What is Dr. Zubair Khalid's research focus?

Dr. Zubair Khalid specializes in molecular virology, mRNA vaccine development, and computational biology, with a focus on avian pathogens like IBDV and Avian Reovirus.

Where is Dr. Zubair Khalid currently working?

Dr. Zubair Khalid is a Postdoctoral Research Associate at the University of Maryland (UMD), specifically within the Department of Animal and Avian Sciences.

Gel Filtration Chromatography (GFC): Principles and Applications

By Dr. Zubair Khalid, DVM, MS, PhD · 2026-08-11

Key Takeaways

Gel filtration chromatography (GFC), also known as size exclusion chromatography (SEC), separates biomolecules based on their hydrodynamic size (Stokes radius) rather than chemical affinity, utilizing a porous stationary phase that excludes larger molecules while allowing smaller ones to enter and be retarded.
The separation mechanism relies on the principle of molecular sieving, where molecules larger than the largest pores elute first in the void volume (V₀), while smaller molecules that can access the pore network elute later, with elution volume (Ve) being inversely proportional to the logarithm of molecular weight for globular proteins.
Key applications include desalting and buffer exchange by separating small molecules from larger proteins, purification of target proteins from contaminants, estimation of native molecular weight, and the study of protein-protein interactions by detecting stable complexes.
Common stationary phases include cross-linked dextrans (Sephadex), agarose (Sepharose), and composite materials (Superdex, Superose), with resin selection dictated by the molecular weight range of interest, required resolution, and compatibility with flow rates and buffers.
GFC offers gentle separation conditions, high recovery (>90%), and preservation of native protein conformation and activity, but is limited by low resolving power, small sample volume capacity for high resolution, and sample dilution during elution.
Critical parameters for successful GFC include proper column packing, buffer selection with adequate salt concentration (100-300 mM NaCl) to prevent non-specific interactions, and adherence to recommended sample volume (1-2% of bed volume for high resolution) and flow rates.

Introduction to Gel Filtration Chromatography

What is GFC?

Gel filtration chromatography (GFC) is a liquid chromatographic technique that separates biomolecules—primarily proteins, nucleic acids, and polysaccharides—on the basis of their hydrodynamic size (Stokes radius) rather than their chemical affinity for a stationary phase. The technique is also known as size exclusion chromatography (SEC) when applied to aqueous systems, and gel permeation chromatography (GPC) when applied to organic solvents. The fundamental distinction between GFC and other chromatographic methods is that the stationary phase is chemically inert: molecules do not bind to the matrix. Separation occurs purely as a function of whether a molecule can physically enter the pores of the chromatographic beads.

The column is packed with a porous gel matrix composed of spherical beads. Each bead contains a network of internal pores of a defined size distribution. When a mixture of molecules of different sizes is applied to the column, molecules larger than the largest pores are excluded from the bead interior and travel only through the mobile phase occupying the space between beads. Smaller molecules diffuse into the bead pores, and their progress down the column is retarded. Consequently, large molecules elute first, and small molecules elute last. This inverse relationship between molecular size and elution order is the defining feature of GFC.

Why use GFC?

GFC serves several distinct purposes in the biochemistry laboratory. Its most common applications include:

Desalting and buffer exchange: Removing small molecules (salts, nucleotides, denaturants) from a protein solution in a single, gentle step.
Purification: Separating a target protein from higher- and lower-molecular-weight contaminants, particularly in the final polishing steps of a purification scheme.
Molecular weight determination: Estimating the native molecular weight of an unknown protein by comparing its elution volume to a set of known standards.
Studying protein-protein interactions: Determining whether two proteins form a stable complex and estimating the stoichiometry of that complex.
Removing aggregates: Eliminating high-molecular-weight aggregates from a protein preparation, which is critical for biopharmaceutical formulations and structural studies.

Because GFC is performed under non-denaturing conditions and does not involve binding to a solid support, it is one of the gentlest chromatographic methods available. Proteins typically retain their native conformation, enzymatic activity, and oligomeric state throughout the separation.

