Glutamate Neurotransmission in OCD: Genetic Study of a Candidate Biological System (2006-2010)

Spit for Science_imageMy project investigating glutamate system genes in OCD has yielded important findings and has formed the basis of an active and productive line of research that will help elucidate the genetic underpinnings of this disorder. This report will summarize progress in recruitment of families and genotyping results. I will conclude with a description of future research directions made possible by [Healthy Minds Canada’s] award.

Recruitment of Families: We were successful in recruiting over 90% of our target sample (233 families) based on individuals from the original collaborative sample involving Toronto and University of Michigan. In addition, we have since considerably expanded our collaborative group to include a total of 6 sites: Toronto, University of Michigan (Ann Arbor), Massachusetts General Hospital (Boston), University of California at San Francisco, Paris, and University of Sao Paulo (Brazil). This collaborative, with a total of 379 nuclear families (1575 individuals, 683 of whom were from our original sites in Michigan + Toronto), represents one of the largest samples assembled for candidate gene studies in the world. All probands and family members have received detailed assessments including semi-structured interviews and age-appropriate versions of the Yale-Brown Obsessive Compulsive Scale (YBOCS) questionnaires. This collaborative group also forms an integral part of a much larger collaboration (including 21 international sites) which has more recently been formed to conduct a genome-wide association study of OCD (as discussed below under Future Research Directions).

Genotyping Results:

Genotyping of Collaborative OCD sample: My role within the collaborative group just described is lead investigator on glutamate candidate genes. To date, I have analyzed a total of 78 single nucleotide polymorphisms in the following candidate genes: Glutamate Receptor Ionotropic N-methyl-D-aspartate 2B (GRIN2B), Solute-Linked Carrier Family 1, Member 1 (SLC1A1), and DLGAP3. Tag single nucleotide polymorphisms (SNPs) were selected using publicly available databases (dbSNP, HapMap) and the Haploview program. In the case of SLC1A1 and DLGAP3 comprehensive coverage of tagging SNPs was achieved. GRIN2B was too large and contained too many common SNPs to be tagged comprehensively within current budget constraints, although we were able to expanded coverage considerably compared to our earlier work (379 vs. 130 families, and 10 compared with 3 SNPs in Arnold et al., 2004). Genotyping was performed on the Sequenom iPlex platform. After applying quality control filters to genotyping data (exclusion of SNPs with low success rates, low minor allele frequencies, or high number of Mendelian errors) I analyzed for association with OCD diagnosis using Family Based Association Test (FBAT) and Haploview software. For DLGAP3, three SNPs were nominally associated with OCD diagnosis at the p<0.01 level. However, no SNPs or haplotypes were significantly associated following permutation testing. For SLC1A1 and GRIN2B we identified gender-specific significant associations with haplotypes based on transmissions to males only. Results have been presented at international scientific meetings (see list of meeting abstracts below) and manuscripts are in preparation regarding DLGAP3 and SLC1A1.

Imaging Genetics Studies: I have continued to study glutamate candidate genes in medication-naïve children who have undergone magnetic resonance imaging (MRI) and magnetic resonance spectroscopy. This project, the first imaging genetics study in OCD, was made possible through collaboration with Dr. David Rosenberg at Wayne State University, where the samples were collected and imaging studies performed. To date, I have identified associations between variants of GRIN2B and size of the orbital frontal cortex (OFC) and anterior cingulate cortex (ACC). SLC1A1 variation was associated with thalamic volume (Arnold et al., 2009a) I also identified an association between a GRIN2B variant known to influence gene expression and glutamate levels in the anterior cingulate cortex. This finding, although based on a small sample (18 individuals) is particularly intriguing as it is the first report of a direct association in humans of a glutamate risk gene and a measure of brain glutamate associated with OCD (Arnold et al., 2009b). I am currently analyzing DLGAP3 and 66 other glutamate candidate genes (identified through) in the Wayne State imaging genetics sample for association with structural MRI and MRS glutamate levels. Glutamate candidate genes were selected using computerized databases such as the UCSC genome browser. This comprehensive analysis of the glutamate system is based on genotyping data from Illumina genome-wide arrays.

