Serotonin Genes and Their Association with OC Trait

‘Long’ Story ‘Short’: Serotonin Genes and Their Association with OC Trait Dimensions and Hoarding in a Community, Pediatric Sample and with Brain Volume Differences in a Clinical, Pediatric OCD Sample

Vanessa M. Sinopoli

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy (PhD) Institute of Medical Science, Faculty of Medicine School of Graduate Studies University of Toronto

© Copyright by Vanessa M. Sinopoli (2019) ‘Long’ Story ‘Short’: Serotonin Genes and Their Association with OC Trait Dimensions and Hoarding in a Community, Pediatric Sample and with Brain Volume Differences in a Clinical, Pediatric OCD Sample

Vanessa M. Sinopoli

Doctor of Philosophy (PhD)

Institute of Medical Science, Faculty of Medicine School of Graduate Studies University of Toronto

2019

ABSTRACT

Background: Serotonin genes are commonly studied in obsessive-compulsive disorder

(OCD), but findings have been inconsistent. OCD is phenotypically heterogeneous, with subgroups reflecting symptom dimensions, sex, and age of onset. The thesis aimed to reduce heterogeneity in pediatric samples, stratifying by sex and studying either homogenous symptom-based subgroups or studying structural brain imaging as putative intermediate OCD phenotypes. For Study 1, we hypothesized that different serotonin gene variants would be associated with different obsessive-compulsive (OC) trait dimensions. In Study 2, we examined hoarding traits more closely and hypothesized that unique serotonin gene variants would be associated with hoarding in the absence of other OC traits. For Study 3, we hypothesized that specific serotonin gene variants would differ in their relationship with regional brain volume between OCD patients and controls.

ii Methods: We genotyped candidate serotonin genes in 5213 pediatric participants in the community (for Studies 1 and 2) and in 200 pediatric participants in the clinic (for Study 3).

In Studies 1 and 2, we assessed the association between gene variants and trait groups in males and females separately. In Study 3, we conducted structural magnetic resonance imaging

(sMRI) to measure regional brain volumes within cortico-striato-thalamo-cortical (CSTC) circuits. In males and females separately, we assessed the effect of genotype-diagnosis interaction on brain volume.

Results: In Study 1, the [LG + S] variant in 5-HTTLPR was significantly associated with hoarding in males (P-value of 0.003). In Study 2, [LG + S] was significantly associated with hoarding alone in males (P-value of 0.009). In Study 3, there was a significant genotype- diagnosis interaction for two single nucleotide polymorphisms (SNPs) in HTR2C, rs12860460

(P-value of 9.70e-8) and rs12854485 (P-value of 2.07e-6) in females.

Conclusions: Study 1 indicated that hoarding may be distinct in its underlying serotonin system genetics and Study 2 more specifically showed that hoarding traits alone were driving our findings in Study 1. Study 3 suggested that sequence variation in HTR2C influences ACC volume in female pediatric OCD. Future genetic association studies in OCD should similarly account for heterogeneity and study homogenous subgroups, where possible, to help further decipher the biological mechanisms driving symptoms.

iii ACKNOWLEDGEMENTS

Research Personnel, Supervisors, and Mentors: Dr. Paul Arnold (supervisor), Dr. Russell

Schachar (Program Advisory Committee member), Dr. Peggy Richter (Program Advisory

Committee member), Dr. Christie Burton (lab personnel and mentor)

Other Individuals: My family, my husband, my son, and my close friends for their patience, support, and encouragement to persevere

GENERAL STATEMENT OF CONTRIBUTIONS

The following individuals have contributed to my published review and/or to at least one of my three studies as coauthors (research personnel, lab technicians, statisticians, colleagues, collaborators):

Lauren Erdman, Christie L. Burton, Sefi Kronenberg, Laura S. Park, Rageen Rajendram, Julie

Coste, Annie Dupuis, Janet Shan, Tara Goodale, S-M Shaheen, Jennifer Crosbie, Phillip

Easter, Gregory Baldwin, Kelli Peterman, Gregory L. Hanna, David R. Rosenberg, Russell J.

Schachar, Paul D. Arnold

Funding:

I received funding from the Canadian Institutes of Health Research (CIHR) Master's Award:

Frederick Banting and Charles Best Canada Graduate Scholarships, Ontario Graduate

Scholarship (OGS), and the Hospital for Sick Children Restracomp Studentship.

iv Published Material:

The LITERATURE REVIEW chapter is modified from the following article and contains the entire article contents, in full:

Sinopoli, V. M., Burton, C. L., Kronenberg, S., & Arnold, P. D. (2017). A review of the role of serotonin system genes in obsessive-compulsive disorder. Neuroscience and Biobehavioral Reviews, 80, 372-381.

