In article <3bjoae$m3f at jhunix1.hcf.jhu.edu>, ejf at welchlink.welch.jhu.edu (Ephraim Fuchs) writes:
|> Regarding question 2, which asked whether the same mechanism preventing
|> autosensitization may be used by tumor cells to escape immunosurveillance:
|>|> Yes. It is now generally accepted that T cells require two signals for
|> efficient activation: signal 1 is ligation of the T cell receptor for
|> antigen, and signal 2 is a "costimulatory" signal delivered by the
|> antigen presenting cell (APC). This model proposes that T cell receptor
|> ligation (signal 1) in the absence of an APC-derived costimulatory signal
|> results in antigen-specific tolerance. Many people believe that
|> autoreactivity to many non-thymic (peripheral) tissue antigens does not
|> occur because peripheral tissues display these tissues in the absence of
|> costimulatory signals. Likewise, tumor cells may express tumor specific
|> antigens (specifically, peptides that are derived from mutated
|> intracellular proteins and presented in the context of MHC molecules) but
|> also fail to deliver costimulatory signals, thereby resulting in T cell
|> tolerance to tumor specific antigens.
Is it reasonable to think about the so-called "co-stimulatory" signal
as the primary trigger of T-cell action with the antigen sticking out
of the MHC acting as a verifier and auxillary info? Are co-stimilatory
signals specific or generic? Do they say "kill-me" and the T cell asks
"why?" by examining the cell for antigens; if non-self antigens are found
it can signal B cells to produce antibodies that will ultimately cause
attack on cells that have the same antigen but have not (yet?) said "kill-me"?
Non-self MHC may be only a sign that the cell is itself fighting some
invader; if successful it can live long and prosper but if it fails its
death can warn the organism of invasion.
In any case, if one has a bottle of "costimulatory" signal and some
tumor and T cells it should be easy to test the idea that its the only
missing ingredient in tumor immunoserveillance. Has this been done and
if not what prevents it? ("It doesn't come in a bottle"?)
|>|> Ephraim Fuchs
|>|>
--
John.
John J. Barton jjb at watson.ibm.com (914)784-6645
H1-C13 IBM Watson Research Center P.O. Box 704 Hawthorne NY 10598