Study of Symptom Dimensions in Toronto sample: Finally, I have performed detailed phenotypic analyses within the Toronto sample. These studies are based on a growing body of evidence that specific symptom dimensions have specific genetic correlates. I have provided preliminary evidence supporting this hypothesis for our glutamate candidate genes within the Toronto sample for which we have the most detailed symptom data. Analyses were performed using quantitative factor scores derived from factor analysis within the Toronto sample (see Table 1 for the 5 factors). Using this method, I identified nominally significant associations between hoarding symptoms (Factor 2 in Table 1), and variants within both DLGAP3 (rs16837122, p=0.014), and SLC1A1 (rs1980943, p = 0.011). I also identified a nominally significant association between a variant of GRIN2B and Factor 1, comprising contamination/somatic obsessions and cleaning/checking compulsions (rs1120905, p=0.012). These results fell just short of correction of multiple comparisons given the 5 symptom dimensions tests (corrected p<0.01). However, these preliminary results suggest that glutamate genetic variants may be specifically associated with specific OCD symptoms. I am following up our results by obtaining symptom dimension data from the other samples that were included in our OCD genetics collaboration (see above for list of collaborating centres).

Future Research Directions: I am continuing to collect DNA samples from children with OCD in order to increase the power of the comprehensive glutamate candidate gene analyses. In collaboration with my colleagues at Wayne State and University of Michigan I am greatly expanding the imaging genetics study with a plan to genotype 200 pediatric OCD cases and 200 normal, medication-free controls. This large and unique sample collection is made possible through a recently awarded collaborative RO1 grant from the National Institutes of Health (NIH) in the US. Obtaining the NIH grant was made possible through the invaluable support from the CPRF award. Since the award was originally obtained there have been rapid advances in genotyping technology enabling novel approaches to be applied at a lower cost than was previously possible. I will be applying one of these approaches, targeted re-sequencing, to follow-up on my initial candidate gene results for glutamate genes and OCD. Specifically, in collaboration with The Centre for Applied Genomics (TCAG) I will use next-generation sequencing methods to sequence SLC1A1, GRIN2B, and DLGAP3 to our sample of children with OCD. This method will enable identification of a broader range of genetic variants, including rare and common SNPs. Inclusion of the Wayne State imaging genetics sample will ensure that any variants thus identified can be tested for their association with neuroimaging phenotypes. The increased volume of data resulting from the larger number of SNPs and greater range or approaches than originally planned has increased the complexity of data analysis considerably. In response to this analytical challenge I have obtained statistical consultation for this study. This enhanced access to statistical expertise will also enable me to examine interesting questions within our data set such as gene-gene interaction and analysis of quantitative phenotypes, which I intend to pursue in the coming months. Finally, as mentioned above I am now an investigator on a genome-wide association study of OCD diagnosis conducted by the OC Foundation Genetics Collaborative (OCFGC-GWAS). The OCFGC-GWAS study is an international collaborative effort involving 21 sites led by a 13 member steering committee, of which I am a member (team leader, study design). A total of 1760 individuals with OCD, 3000 controls, and 477 OCD parent-proband trios are included in this study. Genotyping of cases and trios has recently been completed at the Broad Institute in Boston using Illumina 610Quad genome-wide arrays. Through my role in this consortium I will have access to genotype data from a larger sample. I intend to identify the strongest glutamate candidate gene “hits” from this study and follow them up in our imaging genetics sample and using the sequencing methods described in the previous paragraph. I wish to thank the CPRF for their generosity in funding this project. This award has been invaluable for me in launching my career as an independent investigator. I hope that continued progress in this work will enable my colleagues and I to identify susceptibility genes for OCD that will lead to benefits for individuals with OCD and their families. Furthermore, I hope that these findings prove beneficial for individuals suffering from other neuropsychiatric disorders (such as schizophrenia and autistic disorder) that are believed to be associated with abnormalities in glutamate neurotransmission. Table 1. OCD symptom dimension factor groups within Toronto sample using the Yale-Brown Obsessive Compulsive Scale (Goodman et al. 1989; Scahill et al. 1997).

