UHP-SEC-MALS: Enhanced Biophysical Characterization of Biotherapeutics with Light Scattering and UHPLC

Photo Credit: antishock/Shutterstock.com

Daniel Some, Wyatt Technology Corp., Santa Barbara, California, USA

Faster analyses, better separations, and lower consumption of sample and mobile phase are the primary drivers of size-exclusion chromatography with sub-2-µm beads and ultrahigh-pressures (UHP-SEC). The flip side of these benefits is higher sensitivity to column calibration errors and drift. There is also a relatively small selection of column chemistries available for eliminating non-ideal sample-column interactions. UHP-SEC can be combined with on-line multi-angle light scattering (UHP-SEC-MALS) to overcome these limitations and provide absolute molar mass and size of biomacromolecules, independently of retention time. UHP-SEC-MALS is also necessary for UHP-SEC characterization of proteins and biotherapeutics that have no appropriate reference standards, such as glycoproteins and PEGylated proteins.

Across biopharmaceutical analytical laboratories, process development, and quality control, ultrahigh-pressure size‑exclusion chromatography (UHP-SEC) is rapidly replacing standard high-pressure size-exclusion chromatography (HP-SEC) as the technique of choice for detecting and quantifying degradants of drug substance and drug product, such as fragments and soluble aggregates. In our experience of approximately a 4–6-fold reduction in run times and 10-fold reduction in sample and solvent consumption, the time- and cost‑saving benefits are compelling. Factoring in significantly better resolution in separation and the consequential improvement in detection and identification of impurities, it is hard to understand why cutting-edge R&D and production facilities would not make the switch.

With all the advantages of UHP-SEC, though, there are some drawbacks that need to be addressed. The first is surface chemistry: when optimizing a separation method, it is important to find conditions where both the reference markers and the sample to be tested do not interact non-ideally (for example, via hydrophobic interactions) with the column packing. This is because column calibration depends on the assumption that the reference molecules and sample undergo only ideal (steric) interaction with the column, and any additional “stickiness” will render erroneous results. Compared with HP-SEC columns there are relatively few surface chemistries available in UHP-SEC column format for method optimization, hence a higher probability that the method will not be fully optimized.

The second is column creep and drift in chromatography conditions: with such a fast separation, even a minor difference in elution time as a result of column ageing or change to the pump operation conditions will lead to a relatively large error in estimated molecular weight.

The Solution: Light Scattering

On-line multi-angle light scattering (MALS) (1,2) measures protein and polymer molar mass at each elution volume, absolutely, without reference to retention time. It does so by a first-principles physical relationship between the scattered intensity (Isc), molar mass (M), and concentration (c):  

     [1]

where dn/dc is essentially fixed for most proteins in aqueous buffers, and the constant of proportionality may be calculated from the measurement system properties such as laser wavelength. Measurement of Isc and c therefore directly yields the value of M, contiguously, as the solution passes through the MALS detector’s flow cell. The determination of molecular weight is independent of conformation or shape as well as elution properties. For these reasons, SEC-MALS has long been the de facto standard for rigorous analysis of proteins and polymers in solution separated by standard HP-SEC.

With the addition of an embedded dynamic light scattering (DLS) module, SEC-MALS can add a second, independent determination of molecular size (hydrodynamic radius). The combination of molar mass and molecular size gives an indicator of unfolding or conformation. In addition to yielding an intrinsically valuable biophysical property, this information can also help to understand and troubleshoot problematic SEC separations, as shown below.

MALS detectors for UHP-SEC have recently become available (3), conferring the benefits of MALS and embedded DLS on the newer separation technology. UHP-SEC-MALS maintains the central benefits of UHP-SEC, while preserving chromatographic resolution and providing absolute molecular weight and size of the eluting species.

The importance of UHP-SEC-MALS is illustrated in Figure 1, where the chromatograms of four different monoclonal antibodies (mAbs) are shown. A naive interpretation of the chromatograms based on elution time would suggest that mAb4 has a significantly (and surprisingly) lower molar mass than the other three; yet MALS analysis shows that all four do in fact have the expected molecular weight of 145–150 kDa. With the addition of DLS data, which show that all four also have the same hydrodynamic radii of 5.3±0.3 nm, we can further deduce that the reason for the late elution of mAb4 is not a difference in conformation but “stickiness” or a non‑ideal column interaction.

Aggregates and Fragments, Revealed

While it is well established that degradants of IgG include fragments and aggregates, the excellent resolution of UHP-SEC reveals that the aggregate peak actually includes multiple subpeaks, shown in Figure 2. The use of MALS establishes the most likely identity of those subpeaks as aggregates of monomers and fragments: IgG dimer (307 kDa), dual heavy chain dimer (188 kDa), IgG monomer + dual heavy chain (234 kDa), and IgG monomer + dual heavy chain + single light chain (266 kDa). The low chromatographic dispersion and high sensitivity of a MALS detector means that it can provide reliable analysis, even when soluble aggregates represent a tiny fraction of the main peak.