Principles of Size Exclusion

The Gel Matrix and Pore Size

The stationary phase in GFC consists of spherical beads made from cross-linked polymers. The most common materials are cross-linked dextran (Sephadex), agarose (Sepharose), polyacrylamide, and composite or derivatized forms of these. The beads are porous: they contain an internal network of channels and cavities filled with solvent. The average pore diameter and the distribution of pore sizes are determined by the degree of cross-linking and the concentration of polymer used during manufacture.

The separation mechanism is often described as molecular sieving. A molecule's ability to enter a pore depends on whether its Stokes radius—the effective radius of a sphere with the same hydrodynamic behavior—is smaller than the pore diameter. Molecules with a Stokes radius larger than the largest pore are completely excluded from the bead interior. Molecules with a Stokes radius smaller than the smallest pore can enter all pores freely. Molecules of intermediate size can enter a fraction of the pores, depending on the pore size distribution.

It is critical to understand that GFC separates by hydrodynamic volume, not by molecular weight per se. Two proteins with identical molecular weights but different shapes (e.g., a globular protein versus an elongated rod-shaped protein) will have different Stokes radii and will elute at different volumes. The elongated protein will elute earlier because its larger Stokes radius restricts its access to pores.

Exclusion Limit and Fractionation Range

Two parameters define the useful operating window of a given GFC resin:

Exclusion limit: The molecular weight (or Stokes radius) above which all molecules are completely excluded from the pores. Molecules above this limit all elute in the void volume, with no resolution between them.
Fractionation range: The range of molecular weights over which the resin actually separates molecules. Within this range, there is a linear relationship between elution volume and the logarithm of molecular weight (for globular proteins).

For example, Sephadex G-75 has an exclusion limit of approximately 80,000 Da for globular proteins and a fractionation range of 3,000–80,000 Da. A protein of 100,000 Da and a protein of 500,000 Da will both elute in the void volume and cannot be separated on this resin. Conversely, a protein of 2,000 Da and a protein of 1,000 Da will both elute at the total volume and also cannot be separated.

The choice of resin must therefore be matched to the size range of the molecules you wish to separate. For a detailed comparison of commercially available matrices, see the Gel Filtration Chromatography Matrix reference.

Theoretical Basis: Partition Coefficient (Kav)

Elution Volume (Ve) and Void Volume (Vo)

To interpret a GFC experiment quantitatively, you must understand three volumes:

Void volume (V₀): The volume of mobile phase outside the beads, i.e., the space between beads. This is the elution volume of a molecule that is completely excluded from the pores. V₀ is typically measured using a very large molecule such as Blue Dextran (2,000 kDa), which is too large to enter any pores.
Total volume (Vt): The total geometric volume of the packed column bed. This equals the sum of the void volume, the volume inside the pores, and the volume occupied by the matrix itself.
Elution volume (Ve): The volume of buffer required to elute a given molecule from the column. Ve is measured from the moment of sample application to the moment the peak maximum appears at the detector.

A molecule that can enter all pores freely will elute at the total liquid volume (Vt − Vmatrix), which is the sum of V₀ and the internal pore volume (Vi). Molecules of intermediate size elute between V₀ and V₀ + Vi.

Calculating Kav

The partition coefficient (Kav) describes the fraction of the internal pore volume that is accessible to a given molecule. It is calculated as:

Kav = (Ve − V₀) / (Vt − V₀)

Where:

Ve = elution volume of the molecule
V₀ = void volume
Vt = total bed volume

Kav ranges from 0 to 1:

Kav = 0: The molecule is completely excluded; it elutes at the void volume.
Kav = 1: The molecule can access the entire internal pore volume; it elutes at the total liquid volume.
0 < Kav < 1: The molecule has partial access to the pores.

For a given resin, Kav is inversely related to the logarithm of molecular weight for globular proteins. This relationship forms the basis for molecular weight determination. By running a set of __MASK_2__ of known molecular weight, you can construct a calibration curve of Kav versus log(MW). The molecular weight of an unknown protein is then read from this curve using its measured Kav.

Column and Matrix Selection

Common Resins: Sephadex, Sepharose, Superose

The choice of gel filtration resin is dictated by the molecular weight range of interest, the required resolution, and the chemical compatibility with the sample.

