Biology Reference
In-Depth Information
that by its structure, by the relative position of the genes, DNA
continues to have an effect on cell behaviour. It is not reduced to
the role of a simple passive supplier of protein components or RNA
that the cell might use as it sees fit. At the present time, a great
many researchers are reconsidering genetic determinism. They
include some who not so long ago were staunch supporters of it. In
a kind of pendulum movement, they are coming round to denying
any active role for DNA. We think that is an error.
The model of DNA as a random protein generator (Kupiec,
1983, 1989, 1996) is based on two properties of chromatin molecules:
first, they move by Brownian diffusion to find their target sequences
in the DNA (Berg and von Hippel, 1985; Halford and Marko, 2004),
and secondly their interactions are not specific. Due to these two
properties, the structure of DNA molecules determines gene expres-
sion probabilities during embryonic development. This model can be
summarised by two general principles which explain both the struc-
ture of chromatin and gene expression.
First principle
: The non-specificity of interactions between mole-
cules leads to countless interactions between chromatin molecules
41
and their DNA binding sites. The result is a great many possible dis-
tributions of the molecules on the binding sites. Each of these dis-
tributions corresponds to a different chromatin structure and to
activation of different sets of genes. Due to the diffusion of regula-
tor elements and interactions between molecules being intrinsically
random phenomena, the structure of the chromatin and gene expres-
sion are also random phenomena, each structure having a certain
probability of being fulfilled.
Second principle
: Because the interactions between chromatin mol-
ecules are unstable, the molecules can be randomly redistributed on
their binding sequences, though the transitions between different
distributions are not all equally probable. The probability of tran-
sition between two distributions depends on two main parameters:
on the one hand, the stability of the interactions between the
41
This principle remains valid whatever the chemical nature of these molecules.
