Protein Purification: Strategy, Affinity Tags, and Chromatography Sequence

Protein purification is the process of isolating one target protein from the thousands of other proteins, nucleic acids, lipids, and small molecules present in a cell lysate, until what remains is pure enough for the downstream use — structural work, an activity assay, an antibody production run, or a biophysical characterization. There is no single universal protocol; the right sequence of steps depends on the protein’s properties (size, charge, solubility, stability) and on whether it carries an engineered affinity tag. What follows is the general strategic logic most protein purification schemes share, not a rigid recipe.

The general purification strategy

Most protein purification workflows move through the same five stages, in roughly this order:

• Cell lysis — breaking open the cells (bacterial, insect, yeast, or mammalian) to release the target protein into a lysate, along with everything else the cell contained.
• Clarification — removing insoluble debris (unbroken cells, membrane fragments, inclusion bodies) from the lysate, typically by centrifugation and/or filtration, so what remains is a clear solution suitable for loading onto a chromatography column.
• Capture — the first chromatography step, chosen to bind the target protein selectively (or near-selectively) and concentrate it out of a large, dilute volume, while removing the bulk of contaminating host proteins.
• Intermediate purification — a second chromatography step, usually on a different separation principle than the capture step, to remove the contaminants that co-purified because they happened to share the capture step’s binding chemistry.
• Polishing — a final step aimed at removing closely related impurities (aggregates, degradation products, near-identical variants) and delivering the protein in a buffer suitable for its intended use.

Not every protein needs all five stages run as distinct steps — a well-behaved, highly overexpressed tagged protein might reach acceptable purity after capture and polishing alone — but thinking in these stages helps diagnose where a purification is failing when yield or purity comes up short.

Cell lysis and clarification

Lysis method depends on the expression host and the protein’s location (cytoplasmic, periplasmic, membrane-bound, or secreted). Common methods include sonication, high-pressure homogenization, enzymatic lysis (e.g., lysozyme for gram-negative bacteria), freeze-thaw cycling, and detergent-based lysis for membrane proteins. See What Is a Sonicator? for how ultrasonic disruption works and where it fits among these options — it is one of the most common lysis methods for small-to-moderate volume bacterial preps, but it generates heat that can denature a temperature-sensitive protein if not controlled with cooling and pulsed cycles.

Once cells are broken open, the lysate is clarified — almost always by centrifugation, sometimes followed by a filtration step (typically 0.22-0.45 µm) before loading onto a chromatography column, since particulates can clog resin and column frits. See What Is a Centrifuge? for how spin speed (RCF, not just RPM) and rotor choice affect how completely debris pellets out. Clarification speed and time matter: under-clarifying leaves particulates that foul the column; over-clarifying (excessive speed or time) can pellet soluble aggregates you actually wanted to keep, or waste time on a protein with limited stability at room temperature.

Affinity tags and capture chromatography

Most recombinant protein purification schemes rely on an engineered affinity tag fused to the target protein’s N- or C-terminus, because a generic tag lets one standardized capture step work regardless of the target protein’s own biochemistry. The three most common tags:

• His-tag (polyhistidine tag) with Ni-NTA — a short stretch of consecutive histidine residues (commonly six, a “6xHis-tag”) binds nickel ions immobilized on a chelating resin (nickel-nitrilotriacetic acid, Ni-NTA, or related IMAC resins using cobalt). This is the most widely used capture method for recombinant protein because the tag is small, rarely disrupts folding or function, and the resin is inexpensive and reusable. Elution is typically achieved with an imidazole gradient, which competes with the histidine residues for the immobilized metal.
• GST-tag (glutathione S-transferase) — a larger (~26 kDa) fusion tag that binds glutathione immobilized on agarose or sepharose resin, eluted with free reduced glutathione. GST is bulkier than a His-tag and can affect the fusion protein’s solubility (sometimes favorably, improving expression of otherwise poorly soluble proteins) or its native oligomeric state, since GST itself dimerizes.
• FLAG-tag — a short synthetic peptide (DYKDDDDK) recognized with high specificity by anti-FLAG antibodies, typically immobilized on agarose resin, and eluted competitively with free FLAG peptide or under low pH. Its main advantage is specificity (antibody-based capture has very low nonspecific background) at the cost of resin capacity and cost per purification compared to Ni-NTA.

Whether a tag is removed afterward depends on its intended downstream use. Tags are commonly cleaved when the target protein is going into structural work (crystallography, cryo-EM) where the tag could interfere with packing or introduce flexible, disordered regions, or when the tag risks interfering with a functional assay. Cleavage relies on a protease-recognition site engineered between the tag and the protein — TEV protease, thrombin, and PreScission (HRV 3C) protease are the most common choices, each recognizing a specific short peptide sequence and leaving a defined (ideally minimal) scar. After cleavage, a second pass over the original affinity resin (or a nickel/glutathione subtraction step) removes the free tag and any uncleaved fusion protein, since the cleaved tag and protease itself typically still carry a tag of their own. Tags are often left in place when the downstream use is an activity assay, an antibody, or anything where the small structural risk from the tag doesn’t outweigh the extra time and yield loss cleavage costs.

