pH- and temperature-dependent mechanisms of non-native aggregation of anti-CD40 IgG1
Date
2015
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
A major concern for protein-based therapeutics is degradation that occurs
during in vivo expression, purification and processing, as well as upon commercial
manufacturing, storage and administrations. Non-native aggregation is one of the most
common degradation routes that may affect quality and safety of protein-based
biologics. There remain many unanswered questions regarding mechanisms of
aggregation, and that relates to approaches to control, and predict aggregation rates,
aggregation-prone region(s), and solubility at the formulation conditions and
temperatures of industrial interest. This dissertation addresses some of these
challenges by characterizing thermal unfolding, protein-protein interaction, and rates
of aggregate formation for a human IgG1, its Fab and Fc fragments as well as a
recombinantly expressed Fc domain, and providing a mechanistic analysis of
aggregate formation as a function of solution conditions. In addition, the results and
analysis illustrate the importance of considering multiple competing species and
pathways when attempting to mitigate protein aggregation via either controlling the
solution conditions or utilizing protein engineering to remove aggregation “hot spots”.
Monomeric and aggregated states of a human IgG1 were characterized
under acidic to near neutral conditions as a function of solution pH (4.0 – 6.0) and
temperatures (40 °C – 75 °C). A combination of differential scanning calorimetry,
laser light scattering, size-exclusion chromatography, and capillary electrophoresis was
used to characterize thermal unfolding, native monomer-monomer interactions, and
aggregation kinetics. Lower pH led to larger net repulsive monomer-monomer
interactions, decreased thermal stability of Fc and Fab domains, and increased
solubility of thermally-accelerated aggregates. pH-dependent aggregation kinetics
assessed by size-exclusion chromatography with in-line static laser light scattering
show that aggregation proceeded through a dimer nucleus at a broad range of initial
protein concentrations (0.5 – 8.0 mg/ml) for the selected conditions. However the
mechanism of aggregate growth as well as the size and solubility of formed aggregates
was affected by solution pH. Qualitatively, the global aggregation behavior was
consistent with reduction of charge-charge repulsions as a primary factor in promoting
larger aggregates and aggregate phase separation. Temperature-dependent aggregation
kinetics had two regimes with a breakpoint at intermediate temperatures. At
temperatures near or above the breakpoint (Tm), it was Arrhenius and thermal
unfolding was the rate limiting step. Significantly below this temperature, enzymecatalyzed
fragmentation was observed at a comparable rate as aggregation. In addition,
at these lower temperatures, the temperature-dependent unfolding enthalpy as well as a
convolution of fragmentation resulted in non-Arrhenius kinetics.
The thermal unfolding, colloidal interactions, and isothermal aggregation
rates for a human IgG1 and its corresponding Fab and Fc fragments produced by
enzyme cleavage were compared to illustrate the effects of solution pH on the domainmediated
aggregation pathways of the IgG1. The direct experimental comparison of
the aggregation rates for the IgG1 vs. its Fab and Fc fragments illustrated that there
were at least two competing aggregation pathway, Fab- or Fc-mediated. The results
here demonstrated that improving the intrinsic aggregation propensity of the unfolded
common Fc domains should also be considered when protein engineering is used to
improve aggregation resistance although engineering variable regions in Fab domain is
a common strategy.
In order to focus on aggregation mediated by the conserved region of the
IgG1 as well as to eliminate any convoluting effects due to fragmentation observed for
the IgG1 at those intermediate temperatures, changes in conformational stability,
protein-protein interactions, and aggregation of NSO-produced human Fc1 were
quantified experimentally as a function of pH (4 to 6) and temperature (30 to 77 °C).
Thermal unfolding transitions were significantly affected by changing pH, but there
were much smaller pH effect on electrostatic protein-protein interactions. The
aggregation behavior was qualitatively similar across different solution conditions
with soluble dimers and larger oligomers formed in most cases. Temperaturedependent
aggregation kinetics could be divided into two regimes: (i) Arrhenius,
unfolding-limited aggregation at temperatures near or above the midpoint-unfolding
temperature of the CH2 domain; (ii) a non-Arrhenius regime at lower temperatures.
The most pronounced non-Arrhenius regime at lower temperatures highlights
challenges that are expected for maintaining long-term stability of any human Fcbased
biotechnology products.
Finally, protein-protein interactions and net charges of the Fc1 were
quantified experimentally in citrate, acetate, and aqueous NaCl buffers as a function of
pH and ionic strength to understand if the diminished effect of pH on protein-protein
interactions and net charges of the Fc1 observed in 30 mM sodium citrate resulted
from either the effects of ionic strength and / or preferential interactions of citrate ions.
The data indicated that accumulation of citrate anions at the Fc1 surface as well as
high ionic strength resulted in much less pH-dependence of protein-protein
interactions and net charges. Similar effect was observed for acetate anions but much
weaker than that of multivalent citrate anions. In addition, thermodynamic stability of
the native state of the Fc1 decreased. Taken together, the interactions of the Fc domain
with different solution components should also be considered carefully when
interpreting the behavior of IgG proteins, as well as when utilizing Fc domains for
non-natural antibody platforms based on IgG proteins and Fc fusion constructs.