stormo at boulder.Colorado.EDU (Gary Stormo) writes:
> ...
>Given the large number of chromosome translocations and inversions
>that exist in Drosphilia, one ought to be able to construct some flies
>whose offspring are competent to mate with each other (or at least a
>subset of their sibs) but would have very low fertility in matings to
>any other strains, even the parental. If so, these would represent a
>new species, by the usual definition, in which there has been no change
>in allelic frequencies (at the gene level, I'm not counting a rearranged
>chromosome as a new "allele").
>...
>Certainly allelic changes can occur without speciation and
>significant allelic change can accompany speciation, but it can also
>occur in the absence of significant allelic change. ...
Actually, a speciation event occurring in this way would inevitably be
accompanied by large changes in allele frequencies, simply because many alleles present in the
population as a whole would (by chance) not happen to be present in the individual in whom the
translocation occurred, or its mate, and thus would not be present in the
offspring who interbreed to form the new species. For stochastic reasons
allele frequencies have to change whenever a population passes through a
"bottleneck", and this would be a particularly extreme example of a bottleneck.
My guess is that most speciation events do in fact arise as a result of this sort of
chromosome rearrangement mechanism rather than by accumulation of point mutations.
Certainly they are the most obvious way of causing reproductive isolation.
Moreover, I believe it is correct to say that in essentially all cases where we
have been able to check (e.g. by chromosome
banding studies), even closely related species seem to differ
by several chromosome rearrangements. It will be interesting to learn how
many of these rearrangements may have actually resulted in phenotypic changes (e.g.
by altering expression of genes near the rearrangement breakpoints). In this
vein, does anyone out there know of examples where evolutionarily relevant
translocations are known to lie near genes and possibly to have affected
their expression? (N.B. I'm not talking about somatic translocations, like
those occurring in many leukemias).
Phil Green
Genetics Dept.
Washington Univ. School of Medicine
St. Louis
pg at genome.wustl.edu
