Biomedical Engineering Reference
In-Depth Information
process, since slightly deleterious mutations are more likely to be fixed in small
populations. This is the basis of the nearly neutral model proposed by Ohta [
30
].
Since small populations are more tolerant to mutations, the frequency
z
of effectively
neutral mutations is larger in small populations, and the substitution rate increases.
Finally, if N is not small, multiple mutations, both beneficial and detrimental,
may happen in the same genome. This effect complicates the substitution process
considerably, since neutral and slightly deleterious mutations can hitchhike genomes
with advantageous mutations that are positively selected, and therefore increase
their substitution rate. When the advantageous mutation has been fixed, these
passenger mutations are eliminated by purifying selection.
4
Translation Load
As discussed above, if N is not small, mutational robustness arises even in the
neutral model as a consequence of population dynamics. Mutational robustness is
correlated with stability, since a more stable protein is more tolerant to mutations.
Therefore, a large mutation rate is expected to favor tolerance to mutations and
a larger stability than would be expected in the monomorphic limit. However,
the mutation rate per gene is very small in natural populations of bacteria and
eukaryotes, so that it is doubtful that this mechanism is relevant to explain the high
tolerance to mutations observed in natural proteins. A related explanation for this
mutational robustness has been recently proposed by Wilke and collaborators [
52
].
When proteins are synthesized on the ribosome, translation errors may happen
relatively frequently, since the accuracy of the ribosome is not very high (for a
200 amino-acids protein, a wrong amino acid is incorporated on the average every
few replication cycles in
E. coli
). These translation errors may produce wrongly
folded proteins that are not functional and, moreover, tend to aggregate; therefore,
there is a strong selective pressure enforcing robustness against translation errors.
This selective pressure is expected to be stronger for highly expressed proteins.
This is in agreement with the observations that highly expressed proteins tend to be
codified by optimal codons (which improve the accuracy of translation) and tend
to evolve more slowly (since they are subject to stronger selective pressure) [
52
].
Another element that enforces robustness against translation errors is the genetic
code. It has been shown that the standard genetic code is almost optimal for reducing
the consequences of translation errors on the physiochemical properties of protein
sequences [
53
]. Using the protein folding model that we adopted for evolutionary
studies, we recently verified that the standard genetic code reduces the effects
on protein folding stability of frequent translation errors with respect to existing
alternative codes [
54
], although the advantage of the standard code is sometimes
reduced for extreme mutation bias. Despite the importance of the translation load
for protein evolution, however, this ingredient is seldom taken into account when
modeling the fitness.
Search WWH ::
Custom Search