleshooting: from variables to root cause

9. Patent records by sequence and how to verify them

9.1 Three distinct sets of rights

9.2 Patent records corresponding to tag sequences

9.3 Patent records corresponding to capture ligands and resins

9.4 How to apply this chapter

10. Experimental appendix

10.1 Standard buffer compositions

10.2 Additive and resin compatibility

10.3 Fraction naming and mass balance

10.4 Final pre-run checklist

11. References

1. Purification design starts with protein classification

Many purification failures occur before resin selection. Choosing a tag without checking native localization, transmembrane helices, signal peptides, disulfides, glycosylation, cofactors, pI, oligomeric state, and flexible termini can yield expressed but misfolded, insoluble, inactive, or uncleavable material. The classification below jointly determines the expression host and capture route [1,20,21].

Figure 2. Primary purification-route selection based on physical and biological properties of the target.

1.1 Required project inputs

1. Sequence and construct boundaries: annotate native signal peptide/propeptide, transmembrane helices, low-complexity segments, and domain boundaries.

2. Physicochemical properties: record predicted mass, pI, extinction coefficient, cysteine count, glycosylation sites, cofactors, and metal-binding sites.

3. Function: annotate the active site, ligand-binding surface, oligomer interface, terminal motifs, and essential PTMs.

4. Product criteria: predefine required quantity, tag-free status, acceptable terminal scar, monomer/oligomer state, endotoxin target, and assay.

5. Analytics: designate the release set from SDS-PAGE, intact mass, SEC, DLS, native MS/SEC-MALS, and functional assays.

1.2 Expression-host selection rules

2. Cleavable construct design and tag placement

Tag position is not merely a cloning convenience; it can change expression, membrane insertion, activity, cleavage efficiency, and selection for full-length product. Prefer the opposite terminus when one end is close to an active site or oligomer interface or is structurally buried. Even when a predicted terminus is disordered, the tag and protease site add 15-45 residues, so accessibility and function must be tested experimentally [23,30,31].

Figure 3. Cleavable construct architectures for N-terminal, C-terminal, and dual-end integrity selection.

2.1 Placement checklist

1. Do not place a purification tag before an N-terminal signal peptide. Put the tag on an exposed terminus of the mature protein and preserve signal-peptide processing.

2. Do not mask terminal motifs such as KDEL/HDEL, peroxisomal SKL, prenylation CAAX, GPI-anchor signals, PDZ-binding motifs, or amidation signals.

3. Inspect the biological assembly, not a predicted or deposited monomer alone, to determine whether a terminus lies in a subunit interface.

4. When the N terminus must be exact, prefer His-SUMO/Ulp1; with TEV, design around P1' residue tolerance and the residual scar [13,15].

5. Provide a 2-5-residue flexible spacer around the protease site when needed, and determine in advance whether added GGS residues remain in the final product [30].

6. Compare at least one N-terminal and one C-terminal construct, then compare activity, Tm, and SEC retention with an untagged or native control.

2.2 Protease selection

3. Selection of cleavable purification tags

No tag simultaneously optimizes yield, purity, cost, capacity, mild elution, and solubility enhancement. Address the target's dominant risk first: use His when expression is adequate and economics matter, a tandem SBP tag for dilute supernatants and native complexes, a DYKDDDDK tag for small-scale high-selectivity capture, MBP or SUMO for poor solubility, and His-SUMO when the N terminus must be exact [2-5]. In every case include the separate protease site shown below.

Figure 4. Relative tag-selection scores. Scores are design-oriented comparisons synthesized from the literature, not specifications of any product. On the 'cost efficiency' axis a higher value means a lower cost burden.

3.1 Recommended default combinations

3.2 Tag modes excluded from default protocols

Effectively irreversible or covalent capture modes such as biotin-streptavidin capture, self-labelling enzyme tags, and covalent peptide-protein pairs do not permit mild tag-free recovery unless a separate protease-release module is engineered. Single-epitope antibody purification is useful for detection but is not preferred over the DYKDDDDK route for preparative capacity, economics, or controlled elution. Use these only when immobilization is the final objective or an on-resin cleavage workflow has been independently validated.

4. Platform protocol: expression through storage

Figure 5. Coupling cleavage to subtractive affinity turns tag removal into a second selectivity step.

4.1 Small-scale construct and condition screen

1. Prepare at least two constructs: compare N- and C-terminal placement or a His-family route with a tandem SBP route. For a high insolubility risk, include His-SUMO and His-MBP for a total of three to four constructs.

2. For E. coli, compare overnight induction at 16-20°C with a shorter induction at 25-30°C after growth at 37°C in 1-5 mL cultures. Prioritize a soluble, correctly sized, functional band over total expression.

3. Run total, soluble, and insoluble fractions at equal culture equivalents by SDS-PAGE. Do not judge soluble target from a tag blot alone; combine total-protein staining with activity or ligand binding.

4. Use a 50-200 uL resin microcapture to assess binding, wash, elution, and cleavage together. Retain load, flow-through, wash, eluate, cleavage supernatant, and pellet.

