Self/Not-self discrimination (YORK)

FORSDYKE at QUCDN.QueensU.CA FORSDYKE at QUCDN.QueensU.CA
Fri Aug 12 08:06:38 EST 1994


Ian York has asked me to post this on his behalf.

Date: Thu, 11 Aug 1994 21:10:26 +0500 (EST)
From: Ian York <york at fhs.csu.McMaster.CA>
Sender: Ian York <york at fhs.csu.McMaster.CA>
Reply-To: Ian York <york at fhs.csu.McMaster.CA>
Subject: Antigen presentation
To: FORSDYKE at QUCDN.QueensU.CA
Message-Id: <Pine.3.03+.9408112111.A15465-c100000 at fhs.csu.McMaster.CA>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; CHARSET=US-ASCII

        Whether it's due to an incompetent test editor or an incompetent
mind, I don't seem to be able to post anything to the Bionet.Immunology
newsgroup; but I would like to make a few comments on the antigen
presentation thread: I'm quite dubious about about some of the comments
being made there.  If you wouldn't mind posting this . . .
        This is all with regard to MHC class I antigen presentation.  There are
three separate points I'd like to address: (1)  Cells do
not discriminate  between self and non-self antigens in
presentation; (2) high concentrations of intracellular antigen are not
necessary for presentation (although it undoubtedly helps);  (3)
aggregation is probably unnecessary (and I suspect it might not even
help).
        (1)  A number of papers show very clearly that uninfected, normal
cells contain MHC-associated peptide linked with their MHC I.  These are
peptides derived from normal cellular proteins; some hve been sequenced
and confirmed this.  In fact this should be clear from the normal
cell-surface expression of MHC I; the complex does not reach the surface
efficiently in the absence of bound peptide.  There are normally on the
order of 100,000 - 500,000 MHC molecules on a cell; these have been driven
out to the surface by cellular peptides.
        The whole purpose (if I can be forgiven a little tautology) of the
antigen presentation system appears to be its very lack of discrimination.
 Joe somatic cell is supposed to worry about predicting and forestalling
parasites; that's the function of the professionals - the T cells.  The
somatic cells simply present a representative sample of just about
everything that's going on in their cytoplasm and let the T cells worry
about sorting out what belongs and what doesn't.  That's why thymic
education is so elaborate.
        (2)  There are quite a few examples of intracellular proteins
which are barely detectable by standard biochemical means, but which can
be easily detected by cytotoxic T cells.  Moreover, although I haven't
read the reference you cite (I can't get into the library a this hour of
the night) I disagree that most foreign proteins aggregate when expressed
in cells.  Perhaps if you push them with a killer promoter, but my
experience with foreign proteins in tissue culture cells is that they tend
to be quite soluble.  (That isn't the case in bacterial expression systems
in which you express eukaryotic proteins in a cellular environment quite
alien to them; and perhaps may also not be true of bacterial proteins in
eukaryotic cells; but viral proteins, with which I have the most
experience, seem just fine.)  Moreover, some soluble viral proteins
present in quite low concentrations are readily presented.
        Having said that, high concentrations of any protein will mean
that it is sampled more frequently by the antigen presentation system, and
so has a better chance of reaching a threshold for stimulation on the cell
surface (stimulation, by the way, for a CTL is probably around 100 MHC
complexes on the surface.  Since cell-surface MHC complexes are quite
stable - half lives seem to range from maybe 4 - 13 hours - that means
that if a particular peptide happens to fall into the antigen presentation
pathway maybe 200 times a day it could achieve adequate levels.)
        (3)  The current belief is that the degradation of cytoplasmic
proteins headed to the MHC is via the proteasome complex, and the proteins
are targeted toward the proteasomes by ubiquitination.  While the
factors involved in determining what will and won't be ubiquitinated are
complex and not terribly well understood, aggregation is not (to my
knowledge) an very important factor.

        Hope this helps, or at least sparks some more discussion.

        Ian





More information about the Immuno mailing list