The chapter, STUDY 1, is modified from the following article and contains the entire article contents, in full:

Sinopoli, V. M., Erdman, L., Burton, C. L., Park, L. S., Dupuis, A., Shan, J., Goodale, T., Shaheen, S-M, Crosbie, J., Schachar, R. J., Arnold, P. D. (2019). Serotonin system genes and obsessive-compulsive trait dimensions in a population-based, pediatric sample: A genetic association study. Journal of Child Psychology and Psychiatry, doi: 10.1111/jcpp.13079. [Epub ahead of print]

The chapter, STUDY 2, is modified from the following article and contains the entire article contents, in full:

Sinopoli, V. M., Erdman, L., Burton, C. L., Park, L. S., Dupuis, A., Shan, J., Goodale, T., Shaheen, S-M, Crosbie, J., Schachar, R. J., Arnold, P. D. (under review). Serotonin system genes and hoarding with and without other obsessive-compulsive traits in a population-based, pediatric sample: A genetic association study. Depression and Anxiety.

The chapter, STUDY 3, is modified from the following article and contains the entire article contents, in full:

Sinopoli, V. M., Erdman, L., Burton, C. L., Easter, P., Rajendram, R., Baldwin, G., Peterman, K., Coste, J., Shaheen, S-M, Hanna, G. L., Rosenberg, D. R., Arnold, P. D. (2019). Serotonin system gene variants and regional brain volume differences in pediatric OCD. Brain Imaging and Behavior, doi: 10.1007/s11682-019-00092-w. [Epub ahead of print]