Resin	Matrix Material	Fractionation Range (globular proteins, Da)	Typical Applications
Sephadex G-25	Cross-linked dextran	1,000–5,000	Desalting, buffer exchange
Sephadex G-75	Cross-linked dextran	3,000–80,000	Protein purification, MW determination
Sephadex G-200	Cross-linked dextran	5,000–600,000	Protein purification, complexes
Sepharose 6B	Agarose (6% bead)	10,000–4,000,000	Large proteins, protein complexes, viruses
Sepharose CL-4B	Cross-linked agarose	60,000–20,000,000	Very large complexes, lipoproteins
Superdex 200	Dextran-agarose composite	10,000–600,000	High-resolution FPLC purification
Superose 6	Composite agarose-dextran	5,000–5,000,000	High-resolution separation of large proteins

Sephadex resins are produced by cross-linking dextran with epichlorohydrin. The "G" number roughly correlates with the exclusion limit: G-25 has a low exclusion limit and is ideal for desalting, while G-200 has a much higher exclusion limit and is used for protein fractionation. Sephadex resins are relatively soft and cannot withstand high flow rates or pressures.

Sepharose resins are made from agarose, a linear polysaccharide extracted from seaweed. Agarose beads have larger pores than dextran beads and are suitable for very large proteins, protein-DNA complexes, and viruses. Cross-linked versions (Sepharose CL, Sepharose High Performance) have greater mechanical stability.

Superdex and Superose are composite resins designed for high-resolution separations on FPLC systems. They have rigid bead structures that tolerate higher flow rates and pressures, resulting in faster runs and sharper peaks. Superdex 200 is the workhorse for analytical gel filtration of proteins in the 10–600 kDa range.

Choosing the Right Resin

When selecting a resin, consider the following:

Molecular weight range of your target: The target protein should fall well within the fractionation range, ideally in the middle third where resolution is best. If you are separating a 50 kDa protein from a 200 kDa contaminant, Sephadex G-150 or Superdex 200 would be appropriate. If you are removing ammonium sulfate from a 50 kDa protein, Sephadex G-25 is the correct choice.

Required resolution: For preparative purification where baseline resolution is needed, choose a resin with a narrow fractionation range around your target size. For desalting, resolution is irrelevant—you only need to separate the protein from small molecules, and a coarse resin with a steep calibration curve is ideal.

Flow rate and pressure: Soft resins (Sephadex, Sepharose) operate under gravity flow or low-pressure peristaltic pumps. Rigid composite resins (Superdex, Superose) are compatible with FPLC systems operating at higher pressures.

Chemical compatibility: All standard GFC resins are compatible with aqueous buffers in the pH range 3–12. If you need to run GFC in the presence of detergents (e.g., for membrane proteins), choose a resin that is stable in the detergent of choice. Sepharose and Superose tolerate most non-ionic detergents; Sephadex is less robust.

For a deeper discussion of resin properties and selection criteria, consult the __MASK_3__ entry.

Running a GFC Experiment

Sample Preparation and Buffer

The sample must be clarified before loading onto a GFC column. Particulate matter will clog the column frit and disrupt the bed. Centrifuge the sample at 15,000 × g for 10 minutes at 4°C, or filter through a 0.22 µm syringe filter.

The buffer should be chosen to maintain protein stability and solubility. Typical buffers include 50 mM Tris-HCl, pH 7.5, containing 150 mM NaCl, or 50 mM sodium phosphate, pH 7.4, containing 150 mM NaCl. The salt (usually NaCl or KCl) is included at 100–300 mM to suppress non-specific electrostatic interactions between the protein and the matrix. Without salt, basic proteins may interact with residual negative charges on the resin, causing tailing and anomalous elution.

The sample volume and concentration are critical parameters. For a high-resolution analytical run, the sample volume should not exceed 1–2% of the total bed volume. For a desalting run, the sample volume can be up to 30% of the bed volume. The sample should be in the same buffer as the column; if it is not, the sample must be exchanged first, or the run will be complicated by buffer mismatch peaks.

Column Equilibration and Loading

A GFC column must be thoroughly equilibrated with the running buffer before use. Equilibrate with at least 2–3 column volumes of buffer. The flow rate should be set according to the resin's specifications. For gravity-flow columns, this is typically 0.5–1 mL/min for a 1 cm diameter column. For FPLC systems, follow the manufacturer's recommended flow rate (e.g., 0.5 mL/min for Superdex 200 10/300 GL).

