Biomedical Engineering Reference
In-Depth Information
5
Protein Stability and Mutation Bias
Another key evolutionary variable, which has received little attention, is the
nucleotide composition of the genome. In prokaryotes, it varies from extreme
adenine plus thymine (AT) content in obligatory intracellular bacteria to extreme
guanine plus cytosine (GC) content, for instance in actinobacteria. These differences
in GC content are thought to be prevalently due to mutation bias [
55
,
56
]. They are
strongest at the third codon position, where GC content barely affects the amino
acid composition of the protein, but also influences the coding positions [
57
,
58
].
Due to the structure of the genetic code, a mutation bias favoring thymine at
the nucleotide level favors the incorporation of hydrophobic amino acids in the
translated protein [
35
,
59
]. Hydrophobicity is a key property for protein folding [
60
].
Proteins that are too hydrophobic tend to misfold and aggregate, whereas proteins
that are too hydrophilic tend to be naturally unfolded [
61
]. This qualitative trade-
off between unfolding and misfolding was confirmed by a computational study of
the properties of homologous proteins in the proteomes of several bacterial species,
using a model of protein folding stability that correlates well with experimentally
measured unfolding stabilities [
35
]. The trade-off between unfolding and misfolding
stability is also clear if we consider the unfolding free energy (
3
). In this case,
we can define G
E.
A
;
C
nat
/=2
D
G
u
C
G
m
, with G
u
D
C
k
B
Ts
0
L
E.
A
;
C
nat
/=2
E
=2k
B
T . Using the hydrophobic
and G
m
D
h
E.
A
;
C
/
i C
approximation U.a; b/
"
H
h.a/ h.b/ (see (
9
)and(
10
) below) and writing g
i
D
h.A
i
/=
h
h
i
,where
h
h
i
is the mean hydrophobicity of the protein sequence, we see
2
that G
u
C
sL, with a>0, so that stability against unfolding increases
with hydrophobicity, whereas G
m
a
h
h
i
4
with b>0and c>0,so
that stability against misfolding decreases with hydrophobicity when it is large. We
and coworkers investigated the relationship between unfolding stability, misfolding
stability, and mutation bias using a protein evolution model with a neutral fitness
landscape. We indeed found that the mutation bias modulates the trade-off between
the two kinds of stability, making proteins evolving under AT mutation bias more
stable against unfolding but less stable against misfolding [
62
].
Interestingly, the two aspects discussed above, effective population size and
mutation bias, are correlated in nature. In fact, most bacterial and eukaryotic species
that adopted an intracellular lifestyle, with consequent reduction of their effective
population size, also shifted their mutation spectrum toward AT [
63
], as indicated
by the strong correlation between reduced genome size, which is a signature of
intracellularity, and the AT bias [
32
,
35
]. In order to investigate this relationship,
we have modeled protein evolution in a nonneutral fitness landscape where fitness
smoothly depends on stability against unfolding and against misfolding [
25
].
Mutations are randomly drawn at each step of the simulations according to a
given mutation bias, and they are fixed in the population according to (
5
), since
we assume that the population is monomorphic. Protein stability increases with
effective population size, in agreement with theoretical expectations. Interestingly,
for a given effective population size the fitness that can be achieved depends on the
2
b
h
h
i
C
c
h
h
i
Search WWH ::
Custom Search