v TABLE OF CONTENTS

Page

1. LITERATURE REVIEW...... 1

1.1. Overview of obsessive-compulsive disorder (OCD)...... 3

1.2. History of obsessive-compulsive phenomena...... 3

1.3. Diagnostic criteria...... 5

1.3.1. OCD...... 6

1.3.2. Body dysmorphic disorder (BDD)...... 8

1.3.3. Hoarding disorder...... 9

1.3.4. Trichotillomania (hair-pulling disorder)...... 9

1.3.5. Excoriation (skin-picking) disorder...... 10

1.4. Etiology of OCD: Differences in pediatric etiology...... 10

1.5. OCD and the serotonin system...... 11

1.6. Discovery of serotonin...... 14

1.6.1. The function of serotonin as a neurotransmitter...... 15

1.6.2. Discovery of serotonergic antidepressants...... 16

1.7. OCD treatment...... 17

1.7.1. Pharmacotherapy...... 17

1.7.2. Pharmacological mechanisms of SRIs...... 18

1.7.3. Psychotherapy and other treatment...... 19

1.8. A review of the role of serotonin system genes in OCD...... 19

1.9. Candidate gene studies of serotonin system genes in OCD...... 20

1.10. The serotonin transporter gene (SLC6A4)...... 21

1.10.1. The serotonin transporter-linked polymorphic region (5-HTTLPR)...... 22

vi 1.10.2. Variable number of tandem repeats polymorphism in intron 2

(STin2 VNTR) and other SLC6A4 variants...... 24

1.11. SLC6A4: Genetic association with OCD...... 24

1.11.1. 5-HTTLPR...... 24

1.11.2. STin2 VNTR and other SLC6A4 variants...... 25

1.12. Serotonin 2A receptor gene (HTR2A): Genetic association with OCD...... 27

1.13. Serotonin 1B receptor gene (HTR1B): Genetic association with OCD...... 28

1.14. Serotonin 2C receptor gene (HTR2C): Genetic association with OCD...... 29

1.15. Pharmacogenetics...... 30

1.16. Intermediate phenotypes: Neuroimaging...... 32

1.17. Discussion...... 34

1.17.1. Genetic complexity of OCD...... 34

1.17.2. Animal models of OCD: Genetic mouse models of obsessive-compulsive-like

behaviors...... 37

1.17.3. Phenotypic heterogeneity of OCD...... 39

1.17.4. General conclusion...... 42

2. AIMS AND HYPOTHESES...... 45

2.1. Rationale and general aims...... 45

2.2. Specific aims...... 46

2.3. Hypotheses...... 46

2.4. Research study 1...... 47

2.4.1. Part A...... 47

2.4.2. Part B...... 48

2.5. Research study 2...... 48

vii 2.5.1. Part A...... 48

2.5.2. Part B...... 48

2.6. Research study 3...... 49

2.6.1. Part A...... 49

2.6.2. Part B...... 49

2.6.3. Part C...... 49

3. STUDY 1 – Serotonin system genes and obsessive-compulsive trait dimensions in a

population-based, pediatric sample: A genetic association study...... 51

3.1. Introduction...... 55

3.2. Methods...... 58

3.2.1. Participants...... 58

3.2.2. Obsessive-compulsive features...... 58

3.2.3. DNA collection and extraction...... 59

3.2.4. Selection of candidate genes...... 60

3.2.5. 5-HTTLPR...... 60

3.2.5.1. Direct genotyping...... 60

3.2.5.2. Statistical analyses...... 63

3.2.6. Candidate gene SNPs...... 64

3.2.6.1. Genotyping...... 64

3.2.6.2. Statistical analyses...... 65

3.2.7. Sample size considerations...... 66

3.3. Results...... 66

3.3.1. 5-HTTLPR analyses...... 66

3.3.2. Candidate gene SNP analyses...... 70

viii 3.4. Discussion...... 75

3.5. Conclusion...... 79

4. STUDY 2 – Serotonin system genes and hoarding with and without

other obsessive-compulsive traits in a population-based, pediatric

sample: A genetic association study...... 80

4.1. Introduction...... 84

4.2. Methods...... 87

4.2.1. Participants...... 87

4.2.2. Hoarding/obsessive-compulsive features...... 88

4.2.3. DNA collection and extraction...... 89

4.2.4. Selection of candidate genes...... 89

4.2.5. 5-HTTLPR...... 90

4.2.5.1. Direct genotyping...... 90

4.2.5.2. Statistical analyses...... 91

4.2.6. Candidate gene SNPs...... 91

4.2.6.1. Genotyping...... 91

4.2.6.2. Statistical analyses...... 92

4.2.7. Sample size considerations...... 93

4.3. Results...... 93

4.3.1. 5-HTTLPR analyses...... 93

4.3.2. Candidate gene SNP analyses...... 96

4.4. Discussion...... 100

4.5. Conclusion...... 104

5. STUDY 3 – Serotonin system gene variants and regional brain volume differences in

ix pediatric OCD...... 105

5.1. Introduction...... 109

5.2. Methods...... 113

5.2.1. Subjects...... 113

5.2.2. Imaging...... 115

5.2.3. DNA collection and extraction...... 116

5.2.4. Selection of candidate genes...... 116

5.2.5. Genotyping...... 117

5.2.5.1. 5-HTTLPR...... 117

5.2.5.2. SNPs across SLC6A4, HTR2A, HTR1B, and HTR2C...... 118

5.2.6. Statistical analyses...... 120

5.2.6.1. Assessing the association of serotonin gene variants with OCD...... 120

5.2.6.2. Assessing the effect of serotonin gene variants on brain volume...... 120

5.2.6.3. Assessing the effect of genotype-diagnosis interaction on brain volume...... 121

5.3. Results...... 121

5.3.1. Assessing the association of serotonin gene variants with OCD...... 123

5.3.2. Assessing the effect of serotonin gene variants on brain volume...... 123

