Figure 1. MALS and RI detection traces for size-exclusion chromatography (SEC) of a polymer series, highlighting the ability of MALS to detect large, multi-chain aggregates. (Frank et al., 2015)
1. Define the Aggregate Question before Choosing a Method
Begin by defining the material, concentration, formulation, expected monomer size, known oligomers, and the decision the data must support. A release-oriented method may need precise relative abundance of soluble high-molecular-weight species, whereas formulation screening may prioritize early detection of colloidal instability. A characterization study may need absolute molar mass, reversibility, and evidence about particles larger than the SEC separation range.
Use "aggregate" carefully. Covalent dimers, reversible oligomers, nonspecific clusters, and subvisible particles are not interchangeable. State which size range and physical state each procedure can observe. No single result establishes the complete particle distribution, and a method that reports a monomer percentage does not automatically characterize insoluble or filter-retained material.
Define an orthogonal measurement plan
Assign each method a hypothesis. SEC can test whether discrete soluble species are chromatographically resolvable. MALS can test whether those peaks have the molar masses expected for monomer or oligomer. DLS can compare hydrodynamic size distributions before and after stress without a column. Add a particle-counting or sedimentation method when the question extends beyond these capabilities.
2. Control Sampling before Instrumental Analysis
Sampling can create or remove the very species being measured. Record thawing, mixing, equilibration temperature, time from preparation to injection, container, filtration, centrifugation, and dilution. Vortexing, repeated aspiration, air exposure, or a poorly recovered filter can change low-level aggregate results. Analyze an undisturbed aliquot where practical and compare any clarification treatment by target and aggregate recovery.
Dilution deserves a specific study. Reversible oligomers may dissociate as concentration falls, while dilution into a mismatched buffer can trigger association. Prepare dilutions with formulation-matched matrix and allow a defined equilibration time. For stressed samples, protect the original material from further incubation while the sequence runs.
Bracket the sequence with a reference or control sample to detect column conditioning, autosampler aging, and detector drift. Randomize formulations when feasible. If high-aggregate samples can carry over, place blanks and cleaning injections according to demonstrated risk rather than relying on a generic wash.
3. Develop SEC as a Controlled Separation
Choose pore range so the monomer and expected soluble oligomers occupy the useful fractionation region. Very large species may elute in or near the void volume, where different structures are compressed into one region. Select column dimensions, particle size, and flow to balance efficiency, pressure, analysis time, and protein stability. Relevant size-exclusion chromatography products can support development at analytical or preparative scale.
Keep injection volume and mass within a range that preserves resolution and detector response. Volume overload broadens peaks; mass overload can alter shape or promote nonideal behavior. Evaluate a load series with the actual protein. Report both concentration and injected mass because identical volumes can behave differently across formulations.
Resolve the integration strategy before routine use. Define baseline placement, shoulder treatment, void-region reporting, fragments, and peaks affected by system artifacts. Use representative low- and high-aggregate samples. A visually clean chromatogram can still give unstable percentages when integration boundaries move between analysts or software settings.
4. Distinguish Size Separation from Secondary Interaction
Ideally SEC retention reflects access to pores, but proteins can interact electrostatically or hydrophobically with the stationary phase. Symptoms include concentration-dependent retention, tailing, unexpected recovery, or apparent masses inconsistent with orthogonal evidence. Screen ionic strength, pH, and compatible additives within the protein’s stable range. Do not improve peak shape with a condition that changes the sample state.
Evaluate total recovery by comparing injected and recovered protein or a justified detector-based balance. Loss of a sticky high-molecular-weight species can make the sample appear purer. Conversely, on-column association can create an apparent aggregate peak. A no-column hold control and an alternative stationary-phase chemistry can help separate sample behavior from column interaction.
Use a mobile phase with controlled preparation, filtration, degassing, composition, and age. Confirm that preservatives or salts do not interfere with downstream light-scattering or refractive-index detection. Match sample and mobile-phase matrices closely for sensitive detectors.
5. Use SEC-MALS for Molar Mass across Resolved Peaks
MALS estimates weight-average molar mass from excess scattering together with concentration at each elution slice. Concentration is commonly obtained from refractive-index or absorbance data. The calculation requires an appropriate refractive-index increment or extinction coefficient and accurate detector alignment. MALS is not simply a more sensitive UV detector; assumptions about composition directly affect the mass result.
