APC & Behaviour

Autism Spectrum Disorders
All Autism Spectrum Disorders (ASDs) are developmental disorders, symptoms of
ASD are expressed pervasively and appear early in the development of affected
individuals. The defects that characterize ASD occur in the domains of
communication and social interaction. Together with rigidity, insistence on
sameness and a tendency towards obsessive, repetitive behaviors, these symptoms
often remain manifest throughout the entire life of affected individuals. The
autistic spectrum ranges from severe cases (the core syndrome called autism or
autistic disorder) to milder forms which include pervasive developmental
disorder, not otherwise specified (PDD-NOS) and Asperger syndrome.

ASDs have a major impact on human well-being and represent a large economic
burden to our society. The long term prognosis of autism is unfavorable.
Although there is variability, there is an average tendency for developmental
regression during childhood. Even when restricting the follow-up to autistic
individuals with relatively better cognitive abilities (performance IQ>50),
only about 12% acquire a high level of independence in adulthood, 10 % require
a minimal degree of support in daily living but is able to hold a job, 19%
require support and supervision but have some degree of independence outside a
specialized residence, 46% need a high level of support within a specialized
residential setting and 12% are in need of high level hospital care 16.

To date no treatment for ASDs is available
Despite the huge progress medical science has made over the past few decades,
is it to date still not possible to cure ASD. Pharmacotherapy is sometimes used
to reduce undesired behaviors such as irritability or aggressive behaviors.
However, to date no effective medication is available against the core deficits
of ASDs, such as the social and communicative impairments. The main reason for
the current lack of a safe and effective treatment for ASD is that the
biological processes underlying this group of disorders are not well
understood. Thus, improving the understanding of the biology of ASDs is a
crucial step required for future development of better and effective treatment
methods.

With the aforementioned identification of rare variants with large effect, a
novel strategy has become possible. Instead of starting with the phenotype
(ASD) in order to study the possible role of a genotype (a genetic variant) it
is now feasible to bring together a group of individuals who share the same
genetic variant and subsequently compare phenotypic characteristics of this
group compared to individuals without that particular genotype. In other words;
a reversed approach starting with the genotype in order to study the phenotype.


The fundamental and innovative characteristic of this approach is that any
distinguishing phenotypic characteristic, if reliably and statistically
confirmed, can be directly attributed to the effect of the genetic variant.
Indeed, Abrahams and Geschwind recently argued that the study of subsets of ASD
patients with specific genetic variants is likely to yield valuable information
with regard to specific behavioral effects of particular variants1.

Stratification of the ASD population is likely to increase efficacy of
treatment
Based upon the rationale explained above we propose that:
1. Studying the behavioral phenotype of ASD samples collected on the basis of
shared genetic variants has the potential to dissect the clinical phenotype of
ASD into a number of behavioral components and their underlying genetic
etiology.
2. Studying the neurobiology of ASD samples collected on the basis of shared
genetic variants can identify specific neurobiological pathways relevant to the
(behavioral components of) the ASD phenotype.

For example; it is well known that mutations in the MECP2 gene lead to a
specific subtype of ASD, Rett Syndrome3, 25. Girls with the MECP2 mutation
display characteristic ASD symptoms including stereotypic movements of the
hands. Therefore one can conclude that, amongst others, the MECP2 gene is
relevant with regard to the development of stereotypic motor behavior7. Studies
on the function of MECP2 and the neurobiological pathway in which it is
involved in the brain will therefore likely shed light on the underlying
neurobiology of stereotypic motor behavior. If, in the best of scenarios, this
leads to the development of novel treatment options, this knowledge can be used
to target the intervention to the subpopulation of ASD patients with
predominant stereotyped motor movements.
Similarly we recently studied ASD patients with Klinefelter Syndrome or the
22q11.2 Deletion Syndrome, and reported that ASD patient groups with a specific
underlying genetic etiology demonstrate an autistic symptom profile that can be
delineated from the general ASD population and has less heterogenic
properties9. The fact that specific genetic etiologies translate into
distinguishable ASD sub-phenotypes strongly suggest that these genetic
subpopulations may also benefit from distinctive treatment approaches.



A need for the identification of genetic variants with behavioral effects
The field of autism research, and indeed, of the entire field of behavioral
science would greatly benefit from the identification of genetic variants
associated with specific behavioral phenotypes. This reversed phenotyping
approach will likely yield a better understanding of the biologic underpinning
of ASD and its phenotypic behavioral components. In addition, this could be the
first step towards the development of medication or other interventions that
may be specific for subgroups of ASD patients. For instance, the behavioral and
cognitive impairments found in fragile X patients are caused by a genetic
variant (not a CNV) in the gene FMR1. Subsequent studies have shown that the
biological pathway affected by this gene dysfunction includes the metabotropic
glutamate receptor 5 pathway, resulting in excess signaling. Studies with
antagonist agents of this pathway have shown valuable effects on behavior and
cognition, as well as on seizures, in animals15. Currently, the first trials
with this agent have been initiated in individuals with fragile X syndrome.
The reversed phenotyping approach is hindered by the fact that most variants
with modest to large effect size on (components of) the ASD phenotype occur at
a low rate in the population. This makes it difficult to obtain sufficiently
large samples. For instance, findings of several studies strongly suggest that
certain genetic variations in the Neuroligins genes (NLGNs) contribute to the
causation of ASDs13, 18. However, the rarity of each of these variations in ASD
patients obstructs any attempt to study a sizeable sample of subjects with the
same NLGN gene variant.


