Focus on aggregation: types, causes, characterization, and impact
3 likes1,172 views
This document discusses protein aggregation, including types and sizes of aggregates, mechanisms of aggregation, and challenges in characterization. It summarizes sedimentation velocity and field-flow fractionation as useful analytical methods to cross-check size exclusion chromatography for characterizing aggregates. Sedimentation velocity can resolve multiple aggregate peaks but requires expertise, while field-flow fractionation has less surface interaction but high method development. No single method detects all aggregate types.
Focus on aggregation: types, causes, characterization, and impact
- 1. Focus on aggregation: types, causes, characterization, and impact John Philo V.P. & Director of Biophysical Chemistry © 2007 Alliance Protein Laboratories; all rights reserved
- 2. Outline Why do we care about aggregates in biopharmaceuticals? Review some basic facts about aggregate sizes and types Aggregation mechanisms Utility of sedimentation velocity for analysis of long-lived aggregates A few words about field-flow fractionation (FFF)
- 3. Time won’t permit talking about light scattering techniques today, but… Background and examples can be found on the APL web site, www.ap-lab.com Many articles, talks, and posters on aggregation and comparability studies can be downloaded from our ‘Further Reading’ page
- 4. Protein aggregates: What is all the fuss about? Aggregates (both large and small) often are a major degradation product Hence they often are a major factor limiting shelf life Aggregates in the product may affect its: 1. manufacturability clogged columns or diafiltration membranes 2. bioactivity (potency) 3. serum half-life or absorption rate 4. immunogenicity
- 5. Why is there heightened concern about immunogenicity? In Europe about 200 patients taking Eprex® (one brand of recombinant erythropoietin, EPO) developed antibodies that cross-reacted and neutralized their own internally-produced EPO Consequently those patients made no new red blood cells and require regular transfusions (Pure Red-Cell Aplasia) While the manufacturer has published evidence that this immunogenicity was not due to aggregates, this incident has raised alarm bells about immunogenicity
- 6. Some known cases where aggregates cause immunogenicity 1. Early versions of intravenous immunoglobulin from donor blood (IVIG) had high aggregate levels and caused anaphylaxis similar experience for human serum albumin 2. Aggregate levels in human growth hormone (hGH) correlated with persistence of anti-hGH antibody in patient serum 3. A recent study using interferon-α and transgenic mice confirmed that immunogenicity depends on the type and size of the aggregates S. Hermeling et al. (2006) J. Pharm. Sci. 95, 1084-1096.
- 7. AAPS Protein Aggregation and Immunogenicity Focus Group Sponsored a workshop September 2006 in Colorado to summarize current state-of-the-art and remaining challenges talks, posters, summaries available at http://www.aapspharmaceutica.com/inside/Focus_Groups/ProteinAgg/in dex.asp Goal is to form an industry consortium to sponsor (pay for) new studies to better define how immunogenicity varies with aggregate type and size Please join us!
- 8. The word “aggregate” covers a wide spectrum of types and sizes of associated states 1. rapidly-reversible non-covalent small oligomers (dimer, trimer, tetramer…) 2. irreversible non-covalent oligomers 3. covalent oligomers (e.g. disulfides) 4. “large” aggregates (> 10-mer) could be reversible if non-covalent 5. “very large” aggregates (diameter ~50 nm to 3 μm) could be reversible if non-covalent 6. visible particulates probably irreversible “soluble” “insoluble”
- 9. reversible irreversible Reversible vs. irreversible aggregates reversible irreversible
- 10. Whether aggregates are “irreversible” or “reversible” depends on the context solvent components salts, sugars, other excipients organic modifiers (alcohols, acetonitrile) pH temperature how long you wait
- 11. Aggregates have a spectrum of lifetimes rates of non-covalent association and dissociation (half- times) can vary from milliseconds to days metastable oligomers with dissociation rates of hours to days occur fairly frequently for an antibody example see J.M.R. Moore et al. (1999) Biochemistry 38: 13960-13967 see also Philo, J.S. (2006) AAPS Journal 8(3): E564-E571 many common analytical methods will detect only the longer-lived species it may take hours to days for a protein to re-equilibrate its association after a change in concentration, solvent conditions or temperature
- 12. Aggregation mechanisms (1): reversible association of native protein Native protein Reversible oligomerization Higher oligomers (possibly irreversible) or ↑ + sucrose
- 13. Aggregation mechanisms (2): oligomerization following conformational change Native protein Conformational change or partial unfolding Oligomerization of non-native protein Higher oligomers (probably irreversible) ↓ + sucrose
- 14. Aggregation mechanisms (3): oligomerization driven by covalent modification Native protein Modified protein (oxidation, deamidation, etc.) Oligomerization of modified protein Higher oligomers (possibly irreversible) ↓ ↑ + sucrose
- 15. Aggregation mechanisms (4): nucleation controlled aggregation (“seeding”) Native protein Critical nucleus (aggregate of native or modified protein, or a contaminant) Addition of protein monomers onto surface of nucleus (often with partial unfolding) Visible particulates or precipitation + ↑ + sucrose