5. Score yield, purity, cleavage percentage, SEC monodispersity, and activity, and scale only one condition. Do not select solely from high expression or an intense band.

4.2 Harvest, lysis, and clarification

1. Resuspend the cell pellet homogeneously in chilled lysis buffer. Do not use an EDTA/EGTA-containing inhibitor cocktail when IMAC will follow.

2. If lysozyme is appropriate, treat at 0.1-0.5 mg/mL for 20-30 min in the cold. Dose nuclease by supplier units and meet its Mg2+ requirement.

3. Sonicate with 2-5 s on/off pulses while keeping the sample below 10°C, or use 2-3 passes of high-pressure homogenization at 10-15 kpsi. Record the method and total energy.

4. Clarify at 20,000-40,000 x g for 30-45 min at 4°C. For membrane proteins, recover the membrane pellet by separate ultracentrifugation.

5. Filter through 0.45 um, or 0.22 um when viscosity is low, and capture immediately. Record A280 and sample volume before and after filtration to quantify losses.

4.3 Capture through polishing

1. Load to no more than 50-80% of resin binding capacity. If the target amount is unknown, estimate dynamic capacity by microcapture first.

2. Use 5 CV equilibration, load, 10-20 CV wash, and 3-6 CV elution as defaults, and retain every fraction until SDS-PAGE is complete. CV means packed resin volume.

3. Exchange the eluate immediately into a cleavage-compatible buffer by desalting or dialysis. Do not leave imidazole, glutathione, maltose, desthiobiotin, or low pH in contact for long.

4. Titrate protease ratio and temperature at 0.2-2 mg/mL target. Measure concentration and soluble recovery before and after the reaction so that cleavage percentage and mass recovery are reported separately.

5. Design the released tag and the protease to carry affinity handles, then pass the mixture over the original or an orthogonal resin so the tag-free target appears in the flow-through. Confirm that uncleaved fusion also binds.

6. Select one of IEX, heparin, or HIC to match the dominant impurity, then finish with SEC to remove aggregates and misassembled species. Run SEC in the formulation buffer or something very close to it.

7. Concentrate with a membrane whose MWCO is one third to one half of the expected oligomer size, checking 10-20 uL after every two- to four-fold step to find the onset of precipitation.

8. Filter the final sample only when a 0.22 um low-protein-binding filter is appropriate, and snap-freeze small aliquots. Assess one freeze-thaw cycle and 24-72 h refrigerated stability before release.

5. Default protocols by protein class

This chapter gives the whole process for each protein class and is used together with the tag-specific SOPs in the next chapter. If the soluble-enzyme route selected His8-TEV, for example, combine the expression, lysis, and polishing guidance here with the IMAC SOP in 6.1. Each process assumes a starting scale of 1 L E. coli culture or 100-500 mL eukaryotic supernatant; the actual resin volume follows from the measured target mass and the dynamic binding capacity in the resin specification.

5.1 Soluble cytosolic enzymes and single domains

This is the most predictable class, but proteolysis, cofactor loss, oligomer mismatch, and host-derived histidine-rich contaminants are common. Secure solubility at low temperature first, then judge by whether activity and the SEC peak survive tag cleavage [1,6,20].

1. In a T7-based expression strain, compare overnight induction at 16-20°C with 4-6 h at 25°C. If metals or cofactors are required, supplement the medium or lysis buffer within a nontoxic range.

2. Lyse in 50 mM HEPES pH 7.5, 300 mM NaCl, 10% glycerol, 0.5 mM TCEP, 10 mM imidazole and clarify immediately.

3. On Ni-NTA use a 20-40 mM imidazole wash and 300 mM imidazole elution. If the target appears in the wash, check tag accessibility and pH.

4. Exchange into 20 mM HEPES pH 7.5, 150-300 mM NaCl, 0.5 mM TCEP and start TEV or 3C at 1:50 w/w.

5. Pool the reverse-IMAC flow-through, choose Q or SP IEX according to pI, and finish with SEC. For metalloenzymes avoid EDTA and keep the required metal at low concentration in the SEC buffer.

5.2 Poorly soluble proteins, IDPs, and short peptides

A large solubility tag can mask aggregation during production, so treat cleavage as the most important stress test in the process. His-SUMO is particularly useful for short peptides and an exact N terminus; MBP is used when stronger solubility rescue is needed [9,12,13].

1. Screen expression at 12, 16, 20, and 25°C, and check fusion integrity and proteolysis, not only soluble yield.

2. Use 300-500 mM NaCl, 10% glycerol, and 0.5-1 mM TCEP in the lysis buffer. If the IDP binds nucleic acid, add nuclease and 0.5-1 M salt.

3. After IMAC or amylose capture, run cleavage tests at 0.1, 0.25, 0.5, and 1.0 mg/mL to find concentration-dependent precipitation.