Inspect molar mass across each peak rather than reporting one number. A flat mass trace supports a relatively uniform population, while a slope can indicate unresolved mixtures, band broadening between detectors, or concentration-dependent association. Exclude detector spikes only under predefined rules. Large particles scatter strongly and can dominate a few data points even when their mass fraction is small.
For conjugated or glycosylated proteins, a single protein value for refractive-index increment may bias interpretation. Component-specific analysis may be possible when independent concentration signals and optical constants are justified. Otherwise, state that the reported mass is an apparent value under the selected model.
6. Use DLS as a Sensitive Ensemble Screen
DLS derives translational diffusion behavior and a hydrodynamic size estimate from fluctuations in scattered light. It is fast and consumes little sample, making it valuable for buffer, temperature, concentration, and hold-time screens. However, scattering intensity rises steeply with particle size, so a small number of large particles or dust can dominate an intensity-weighted distribution.
Control cuvette cleanliness, equilibration time, temperature, measurement position, viscosity input, and replicate acquisition. Report the analysis model, derived diameter or radius, polydispersity indicator, and quality flags. Converting intensity distributions into volume or number distributions introduces assumptions; it does not create a direct mass fraction equivalent to SEC peak area.
Interpret a change in apparent size with context. A larger value may reflect oligomerization, non-spherical shape, viscosity error, or a trace particle population. A stable average can also conceal a changing minority. Inspect correlation functions and replicate consistency rather than accepting a single fitted number.
7. Reconcile Results without Forcing Agreement
SEC physically separates species and usually dilutes them during passage through the column. DLS measures the unseparated sample at its test concentration. Consequently, a reversible oligomer may appear in DLS but dissociate during SEC, while an irreversible dimer may give a discrete SEC peak with only a small effect on the DLS average. SEC-MALS reports the species that survive and resolve during chromatography.
| Observation | Likely explanations to test | Focused experiment |
|---|---|---|
| SEC high-mass peak; DLS unchanged | Low mass fraction or modest size difference | Enrich the SEC fraction and remeasure |
| DLS large-size signal; SEC little aggregate | Dust, very large particles, or reversible association | Centrifugation control and concentration series |
| SEC retention shifts with load | Secondary interaction or self-association | Load series in alternative mobile phases |
| MALS mass rises across a peak | Unresolved mixture, dispersion, or association | Detector-alignment check and slower or higher-resolution SEC |
| Poor SEC recovery | Adsorption, precipitation, or excluded material | Mass balance including filters and column wash |
Document unresolved discrepancies. Orthogonality is valuable because methods respond to different properties; identical-looking answers are not the goal. The goal is a model of sample behavior that predicts changes under relevant conditions.
8. Build System Suitability around the Failure Modes
SEC suitability can include retention-time precision, efficiency, peak symmetry, resolution between relevant species, area precision, and carryover. Select acceptance criteria from method capability and decision risk rather than copying a standard-protein target. A reference protein may monitor the system, but a product-specific control is needed when nonideal interaction or aggregate integration is central.
For MALS, monitor normalization, detector alignment, baseline quality, concentration-detector agreement, and a suitable molar-mass control. For DLS, use a size standard for instrument checks and a sample control for preparation variability. Trending is often more informative than a single pass/fail value.
Challenge minor changes in mobile phase, temperature, flow, injection mass, equilibration, and data processing during robustness work. The method should retain its ability to detect and quantify the species that drive the intended decision.
9. Quantify with Clear Denominators and Uncertainty
SEC aggregate percentage is commonly based on corrected chromatographic area, but detector response may differ among species and material can be lost before detection. State the detector, wavelength, integration basis, and treatment of excluded or fragment peaks. When an aggregate reference is unavailable, relative area can remain useful if its limitations are explicit and performance is demonstrated.
Use dilution linearity, spike or blend studies, repeatability, intermediate precision, and relevant stressed samples to evaluate quantitation. The lower range should reflect the specification or development threshold. Avoid claiming accuracy from peak-area precision alone; recovery and species response matter.
Report MALS molar mass with model inputs and uncertainty appropriate to signal quality. Report DLS as a hydrodynamic measurement, not percentage aggregate. Align all result units with the scientific question so readers do not compare incompatible percentages and diameters.