5.3.3. Assessing the effect of genotype-diagnosis interaction on brain volume...... 125

5.4. Discussion...... 129

5.5. Conclusion...... 134

6. GENERAL DISCUSSION, CONCLUSIONS AND FUTURE DIRECTIONS...... 136

6.1. General Discussion...... 136

6.1.1. Summary of key findings...... 136

6.1.2. Candidate gene approach in homogenous subgroups...... 141

x 6.1.3. Community versus clinic-based samples in the study of OCD and other

complex disorders...... 144

6.1.4. Intermediate phenotypes...... 145

6.1.5. Limitations...... 148

6.1.6. Thesis findings and the literature...... 149

6.2. Conclusions...... 151

6.2.1. Summary...... 151

6.2.2. Integrating information across genetic, functional, and behavioral levels...... 151

6.2.3. Translation to clinic...... 155

6.3. Future Directions...... 156

6.3.1. Genetics and environmental risk in neurodevelopment...... 158

6.3.1.1. Environmental risk factors in OCD...... 158

6.3.1.2. Effect of environmental risk on the genome...... 159

6.3.1.3. Epigenomic changes and OCD...... 160

6.3.1.4. Conclusion...... 160

6.3.2. Proposal for Future Experiment...... 161

REFERENCES...... 164

xi LIST OF TABLES

Page

Table 1. OCD obsessions and compulsions...... 7

Table 2. Summary of serotonin gene findings...... 21

Table 3. SNP counts across 3 candidate genes, including all genotyped and

imputed SNPs...... 65

Table 4. 5-HTTLPR-genotyped individuals...... 67

Table 5. Trait dimension group counts for 5-HTTLPR-genotyped individuals...... 68

Table 6. Individuals genotyped for candidate gene SNPs...... 71

Table 7. Trait dimension group counts for individuals genotyped for candidate

gene SNPs...... 72

Table 8. Association between serotonin gene SNPs and OC traits overall/OC

trait dimensions in males...... 73

Table 9. Association between serotonin gene SNPs and OC traits overall/OC

trait dimensions in females...... 74

Table 10. SNP counts across 3 candidate genes, including all genotyped and

imputed SNPs...... 92

Table 11. 5-HTTLPR-genotyped individuals...... 94

Table 12. Trait group counts for 5-HTTLPR-genotyped individuals...... 95

Table 13. Individuals genotyped for candidate gene SNPs...... 97

Table 14. Trait group counts for individuals genotyped for candidate gene SNPs...... 98

Table 15. Association between serotonin gene SNPs and trait groups in males...... 99

Table 16. Association between serotonin gene SNPs and trait groups in females...... 100

Table 17. SNP counts across 4 candidate genes...... 119

xii Table 18. Patient and control demographics...... 122

Table 19. Effect of serotonin gene variants on brain volume in males...... 124

Table 20. Effect of serotonin gene variants on brain volume in females...... 125

Table 21. Effect of genotype-diagnosis interaction on brain volume in males...... 127

Table 22. Effect of genotype-diagnosis interaction on brain volume in females...... 128

Table 23. Summary of thesis results...... 138

xiii LIST OF FIGURES

Page

Figure 1. Serotonin signaling...... 14

Figure 2. OCD as a complex disorder...... 42

Figure 3. 5-HTTLPR; L and S variants and rs25531...... 62

Figure 4. Association between 5-HTTLPR and OC traits overall/OC trait dimensions...... 69

Figure 5. Association between 5-HTTLPR and trait groups...... 96

Figure 6. Brain volume versus genotype in females graphed...... 129

Figure 7. Study 3 SNP alleles explained...... 147

Figure 8. A multi-dimensional approach to OCD and related disorders...... 154

Figure 9. Potential neurodevelopmental changes in OCD...... 161

xiv LIST OF ABBREVIATIONS

3D – three-dimensional

5-HT – 5-hydroxytryptamine; serotonin

5-HT1A – serotonin 1A

5-HT1B – serotonin 1B

5-HT2A – serotonin 2A

5-HT2C – serotonin 2C

5-HTTLPR – serotonin transporter-linked polymorphic region

ABCD – Adolescent Brain Cognitive Development

ACC – anterior cingulate cortex

ADHD – attention-deficit/hyperactivity disorder

APA – American Psychiatric Association

ASD – autism spectrum disorder

BDD – body dysmorphic disorder

BDNF – brain-derived neurotrophic factor gene bp – base pair

CBT – cognitive behavioral therapy cc – cubic centimeters

COMT – catechol-O-methyltransferase gene

CS – cesarean section

CSTC – cortico-striato-thalamo-cortical

CY-BOCS – Children’s Yale-Brown Obsessive-Compulsive Scale

DAT – dopamine transporter

xv DSM – Diagnostic and Statistical Manual of Mental Disorders

FDA – Food and Drug Administration

GxE – gene-environment

GABA – γ-aminobutyric acid

GAD – generalized anxiety disorder

GEC – Genetic Type 1 error calculator

GWAS – genome-wide association study

HTR1B – serotonin 1B receptor gene

HTR2A – serotonin 2A receptor gene

HTR2C – serotonin 2C receptor gene

ICC – intraclass correlation coefficient

ICV – intracranial volume

K-SADS-PL – Schedule for Affective Disorder and Schizophrenia for School-Age Children- Present and Lifetime Version