4. For SUMO, compare Ulp1 at 1:1000, 1:300, and 1:100 w/w at 4°C and 20-26°C. For MBP, compare TEV at 1:50 w/w overnight at 4°C and for 2 h at 25°C.

5. Perform reverse IMAC immediately after cleavage, then polish short peptides by RP-HPLC or a suitable SEC and IDPs by ion exchange. Check adsorptive peptide loss using low-binding tubes.

5.3 Secreted, disulfide-rich, and glycosylated proteins

Correct disulfides and glycoforms govern function more than high cytosolic yield. Preserve signal-peptide and propeptide processing and C-terminal trafficking motifs first, and capture mildly and directly from the supernatant wherever possible [21].

1. Run a 30-100 mL pilot by HEK293 transient expression and determine the target distribution between supernatant and cell lysate. Keep a mammalian host if complex glycans are required.

2. Clarify the supernatant at 3,000-10,000 x g and filter at 0.22 um. Adjust to pH 7.5-8.0 and 150 mM NaCl, and remove biotin-rich medium supplements before SBP capture.

3. Capture by column binding on the engineered streptavidin resin using 5 CV equilibration, 10 CV wash, and 5-6 CV elution with 2.5 mM desthiobiotin.

4. Exchange into a 3C-compatible buffer and cleave at 1:50 w/w at 4°C. Pass again over the streptavidin resin, then remove His-tagged 3C by IMAC.

5. Separate aggregates and oligomers by SEC, then run intact and reduced mass, nonreducing SDS-PAGE, and a glycan profile or ligand-binding assay.

5.4 Intact IgG, Fab, and Fc fusions

Intact IgG uses the native Fc-Protein A interaction and needs no foreign tag. For Fab, a light-chain-binding protein ligand, a CH1-specific ligand, or a cleavable tandem SBP tag can be used. When Fc is not part of the required product in an Fc fusion, remove it through a hinge-proximal 3C or TEV site [26-28].

1. Adjust the clarified mammalian supernatant to PBS or 20 mM sodium phosphate, 150 mM NaCl, pH 7.2-7.4.

2. Equilibrate the Protein A resin with 5 CV and load at the residence time given in the resin specification. Wash with 10 CV, then elute over 3-5 CV with glycine or citrate at pH 3.0-3.6.

3. Collect each fraction directly into one tenth volume of 1 M Tris pH 8.0-9.0 prepared in advance, neutralizing immediately to pH 6.5-7.5. Record the low-pH residence time.

4. For IgG, reduce HCP, DNA, leached ligand, aggregates, and charge variants by AEX flow-through or CEX bind-elute, and analyse by SEC.

5. When cleaving an Fc fusion, pass the 3C or TEV reaction back over Protein A so that Fc and uncleaved fusion bind, recovering the tag-free protein in the flow-through.

5.5 Membrane proteins

For membrane proteins, topology, detergent/lipid choice, and stability screening matter more than the tag. Long-chain mild nonionic detergents usually help post-extraction stability, but no single detergent is a universal answer. Detergent micelles can weaken streptavidin binding and protease activity, so test in the actual formulation [14,17,18,22,23].

1. Screen expression host and tag position by small-scale fluorescence-detection SEC and scale only combinations that give a monodisperse peak.

2. After lysis, recover the membrane pellet at 100,000 x g for 45-60 min and resuspend in 20 mM HEPES pH 7.5, 150 mM NaCl, 10% glycerol.

3. Screen at least three mild detergent classes, such as maltoside-type, neopentyl-glycol-type, and steroid-based, plus sterol or lipid if needed, at target-specific concentrations for 1-2 h at 4°C. Evaluate extraction yield and SEC peak together.

4. Re-centrifuge at 100,000 x g and capture in detergent-containing buffer. Keep wash and elution detergent above the CMC and close to the stabilizing concentration.

5. Titrate HRV 3C at 1:20-1:50 w/w at 4°C in the actual detergent-containing buffer. Run subtractive affinity and SEC promptly after cleavage.

6. Keep the lipid and detergent composition constant during SEC, and evaluate monomer/oligomer peaks, thermal shift, ligand binding, or functional reconstitution.

5.6 DNA/RNA-binding proteins

Host nucleic acid distorts viscosity, A260, apparent oligomerization, column binding, and activity. High salt alone may not remove it, so apply nuclease early in lysis and use heparin or IEX as an orthogonal step [25].

1. Lyse in 0.5-1.0 M NaCl and 1-2 mM MgCl2 with a DNase, RNase, or broad nuclease appropriate to the target.

2. Check A260/A280, viscosity, and a nucleic-acid-stained agarose gel of the clarified lysate. If the target is a complex that requires RNA, design a defined ligand exchange instead of nuclease treatment.

3. Start affinity capture at 0.5 M NaCl and, if binding holds, screen washes up to 0.75-1.0 M NaCl.

4. After cleavage and subtraction, bind to heparin at 0.1-0.2 M NaCl and elute with a 0.1-1.0 M NaCl gradient, or use Q or SP IEX according to the calculated pI.

