Psychiatric Neurogenetics: Research Annual Report 2003
Section Head: Dr. James L. Kennedy
Schizophrenia
The neurodevelopmental hypothesis of schizophrenia suggests that structural abnormalities of the brain, acquired during the
development of the central nervous system, are responsible for the susceptibility to develop schizophrenia later in life.
These structural abnormalities lead to the dysfunctional connectivity ("wiring") of several areas of the brain affected in
schizophrenia.
The brain-derived neurotrophic factor (BDNF) is a protein molecule that plays an important role in the development and survival
of dopaminergic and serotonergic neurons. Animal models of schizophrenia show altered expression of the BDNF gene: mice that
were deprived of BDNF exhibit abnormalities of dopaminergic and serotonergic systems. Antipsychotic and antidepressant medications
appear to affect the activity of the BDNF gene. All these pieces of evidence have led us to study BDNF as a genetic risk factor
in schizophrenia.
Our first analysis of a study on families, consisting of a schizophrenia patient and both parents, collected from Italy and
the Toronto area, shows an association between a DNA variant in the BDNF gene and schizophrenia. We then extended this study
to additional variations of the BDNF gene and to a larger collection of these triad families, and also to families that consist
of subjects with a mood disorder and their parents.
We are excited that our new analyses on new samples of patients and families continue to show association between two BDNF
gene variations (and their combination, i.e. haplotype) and both schizophrenia and bipolar disorder. These findings suggested
that BDNF could be a common genetic risk factor both for schizophrenia and for mood disorders.
The investigation of BDNF in schizophrenia has led to a publication in Molecular Psychiatry (Muglia et al., 2002) and our
strong results for BDNF in bipolar disorder were published in The American Journal of Human Genetics (Neves-Pereira et al.,
2002).
We are now investigating subtypes of schizophrenia and bipolar disorder to understand whether the BDNF gene is a risk factor
for specific clinical symptoms common to these disorders.
Suicide
In the past year, we submitted funding proposals for genetic studies of suicide ideas and attempts in people who participated
as research subjects in our schizophrenia and mood disorder investigations. We will investigate to see if genetic variants
can predict risk for suicide in these disorders.
Dr. Vincenzo De Luca received a three-year award from the American Foundation for Suicide Prevention for his genetic investigations
of suicidal behaviour in schizophrenia and bipolar disorder. Dr. John Strauss also received a similar award from the same
foundation to examine genetics of suicide ideas and attempts in child-onset depression.
Also, our staff (Dr. Xingqun Ni and colleagues) are working on a proposal to study suicidal behaviour in the context of borderline
personality disorder.
Pharmacogenetics and Pharmacogenomics
Strong evidence suggests that genetic variation plays an important role in inter-individual differences in medication response
and toxicity. The rapidly evolving disciplines of pharmacogenetics and pharmacogenomics seek to uncover this genetic variation
in order to predict treatment outcomes. The goal of these disciplines is to be able to tailor the therapy to the individual,
by using the person's genetic make-up to select the drugs with the greatest likelihood of benefit and the least likelihood
of harm.
In our laboratory, we investigated the dopamine D1 receptor gene's ability to predict response to clozapine treatment. So
far, we have found that the D1 gene predicts improvement in memory and attention during clozapine treatment (Masellis et al.,
2002, New York Pharmacogenetics Meeting presentation).
In addition, we studied candidate genes to try to determine the cause of clozapine-induced weight gain (Basile, Masellis,
Potkin, & Kennedy, Human Molecular Genetics,, 2002; Basile et al., Lancet,, 2002). Using large, combined datasets, we have
further examined antipsychotic-induced tardive dyskinesia and its association with the dopamine D3 receptor, first reported
by our group in 1996.
We have also found that the pharmacogenetic principles developed in schizophrenia can be used to study many other psychiatric
and addiction disorders (Masellis et al., 2002).
Obsessive-Compulsive Disorder
In collaboration with Drs. Peggy Richter and Emanuela Mundo, over the last year we made an important step forward in understanding
the genetic basis of obsessive-compulsive disorder (OCD). Our group detected and, most importantly, replicated the finding
that a serotonin system gene encoding the 5HT1Dbeta receptor (a.k.a. 5HT1B) can be a predisposing factor to OCD (Mundo et
al., 2002).
The clue to this gene came from the unusual finding by Zohar et al. that the migraine drug sumatriptan, which binds to 5HT1B,
increases OCD symptoms in the short term. Our genetic finding, also reported in the lay press (National Post, Sept. 4, 2002),
could have important implications for the molecular diagnostics and design of new therapeutic strategies in OCD.
Another interesting investigation, the first ever of OCD risk genes in children, is being led by Dr. Paul Arnold, who has
reported a significant role for the NMDA receptor gene in OCD. He is now examining this gene in a sample of children from
Wayne State University who have OCD and have undergone functional MRI studies testing glutamate activity in the living brain.