L variant – long variant

LA – L variant with an A allele at rs25531

LG – L variant with a G allele at rs25531

MAF – minor allele frequency

MAOA – monoamine oxidase A gene mCPP – meta-chlorophenylpiperazine

MDD – major depressive disorder miRNA – microRNA

MOG – myelin oligodendrocyte glycoprotein gene

MRI – magnetic resonance imaging

xvi N – sample size

NIH – National Institutes of Health

NIMH – National Institute of Mental Health

OC – obsessive-compulsive

OCD – obsessive-compulsive disorder

OCPD – obsessive-compulsive personality disorder

OCSD – obsessive-compulsive spectrum disorder

OFC – orbitofrontal cortex

OR – odds ratio

PANDAS – pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections

PANS – pediatric acute-onset neuropsychiatric syndrome

PC – principal component

PCA – principal component analysis

PCR – polymerase chain reaction

PDS – Pubertal Development Scale

PET – positron emission tomography

POND – Province of Ontario Neurodevelopmental Disorders

QC – quality control

RDoC – Research Domain Criteria

ROI – region of interest

S variant – short variant

SD – standard deviation

SERT – serotonin transporter protein

xvii SIDS – sudden infant death syndrome

SLC1A1 – glutamate transporter gene

SLC6A3 – dopamine transporter gene

SLC6A4 – solute carrier family 6 member 4; serotonin transporter gene sMRI – structural magnetic resonance imaging snoRNA – small nucleolar RNA

SNP – single nucleotide polymorphism

SNRIs – serotonin-norepinephrine reuptake inhibitors

SOCOBS – Schedule for Obsessive-Compulsive and Other Behavioral Syndromes

SRIs – serotonin reuptake inhibitors

SSRIs – selective serotonin reuptake inhibitors

STin2 VNTR – variable number of tandem repeats polymorphism in intron 2

TAG – Thoughts, Actions and Genes

TCAG – The Centre for Applied Genomics

TOCS – Toronto Obsessive-Compulsive Scale

TS – Tourette syndrome

Y-BOCS – Yale-Brown Obsessive-Compulsive Scale

xviii 1. LITERATURE REVIEW

This chapter is modified from the following article and contains the entire article contents, in full:

Copyright permission was obtained from journal publisher, Elsevier, for full use of the article and its contents in this dissertation.

Authors

Vanessa M. Sinopolia,b, Christie L. Burtonb,e, Sefi Kronenbergc,d, Paul D. Arnoldb,f,g,*

Affiliations aInstitute of Medical Science, University of Toronto, Canada bProgram in Genetics & Genome Biology, The Hospital for Sick Children, Toronto, Canada cDepartment of Psychiatry, The Hospital for Sick Children, Toronto, Canada dDepartment of Psychiatry, University of Toronto, Canada eProgram in Neurosciences and Mental Health, The Hospital for Sick Children, Toronto, Canada fMathison Centre for Mental Health Research & Education, Hotchkiss Brain Institute, University of Calgary, Canada gDepartments of Psychiatry and Medical Genetics, Cumming School of Medicine, University of Calgary, Canada * Corresponding author at: Mathison Centre for Mental Health Research and Education, 4th floor, Teaching, Research and Wellness (TRW) Building, 3280 Hospital Dr NW, Calgary AB, T2N 4Z6, Canada; E-mail address: [email protected] (P.D. Arnold, MD, PhD, FRCPC)

Statement of Contributions

The following authors have contributed to my published review as follows:

Christie L. Burton – coauthor, insight, editing

Sefi Kronenberg – coauthor, insight, editing

Paul D. Arnold – editing, supervision

Conflicts of Interest

The authors have no conflicts of interest to disclose.

Acknowledgements

Dr. Paul D. Arnold receives funding from the Alberta Innovates Health Solutions (AIHS)

Translational Health Chair in Child and Youth Mental Health. Vanessa Sinopoli received funding from the Canadian Institutes of Health Research (CIHR) Master's Award: Frederick

Banting and Charles Best Canada Graduate Scholarships, Ontario Graduate Scholarship

(OGS), and the Hospital for Sick Children Restracomp Studentship.

2 1.1. Overview of obsessive-compulsive disorder (OCD)

OCD is a neuropsychiatric disorder characterized by recurring, intrusive thoughts and/or repetitive and often ritualized behaviors that are carried out in response to obsessions or set rules and usually intended to reduce distress (American Psychiatric Association [APA], 2013).