5. Repeat A260/A280 and the nucleic-acid gel before SEC, and analyse ligand-free and defined-ligand states separately.

5.7 Multisubunit complexes and oligomers

Overexpressing single subunits and mixing them afterwards can give failed assembly and nonstoichiometric species. Co-express interdependent subunits and place orthogonal tags on different subunits so that only complexes containing both components are selected [31].

1. Screen subunit ratios using bicistronic or multicistronic vectors, baculovirus, or mammalian co-expression.

2. Lyse under the salt, detergent, nucleotide, or cofactor conditions that preserve the weakest interaction. Add protease inhibitors immediately.

3. Move to the second orthogonal affinity step without changing conditions drastically after the first. Check subunit stoichiometry by densitometry at each step.

4. Cleave the tags sequentially or together, but follow subunit dissociation by time-course SEC. If a tag was stabilizing the assembly, do not cleave at low concentration.

5. Confirm molecular mass and stoichiometry by one of native SEC-MALS, native MS, or AUC.

5.8 Inclusion bodies and refolded proteins

Inclusion bodies can give high apparent purity but guarantee neither native folding nor activity. Use them only after testing soluble constructs, low-temperature expression, alternative hosts, and solubility tags. Refolding is a protein-by-protein screening problem with no universal recipe [24].

1. Recover the insoluble pellet at 10,000-20,000 x g after disruption and wash twice with 20 mM Tris pH 8, 0.5 M NaCl, 1% nonionic detergent and once with detergent-free buffer.

2. Solubilize for 1-2 h in 6 M guanidine-HCl or 8 M urea, 20-50 mM Tris pH 8.0, 10-20 mM reducing agent, then clarify.

3. Concentrate the target by denaturing IMAC, then compare direct dilution, step dialysis, and on-column gradient refolding in 24-96 well format at 0.1-0.5 mg/mL.

4. Run a factorial screen over 0.2-0.5 M arginine, glycerol, salt, pH, and oxidized/reduced glutathione ratio. Include a redox pair for disulfide proteins, matched to the target cysteine chemistry.

5. Recover only the soluble refolded fraction, remove aggregates by SEC or IEX, then cleave the tag and run SEC again.

5.9 Protease-sensitive, toxic, and cofactor-dependent proteins

For this class, first locate the time at which degradation begins and the boundary at which the cofactor is lost. Use slower expression, a shorter process, co-expression partners, and required ligands, while checking compatibility between the affinity resin and the protease.

1. Determine the full-length to fragment ratio over a 0, 1, 2, 4, 8 h induction time course and choose the earliest sufficient time point.

2. Record the elapsed time from harvest to affinity load and process continuously at 2-8°C. Add protease inhibitors and ligands or cofactors immediately before lysis.

3. Use a C-terminal tag so truncated species leave in the flow-through. If N-terminal fragments dominate the impurity profile, use N/C dual-tag selection.

4. If nickel leaching or imidazole interferes with metalloenzyme activity, switch to a Co resin or the SBP route, and keep the required metal or cofactor at low concentration in the SEC buffer.

5. Complete cleavage and SEC on the same day and measure activity, metal occupancy, or the cofactor spectrum immediately.

6. Detailed protocols by purification tag

The SOPs below are written per 1 mL of packed resin, that is 1 CV. Binding capacity, permissible flow rate, pressure limits, and reducing-agent tolerance differ between resins, so the product specification takes precedence. No option that ends without cleaving the purification tag is included. In every SOP, retain load (L), flow-through (FT), wash (W), eluate (E), cleavage (C), subtractive flow-through (SFT), and bound fraction (SB). Representative patent links for each tag are collected in chapter 9.

6.1 His6/8/10 - IMAC - TEV/3C

This is the most economical and most easily scaled general route. Ni-NTA gives high capacity while Co-based resins usually give higher selectivity at lower capacity. Do not stop at a single IMAC; follow cleavage with reverse IMAC and orthogonal polishing [6,7].

1. Transfer 1 mL of resin out of water or storage solution and equilibrate with 5 CV of A. Never let the resin run dry.

2. Adjust the clarified lysate to pH 7.5-8.0 and 10 mM imidazole. Load no more than 50-80% of the nominal resin capacity.

3. Retain the FT and wash with 5 CV of A and 10-20 CV of B. Define the wash endpoint from the A280 baseline and SDS-PAGE.

4. Elute 5 CV of C in 0.5-1 CV fractions. Pool only target fractions and desalt into D immediately.

5. Start His-tagged TEV or 3C at 1:50 w/w and react overnight at 4°C or for 1-4 h at 20-25°C. Use a time course to find the point of at least 90% cleavage.

6. Bring the cleavage reaction to 10 mM imidazole or less and pass it over 0.2-0.5 mL of fresh Ni-NTA or the original resin. The His-tagged tag, the protease, and uncleaved fusion bind, while the tag-free target is recovered in the SFT.