Attention-Deficit/Hyperactivity Disorder
In collaboration with Dr. Cathy Barr's lab at the Toronto Western Hospital, we investigated the role of catecholamine system
genes in children with attention-deficit/hyperactivity disorder (ADHD) (Wigg et al., American Journal of Psychiatry, 2002;
Barr et al., 2002). Our adult ADHD studies in collaboration with Dr. Umesh Jain (Muglia et al., 2002) continue to show a role
for the dopamine D4, but not D3, receptor genes.
Problem Gambling
In collaboration with Drs. Umesh Jain, Nigel Turner, Michael Bagby and others, we obtained a grant from the Ontario Problem
Gambling Research Centre to investigate genetic risk factors for gambling behaviour in people who have symptoms of ADHD. Early
results suggest that dopamine system genes may be involved in predicting risk for gambling behaviour, but more work is needed
to establish these findings.
Seasonal Affective Disorder
Dr. Robert Levitan is leading the effort to study the role of serotonin system genes in people who eat larger amounts of food
in the autumn, become depressed and slowed-down in the winter, then return to normal mood and activity in the spring and summer.
Dr. Levitan showed that the serotonin 2A receptor gene was associated with a history of childhood attention-deficit/ hyperactivity
disorder in adult women who have seasonal affective disorder (Levitan et al., 2002). The next steps in this project include
examining the interaction between the serotonin 2A and the dopamine D4 receptor genes.
Childhood Onset Depression
Dr. John Strauss, in collaboration with investigators at the University of Pittsburgh, reported a strong effect of a variant
of the BDNF gene in predicting severe mood disorder in children. This finding has led us to study the role of this gene in
suicidal behaviour and the tendency of depressed children to develop bipolar disorder in adulthood.
Bipolar Disorder
We have dedicated significant effort over the last year to genetic studies of bipolar disorder, which, together with schizophrenia,
represents the group of major psychoses.
We investigated 300 small families (consisting of the person with bipolar disorder and his or her parents), first examining
the dopamine D1 receptor gene. We found that a particular combination of genetic variants of D1 create increased risk for
bipolar disorder (Ni et al. 2002). Dr. Ni also showed that a suspect gene, for the serotonin 2A receptor, was not involved
in bipolar disorder.
In these 300 small families, we also examined how specific variants of the dopamine D4 receptor gene in the parents were transmitted
to the affected child. Interestingly, we found that having the D4 gene transmitted from the mother increased risk for bipolar
disorder, as opposed to having the same genetic variant coming from the father (Muglia et al., 2002). This finding suggests
that regulation of this gene is complicated and likely to be subjected to genomic imprinting, which links the traditional
DNA sequence-based studies to epigenetic developments.
Epigenetics
The epigenetic theory of major psychosis puts the emphasis on looking at possible dysregulation of gene activity, rather than
changes in DNA sequence. Psychiatric epigenetics is a relatively new field in psychiatric research. The great advantage of
the epigenetic theory is that epigenetic mechanisms, unlike the traditional genetic ones, can explain unclear issues in major
psychosis, such as age of disease onset, fluctuating course (remissions and relapses), common major differences in identical
twins. To our knowledge, we represent the only group in the world fully dedicated to this development.
Dr. Art Petronis, head of the Epigenetics Laboratory, received a prestigious OMHF Special Project award to study epigenetic
mechanisms in schizophrenia and bipolar disorder.
Over the last year, the Epigenetics group performed a series of laboratory experiments looking for molecular differences in
identical twins, and compared epigenetic regulation of non-coding regions of the genomes of people who have schizophrenia
or bipolar disorder with controls. Two manuscripts that describe the differences in DNA methylation patterns between ill and
healthy control individuals are currently submitted to peer-reviewed journals.
Developmental Neuropsychiatry and Autism
Dr. John Vincent, head of the Laboratory for Molecular Studies in Developmental Neuropsychiatry, has reported the cloning
and characterization of a gene, called RAY1, that spans an area of breakage on the long arm of chromosome 7 in a person who
has autism (Vincent et al., 2002). This gene has recently, and controversially, been reported to be a tumour-suppressor gene,
although several reports have been unable to support this finding.
Dr. Vincent has continued to further characterize this gene and its surrounding DNA and has established the surprising situation
that there are at least five other genes within RAY1 (Vincent et al., 20022). Mutations in RAY1 were identified in two families
that have several members affected with autism. This work was recently published in a leading journal in the field, Genomics.
Dr. Vincent's work has also established the location of breakage regions within the suspected autism region on chromosome
7 for several other people who have autism. In one of these cases, the breakpoint maps to the same point identified in an
unrelated person with autism. Both translocations are likely to disrupt a new gene that we have identified. This gene thus
represents a very strong candidate for autism, and further characterization is under way.