- 16. Aggregation mechanisms (5): surface-induced aggregation Native protein Container surfaces and air-liquid interfaces Adsorption of protein monomers onto surfaces promotes partial unfolding + Aggregation of altered protein (as in mechanism 2) ↑ + sucrose ↓ + detergent
- 17. Our analytical challenge 1. Any protein sample may contain aggregates with a wide range of sizes, types, and lifetimes 2. Any one analysis method may not detect all the aggregate sizes or types that are present 3. The measurement itself may perturb the aggregate distribution that was initially present
- 18. The measurement itself may create or destroy aggregates dissociation or loss of aggregates can be caused by: SEC SV FFF dilution +++ + +++ change of solvent conditions +++ - ++ adsorption to surfaces +++ + ++ physical filtration (e.g. column frit) +++ - - physical disruption (e.g. shear forces) ++ - - creation of new aggregates can be caused by: change of solvent conditions +++ - ++ surface or shear-induced denaturation ++ - + concentration on surface - - ++
- 19. Regulatory concerns about analytical methods for aggregates Although SEC is usually the primary method of aggregate analysis the regulatory agencies are well aware that SEC columns can act as filters and that the SEC mobile phase can change the distribution of non-covalent aggregates but SEC is often the only qualified method that can be validated for lot release Therefore by phase 3 (and sometimes earlier) they will now nearly always ask for cross-validation of SEC methods by orthogonal methods
- 20. Methods typically used to cross-check SEC analytical ultracentrifugation (AUC) sedimentation velocity (primarily) sedimentation equilibrium (occasionally) light scattering flow mode classical scattering used after SEC (SEC- MALLS) dynamic light scattering (DLS) batch mode classical scattering field-flow fractionation (FFF) usually used with MALLS to measure true MW ← has been validated ← has been validated
- 22. The fundamentals of sedimentation velocity 6.0 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 6.9 7.0 Radius (cm) -0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 Absorbance 6.0 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 6.9 7.0 Radius (cm) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 Absorbance centrifugal force diffusion ← meniscus The sedimentation coefficient is determined from the boundary motion over time. It depends on both molecular weight and molecular shape. cell base → friction ←regionofsolute depletion boundary
- 23. High resolution analysis of a highly stressed antibody sample resolves 6 aggregate peaks plus 2 fragments 0 2 4 6 8 10 12 14 16 18 20 22 24 0.0 0.2 0.4 0.6 0.8 1.0 heptamer,0.1% hexamer,0.4% pentamer1.4% tetramer5.3% trimer14.6% dimer30.6% main peak (monomer), 45.5% ?HLhalfmolecule,0.8% ?freelightchain,1.4% c(s),normalized(totalarea=1) sedimentation coefficient (Svedbergs)
- 24. This interferon-β sample is 13.7% non-covalent aggregate; by the standard SEC method it would be pure monomer 0 2 4 6 8 10 0 1 2 3 4 5 6 7 0 2 4 6 8 10 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 IFN-β in 5 mM glycine, pH 3, 86.3% main peak c(s) sedimentation coefficient (Svedbergs)
- 25. 0 1 2 3 0.0 0.5 1.0 1.5 0 8 16 24 32 40 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 no salt c(s) +50 mM NaCl c(s) +150 mM NaCl c(s) sedimentation coefficient (Svedbergs) 0 2 4 6 8 10 0.00 0.05 0.10 0.15 20X expanded 0 2 4 6 8 10 0.00 0.05 0.10 0.15 20X expanded Adding NaCl to interferon-β formulations leads to a broad distribution of non-covalent aggregates out to ~100-mers
- 26. SV is very useful for comparability studies, giving comparability of conformation as well as aggregation 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 0 1 2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 0.00 0.01 0.02 normalizedc(s) sedimentation coefficient (Svedbergs) 0.42% 0.10% 0.03% 0.30% lot 1 lot 2 X 100 0.95% 0.05% 0.07%
- 27. The peril: c(s) distributions are often misunderstood 1. the effective resolution goes down as the fraction of minor peaks goes down 2. the resolution you can achieve for a 150 kDa antibody is much greater than for a 20 kDa cytokine 3. in general it is not possible to uniquely assign a stoichiometry to each aggregate peak 4. the nature of the noise (variability) is very different than in chromatography 5. for reversibly associating proteins the peaks probably do not represent individual molecular species
- 29. Principles of cross-flow FFF figure courtesy Wyatt Technology
- 30. FFF of acid-exposed IgG (2 hr at pH 2.9, 5 °C) (courtesy K. Tsumoto and D. Ejima) -0.02 0.00 0.02 0.04 0.06 0.08 0.10 0 5 10 15 20 LS,AUX(volts) Time (min) Chromatograms 070508-120min_01 FFF using 0.1 M citrate, pH 2.9 -0.01 0.00 0.01 0.02 0.03 0.04 0.05 0 5 10 15 20 LS,AUX(volts) Time (min) Chromatograms CF15 XF23 FF20-3_01 FFF after titration to neutral pH, elute using 0.1 M phosphate, pH 6.8
- 31. Advantages & drawbacks of FFF main advantages 1. much less surface area for absorption of sticky aggregates than SEC columns 2. can separate a much wider range of aggregate sizes than SEC drawbacks 1. some proteins stick to all the available membranes 2. many parameters need to be optimized during method development 3. high dilution may dissociate reversible aggregates
- 32. Summary 1. Aggregation is a complex phenomenon! 2. No single analytical method is optimal for all types and sizes of aggregates 3. Sedimentation velocity has many advantageous properties it is the primary tool we use at APL to cross-check SEC methods (and help improve them) it suffers from low throughput and requires a very highly trained operator 4. Our ability to characterize aggregates unfortunately far exceeds our knowledge of how specific aggregate types affect product safety or efficacy