7. Polish the SFT by IEX or SEC. If nickel leaching is a concern, run a metal analysis or colorimetric check.

6.2 SBP tag (WSHPQFEK) - streptavidin mutein resin - 3C/TEV

The strengths of this route are mild near-physiological conditions and high selectivity. A tandem SBP tag binds dilute supernatants and complexes more stably through avidity [10,11]. Two resin generations are in circulation and they use different eluents. First-generation streptavidin mutein resin is eluted competitively with desthiobiotin, whereas the later high-affinity mutein binds far more tightly and is eluted with biotin. The two generations also differ in patent status, so consult the comparison table in chapter 9 before selecting a resin.

1. Equilibrate 1 mL of engineered streptavidin resin with 5 CV of W. Prefer column binding over batch pre-incubation.

2. Remove free biotin and desthiobiotin from the sample, adjust to the W composition, and load. Use a tandem SBP tag for dilute supernatants and allow sufficient residence time.

3. Wash with 10-15 CV of W and retain FT and W. If detergent is present, verify binding recovery on a microcolumn at the actual concentration.

4. Elute 5-6 CV of E in 0.5-1 CV fractions and remove desthiobiotin by desalting.

5. Cleave with His-tagged 3C or TEV at 1:50 w/w. Prefer 3C when 4°C stability matters.

6. Pass the cleavage reaction back over the streptavidin resin so the released tag and uncleaved fusion bind, and collect the SFT. Then remove the His-tagged protease by Ni-NTA.

7. Finish with SEC or a target-specific IEX and check whether desthiobiotin or biotin carryover affects downstream assays.

6.3 DYKDDDDK tag - antibody resin - 3C/TEV

High selectivity favours low-expressing eukaryotic proteins and complexes, but resin capacity and cost make this a poor first choice for large preparative work. The enterokinase site inside the DYKDDDDK sequence can be used directly, but a separate 3C or TEV site is the default because it lowers off-target risk [3,29].

1. Equilibrate 1 mL of anti-DYKDDDDK resin with 5 CV of TBS and bind the clarified sample by slow column flow or gentle batch incubation for 1-3 h.

2. Wash with 10-20 CV of TBS. To preserve complexes, minimize salt and detergent and run a negative-control resin in parallel.

3. Elute competitively for 30 min with 3-5 CV of 100-150 ug/mL competitor peptide. If low-pH elution is used, neutralize fractions immediately.

4. Desalt, then cleave with His-tagged 3C or TEV at 1:20-1:50 w/w.

5. Pass back over the antibody resin to remove the released tag and uncleaved fusion, then remove the protease by Ni-NTA.

6. Polish by SEC or IEX and check residual competitor peptide by LC or a functional-assay interference test.

6.4 GST - glutathione - 3C/TEV

GST captures easily and improves the solubility of some targets, but it is about 26 kDa and dimerizes naturally, which can distort the apparent oligomeric state. For new constructs, prefer a 3C or TEV site over thrombin or Factor Xa [8].

1. Equilibrate 1 mL of glutathione resin with 5 CV of binding buffer and load the clarified lysate.

2. Wash with 10-20 CV while checking whether target or GST leaks into the wash. Overloading degrades purity.

3. Elute with 5 CV of 10-20 mM reduced glutathione, then pool and exchange into cleavage buffer immediately.

4. React with His-tagged 3C or TEV at 1:50 w/w. If cleavage is slow, check accessibility around the protease site.

5. Pass back over the glutathione resin to remove released GST and uncleaved fusion, then remove the His-tagged protease by Ni-NTA.

6. Use SEC to distinguish the target oligomer from GST-induced oligomerization.

6.5 MBP - amylose - TEV

MBP gives among the strongest solubility enhancement of the fusion tags, but it is about 42.5 kDa and the target can precipitate again after cleavage [9]. A His-MBP-TEV architecture makes it easy to remove released MBP and TEV together by reverse IMAC.

1. Equilibrate 1 mL of amylose resin with 5 CV of column buffer. Do not include glucose or maltose in the load.

2. Load the clarified lysate and wash with at least 12 CV. Do not exceed the nominal resin capacity.

3. Elute 5-10 CV of 10 mM maltose in 0.5 CV fractions and pool the target fractions.

4. Exchange into TEV buffer and cleave with His-TEV at 1:50 w/w over 0.1-1 mg/mL. Compare precipitation across concentrations.

5. Bind His-MBP and His-TEV to Ni-NTA and take the tag-free target as the SFT. If MBP carries no His tag, pass back over amylose and remove the protease separately.

6. Confirm monodispersity by SEC and compute post-cleavage target recovery by mass balance.

6.6 His-SUMO - IMAC - Ulp1

His-SUMO combines IMAC capture, solubility enhancement, and generation of an exact native N terminus. Ulp1 recognizes the folded SUMO surface rather than a short linear motif, which gives high specificity [12,13].

1. Capture the His-SUMO fusion by the same IMAC described in 6.1 and elute with 300 mM imidazole.

2. Exchange into cleavage buffer and compare His-Ulp1 at 1:1000, 1:300, and 1:100 w/w at 0.2-2 mg/mL target on a small scale.