The worldwide prevalence of OCD is 2 to 3% and the World Health Organization lists the disorder as one of the top ten most debilitating illnesses (Murray & Lopez, 1996; Angst et al.,

2004; Kessler et al., 2005; Murphy et al., 2013). OCD is a phenotypically heterogeneous disorder with multiple symptom dimensions that sometimes overlap (Bloch et al., 2008a).

These symptom dimensions are captured in a widely-used clinical measure, the Yale-Brown

Obsessive-Compulsive Scale (Y-BOCS), which provides a checklist of OCD symptoms and yields a severity score for obsessive, compulsive, and total symptoms (Goodman et al., 1989a;

Goodman et al., 1989b). OCD patients often present with comorbid disorders including a lifetime tic disorder in up to 30% of OCD patients, anxiety disorders, and major depressive disorder. Tic disorders are particularly common in males with early-onset OCD (onset typically in childhood or early adolescence), and children sometimes present with a combination of OCD, tic disorder, and attention-deficit/hyperactivity disorder (ADHD) (APA,

2013).

1.2. History of obsessive-compulsive phenomena

Hundreds of years ago, among other cultural beliefs, OCD-like symptoms were once believed to be a result of possession by outside, evil forces and were treated by witch doctors or religious leaders (Jenike, 1983). Throughout the 17th century, symptoms were commonly

3 described in religious contexts. In 1621, for example, Burton of Oxford University explained symptoms whereby the individual was afraid of blurting out something indecent while sitting through a sermon (Stein et al., 2009).

In the 19th century, modern theories of OCD phenomena surfaced and mirrored theoretical shifts in psychiatry (Berrios, 1989). In the first half of the 19th century, Esquirol described

OCD-like symptoms as a form of monomania or type of insanity with aspects of “involuntary, irresistible, and instinctive activity” and also noted that affected individuals seemed to have insight into their symptoms (Berrios, 1989; Stein et al., 2009). Around the 1850s, French psychiatrists like Morel, Dagonet, and Magnan described OCD symptoms as a neurosis, and not as an insanity, and more so as a disease of the emotions. OCD phenomena were also described as impulses and the disease was considered to reflect degeneration and cerebrospinal pathology. It was also believed to have a hereditary component (Berrios, 1989; Stein et al.,

2009). Meanwhile in Germany, the prominent belief of psychiatrists like Griesinger and

Westphal was that obsessional presentations were a result of intellectual dysfunction. The terminology Westphal used to describe the disorder would later translate into “obsessive- compulsive disorder” (Berrios, 1989; Stein et al., 2009).

In the early 20th century, Freud and Janet were among the first to present early psychodynamic theories of OCD (Jenike, 1983). Freud then described OCD symptoms using the term

“anxiety neurosis” (a term credited to Hecker) (Stein et al., 2009). Freud first hypothesized that obsessions were a result of a genital experience during childhood and later hypothesized that obsessions were a way of defending oneself against unconscious impulses relating to the mother-infant bond and ideas of aggression surrounding the anal-sadistic era. The patient was

4 thought to regress back to this earlier era and act out obsessions to hide the more distressing ideas or state of anxiety (Jenike, 1983). Janet attributed the symptoms to mental fatigue, which led to a lack of control over one’s thoughts and resulted in obsessive and compulsive behaviors (Jenike, 1983; Stein et al., 2009).

In the late 20th century, epidemiologic studies revealed that OCD was much more prevalent than once thought (Stein et al., 2009). We still strive to understand its etiology and turn to advances in psychology, pharmacology, neuroanatomy, physiology, genetics/genomics, and functional biology to better understand the disorder.

1.3. Diagnostic criteria

In earlier versions of the Diagnostic and Statistical Manual of Mental Disorders (DSM)

(beginning with the DSM-III), OCD was classified as an anxiety disorder. In the 5th and newest edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5), OCD is no longer listed as an anxiety disorder and has now been assigned its own chapter with the heading, “Obsessive-Compulsive and Related Disorders” (APA, 2013).

The concept of an obsessive-compulsive spectrum was proposed in the 1990s and was used to describe disorders thought to be separate from but related to OCD based on similar underlying repetitive thoughts and/or behaviors (Stein & Hollander, 1993; Phillips et al., 2010). A number of these obsessive-compulsive spectrum disorders (OCSDs) are now classified under the new “Obsessive-Compulsive and Related Disorders”. The cluster of disorders under this new designation includes the following:

1.3.1. OCD

OCD is defined by the presence of obsessions and/or compulsions, which cause the patient significant distress and take up a significant amount of time, impairing their ability to function in society. Obsessions are further described as “recurrent and persistent thoughts, urges, or images that are experienced, at some time during the disturbance, as intrusive and unwanted” whereby the patient “attempts to ignore or suppress such thoughts, urges, or images, or to neutralize them with some other thought or action (i.e., by performing a compulsion)”.

