Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Beforesharing sensitive information, make sure you’re on a federalgovernment site.

Https

The site is secure.
Thehttps://ensures that you are connecting to theofficial website and that any information you provide is encryptedand transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
.2015 Oct 9:1415:83-90.
doi: 10.1016/j.chroma.2015.08.068. Epub 2015 Sep 2.

Protein A chromatography increases monoclonal antibody aggregation rate during subsequent low pH virus inactivation hold

Alice R Mazzer 1, Xavier Perraud 2, Jennifer Halley 2, John O'Hara 2, Daniel G Bracewell 3
Affiliations

Protein A chromatography increases monoclonal antibody aggregation rate during subsequent low pH virus inactivation hold

Alice R Mazzer et al. J Chromatogr A. .

Abstract

Protein A chromatography is a near-ubiquitous method of mAb capture in bioprocesses. The use of low pH buffer for elution from protein A is known to contribute to product aggregation. Yet, a more limited set of evidence suggests that low pH may not be the sole cause of aggregation in protein A chromatography, rather, other facets of the process may contribute significantly. This paper presents a well-defined method for investigating this problem. An IgG4 was incubated in elution buffer after protein A chromatography (typical of the viral inactivation hold) and the quantity of monomer in neutralised samples was determined by size exclusion chromatography; elution buffers of different pH values predetermined to induce aggregation of the IgG4 were used. Rate constants for monomer decay over time were determined by fitting exponential decay functions to the data. Similar experiments were implemented in the absence of a chromatography step, i.e. IgG4 aggregation at low pH. Rate constants for aggregation after protein A chromatography were considerably higher than those from low pH exposure alone; a distinct shift in aggregation rates was apparent across the pH range tested.

Keywords:Affinity adsorption; Aggregation; IgG; Size exclusion chromatography; Unfolding pH.

PubMed Disclaimer

Figures

Fig. 1
Fig. 1
Schematic of experimental design where incubation at low pH follows protein A chromatography (a). Incubation is assumed to begin when the elution peak begins, denotedt0. Fractions containing approximately the first two thirds of the elution peak were at pH conditions considerably higher than the elution buffer pH. IgG concentration and pH in fraction 3 were the conditions reproduced in solution experiments for comparability with column experiments. A precise representation of the experimental outcome can be seen in (b), a chromatogram from a protein A chromatography run. The solid line (lefty-axis) represents the IgG elution peak and the dotted line (righty-axis) indicates the pH of column effluent, as measured by the in-line pH probe. Here, flow of elution buffer begins at 7 mL elution volume. The last two fractions of the elution peak, indicated within vertical lines in (b), were at pH conditions low enough to induce aggregation.
Fig. 2
Fig. 2
SEC chromatograms for IgG4 incubated under a range of low pH conditions. Top: pH 2.78 in solution. Middle: pH 3.05 after elution from column loaded with 12 mg IgG. Bottom: pH 2.92 after elution from column loaded with 25 mg IgG. Legends show incubation times to the nearest minute.
Fig. 3
Fig. 3
IgG monomer loss over time in solution at (a) 0.9 mg/mL, (b) 2.7 mg/mL and (c) 4.5 mg/mL. Different symbols represent different incubation pH conditions: circles, pH 3.03; open squares, pH 2.95; filled squares, pH 2.78. Error bars show the standard deviation for each point based on full experimental repeats,n = 2. Exponential decay curves were fitted to the data using the equationy = y0 + AeR0.x(see Section 3.2 for equation specifics). In (a), for pH 2.78 the last time point (2.18 h) was excluded from the curve fit (see Section 3.3.1). All fits were significant with adjustedr2 > 0.99 andP < 0.01.
Fig. 4
Fig. 4
Semi-log plot of monomer decay rate,R0, against pH in solution at three different IgG concentrations.Y-error bars show the standard error for theR0parameter (Eq. (2)) obtained from least squares fitting;x-error bars represent pH measurement error. Linear regression was applied for concentrations 2.7 mg/mL and 4.5 mg/mL; for clarity, 95% confidence intervals only are shown for these fits. A trend has not been fitted to the 0.9 mg/mL data because there are few data points.
Fig. 5
Fig. 5
Rates of monomer loss at 2.7 mg/mL IgG under different pH conditions in solution only (filled symbols and solid lines) and after elution from a protein A chromatography column (open symbols, dashed lines and dotted lines) (a–c). Different pH conditions are shown in the plot legends. Curves for column runs across the full pH range are shown together in (d). Error bars show the standard deviation for each time point based on full experimental repeats,n = 2. All data sets were fitted with exponential decay curves as was done for initial solution only data. All fits were significant with adjustedr2 > 0.98 andP < 0.01.
Fig. 6
Fig. 6
Y0(plateau point parameter) from exponential decay curve fits plotted against pH at 0.9 mg/mL, 2.7 mg/mL and 4.5 mg/mL in solution, and at approximately 2.7 mg/mL after protein A chromatography.Y-error bars show the standard error for theY0parameter (Eq. (2)) obtained from least squares fitting;x-error bars represent pH measurement error. The data was pooled and fitted with a linear trend (solid line); error weighting was not used for the fitting. The 95% confidence interval is shown as dotted lines. The fit was significant with an adjustedr2of 0.75 andP < 0.01.
Fig. 7
Fig. 7
Semi-log plot of monomer decay rate,R0, against pH for solution-only experiments (filled triangles) and experiments including a protein A chromatography step (open circles). For column experiments, the typical concentration of the incubated elution fraction was 2.7 mg/mL. Solution-only data seen in Fig. 4 was pooled due to overlapping 95% confidence intervals for the linear trends. For solution-only data points,xerror bars represent the measurement error of the laboratory pH probe/metre, as reported by the manufacturer. For chromatography experiments,xerror bars correspond to measured error based on three replicate measurements (with a micro pH probe). For both solution-only and column experiments, y error bars represent the standard error for theR0parameter (Eq. (2)) obtained from least squares fitting. A linear trend was fitted to each data set, as shown by solid lines; dashed/dotted lines represent 95% confidence limits for each fit. Both fits were significant withP < 0.01 and adjustedr2of 0.97 and 0.90 for solution-only and column experiments, respectively.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Reichert J.M. Antibodies to watch in 2013: mid-year update. MAbs. 2013;5:513–517.-PMC-PubMed
    1. Chi E.Y., Krishnan S., Randolph T.W., Carpenter J.F. Physical stability of proteins in aqueous solution: mechanism and driving forces in nonnative protein aggregation. Pharmaceut. Res. 2003;20:1325–1336.-PubMed
    1. Cromwell M.E., Hilario E., Jacobson F. Protein aggregation and bioprocessing. AAPS J. 2006;8:E572–E579.-PMC-PubMed
    1. Joubert M.K., Luo Q., Nashed-Samuel Y., Wypych J., Narhi L.O. Classification and characterization of therapeutic antibody aggregates. J. Biol. Chem. 2011;286:25118–25133.-PMC-PubMed
    1. Fekete S., Beck A., Veuthey J.L., Guillarme D. Theory and practice of size exclusion chromatography for the analysis of protein aggregates. J. Pharm. Biomed. Anal. 2014;101C:161–173.-PubMed

Publication types

MeSH terms

LinkOut - more resources