I don't have a reference handy, but the pH dependence of binding
of IgG to protein A is apparently well known. An ex-coworker in
my lab was aware of it and succeeded in separating four type of
monoclonal IgGs in a pH gradient run on silica immobilized Prot-A
The antibodies were the well known MOPC21, UPC10, FLOPC21, and
MOPC141. The increased binding at higher pH for many murine IgGs
is also well known, as is increased binding with higher salt
concentrations. I will look for a reference if you are still
interested.
