Protein Aggregate Analysis for Biologics: SEC-HPLC, DLS, and AUC Methods Compared

Why Protein Aggregates Matter in Biologics

Protein aggregation is the formation of multimeric species from native monomers, and it represents one of the most critical quality attributes in biologics manufacturing. Aggregates can trigger unwanted immunogenic responses, reduce drug potency, and cause adverse clinical events. Regulatory agencies including the FDA and EMA require comprehensive protein aggregation analysis across the full size spectrum from soluble dimers to visible particles.

The challenge is that no single analytical method covers the entire range. Soluble aggregates in the 1–100 nm range require chromatographic or hydrodynamic separation. Submicron particles from 100 nm to 1 µm sit in an analytical gap. Subvisible particles from 1 to 100 µm are counted by compendial methods under USP <787> and <788>. A robust aggregate control strategy combines multiple orthogonal techniques, each covering a defined size window with known sensitivity limits.

Protein aggregation analysis is required at every stage of development: during formulation screening, in-process control, lot release, and stability testing. The methods differ in throughput, sample consumption, quantitative accuracy, and regulatory acceptance. This guide compares the five core techniques — SEC-HPLC, SV-AUC, DLS, MFI, and light obscuration — with practical guidance on when to use each one.

The Aggregate Size Spectrum and Method Coverage

Protein aggregates in biologics span more than five orders of magnitude in size. The analytical challenge is that each technique addresses only a portion of this range, and the most immunogenically concerning particles — those between 0.1 and 10 µm — sit precisely where most methods lose sensitivity.

The sub-micron range (100 nm to 1 µm) has historically been called the “analytical gap” because SEC-HPLC and SV-AUC lose resolution above approximately 50–200 nm, while compendial light obscuration methods only count particles above 10 µm. Nanoparticle tracking analysis (NTA) and resonant mass measurement have partially closed this gap, but neither is yet compendial for product release.

SEC-HPLC: The QC Release Workhorse

Size exclusion chromatography coupled with HPLC (SEC-HPLC) is the industry-standard method for quantifying soluble protein aggregates in biologics QC release and stability testing. It separates species by hydrodynamic radius through a porous stationary phase, with larger aggregates eluting first and the native monomer last within the permeation range.

How SEC-HPLC works for aggregate quantitation

The protein sample is injected onto a column packed with hydrophilic silica or polymer beads with a defined pore size (typically 200–500 Å for mAb analysis). Larger species are excluded from pores and elute earlier. UV detection at 280 nm (or 214 nm for low-concentration samples) records the chromatogram. Integration of peak areas gives the relative percentage of monomer, dimers, and higher molecular weight (HMW) species.

Column selection and method parameters

• TSKgel G3000SWxl (7.8 mm × 300 mm, 5 µm): fractionation range 10–500 kDa, the most widely used column for mAb aggregate analysis
• Agilent AdvanceBio SEC (4.6 mm × 300 mm, 2.7 µm): hydrophilic bonding minimises secondary interactions, faster runs at higher resolution
• Superdex 200 Increase (10 mm × 300 mm): FPLC-scale for preparative or large-volume analytical work, range 10–600 kDa

Typical mobile phase: 200 mM potassium phosphate with 250 mM KCl at pH 6.5–7.0, flow rate 0.5–1.0 mL/min, injection volume 10–50 µL, run time 20–30 minutes. The high-ionic-strength buffer suppresses electrostatic secondary interactions between the protein and the column matrix that would otherwise distort elution profiles.

Known limitations of SEC-HPLC

SEC-HPLC has three well-documented blind spots that make orthogonal methods essential:

1. Large aggregate filtration: aggregates above approximately 0.1 µm may be filtered by the column inlet frit (0.5 µm pores), removed before they reach the detector.
2. Reversible aggregate dissociation: the 10–100-fold dilution during chromatography, combined with shear forces through the column bed, can dissociate non-covalent oligomers back to monomer, under-reporting the true aggregate level in the undiluted sample.
3. Void volume co-elution: all species above the column exclusion limit elute in a single unresolved peak at the void volume, making it impossible to distinguish dimers from higher-order aggregates without multi-angle light scattering (MALS) detection.

