Biology Reference
In-Depth Information
that can modify their structure and function through such a struc-
tural interaction mechanism. In very many instances, these interac-
tions seem to be determined only by the probability of encounters
between partner proteins (Beckett, 2004). Their amino acid com-
position, hydrophobic nature and electrical charge give disordered
proteins a characteristic signature which really differentiates them
from structured proteins. It is possible, with appropriate algo-
rithms, to analyse entire genomes or banks of protein sequences and
determine the proportion representing disordered proteins, and we
have thus been able to measure the overall involvement of disor-
dered proteins in various cell functions. They make up 36% to 63%
of genomes in eukaryotes but only 7% to 33% in prokaryotes and
archaebacteria. Protein disorder is therefore positively correlated
with multicellularity (Dunker
et al.,
2000). It is also significantly
increased in signalling proteins and those implicated in cancer
(Iakoucheva
et al
., 2002), in transcription factors (Liu
et al
., 2006),
and in the “hub proteins” of protein networks (Haynes
et al
., 2006).
These studies demonstrate that protein disorder is not a marginal
phenomenon: it is surprisingly present even in cell signalling and
gene transcription.
We have to acknowledge, therefore, that the existence of these
proteins radically challenges the conventional idea we have of the
relationship between a gene and the structure and function of a pro-
tein. Their structure does not depend in a deterministic way on
their sequence coded in the DNA, but on their encounters within the
cell. Their structure and function are not therefore written,
pre-existing and unalterable, in the genome, but are produced by
cellular processes in real time. Now, it is not possible to envisage
genetic programming as precisely determining intermolecular
encounters. Certain data even strongly suggest that a certain ran-
domness inevitably comes into play here. In an extreme case, the
same intermolecular encounter can produce different effects because
the two partner molecules may interact in a variety of ways induc-
ing different conformations and functions. The choice between these
options seems then, to be probabilistic (Haarmann
et al
., 2003).