To load the sample:

Stop the flow and allow the buffer to drain to the top of the bed.
Carefully apply the sample to the top of the bed using a pipette or sample loop, taking care not to disturb the bed surface.
Open the column outlet and allow the sample to enter the bed.
Once the sample has fully entered the bed, wash the walls of the column with a small volume of buffer (100–200 µL) and allow this to enter the bed as well.
Connect the column to the buffer reservoir and begin elution.

Elution and Detection

Elute the column with the running buffer at a constant flow rate. Collect fractions of a defined volume (e.g., 0.5–1 mL for analytical columns, 2–5 mL for preparative columns). Monitor the eluate with a UV detector at 280 nm for proteins (or 260 nm for nucleic acids). If the protein lacks tryptophan and tyrosine residues, detection at 214 nm (peptide bond absorbance) is more sensitive.

The entire run should be completed within 1–2 column volumes for analytical separations. For desalting, the protein elutes in the void volume and the salt elutes at the total volume, so the run is complete in one column volume.

Applications of GFC

Desalting and Buffer Exchange

Desalting is the most straightforward application of GFC. A column of Sephadex G-25 (fractionation range 1,000–5,000 Da) is equilibrated in the desired final buffer. The protein sample (containing salt, urea, or other small molecules) is applied, and the column is eluted. The protein, being larger than the exclusion limit, elutes in the void volume. The salt and other small molecules enter the pores and elute later, at the total volume.

This method is used to:

Remove ammonium sulfate after protein precipitation.
Remove imidazole after __MASK_4__.
Exchange a protein from a high-salt buffer into a low-salt buffer prior to ion exchange chromatography.
Remove reducing agents or denaturants such as β-mercaptoethanol, dithiothreitol, or urea.

The process is rapid (5–10 minutes for a small column), gentle, and achieves near-quantitative recovery. Unlike dialysis, which requires hours and large volumes of buffer, a desalting column can process a sample in minutes.

Molecular Weight Determination

GFC provides a convenient method for estimating the native molecular weight of a protein. The approach requires:

A column with a known fractionation range (e.g., Superdex 200).
A set of protein standards with known molecular weights covering the fractionation range. Typical standards include thyroglobulin (670 kDa), ferritin (440 kDa), aldolase (158 kDa), conalbumin (75 kDa), ovalbumin (44 kDa), and ribonuclease A (13.7 kDa).
Blue Dextran (2,000 kDa) to measure the void volume.

The procedure is as follows:

Run the standards individually or as a mixture and record their elution volumes.
Calculate Kav for each standard using the formula Kav = (Ve − V₀) / (Vt − V₀).
Plot Kav versus log10(MW) for the standards. This should yield a linear relationship within the fractionation range.
Run the unknown protein under identical conditions and calculate its Kav.
Read the molecular weight from the calibration curve.

This method estimates the native molecular weight, which includes any oligomeric state. A protein that is a homodimer in solution will appear at twice its monomeric molecular weight. This is a key advantage over SDS-PAGE, which measures the denatured monomer. However, the accuracy depends on the protein being globular. Elongated or intrinsically disordered proteins will have larger Stokes radii than globular proteins of the same molecular weight and will elute earlier, leading to an overestimate of molecular weight.

Protein-Protein Interactions

GFC is a powerful tool for studying protein-protein interactions because it preserves native conditions. If two proteins form a stable complex, they will co-elute as a single peak at a volume corresponding to the complex's molecular weight. If they do not interact, they will elute as separate peaks at their individual molecular weights.

For example, to test whether the bacterial proteins CheA (72 kDa) and CheW (18 kDa) form a complex, you would run each protein individually to determine their elution volumes, then run a mixture. If a complex forms, the mixture will show a peak at approximately 90 kDa (or higher if the stoichiometry is 2:1 or 2:2). The stoichiometry can be estimated from the molecular weight of the complex and the known monomer molecular weights.

This approach is also used to study protein-DNA interactions, antibody-antigen complexes, and the assembly of multi-subunit enzymes. For quantitative binding studies, GFC can be combined with analytical techniques such as multi-angle light scattering (SEC-MALS), which provides absolute molecular weights without calibration standards.