SV-AUC: The Orthogonal Gold Standard

Sedimentation velocity analytical ultracentrifugation (SV-AUC) separates protein species in free solution by applying a centrifugal field of 20,000–50,000 rpm and tracking the sedimentation boundary over time using absorbance or interference optics. Because there is no stationary phase, matrix, or dilution, SV-AUC preserves the native association state and detects aggregates that SEC may miss.

FDA and EMA guidance documents recommend using at least two orthogonal methods for aggregate assessment during biologics characterisation, and SV-AUC is the most widely accepted orthogonal technique alongside SEC-HPLC. It provides absolute sedimentation coefficients (in Svedberg units, S) and can resolve monomer, dimer, trimer, and higher-order species without calibration standards.

Practical considerations

• Sample volume: 400–450 µL per cell, typically run in a 4-cell or 8-cell rotor (1.6–3.6 mL total per experiment)
• Run time: 4–8 hours at 40,000–50,000 rpm
• Data analysis: continuous c(s) distribution using SEDFIT or SEDPHAT software; modern approaches use two-dimensional spectrum analysis (2DSA) in UltraScan
• Sensitivity: detects aggregates down to approximately 0.5–1% of total protein depending on the optical system and sample concentration
• Throughput: low (4–8 samples per run), making it unsuitable for routine QC release but essential for development characterisation

A key advantage of SV-AUC is its ability to detect large soluble aggregates (up to approximately 100 S, corresponding to species of several million Daltons) that are invisible to SEC-HPLC. Studies comparing the two methods consistently find that SV-AUC reports equal or higher aggregate levels than SEC, confirming that SEC under-reports due to on-column dissociation and frit filtration.

DLS: Rapid Formulation Screening

Dynamic light scattering (DLS) measures the Brownian motion of molecules in solution by analysing the temporal fluctuations of scattered laser light. Smaller particles diffuse faster, producing rapid intensity fluctuations, while larger aggregates diffuse slowly. The autocorrelation function of the scattered intensity is fit to extract the translational diffusion coefficient (D_t), which is converted to hydrodynamic diameter (D_h) via the Stokes-Einstein equation:

D_h = k_BT / (3πηD_t)

where k_B is the Boltzmann constant, T is absolute temperature, and η is solvent viscosity.

Strengths for formulation development

• Speed: a single measurement takes 2–5 minutes, enabling high-throughput screening of formulation conditions
• Sample consumption: 2–5 µL in plate-based formats (e.g., Unchained Labs UNcle, Wyatt DynaPro Plate Reader III)
• Thermal ramps: DLS instruments measure T_onset (temperature at which aggregation begins) and T_m (melting temperature), providing colloidal and conformational stability rankings in a single experiment
• Polydispersity index (PdI): values below 0.1 indicate monodisperse samples; PdI above 0.3 suggests significant polydispersity or aggregation

Key limitations

DLS signal intensity scales with the sixth power of particle diameter (I ∝ d⁶), meaning a single 100 nm aggregate produces the same scattering intensity as 10⁶ native mAb monomers (~10 nm). This extreme bias toward large particles means DLS is highly sensitive to the presence of aggregates but cannot quantify them accurately. It also cannot resolve species closer than approximately 3-fold in diameter, so monomer, dimer, and trimer appear as a single peak.

How SEC-HPLC Tracks Aggregate Growth During Stability

SEC-HPLC chromatogram overlays are the standard way to visualise aggregate growth during accelerated and real-time stability studies. A fresh mAb drug substance typically shows >99% monomer with trace HMW shoulder. After 6 months at 40 °C (accelerated conditions), the HMW aggregate peak grows while the monomer peak decreases, and low molecular weight (LMW) fragments may appear at later retention times.

MFI, NTA, and Light Obscuration for Subvisible Particles

Subvisible particles (1–100 µm) are regulated under USP <787> (therapeutic protein injections), USP <788> (general parenteral injections), and USP <789> (ophthalmic solutions). The compendial limits are ≤6,000 particles ≥10 µm and ≤600 particles ≥25 µm per container for small-volume injections.

