3
general, pH- and salt-dependence was most
pronounced for the first unfolding transition.
Parallel analysis of antibody aggregation was
performed by measurement of light exctinction
using backreflection detection. Interestingly, the
aggregation data revealed that aggregation was
most pronounced under buffer conditions with
higher pH values (Figure 3B and 4C). These
results indicate that shifts in T
agg
do not correlate
with the large, pH-dependent shifts in the first
unfolding transition temperature T
m
1 (Figure 3 and
4B), suggesting that unfolding of the corresponding
mAb domain is not responsible for aggregation.
In contrast, no aggregation was detected in acetate
buffer pH 4 in absence of salt, showing that
aggregation of the unfolded state is largely
suppressed under these conditions (Figure 3B, 4B
and C). Since an increase in pH resulted in an
increase of the overall aggregation signal,
increasing the pH-value seems to promote the
formation of unfolded-state aggregation. In addition
to sample pH, sodium chloride greatly affected
temperature-induced mAb aggregation: The
addition of 130 mM NaCl led to strong aggregation
under all conditions tested, as well as to a shift of
the aggregation onset temperature to lower values.
Figure 3: Conformational stability and aggregation of a mAb under different buffer conditions. (A) Thermal unfolding monitored by
detection of shifts in the fluorescence ratio (F350/F330) in dependence of different buffer pH values and NaCl concentrations. (B)
Aggregation is detected by changes in backreflection.
Since partial unfolding of proteins can be reversible
under certain conditions, we next tested for
reversibility of the first unfolding transition. For this,
the samples were heated to temperatures either
near at or ~10 °C beyond T
m
1, respectively, and
subsequently cooled back to 20 °C at 1 °C/min