3. Compare overnight at 4°C with 1-3 h at 20-26°C and choose the shortest condition with the highest recovery.

4. Pass the cleavage reaction over Ni-NTA so that His-SUMO, His-Ulp1, and uncleaved fusion bind, taking the native-N-terminus target as the SFT.

5. Polish short peptides by RP-HPLC or LC-MS and ordinary proteins by IEX or SEC.

6. Confirm the exact cleavage junction by intact mass or N-terminal sequencing.

6.7 His-thioredoxin - IMAC - TEV/3C

Thioredoxin (Trx) is smaller than MBP and can serve as a redox-active soluble fusion partner. Because it can perturb target disulfide chemistry or redox assays, compare function after removal [4,5].

1. Capture the His-Trx fusion by IMAC using a 20-40 mM imidazole wash and 300 mM elution.

2. Exchange into cleavage buffer and use His-tagged TEV or 3C at 1:50 w/w.

3. Track the target oxidation state alongside the time course by nonreducing gel or intact mass.

4. Recover the reverse-IMAC SFT and confirm that Trx, protease, and uncleaved fusion are in SB.

5. Polish by IEX or SEC and measure redox-sensitive activity.

6.8 Intein-CBD self-cleavage

An intein-CBD system allows on-column tag release without an external protease, but it is sensitive to cleavage rate, junction residues, pH, temperature, and thiol. It can deliver a high-purity tag-free product, yet it is a specialist route chosen when protease cost or a native terminus dominates, not a general first choice [4].

1. Design the CBD-intein-POI or POI-intein-CBD junction with the recommended extein residues and check for premature cleavage in a pilot.

2. Equilibrate 1 mL of chitin resin with 10 CV of column buffer and load the clarified lysate.

3. Wash with 10-20 CV, displace with 3 CV of cleavage buffer, close the column, and run an 8-24 h time course at 4-25°C.

4. Open the column, elute the tag-free target over 3-5 CV, and check whether released-tag contamination rises in later fractions.

5. Remove the thiol or trigger immediately, polish by SEC or IEX, and confirm the junction by intact mass.

7. Polishing, formulation, and quality control

Affinity capture and tag cleavage are only the beginning. High purity comes from applying different separation principles in sequence and checking mass balance and function at every step. IEX or heparin addresses charge and nucleic-acid contamination, HIC addresses hydrophobic variants, and SEC removes aggregates and misassembled species. All values are research-grade starting points and must be predefined for the target and its intended use [1,20,25,28].

7.1 Selecting a polishing step

7.2 Running SEC and concentration

1. Choose an SEC medium that places the target oligomer in the middle of its fractionation range. Equilibrate with at least 1.5 CV and record blank pressure and UV baseline.

2. Start preparative SEC loads at 1-5% of column volume and reduce to 0.5-2% when higher resolution is required. Record sample concentration and viscosity.

3. Centrifuge the sample for 10-20 min at 15,000-20,000 x g and 4°C before injection. If filter adsorption is a concern, compare small-scale recovery first.

4. Assign void aggregates, target oligomer, and fragments using conductivity, fraction SDS-PAGE, and where possible MALS or calibration standards, not the A280 peak alone.

5. Pool the central fractions of the main peak first and store shoulders separately. Calculate both pooled mass and total recovered mass to document the purity-recovery trade-off.

6. Concentrate in short cycles at 4-10°C, mixing between cycles. Recover membrane-adsorbed target with a small volume of formulation buffer and analyse it separately.

7.3 Formulation screen

7.4 Research-grade release criteria and analytical panel

8. Troubleshooting: from variables to root cause

When something fails, do not immediately change the resin or the protease. First check concentration, volume, pH, conductivity, temperature, and time on both input and output, then trace the material on a single mass basis from load through flow-through, wash, elution, cleavage supernatant and pellet, subtractive flow-through and bound fraction, and SEC fractions. This variable-transfer-step order isolates the cause fastest.

9. Patent records by sequence and how to verify them

9.1 Three distinct sets of rights

Three independent sets of rights must be checked when adopting a purification tag, and resolving one leaves the others untouched. Trademark protects the name; naming by generic term and sequence instead of product name, as this document does, addresses only that. Patent protects sequence, composition, and method claims and is unaffected by what you call the thing. Contract arises from purchase terms, MTAs, and vector licences, and binds independently even after a patent has expired.

9.2 Patent records corresponding to tag sequences

The sequences below are those the reader encodes directly into a construct. Within the scope checked on the review date, every representative foundational patent in this table was displayed as expired. That is a record of the display only; it does not constitute permission to use, and later improvement patents and contractual terms may exist separately.

9.3 Patent records corresponding to capture ligands and resins

A capture ligand is not a sequence encoded into the construct; it is the stationary phase. The only record in this document displayed as unexpired sits here. Even with the same SBP tag, what you must check depends on which resin generation you select.