Factor Group YBOCS Categories
1 Contamination obsessions, Somatic obsessions, Cleaning compulsions, Checking compulsions
2 Hoarding obsessions, Hoarding compulsions
3 Aggressive obsessions, Sexual obsessions, Religious obsessions
4 Symmetry obsessions, Ordering compulsions
5 Repeating compulsions, Counting compulsions

Publications Resulting from CPRF Award (July 2006 – present): Referred Publications (In Preparation): Arnold PD, Stewart SE, Shaheen S-M, Fagerness J,  Taillefer S, Doan B, Hounie AG,  Kennedy JL, Cook E, Pauls DL, Richter MA, Mathews CA, Hanna GL: DLGAP3 and obsessive-compulsive disorder.

Refereed Abstracts (Published/Accepted) from International/National Conferences:
  1. Arnold, P.D; MacMaster, F.P.; Hanna, G.L.; Richter, M.A..; Sicard, T.; Burroughs, E.; Mirza Y,; Easter, P.C.; Rose, M.; Kennedy, J.L.; Rosenberg, D.R: Combined Imaging – Genetic Approaches in OCD. Presented at American College of Neuropharmacology (ACNP), Hollywood, FL as part of panel presentation entitled “Examining the Glutamate Hypothesis of Pediatric OCD with Neuroimaging and Genetics” (Chair DR Rosenberg), Dec 6-11, 2009.
  2. Arnold PD, Stewart SE, Shaheen S-M, Fagerness J,  Taillefer S, Doan B, Hounie AG,  Kennedy JL, Cook E, Pauls DL, Richter MA, Mathews CA, Hanna GL: DLGAP3 and obsessive-compulsive disorder. Presented at World Congress of Psychiatric Genetics (WCPG) XVII, San Diego CA as part of symposium entitled “Obsessive-compulsive disorder and tourette disorder: phenotype, genotype and genome-wide association studies (Chair: DL Pauls), November 7, 2009.
  3. Zai G, Likhodi O, Arnold PD, Sicard T, Burroughs E, Kennedy JL, Richter MA: A role for the serotonin receptor gene 5HTR2A in predicting SRI response in obsessive-compulsive disorder (OCD). Pharmacogenetics in Psychiatry Meeting, New York, NY, April 16-17, 2009.
  4. Arnold PD, Beneteau J, Sicard T, Burroughs E, Kennedy JL, Richter MA: The association between sequence variation in the serotonin transporter gene and obsessive-compulsive symptom dimensions. American Society of Human Genetics 58th Annual Meeting, Philadelphia, PA, Nov 11-15, 2008.
  5. Arnold PD, Stewart SE, Fagerness JA, Moorjani P, Scharf J, Hounie AG, Sampaio A, Miguel E, Delorme R, Leboyer M, Kennedy JL, Cook EH Jr., Pauls DL, Richter MA, Mathews Ca, Hanna GL: Sequence variation in DLGAP3 and obsessive-compulsive disorder. Human Genome Variation Meeting, Toronto, ON, Oct 15-17, 2008.
  6. Arnold PD, Stewart SE, Fagerness JA, Moorjani P, Scharf J, Hounie AG, Sampaio A, Miguel E, Delorme R, Leboyer M, Kennedy JL, Cook EH Jr., Pauls DL, Richter MA, Mathews Ca, Hanna GL: DLGAP3, a candidate gene derived from a mouse knockout model, may be associated with obsessive-compulsive disorder. 31st Annual Meeting of the Canadian College of Neuropsychopharmacology (CCNP), Toronto, ON, June 8, 2008
  7. Arnold PD, MacMaster FP, Richter MA, Hanna GL, Sicard T, Burroughs E, Mirza Y, Easter PC, Rose M, Kennedy JL, Rosenberg DR:  Obsessive-Compulsive Disorder: Study of a Complex Psychiatric Trait Using Putative Neuroimaging Intermediate Phenotypes. [Oral presentation] 7th Annual Meeting of the Complex Trait Consortium, Montreal, Quebec, June 3, 2008.