As mentioned above, in order to determine molar mass, both light scattering intensity and concentration values are required. While a UV detector suffices for determining concentration when the protein species and its extinction coefficient are known, impurities such as host cell proteins and their associated extinction coefficients may be unknown. In those cases, a differential refractive index (dRI) detector, which is a universal concentration detector useful for proteins, biopolymers, nucleic acids, sugars, and most other macromolecules, even in the absence of chromophores or fluorophores, can provide the requisite data. As the overwhelming majority of proteins have a universal dRI response of 1.8510-4 mL/g at the MALS laser’s wavelength, the molecular weight of unknown proteins may also be determined by MALS-dRI.

When new peaks resulting from impurities are found, it can be challenging to identify the species behind them. The UHP-SEC chromatogram of Figure 3 shows the utility of MALS-dRI for identifying unknown proteins. In addition to the IgG monomer at 2.5 mL elution volume with molecular weight of 145 kDa, the additional peaks following the monomer are tentatively identified as dual heavy chain (95 kDa), single heavy chain (45 kDa), and light chain (24 kDa). Furthermore, analyzing the combined dRI and UV signals permits the extinction coefficients of each of the species to be determined. The nearly identical values clinch the identification of these peaks as fragments of IgG rather than foreign impurities.

Where Column Calibration Dare Not Go: Protein Conjugates

One of the most challenging tasks for biophysical characterization is the analysis of protein conjugates, which include glycoproteins, PEGylated proteins, protein‑polysaccharide antigens, and antibody-drug complexes. No well‑characterized, narrow reference standards exist for these often‑heterogeneous complexes, so analysis by SEC with column calibration is impossible.

As an absolute technique, which does not depend on elution time and reference markers, SEC-MALS is the primary biophysical method for analyzing the molecular weight and heterogeneity of protein conjugates. However, since they are typically binary complexes, their concentration at each elution volume cannot be determined with a single concentration detector: two distinct concentration measurement signals must be analyzed, to determine not just overall concentration but also the weight‑averaged dn/dc value of the complex. These are plugged back into the light scattering equation for the final molar mass determination.

Triple detection combining MALS, UV, and dRI is the most common method for doing so (4). The simplest scenario is presented in Figure 4: a glycoprotein and its aggregates. In the MALS-UV-dRI analysis of a glycoprotein, the UV detector measures the concentration of just the proteinaceous part of the complex since glycans do not absorb UV. On the other hand, dRI is a universal concentration detector, so it measures the combined concentration of protein and modifier. This information can be broken down to give the fraction of protein and fraction of modifier, from which the overall dn/dc of the complex is calculated.

The final light scattering result provides not just the overall molecular weight of the complex, but also that of the individual species making up the complex (protein and modifier) at each elution volume. These are plotted in Figure 4. Depending on the species comprising the complex, other pairs of concentration signals may be utilized, such as fluorescence + UV or fluorescence + dRI.

Experimental
The data presented here were acquired using an Acquity UPLC system with BEH200 SEC columns (Waters Corp.), a µDAWN MALS detector (Wyatt Technology Corp.), and an Optilab UT-rEX refractive index detector for UHPLC (Wyatt Technology), and were analyzed with the ASTRA software (Wyatt Technology). Typical conditions included 10-µL injections and 0.35 mL/min flow rate. Additional tests have shown that the elution volume of any given species may vary with flow rate, but molar mass and size as determined by light scattering do not, at least down to 0.05 µL/min and up to 1 mL/min. All data presented were drawn from reference (5).

Conclusions

UHP-SEC provides many benefits for the biophysical characterization of proteins and other biotherapeutic macromolecules, including assessment of degradants such as aggregates and fragments. However, these can only be fully realized with the addition of an on-line light scattering detector for absolute determination of molar mass and size to overcome the inherent limitations of SEC, which may be exacerbated in the context of UHPLC. In addition, UHP‑SEC‑MALS characterizes complex molecules, such as protein conjugates, which do not have available reference markers. Light scattering is therefore an essential tool for analytical, process development, and QC labs implementing UHP-SEC.

References

1. P.J. Wyatt, Analytica Chimica Acta272, 1–40 (1993).
2. For brief overview of light scattering as applied to macromolecules and size-exclusion chromatography, see http://www.wyatt.com/Theory
3. E.S.P. Bouvier and S. Koza, TrAC Trends in Analytical Chemistry63, 85–94 (2014).
4. B.S. Kendrick, B.A. Kerwin, B.S. Chang, and J.S. Philo, Analytical Biochemistry299, 136–146 (2001).
5. M. Chen, S. Kenrick, J. Lin, W. Hassouneh and D. Some, “Faster, smaller, deeper: applications of UHPLC-SEC-MALS for well-characterized biologics,” paper presented at the 2016 Conference on Well-Characterized Biological Products, Washington DC, USA.

Daniel Some is Principal Scientist and Director of Marketing at Wyatt Technology Corporation, where he has spent the last 12 years in various roles including R&D, software development, product management, and marketing. Previously Some was involved in the development of light-scattering–based patterned wafer inspection tools for the semiconductor industry.

E-mail: dsome@wyatt.com

Website:www.wyatt.com