Self/nonself discrimination

mdcabl at mdcabl at
Mon Jul 18 03:24:11 EST 1994

In article <94198.164503FORSDYKE at QUCDN.QueensU.CA>, <FORSDYKE at QUCDN.QueensU.CA> writes:
> (Ephraim Fuchs) says:
>>Please help me with this question:
>>Since antigen presenting cells take in antigens randomly from their
>>environment, how is their delivery of signal two regulated so as to
>>achieve a self/nonself discrimination?
>     Does an antigen-presenting cell take in antigens randomly? Surely, an
> antigen has first to be labelled as "foreign" by reacting with an antibody
> (either free or cell-bound)? The pre-existing repertoire of antibodies
> then determines which antigens will be taken up. The antibody repertoire is not
> random, so why should antigen-uptake be random?

One place where antigen uptake may be truly random is in the gut with
by the M cell in the gastrintestinal epithilum.  Here new antigens are sampled
from the gut and presented to the Peyer's Patch lymphoid aggregate.  I am not 
sure if M cells have an FcAlpha receptor, but even if they did Fc Receptor
uptake could not account for the presentation of new antigens to which little
IgA would be able to bind.  The other interesting point is that you can get
clonal proliferation in vitro with only dendritic cells and T cells.  Clearly,
no antibody or B cells for that matter is necessary for antigen processing. 
Dendritic cells have been shown to be fantastically effective at antigen
presentation.  However, uptake of antigen has been described as pinocytosis
which I find lacking.  As I understand it from Anne O'Garra's work the DC is
involved in priming naive T cells, whereas the B cell and macrophage are
involved in priming memory cells.  I think it would be to the advantage of the
individual to sample the surroundings randomly as well as having targeted
uptake like binding surface IgM especially when antigenic load is low. 

Allen Black
Dept. of Pathology
Univ. of Newcastle
mdcabl at

proliferation in

More information about the Immuno mailing list