  8. Arnold PD, MacMaster FP, Hanna GL, Richter MA, Sicard T, Burroughs E, Mirza Y, Easter PC, Rose M, Kennedy JL, Rosenberg DR: Obsessive-Compulsive Disorder: Combined Genetic and Neuroimaging Analysis of a Complex Psychiatric Trait. 3rd Annual Canadian Genetic Epidemiology and Genetic Statistics Meeting, Toronto, ON, April 29, 2008.
  9. 9.     Arnold PD, Richter MA, Nicolini H, Nestadt G, Kennedy JL: What can genetics tell us regarding the nosology of OCD and spectrum disorders? American College of Neuropsychopharmacology 2007 meeting as part of Study Group entitled “How Can Translational Neuroscience Inform DSM-V? The Example of the OCD Spectrum”, K Phillips (chair), Boca Raton, FL, December 10, 2007.
  10. Arnold PD, MA Richter, T Sicard, GL Hanna, C Pato, M Pato, DR Rosenberg, JL Kennedy: The glutamate system in OCD:  A combined genetics-imaging approach. American Academy of Child and Adolescent Psychiatry (AACAP), 2007 meeting as part of symposium entitled, “An Update on the Genetics of OCD and Tourette Disorder”, SE Stewart (chair),  Boston MA, October 25, 2007.
  11. Arnold PD, Sicard T, Burroughs E, MacMaster FP, Mirza Y, Taormina P,  Boyd C, Lynch M, Ivey J, Rose M, Richter MA, Kennedy JL, Rosenberg DR: Glutamate system genes in pediatric obsessive-compulsive disorder: an imaging genetics study using magnetic resonance spectroscopy. World Congress of Psychiatric Genetics (WCPG) XV, New York, NY, October 7-11, 2007, 130:396.
  12. Arnold PD, Sicard T, Burroughs E, Richter MA, Kennedy JL: The glutamate transporter gene (SLC1A1) in obsessive-compulsive disorder. 14th Annual Obsessive-Compulsive Foundation (OCF) International Conference as part of symposium entitled “ An Update on Genetics Research in OCD: A Reports from Members of the OCF Genetics Collaborative”, SE Stewart (chair), Spring, TX, July 21, 2007.
  13. Arnold PD, Sicard T, Burroughs E, MacMaster FP, Mirza Y, Taormina P,  Boyd C, Lynch M, Ivey J, Rose M, Richter MA, Kennedy JL, Rosenberg DR: A candidate gene study of thalamic volume in pediatric obsessive-compulsive disorder.  Neuropharmacology 31:S225 [Poster presentation at the American College of Neuropsychopharmacology Meeting, Hollywood, FL, Dec 6, 2006]
  14. 14.  Arnold P, Sicard T, Burroughs E, MacMaster F, Mirza Y, Smith J, Banerjee S, Bhandari R, Boyd C, Lynch M, Ivey J, Rose M, Hanna G, Richter M, Kennedy J, Rosenberg D: Glutamate system genes and neuroimaging phenotypes in pediatric obsessive-compulsive disorder. [Poster presentation at Canadian Academy of Child and Adolescent Psychiatry Annual Meeting, November 13, 2006]
  15. Arnold PD, Sicard T, Burroughs E, Hanna, G.L., Pato M, Pato C, Richter MA, Kennedy JL: A candidate gene study of the glutamate receptor gene GRIN2A in obsessive-compulsive disorder. [Poster presentation at Canadian Psychiatric Association Annual Meeting, November 10, 2006].
  16. Arnold PD, Sicard T, Burroughs E, Hanna, G.L., Pato M, Pato C, Richter MA, Kennedy JL: An association study of a glutamate receptor gene (GRIN2A) in obsessive-compulsive disorder. [Presented at American Society of Human Genetics, New Orlean, LA; Oct 9-13, 2006].