Mutations in the APC gene, a pivotal player in the Wnt pathway, lead to FAP
Familial adenomatous polyposis coli (FAP) is a heritable autosomal dominant
intestinal disease characterized by the development of large bowel adenomatous
polyps in the large intestine. Without treatment a progression to colorectal
cancer will occur in nearly all cases12. The genetic etiology of FAP is
autosomal dominant and consists of a mutation in a gene located on 5q21 which
has been named the Adenomatous Polyposis Coli (APC) gene12. The APC gene plays
a pivotal role in the Wnt pathway, a regulatory pathway that is important for
numerous developmental processes. APC forms a multiprotein complex which
downregulates β-catenin, an activator of the transcription of several target
genes in the Wnt cascade. When APC is impaired in its function, β-catenin can
accumulate and generate abnormal expression of various genes affecting key
developmental processes, including apoptosis, migration and differentiation14.


The Wnt pathway is thought to be a central actor in cell development
Interestingly, the Wnt pathway also plays a role in the development of the
central nervous system28 as as well as in the maintenance and function of the
adult brain17.
Several genes within this pathway have been associated with ASD susceptibility;
notably Wnt2 and EN2 11, although the association in these genes was not
consistently replicated, as is the case in most candidate gene studies in
neuropsychiatric disorders. Interestingly, the genes TSC1 and TSC2, of which
variations can cause tuberous sclerosis, alter the function of β-catenin and
the subsequent Wnt cascade22. In 20 - 25% of patients with tuberosclerosis
autistic behavioural features can be identified29. Importantly, the APC gene is
highly expressed in the fetal and adult human brain (see APC gene expression
figure: adapted from: http://smd.stanford.edu/cgi-bin/source/sourceSearch).
Recently it was shown that APC is a key coordinator for presynaptic and
postsynaptic maturation by providing anchorage to Neurexin and Neuroligin,
providing further evidence for an important role of APC in brain development23.

The above anounced studies suggest that some abnormalities in the Wnt pathway
can be expected to affect development and function of the brain, potentially
leading to recognizable behavioural phenotypes. Given the key role of APC in
the Wnt pathway, and more specifically, in synaptic maturation, it is not
unlikely that impaired function of the APC gene can be associated with
neurodevelopmental disorders. Indeed, among the first papers indicating the
role of the APC gene in FAP was a case report in which an autistic FAP patient
with a deletion of the APC gene was described 6. In addition, a recent
retrospective case-control study (autistic individuals versus controls)
reported a significant association of the APC gene variation in cases30. Given
these associations, the aim of the current study is to investigate whether the
APC mutation, associated with FAP, also leads to a behavioural phenotype

Hypothesis: Carriers of FAP-associated APC mutations display a higher rate of
behavioural abnormalities in the autistic spectrum compared to controls with
the APC wild type alleles.

Statistical methods, Sample size, power calculation
Normative and clinical threshold scores recently assessed in a German
population8 will be used as set points for this study. The average SRS score in
a typically developing childhood sample was found to be 25 with standard
deviations of 15. The threshold for clinical relevance is 59. Above this
threshold the presence of social behavioural pathology (autism spectrum
disorders in particular) becomes more likely.
By including 100 cases and two groups of 100 controls we will have a power of
0.81 to detect a difference between case and control groups means of 6 points.
Differences in mean scores of 10 points or larger have greater clinical
relevance and can be detected with a power of 0.99. Type I error probability
rate is set at 0.05, two tailed.


A two-step behavioral assessment protocol
The first stage of behavioural assessment will be performed using a
questionnaire sent by letter mail and to be filled out by parents or
caregivers, as well as the proband. After completion, the questionnaires are to
be returned using a prestamped return envelop.

For the behavioural assessment of the first stage will be used:
1. Social Responsiveness Scale (SRS10), a written 65 item questionnaire. A self
report will be filled out by the participant and a reference report will be
filled out by the parents or caregivers of the participants. The SRS is
well-validated against the gold standard clinical assessment for autism
diagnosis, the ADI-R and the ADOS20, 21.
2. Child Behavior Checklist (CBCL), a written 103 item questionnaire. Again a
self report (Youth Self Report, YSR) as well as a parental report (CBCL) will
be used2
3. Global enquiry of social economic and educational level of the child and
their parents. Age of each parent at birth of the child will be asked.
Assessment of clinical burden to be filled out by parents and children.


The second stage of the assessment protocol concerns only those probands that
show SRS scores in the clinical range, suggestive of the presence of an autism
spectrum disorder. At this stage it is difficult to predict what exact
proportion of the participants will have clinical range SRS scores and thus
will be invited to participate in the second assessment. At any rate, this
proportion is unlikely to be more than 20%, as higher prevalence of autism
spectrum disorders amongst FAP children would certainly have been picked up and
reported in the literature.

The behavioural assessment of the second stage will consist of a standard
clinical evaluation, according to the guidelines of good clinical practice.
This assessment will be performed by a Child and Adolescent Psychiatrist in
those cases with SRS scores within the clinical range. If applicable, a DSM-IV
based diagnosis will be made and adequate medical care will be provided. Zie
Flow chart bijlage ...

not applicable