Why will MHC indicate graft as foreign?

Ian A. York iayork at panix.com
Tue Jul 8 06:41:16 EST 2003

In article <noemail-42FD4D.23464007072003 at cnews.newsguy.com>,
Hobo  <noemail at noemail.com> wrote:
>strings they may sample. This means the total number of MHC/protein 
>possibilities is extraordinarily large. How do the white blood cells 
>recognize such a large number of possibilities?

Lymphocyte receptors (antibody and T cell receptor) are extraordinarily 
diverse.  This diversity is produced, in general, by shuffling multiple 
DNA segments during their differentiation.  The various potential 
combinations add up to a number that is large enough to handle all the 
possible MHC/peptide combinations in the universe.

It's also worth noting that in recent years, we've come to appreciate that 
there's a significant amount of overlap.  The classic concept of one T 
cell receptor, one peptide/MHC complex, is not correct; a single TcR can 
see many peptide/MHC complexes, albeit not always with equal efficiency.

>I've read that cancer cells downregulate MHC expression, often to zero. 
>NK cells are programmed to kill cells with zero MHC expression. I've 
>read that the cancer cells avoid this by expressing MHC homologs at the 
>surface, fooling the NK cells. If the immune system can be fooled so 
>easily by MHC homologs, why is it so sensitive to differences in MHC 
>composition between native and donated tissue?

Cancer cells probably don't express MHC homologues.  However, your general 
point may be right, because it's possible that some viruses that 
downregulate MHC class I do express MHC class I homologues (e.g. 
cytomegalovirus).  The function and mechanism of these homologues isn't 
quite clear, and that's true for viral immune evasion in general.  Viruses 
(as opposed to cancer cells) *may* only have to delay recognition and 
killing, rather than avoid it forever, so that the virus can complete 
replication within the cell--from the virus's viewpoint, postponing 
recognition by a few hours may be all it needs to do.

These mechanisms, whether viral or tumour, don't operate in isolation.  
Cancer cells are the product of many rounds of selection, so that the 
established tumours you see are the best survivors out of billions or 
trillions of precursors clones that were killed by immune responses.  Even 
viruses, with their much-simpler genomes, have subtle and effective 
combinations of immune evasion approaches, all working together.

Or, in other words, we don't know.

    Ian York   (iayork at panix.com)  <http://www.panix.com/~iayork/>
    "-but as he was a York, I am rather inclined to suppose him a
     very respectable Man." -Jane Austen, The History of England

More information about the Immuno mailing list