The chromatography sequence: affinity, ion exchange, size exclusion

A common logic — not a rigid rule — runs affinity chromatography first (capture), ion exchange chromatography second (intermediate purification), and size exclusion chromatography last (polishing):

• Affinity chromatography (capture) separates on the tag-resin interaction described above, which has little to do with the target protein’s own size or charge, so it captures the tagged protein out of a large volume of dilute, mixed lysate efficiently.
• Ion exchange chromatography (intermediate) separates based on surface charge at a given pH, using either an anion exchanger (binds negatively charged proteins) or a cation exchanger (binds positively charged proteins), eluted with a salt gradient. Because charge is largely unrelated to the tag-binding chemistry used in the capture step, this step removes a different population of contaminants than affinity chromatography did — including proteins that bound the affinity resin nonspecifically.
• Size exclusion chromatography (polishing, also called gel filtration) separates by hydrodynamic size, with larger species eluting first as they are excluded from the porous resin beads and smaller species retained longer as they diffuse into the pores. Run last, it removes aggregates and any remaining size-distinct contaminants, and it exchanges the protein into its final storage buffer in the same step, since the column is equilibrated in whatever buffer the eluted protein comes out in.

This sequence works because each step separates on a different physical property (tag affinity, then charge, then size), so contaminants that survive one step are unlikely to survive the next by coincidence. Real workflows deviate from it constantly: a highly pure, highly expressed tagged protein might skip straight from affinity capture to a size exclusion polishing step with no ion exchange step at all; a poorly behaved protein might need an extra ion exchange pass, a hydrophobic interaction chromatography (HIC) step, or a different order entirely if the target is unstable at the pH or salt conditions a particular step requires.

Assessing purity and yield

Three complementary checks are standard at each stage of a purification, not just at the end:

• SDS-PAGE — denaturing gel electrophoresis separates proteins by molecular weight regardless of native charge or shape, giving a visual read on purity (a single band at the expected molecular weight, versus a lane full of contaminating bands) and a rough sense of relative abundance across fractions. See What Is Gel Electrophoresis? for the underlying separation principle, and SDS-PAGE: How Protein Gel Electrophoresis Works, Gel Selection, and Troubleshooting for gel percentage selection and troubleshooting smeared or missing bands.
• A280 concentration — absorbance at 280 nm, driven mainly by tryptophan and tyrosine residues, gives a fast concentration estimate when the protein’s extinction coefficient is known (calculable from its amino acid sequence). It doesn’t distinguish the target protein from a contaminant absorbing at the same wavelength, which is why it’s paired with SDS-PAGE rather than used alone. Where a sequence-based extinction coefficient isn’t reliable (unusual folding, added chromophores) or a total-protein number across a mixed sample is needed, a colorimetric assay is used instead — see BCA Protein Assay: Working Reagent, Standard Curve and Calculating Concentration and Bradford Protein Assay: Standard Curve, Dilution Series and Calculating Concentration.
• Activity assays — where applicable (an enzyme, a binding protein, a receptor), a functional assay confirms the protein wasn’t just concentrated and purified but also stayed correctly folded and active through the process. Specific activity (activity per unit mass) tracked across purification steps is a standard way to see whether a step enriched for functional protein or merely concentrated a mix of active and misfolded/aggregated material.

Tracking yield and purity at every step, not just the final pool, is what actually diagnoses a failing purification: a large yield loss between capture and intermediate purification points at that specific step (wrong pH, competing binder, protein instability under those buffer conditions) rather than requiring a re-optimization of the whole scheme from scratch.

Frequently asked questions

What is the general strategy behind protein purification?

Lysis to release the protein, clarification to remove insoluble debris, a capture step (usually affinity chromatography) to isolate and concentrate the target, intermediate purification (commonly ion exchange) to remove co-purifying contaminants, and polishing (commonly size exclusion) to remove aggregates and finalize the buffer. Not every protein needs all five as distinct steps.

Do you always need to remove an affinity tag after purification?

No. Tag removal is common before structural work or when the tag risks interfering with a functional assay, using an engineered protease-cleavage site (TEV, thrombin, or PreScission are the most common). Many purifications leave the tag in place, particularly for antibody production or assays unaffected by a small terminal tag, since cleavage costs time and yield.

Does chromatography always have to run affinity, then ion exchange, then size exclusion, in that order?

That order is a common logic, not a rule, because each step separates on a different physical property (tag affinity, charge, size), which keeps contaminants surviving one step from also surviving the next by coincidence. Real workflows regularly skip a step, add an extra pass, substitute hydrophobic interaction chromatography, or reorder steps to protect a protein that’s unstable under a given step’s buffer conditions.

How is protein purity actually measured?

SDS-PAGE gives a visual, size-based purity read (ideally a single band at the expected molecular weight); A280 or a colorimetric assay (BCA, Bradford) gives concentration; and, where the protein has a measurable function, an activity assay confirms the purified protein is still correctly folded and active, not just concentrated.

Related equipment and technique guides

Protein purification workflows draw on several instrument- and technique-level guides covered separately on this site: What Is a Sonicator? and What Is a Centrifuge? for the lysis and clarification stages, What Is Gel Electrophoresis? and SDS-PAGE: How Protein Gel Electrophoresis Works for purity assessment, and BCA Protein Assay / Bradford Protein Assay for concentration determination. For downstream identity confirmation, see Protein Identification by Mass Spectrometry.