The monoclonal antibody (mAb) 31D8 has previously been described to bi
nd more avidly to functionally active neutrophils and proved to be use
ful as a differentiation marker of neutrophils. However, attempts to f
urther characterize the antigen recognized by 31D8 have not been succe
ssful. Studying the altered Fc gamma-receptor expression of human neut
rophils induced by granulocyte colony-stimulating factor (G-CSF) in vi
vo, we could demonstrate a parallel decrease in the expression of 31D8
and the CD16 antigen. Furthermore, 31D8 showed a binding pattern on l
eukocyte subsets similar to that of clustered CD16 antibodies, exhibit
ing identical cells in double-staining experiments. Preincubation of n
eutrophils with 31D8 resulted in a dose-dependent inhibition of the bi
nding of immune complexes. A decreased expression of the 31D8 antigen
was found on the same cell clones to the same extent as found for the
3G8 antigen on neutrophils from patients with paroxysmal nocturnal hem
oglobinuria (PNH). Treatment of polymorphonuclear leukocytes (PMN) wit
h PI-PLC resulted in a dose-dependent decrease of mAb 31D8 binding, sh
owing that the 31D8 antigen is phosphatidylinositolglycan (PIG)-anchor
ed. Moreover, 31D8 competed with the binding of antibodies (such as mA
b 3G8) directed against the binding site of Fc gamma RIII for the Fc-p
art of IgG. However, this mAb did not influence the binding of a CD16
antibody (mAb B73.1) which recognizes an epitope elsewhere on the CD16
antigen. We conclude from our experiments that the mAb 31D8 binds wit
h high avidity to Fc gamma RIII (CD16 antigen). Furthermore, our data
indicate that its binding site is most probably located at the binding
site for the Fc-part of IgG.