Advantages and Limitations

Advantages

Gentle conditions: GFC does not involve binding to a solid support, so proteins are not subjected to the harsh conditions (extreme pH, high salt, competing ligands) often required for elution in affinity or ion exchange chromatography. This preserves enzymatic activity and native conformation.
High recovery: Recovery is typically greater than 90%, even for labile proteins.
Predictable separation: The separation mechanism is well understood, and elution behavior can be predicted from molecular size.
Buffer flexibility: Any buffer compatible with the resin can be used, and the protein is eluted in the same buffer, so no post-column buffer adjustment is needed.
No sample loss from binding: Unlike __MASK_5__, where the target may bind irreversibly or denature during elution, GFC has no binding step.
Simultaneous separation and analysis: GFC can be used preparatively (to purify a protein) and analytically (to determine molecular weight or assess aggregation) in the same run.

Limitations

Low resolving power: GFC separates only on the basis of size. Two proteins of similar size cannot be resolved, even if they are chemically very different. The peak capacity is limited compared to reverse-phase or ion exchange chromatography.
Sample volume limitation: For high-resolution separations, the sample volume must be small (1–2% of bed volume). This limits the amount of protein that can be processed per run.
Dilution of sample: As the protein travels down the column, it diffuses and dilutes. The eluted peak is typically 2–5 times more dilute than the applied sample.
Speed: GFC is slower than many other chromatographic methods, particularly if high resolution is required. A high-resolution run on Superdex 200 takes 30–60 minutes.
Limited capacity: GFC columns have a lower loading capacity than ion exchange or affinity columns. You cannot overload a GFC column without destroying resolution.
Non-specific interactions: Some proteins interact weakly with the matrix, causing tailing or anomalous elution. This is minimized by including salt in the buffer but cannot always be eliminated.

Common Pitfalls and Troubleshooting

Column Packing and Flow Rate

Pitfall: Air bubbles in the column. Air bubbles disrupt the bed and create channels that allow proteins to bypass the matrix, destroying resolution. Always degas buffers under vacuum before use. If a bubble forms, the column must be repacked.

Pitfall: Uneven bed surface. A slanted bed surface causes sample to enter the bed unevenly, resulting in skewed peaks. The bed surface should be perfectly flat. If it is disturbed during sample loading, the top few millimeters of resin should be resuspended and allowed to settle.

Pitfall: Flow rate too high. Exceeding the recommended flow rate compresses soft resins (Sephadex, Sepharose), reducing the bed volume and destroying resolution. For gravity columns, the flow rate is set by the hydrostatic pressure (the height of the buffer reservoir above the column outlet). A typical height is 30–50 cm. For FPLC columns, follow the manufacturer's maximum pressure specification.

Sample Volume and Concentration

Pitfall: Sample volume too large. For analytical GFC, the sample volume should be 1–2% of the bed volume. A 24 mL Superdex 200 column (10/300 GL) should receive no more than 200–500 µL of sample. Loading 2 mL will result in severe peak broadening and loss of resolution.

Pitfall: Sample too concentrated. High protein concentrations (>10–20 mg/mL) increase viscosity, which distorts the flow profile and causes peak tailing. Dilute the sample to 1–5 mg/mL if possible.

Pitfall: Sample viscosity mismatch. If the sample contains high concentrations of glycerol (e.g., 50% glycerol used for protein storage), it will be more viscous than the running buffer and will not enter the bed evenly. Dilute the sample or exchange it into the running buffer before loading.

Peak Broadening and Resolution

Pitfall: Interpreting a single broad peak as a single species. A broad peak may contain multiple unresolved species. Reduce the flow rate, use a longer column, or choose a resin with a narrower fractionation range to improve resolution.

Pitfall: Ignoring non-specific interactions. Proteins that interact with the matrix (e.g., basic proteins interacting with residual carboxyl groups on dextran) elute later than expected and show tailing. Increase the salt concentration to 300–500 mM NaCl to suppress these interactions. If tailing persists, try a different resin.

Pitfall: Misidentifying the void volume peak. The first peak in a GFC run is not necessarily the void volume. It may be a large aggregate or a very large protein complex. Always measure the void volume separately using Blue Dextran (2,000 kDa) to establish V₀ for the column.