Light obscuration (LO)

Light obscuration is the primary compendial method. The sample flows through a sensor zone where each particle casts a shadow proportional to its cross-sectional area. It counts and sizes particles from approximately 2 to 150 µm. Limitations: translucent proteinaceous particles (refractive index close to the surrounding buffer) are under-counted or missed entirely, and the method requires 25–100 mL of sample.

Micro-flow imaging (MFI)

MFI captures high-resolution images of every particle as the sample flows through a thin flow cell. Each particle is sized, counted, and classified by morphology (circularity, aspect ratio, transparency). MFI detects translucent protein particles that light obscuration misses, with a lower size limit of approximately 1 µm and sample volumes of 0.5–1.0 mL. MFI is not yet compendial but is widely used as a complementary method and is referenced in USP <787> guidance.

Nanoparticle tracking analysis (NTA)

NTA bridges the gap between DLS and MFI by tracking individual particles from approximately 30 nm to 1 µm. A laser illuminates the sample, and a camera records the Brownian motion of individual particles. Software tracks each particle's diffusion path to calculate its hydrodynamic diameter. NTA provides particle-by-particle data (number-weighted distribution) rather than the intensity-weighted average of DLS, making it more representative of the true size distribution in polydisperse samples.

How to Choose the Right Method for Your Stage

Method selection depends on the development stage, the question being asked, available sample volume, and regulatory requirements. Early formulation screening prioritises speed and sample efficiency. GMP release requires compendial methods with validated acceptance criteria. Full characterisation studies during BLA filing need orthogonal coverage across the complete size spectrum.

Worked Example: SEC-HPLC Stability Study for a mAb

Why SV-AUC Detects Aggregates That SEC-HPLC Misses

SV-AUC consistently reports equal or higher aggregate content than SEC-HPLC for the same sample, and the discrepancy increases with aggregate size and reversibility. In a 2017 study comparing SEC and SV-AUC on 15 mAb drug substances, SV-AUC detected 1.2–3.8-fold more aggregate than SEC in samples with known large-aggregate populations. The gap arises because SEC-HPLC dilutes the sample 10–100-fold during injection and elution, dissociating reversible aggregates, while column frits physically remove particles above 0.1–0.5 µm before they can be detected.

This is why regulatory agencies require orthogonal methods. If SEC-HPLC is the only aggregate assay, reversible and large aggregates will be systematically under-reported, and the true in-vivo exposure to aggregates may be higher than the release specification suggests. SV-AUC closes this gap by measuring in free solution at the formulated concentration.

Related Tools

• Chromatography Calculator — column sizing, resin selection, and capacity estimation for protein purification
• HPLC Column Volume Calculator — dead volume, gradient delay, and theoretical plate calculation
• Protein MW & Extinction Coefficient Calculator — molecular weight and ε280 from sequence

References

1. den Engelsman J, Garidel P, Smulders R, et al. (2011). Strategies for the Assessment of Protein Aggregates in Pharmaceutical Biotech Product Development. Pharmaceutical Research, 28(4), 920–933. doi:10.1007/s11095-010-0297-1
2. Mahler HC, Friess W, Grauschopf U, Kiese S (2009). Protein Aggregation: Pathways, Induction Factors and Analysis. Journal of Pharmaceutical Sciences, 98(9), 2909–2934. doi:10.1002/jps.21566
3. Carpenter JF, Randolph TW, Jiskoot W, et al. (2009). Overlooking Subvisible Particles in Therapeutic Protein Products: Gaps That May Compromise Product Quality. Journal of Pharmaceutical Sciences, 98(4), 1201–1205. doi:10.1002/jps.21530
4. Berkowitz SA, Engen JR, Mazzeo JR, Jones GB (2012). Analytical Tools for Characterizing Biopharmaceuticals and the Implications for Biosimilars. Nature Reviews Drug Discovery, 11(7), 527–540. doi:10.1038/nrd3746
5. Philo JS (2006). Is Any Measurement Method Optimal for All Aggregate Sizes and Types? The AAPS Journal, 8(3), E564–E571. doi:10.1208/aapsj080365

Frequently Asked Questions