10. Design Stress Studies that Produce Interpretable Material
Apply heat, agitation, freeze-thaw, light, pH, concentration, or interface stress according to plausible product risks. Aim for informative change without destroying the sample into an unrelated mixture. Use time points that show progression, retain an unstressed control, and stop or freeze the reaction consistently before analysis.
Analyze stress samples by the same orthogonal panel. SEC can resolve soluble products; MALS can test their masses; DLS can detect a shift toward larger hydrodynamic species. Include visible appearance and total recovery. A reduction in SEC monomer with no increase in soluble aggregate may represent precipitation or material trapped upstream.
Use stress results to support specificity and mechanistic understanding, not to assert that every forced species will appear during shelf life. Compare relevant long-term or accelerated samples when available.
11. Transfer and Trend the Aggregate Method
Transfer the complete sample-handling procedure, mobile-phase preparation, column conditioning, injection sequence, integration method, and detector configuration. Use shared stressed and unstressed samples to separate laboratory bias from material instability. Compare raw chromatograms, recovery, retention, resolution, molar-mass traces, and DLS correlation functions rather than only final percentages.
During routine use, trend monomer recovery, high-molecular-weight area, void-region events, retention, pressure, reference mass, and DLS quality metrics. A gradual loss of SEC recovery or more frequent scattering spikes may identify column fouling, sample particulates, or handling drift before a formal suitability failure. Investigate trends with retained controls and column history.
12. Record the Full Measurement Context
In reports, include protein concentration, formulation, sample treatment, injection load, column and mobile phase, detector configuration, recovery, integration rules, MALS optical inputs, DLS viscosity and temperature, and any excluded data. Present representative raw traces beside summary tables. These details allow a reviewer to distinguish a sample change from a method-dependent observation and make future bridging practical.
Archive representative unstressed, stressed, and anomalous samples when stability permits. These materials help bridge new columns, detectors, software, or laboratories and show whether a future discrepancy reflects the sample or the measurement system.
13. Diagnose an Apparent High-Molecular-Weight Increase
Consider a protein whose SEC high-molecular-weight area doubles after a formulation change, while DLS and visual appearance remain unchanged. Re-injection of the isolated early SEC fraction produces mostly monomer, suggesting the peak is not a stable aggregate. The new formulation has lower ionic strength, raising the possibility of reversible column interaction or association.
A load series shows that the early peak grows at higher injected concentration and diminishes after formulation-matched dilution. Increasing mobile-phase ionic strength improves recovery and removes the anomalous peak without changing an independently stressed aggregate control. SEC-MALS measures a mass near monomer across much of the suspect region, although data quality is limited at the leading edge.
The evidence supports a method-induced or reversible phenomenon rather than a new stable aggregate population. The team revises the mobile phase, validates recovery and specificity, and keeps DLS as a formulation-state screen. The investigation records why the original area result was not accepted at face value and which experiments discriminate the mechanisms.
For external study support, provide molecule type, formulation, concentration, expected oligomers, stress history, chromatograms, and existing particle data through protein aggregate analysis services. A useful plan specifies the question each method will answer and preserves enough sample for discrepancy testing.
Build an orthogonal aggregate-analysis plan
Share the protein, formulation, concentration, expected species, stress conditions, and decision threshold to align SEC, SEC-MALS, and DLS.
References
- ICH. Q2(R2): Validation of Analytical Procedures. 2023; error correction 2025.
- ICH. Q6B: Specifications—Test Procedures and Acceptance Criteria for Biotechnological/Biological Products.
- Fekete S, Beck A, Veuthey JL, Guillarme D. Theory and practice of size exclusion chromatography for the analysis of protein aggregates. Journal of Pharmaceutical and Biomedical Analysis. 2014;101:161-173. doi:10.1016/j.jpba.2014.04.011
- Ogawa T, Hirokawa N. Multiple analyses of protein dynamics in solution. Biophys Rev. 2018;10(2):299-306. doi:10.1007/s12551-017-0354-7
- Frank P, Prasher A, Tuten B, Chao D, Berda E. Characterization of single-chain polymer folding using size exclusion chromatography with multiple modes of detection. Appl Petrochem Res. 2015;5(1):9-17. doi:10.1007/s13203-014-0046-1