The same structure applies to other affinity routes. In the Protein A route of section 5.4, for example, native Protein A and alkali-tolerance-engineered ligands may sit in different rights situations, and the same is true of wild-type versus engineered proteases. This document does not designate any such engineered variant as a default route, but when selecting an actual resin or enzyme, first establish its generation and whether it is engineered, in the same way.

9.4 How to apply this chapter

1. During research, record the tag sequence, resin generation, protease, vector, and supplier terms you intend to use as a single configuration, and verify at that level.

2. Even when a foundational patent is displayed as expired, separate live rights may exist for a particular high-affinity ligand, engineered tag, protease mutant, resin chemistry, or kit.

3. Do not infer rights in other jurisdictions from a US number alone. Check the patent family and the register of the relevant jurisdiction separately.

4. Reading this document and understanding a protocol is not the same act as practising the technology. Verification is required at the point where actual production, service provision, or sale begins.

5. Feasibility in a published experiment and permission for commercial practice are different judgements. For commercial purposes, have your legal and patent function review a claim chart and freedom to operate.

10. Experimental appendix

10.1 Standard buffer compositions

The compositions below are starting points for a 1 L final volume. Adjust pH at the working temperature, and add reducing agents, protease inhibitors, detergents, and cofactors immediately before use. Filter every buffer through an appropriate membrane and follow the preservative-removal instructions in the resin specification.

10.2 Additive and resin compatibility

10.3 Fraction naming and mass balance

Step yield (%) = target mass at that step divided by target mass at the previous step, times 100. Overall yield (%) = final tag-free target mass divided by the estimated target mass in the starting sample, times 100. Do not treat total protein by A280 as target mass; correct it by densitometry, quantitative blot, or a target-specific assay.

10.4 Final pre-run checklist

The protein class, host, cellular location, and essential PTMs are documented.

At least two constructs and tag positions are defined, and every foreign affinity tag has a removal route.

The terminal residues produced by protease cleavage and the acceptable scar are confirmed at sequence level.

Resin binding capacity, sample target mass, and required resin volume are calculated from pilot results.

A tube map exists for retaining L, FT, W, E, C-S/C-P, SFT/SB, and SEC fractions.

Acceptance criteria are predefined for identity, purity, oligomeric state, activity, tag and protease residuals, and storage stability.

Supplier terms for the vector, resin, and tag platform to be used have been reviewed.

For animal, cell, or diagnostic applications, endotoxin limits and additional purpose-specific quality criteria have been separately approved.

11. References

The bibliographic details and links below were checked against the source records on 2026-08-12. The link text of each entry is a DOI or PubMed identifier. Bracketed numbers in the text correspond to the numbering of this list.

[1] Structural Genomics Consortium, et al. Protein production and purification. Nature Methods. 2008;5(2):135-146. doi:10.1038/nmeth.f.202

[2] Waugh DS. Making the most of affinity tags. Trends in Biotechnology. 2005;23(6):316-320. doi:10.1016/j.tibtech.2005.03.012

[3] Lichty JJ, Malecki JL, Agnew HD, Michelson-Horowitz DJ, Tan S. Comparison of affinity tags for protein purification. Protein Expression and Purification. 2005;41(1):98-105. doi:10.1016/j.pep.2005.01.019

[4] Kimple ME, Brill AL, Pasker RL. Overview of affinity tags for protein purification. Current Protocols in Protein Science. 2013;73:Unit 9.9. doi:10.1002/0471140864.ps0909s73

[5] Young CL, Britton ZT, Robinson AS. Recombinant protein expression and purification: a comprehensive review of affinity tags and microbial applications. Biotechnology Journal. 2012;7(5):620-634. PMID 22442034

[6] Bornhorst JA, Falke JJ. Purification of proteins using polyhistidine affinity tags. Methods in Enzymology. 2000;326:245-254. PMID 11036646

[7] Hochuli E, Doebeli H, Schacher A. New metal chelate adsorbent selective for proteins and peptides containing neighbouring histidine residues. Journal of Chromatography. 1987;411:177-184. PMID 3443622

[8] Smith DB, Johnson KS. Single-step purification of polypeptides expressed in Escherichia coli as fusions with glutathione S-transferase. Gene. 1988;67(1):31-40. doi:10.1016/0378-1119(88)90005-4

[9] Kapust RB, Waugh DS. Escherichia coli maltose-binding protein is uncommonly effective at promoting the solubility of polypeptides to which it is fused. Protein Science. 1999;8(8):1668-1674. doi:10.1110/ps.8.8.1668

[10] Schmidt TGM, Skerra A. The Strep-tag system for one-step purification and high-affinity detection or capturing of proteins. Nature Protocols. 2007;2(6):1528-1535. doi:10.1038/nprot.2007.209

[11] Schmidt TGM, Batz L, Bonet L, et al. Development of the Twin-Strep-tag and its application for purification of recombinant proteins from cell culture supernatants. Protein Expression and Purification. 2013;92(1):54-61. doi:10.1016/j.pep.2013.08.021

