12-10-2004, 07:21 PM
Here are extracts and a link to the paper-
The search for susceptibility genes
Although the strong genetic predisposition to autism is no longer in
doubt, identifying susceptibility loci is not straightforward (Box 2). In
the absence of specific drug effects and detailed knowledge of the underlying pathophysiology, there are no very strong pointers to functional candidate genes. Of course, the well-established association
with various medical disorders has been pursued but, for instance, no
linkage has been found with the fragile X mental retardation 1 gene
(FMR1) (Ref. 5). The finding of elevated platelet serotonin levels in
perhaps a quarter of cases has prompted association studies of the serotonin transporter and various serotonin receptor genes with, so far, no clear pattern of findings 6–8 . Indeed, as yet, there are no particularly
strong leads from any of the case-control or family-based candidate-gene
association studies, although the likelihood of genetic hetero-geneity
has not been systematically assessed.
With regard to positional candidates and regions, most attention has
recently focused on the proximal portion of the long arm of chromosome
15 (Ref. 1): a region in which chromosomal duplications and
other rearrangements are sometimes associated with an autistic phenotype with apparent imprinting effects. The presence of a cluster of
g-amino butyric acid (GABA A ) receptor genes in this area has partially
fuelled this interest, but association studies in this region have so far
yielded inconsistent results .
A complication for molecular genetic approaches is that the substantial
difference between the concordance rates for autism in MZ and
DZ twin pairs suggests that autism does not represent a single-gene
disorder. The twin and family findings indicate a multi-locus disorder,
with one model suggesting that 2–10 interacting genes are likely to
underlie susceptibility. Although some of the data support a model of
epistatic interactions between a small number of loci, there is not really
sufficient evidence to strongly support any particular hypothesized
complex mode of inheritance. As autism is associated with several distinct
medical disorders, genetic heterogeneity among idiopathic cases
also seems likely, although phenotypic markers of potential heterogenity
are limited.
Pre-natal and pre-symptomatic testing
At present, it is unclear whether the current interest in pre-natal testing
for monogenic disorders will eventually apply also to autism and related
complex disorders. Obviously, the potential to test for the presence
of a high-risk genotype could significantly influence genetic
counselling to affected families. Nevertheless, the complex gene interactions that are suspected to underlie autism suggest that testing is
unlikely to be a straightforward issue. In particular, the apparent non-deterministic nature of the genetic influences – as evidenced by the
substantial degree of phenotypic variability – means that knowledge
that an individual carries one or more of the susceptibility genes for
autism might not strongly predict a particular outcome. Moreover, it is
possible that susceptibility loci also contribute to traits that are beneficial
– for instance, elevated IQ or single-mindedness.
There might be genetic variants of autism for which testing could be
viewed as potentially more useful. That situation might arise if a
particular genetic predisposition was strongly associated with a severely
disabling disorder, a particular disease course, or the development
of complications such as epilepsy. Genetic knowledge might then
predict the need for particular environmental, behavioural or medical
interventions.
link
The search for susceptibility genes
Although the strong genetic predisposition to autism is no longer in
doubt, identifying susceptibility loci is not straightforward (Box 2). In
the absence of specific drug effects and detailed knowledge of the underlying pathophysiology, there are no very strong pointers to functional candidate genes. Of course, the well-established association
with various medical disorders has been pursued but, for instance, no
linkage has been found with the fragile X mental retardation 1 gene
(FMR1) (Ref. 5). The finding of elevated platelet serotonin levels in
perhaps a quarter of cases has prompted association studies of the serotonin transporter and various serotonin receptor genes with, so far, no clear pattern of findings 6–8 . Indeed, as yet, there are no particularly
strong leads from any of the case-control or family-based candidate-gene
association studies, although the likelihood of genetic hetero-geneity
has not been systematically assessed.
With regard to positional candidates and regions, most attention has
recently focused on the proximal portion of the long arm of chromosome
15 (Ref. 1): a region in which chromosomal duplications and
other rearrangements are sometimes associated with an autistic phenotype with apparent imprinting effects. The presence of a cluster of
g-amino butyric acid (GABA A ) receptor genes in this area has partially
fuelled this interest, but association studies in this region have so far
yielded inconsistent results .
A complication for molecular genetic approaches is that the substantial
difference between the concordance rates for autism in MZ and
DZ twin pairs suggests that autism does not represent a single-gene
disorder. The twin and family findings indicate a multi-locus disorder,
with one model suggesting that 2–10 interacting genes are likely to
underlie susceptibility. Although some of the data support a model of
epistatic interactions between a small number of loci, there is not really
sufficient evidence to strongly support any particular hypothesized
complex mode of inheritance. As autism is associated with several distinct
medical disorders, genetic heterogeneity among idiopathic cases
also seems likely, although phenotypic markers of potential heterogenity
are limited.
Pre-natal and pre-symptomatic testing
At present, it is unclear whether the current interest in pre-natal testing
for monogenic disorders will eventually apply also to autism and related
complex disorders. Obviously, the potential to test for the presence
of a high-risk genotype could significantly influence genetic
counselling to affected families. Nevertheless, the complex gene interactions that are suspected to underlie autism suggest that testing is
unlikely to be a straightforward issue. In particular, the apparent non-deterministic nature of the genetic influences – as evidenced by the
substantial degree of phenotypic variability – means that knowledge
that an individual carries one or more of the susceptibility genes for
autism might not strongly predict a particular outcome. Moreover, it is
possible that susceptibility loci also contribute to traits that are beneficial
– for instance, elevated IQ or single-mindedness.
There might be genetic variants of autism for which testing could be
viewed as potentially more useful. That situation might arise if a
particular genetic predisposition was strongly associated with a severely
disabling disorder, a particular disease course, or the development
of complications such as epilepsy. Genetic knowledge might then
predict the need for particular environmental, behavioural or medical
interventions.
link