Pitfall: Using the wrong standards. Molecular weight standards must be globular proteins. If you use a non-globular standard (e.g., a fibrous protein or an intrinsically disordered protein), the calibration curve will be inaccurate. Use well-characterized globular standards such as those listed in the Gel Filtration Standards entry.

Practical Summary: Key Takeaways

Gel filtration chromatography separates molecules by hydrodynamic size (Stokes radius), not by molecular weight or chemical affinity.
Larger molecules elute first because they are excluded from the pores of the beads; smaller molecules elute later because they enter the pores and are retarded.
The partition coefficient Kav = (Ve − V₀) / (Vt − V₀) quantifies the fraction of pore volume accessible to a molecule and is inversely related to log(MW) for globular proteins.
Resin selection is determined by the molecular weight range of interest: Sephadex G-25 for desalting, Sephadex G-75/G-200 for mid-range proteins, Sepharose for very large complexes, and Superdex/Superose for high-resolution FPLC.
GFC is gentle, gives high recovery, and preserves native protein structure, making it ideal for studying protein complexes and determining native molecular weights.
The technique has low resolving power, requires small sample volumes, and dilutes the sample during elution.
Common pitfalls include air bubbles, excessive flow rates, oversized sample volumes, and misinterpretation of peaks due to non-specific interactions or incorrect void volume determination.

Frequently Asked Questions

What is the principle of gel filtration chromatography?

Gel filtration chromatography separates molecules based on their size (specifically, their hydrodynamic radius or Stokes radius). The column is packed with porous beads. Molecules larger than the bead pores are excluded and travel only through the space between beads, eluting first. Smaller molecules enter the pores, are retarded, and elute later. There is no binding between the sample and the stationary phase; separation is purely a physical sieving process.

How does gel filtration chromatography separate proteins?

Proteins are separated according to their Stokes radius. A protein's Stokes radius depends on its molecular weight, shape, and degree of hydration. Globular proteins of higher molecular weight have larger Stokes radii and elute earlier. The relationship between elution volume and log(molecular weight) is linear within the fractionation range of the resin, allowing molecular weight estimation.

What is the difference between gel filtration and size exclusion chromatography?

There is no fundamental difference. Gel filtration chromatography (GFC) is the term used when the mobile phase is aqueous and the samples are biological molecules (proteins, nucleic acids). Size exclusion chromatography (SEC) is the broader term encompassing all separations by size exclusion, including gel permeation chromatography (GPC), which uses organic solvents for synthetic polymers. In practice, GFC and SEC are often used interchangeably in the protein biochemistry literature.

Why do larger molecules elute first in gel filtration?

Larger molecules cannot enter the pores of the chromatography beads. They are confined to the mobile phase in the space between beads, which is the void volume. This is the shortest path through the column. Smaller molecules can diffuse into the pores, where the mobile phase is stagnant. They spend additional time inside the beads before diffusing back out, so their average velocity down the column is slower. Thus, larger molecules elute first.

What is the void volume in gel filtration?

The void volume (V₀) is the volume of mobile phase outside the beads—the space between the beads in the packed column. It is the elution volume of a molecule that is completely excluded from the pores. V₀ is measured by applying a very large molecule, such as Blue Dextran (2,000 kDa), which cannot enter any pores and elutes at the void volume. V₀ is typically 30–40% of the total bed volume.

How do you determine the molecular weight of a protein using gel filtration?

You construct a calibration curve using protein standards of known molecular weight. Run the standards and record their elution volumes (Ve). Calculate Kav for each standard using Kav = (Ve − V₀) / (Vt − V₀). Plot Kav versus log10(molecular weight). Then run the unknown protein under identical conditions, calculate its Kav, and read the molecular weight from the calibration curve. This gives the native molecular weight, which includes any oligomeric state.

What are common mistakes in gel filtration chromatography?

Common mistakes include: (1) loading too large a sample volume, which destroys resolution; (2) using a flow rate that is too high, which compresses soft resins; (3) failing to degas buffers, leading to air bubbles in the bed; (4) using non-globular proteins as molecular weight standards, which skews the calibration curve; (5) misidentifying the void volume peak; and (6) ignoring non-specific interactions between the sample and the matrix, which cause tailing and anomalous elution.