[12] 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. Journal of Structural and Functional Genomics. 2004;5(1-2):75-86. PMID 15263846

[13] Lee CD, Sun HC, Hu SM, et al. An improved SUMO fusion protein system for effective production of native proteins. Protein Science. 2008;17(7):1241-1248. doi:10.1110/ps.035188.108

[14] Zuo X, Li S, Hall J, et al. Enhanced expression and purification of membrane proteins by SUMO fusion in Escherichia coli. Journal of Structural and Functional Genomics. 2005;6(2-3):103-111. doi:10.1007/s10969-005-2664-4

[15] Kapust RB, Tozser J, Copeland TD, Waugh DS. The P1' specificity of tobacco etch virus protease. Biochemical and Biophysical Research Communications. 2002;294(5):949-955. doi:10.1016/S0006-291X(02)00574-0

[16] Waugh DS. An overview of enzymatic reagents for the removal of affinity tags. Protein Expression and Purification. 2011;80(2):283-293. PMID 21871965

[17] Vergis JM, Wiener MC. The variable detergent sensitivity of proteases that are utilized for recombinant protein affinity tag removal. Protein Expression and Purification. 2011;78(2):139-142. doi:10.1016/j.pep.2011.04.011

[18] Ullah R, Shah MA, Tufail S, et al. Activity of the human rhinovirus 3C protease studied in various buffers, additives and detergent solutions. PLOS ONE. 2016;11(4):e0153436. doi:10.1371/journal.pone.0153436

[19] Arnau J, Lauritzen C, Petersen GE, Pedersen J. Current strategies for the use of affinity tags and tag removal for the purification of recombinant proteins. Protein Expression and Purification. 2006;48(1):1-13. doi:10.1016/j.pep.2005.12.002

[20] Wingfield PT. Overview of the purification of recombinant proteins. Current Protocols in Protein Science. 2015;80:6.1.1-6.1.35. doi:10.1002/0471140864.ps0601s80

[21] Schuetz A, Bernhard F, Berrow N, et al. A concise guide to choosing suitable gene expression systems for recombinant protein production. STAR Protocols. 2023;4(4):102572. doi:10.1016/j.xpro.2023.102572

[22] Krishnarjuna B, Ramamoorthy A. Detergent-free isolation of membrane proteins and strategies to study them in a near-native membrane environment. Biomolecules. 2022;12(8):1076. doi:10.3390/biom12081076

[23] Pandey A, Shin K, Patterson RE, Liu XQ, Rainey JK. Current strategies for protein production and purification enabling membrane protein structural biology. Biochemistry and Cell Biology. 2016;94(6):507-527. doi:10.1139/bcb-2015-0143

[24] Vallejo LF, Rinas U. Strategies for the recovery of active proteins through refolding of bacterial inclusion body proteins. Microbial Cell Factories. 2004;3:11. doi:10.1186/1475-2859-3-11

[25] Kang J, Lee MS, Gorenstein DG. Application of RNase in the purification of RNA-binding proteins. Analytical Biochemistry. 2007;365(1):147-148. doi:10.1016/j.ab.2007.03.003

[26] Hober S, Nord K, Linhult M. Protein A chromatography for antibody purification. Journal of Chromatography B. 2007;848(1):40-47. doi:10.1016/j.jchromb.2006.09.030

[27] Ramos-de-la-Pena AM, Gonzalez-Valdez J, Aguilar O. Protein A chromatography: challenges and progress in the purification of monoclonal antibodies. Journal of Separation Science. 2019;42(9):1816-1827. doi:10.1002/jssc.201800963

[28] Liu HF, Ma J, Winter C, Bayer R. Recovery and purification process development for monoclonal antibody production. mAbs. 2010;2(5):480-499. doi:10.4161/mabs.2.5.12645

[29] Schmidt PM, Sparrow LG, Attwood RM, et al. Taking down the FLAG! How insect cell expression challenges an established tag-system. PLOS ONE. 2012;7(6):e37779. doi:10.1371/journal.pone.0037779

[30] Cantrell MS, Wall JD, Pu X, et al. Expression and purification of a cleavable recombinant fortilin from Escherichia coli for structure activity studies. Protein Expression and Purification. 2021;189:105989. doi:10.1016/j.pep.2021.105989

[31] Raducanu VS, Raducanu DV, Ouyang Y, Tehseen M, Takahashi M, Hamdan SM. TSGIT: an N- and C-terminal tandem tag system for purification of native and intein-mediated ligation-ready proteins. Protein Science. 2021;30(2):497-512. doi:10.1002/pro.3989

Document information and contact

Every tag, ligand, and resin in this document is named generically. The sequences, protein names, and chemical names used are those in common scientific use and do not designate any company's product. Third-party trademarks are the property of their respective owners.

For questions about construct design, tag placement, purification-route selection, or applying this protocol, contact geneorder@bioneer.co.kr. If you send the target sequence together with your product requirements, we will propose a starting construct and purification route based on the classification framework in this document.