Ken Frauwirth frauwirt at mendel.Berkeley.EDU
Tue Dec 7 17:53:55 EST 1993

In article <199312071711.AA26609 at wugate.wustl.edu>, brett at BORCIM.WUSTL.EDU writes:
|> Yes, the recent findings on IL-2 have given us much to think about. Ken
|> Frauwith and anyone else - you offer the much overused redundancy argument:
|> that is, if IL-2 knockouts don't behave as we would predict, then some other
|> mechanism must be compensating. While this argument may be insightful, it
|> could just be handwaving unless you propose some other mechanism. It was
|> interesting to note in the literature earlier this year (Cell, V73, pp147-57)
|> that human XSCID diease maps to the IL-2R gamma gene, and mutations at this
|> locus were found in these patients. Thus, humans with defective IL-2 signaling
|> can be immunodeficient. Although the murine knockouts were of different genes,
|> either IL-2 is functioning in ways other than we have previously thought, or
|> human and murine IL-2 are not entirely functionally homologous. ANY COMMENTS?
|> brett at borcim.wustl.edu

Whether or not the "redundancy" argument is overused, there are only
three explanations for a lack of any visible phenotype in a knockout

1) The gene is not completely disrupted, and still has some function.
2) The gene is not necessary under the tested conditions, but might be
   critical under other specific conditions (e.g. to combat a specific
   class of parasites or viruses).
3) Some other gene product can compensate for the loss, at least
   enough to give apparently normal phenotype.  Under more stressful
   conditions, this compensation may not be enough for normal function,
   similar to explanation 2).
(A fourth explanation is that the gene is truly dispensable, with no real
function, but let's discount that.)

One way to explain an apparent redundancy, especially in light of the
putative IL-2R/XSCID connection, is the possibility of multiple ligands
for the IL-2 receptor.  I do not know if a search for other IL-2-like
genes has been undertaken, but it is not unreasonable to think that there
may be related genes.  However, it is not necessary that the the two
ligands be very closely related in sequence.  Many receptors are capable
of binding multiple ligands, even unrelated molecules (this is especially
true of neurotransmitter receptors).  In fact, the EGF receptor has two
known ligands, EGF and TGF-alpha, which are the products of distinct
genes.  I do not know the degree of homology, if any, between the two
growth factors, but the existence of multiple ligands for grwoth factor
receptors (including IL-2) could help explain the lack of a dramatic
phenotype for growth factor gene knockouts.  One way of addressing this
is to disrupt the gene(s) for the receptor, rather than the ligand.

Ken Frauwirth                  BBB   IIIII    OO    K  K   EEEEE   N   N
frauwirt at mendel.berkeley.edu   B  B    I     O  O   K K    E       NN  N
Dept. of Molec. & Cell Bio.    BBB     I     O  O   KK     EEEE    N N N
Immunology Division            B  B    I     O  O   K K    E       N  NN
Univ. of Cal., Berkeley        BBB   IIIII    OO    K  K   EEEEE   N   N

  "When a rocket loves a space station *very very* much..." - T. Servo 

More information about the Immuno mailing list

Send comments to us at biosci-help [At] net.bio.net