Biology Reference
In-Depth Information
and Cooper, 1996b). This class is followed by chicken, dog, cow, pig, rabbit and
sheep in order of distance from equilibrium. There is thus some correlation
between distance from equilibrium and generation time and, with the exception
of humans, the ranking of species reflects the total synonymous substitution rates
estimated by Bulmer
et al
. (1991). These data are therefore consistent with a
model of DNA sequence evolution which, after species divergence, allows ances-
tral gene sequences to approach equilibrium codon usage faster in one species
than in another.
If current codon usage in different species were indeed the result of the diver-
gent evolution of common ancestral sequences progressing at different absolute
(albeit equal relative) substitution rates, then one species should always be closer
to equilibrium than the other in all codon groups. This, however, is not the case.
For example, hamster is closer to equilibrium than dog for the tyrosine (TAT,
TAC) and histidine (CAT, CAC) encoding triplets, but not for glutamine (CAA,
CAG). Furthermore, the codon frequencies for lysine (AAA, AAG), aspartic acid
(GAT, GAC) and glutamic acid (GAA, GAG) in
Xenopus
are on opposite sites of
the equilibrium when compared to all other vertebrates (Krawczak and Cooper,
1996b).
One may therefore surmise that codon usage has not evolved in a strictly uni-
form way. Although it cannot be excluded that relative mutation rates differ
between species thereby resulting in different equilibria, the fact that
Xenopus
and
rodents are very close to an equilibrium which is itself based upon human genetic
disease data, argues strongly against this objection. It is thus more likely that
species divergence has been accompanied by substantial changes in codon usage,
allowing some species to manifest sudden changes with respect to their distance
from equilibrium. This would also be consistent with the finding that differences
in synonymous codon usage between vertebrates is not explicable merely by dif-
ferential absolute DNA repair efficiency (Eyre-Walker, 1994).
Ikemura and Wada (1991) were able to demonstrate through an analysis of
approximately 2000 human gene sequences that codon usage differs dramatically
between different genomic regions. The major proportion of GC-rich genes was
observed in T-bands whereas AT-rich genes were located mainly in G-bands.
Further, the average G+C percentage at the third position of codons was found to
be related to the quinacrine dullness and the mitotic chiasma density of a partic-
ular chromosome. Since species divergence has almost always been characterized
by gross structural chromosomal rearrangements, and since different genes are
known to evolve at different absolute rates even in one and the same species
(Bulmer
et al
., 1991), the piece-wise reconstitution of new genomes from their
common ancestors is likely to have altered codon frequencies substantially.
7.5 Single base-pair substitutions in gene regions
The smallest changes that add to or subtract from a part in the smallest mea-
surable degree may also arise by mutation. We identify these smaller muta-
tional changes as the most probable variants that make a theory of evolution
possible both because they do transcend the original types, and because they
are inherited.
T. H. Morgan (1925)
Evolution and Genetics