Compulsions are further described as “repetitive behaviors (i.e., hand washing, ordering, checking) or mental acts (i.e., praying, counting, repeating words silently) that the individual feels driven to perform in response to an obsession or according to rules that must be applied rigidly” and where “the behaviors or mental acts are aimed at preventing or reducing anxiety or distress, or preventing some dreaded event or situation” which is identified as unreasonable or unrealistic (APA, 2013). There are a number of common types of obsessions and compulsions and these are well described in the Y-BOCS symptom checklist which, since it was published in 1989, has been accepted as the “gold standard” instrument for assessing

OCD symptoms in terms of their nature and severity (Goodman et al., 1989a; Goodman et al.,

1989b). Some common obsessions and compulsions in OCD are listed in Table 1.

Table 1. OCD obsessions and compulsions: Common obsessions and compulsions as listed in the Y-BOCS symptom checklist. Adapted from Goodman et al., 1989b.

OBSESSIONS pertaining to: COMPULSIONS pertaining to:

Aggression Cleaning/Washing (i.e., fear of harming oneself or others, doing (i.e., excessive or ritualized hand washing, something wrong or obscene, unwanted showering, bathing, grooming, etc., excessive violent or horrific images, fear of being or ritualized cleaning of household items or responsible for terrible or harmful things objects, attempts to eliminate contact with befalling others, fear of not being careful contaminants) enough, fear of thinking, saying, or doing obscene or horrible things or acting on unwanted impulse)

Contamination Checking (i.e., excessive concern with germs, (i.e., repeatedly checking locks, stoves, contaminants, chemicals, bodily fluids, fear appliances, checking that did not or will not of becoming ill or getting others ill as a harm others, self, or that nothing bad did or result) will happen, checking that did not make mistake, repeated checking in response to somatic obsessions)

Sexual Repeating (i.e., forbidden sexual thoughts or images or (i.e., re-reading or re-writing for fear of not impulses that are perverse or aggressive in understanding something or to make “perfect”, nature) obsessive thoughts about the shape of certain letters, repeating routines like turning appliances on and off, entering and exiting though a doorway, or brushing hair, having to do things the “right” number of times)

Hoarding Counting (i.e., concern with saving things, worry (i.e., a need to count objects or activities over about discarding seemingly unimportant and over) things that may be needed in the future, urge to collect useless items)

Religious Ordering/Arranging (i.e., excessive preoccupation with (i.e., a need to arrange and order things right/wrong and morality, scrupulosity or repeatedly and experience of distress if this fear of being sacrilegious or blasphemous) order is perturbed)

Symmetry or Exactness Hoarding (i.e., concern with things being symmetrical, (i.e., hoarding/collecting/saving useless things exact, or in the right place – either like old newspapers, empty bottles, wrappers accompanied or unaccompanied by magical for fear of throwing away things that may be thinking like a fear that something bad will needed one day, picking useless objects from happen to the mother otherwise) the street/garbage)

Miscellaneous Miscellaneous (i.e., a need to remember or know, fear of (i.e., carrying out mental rituals like thinking a forgetting or losing something, fear saying “good” thought to undo a “bad” thought, a certain things, superstitions, fear of not need to touch or tap objects believing it will saying things just right, bothered by prevent illness in family members, a need to intrusive images, sounds, words, lucky or confess things that one may not have even unlucky numbers, colors) done or to ask for reassurance, superstitious behaviors like avoiding certain numbers

Somatic (i.e., concern with illness or disease like cancer, excessive concern with appearance or a body part)

Abbreviations: Y-BOCS, Yale-Brown Obsessive-Compulsive Scale.

1.3.2. Body dysmorphic disorder (BDD)

The DSM-IV previously categorized BDD as a somatoform disorder (APA, 1994), despite being conceptualized as an OCSD along with trichotillomania, excoriation, Tourette syndrome

(TS), bulimia, and anorexia (Phillips et al., 2010). Now categorized as an obsessive- compulsive related disorder, BDD is described by “preoccupation with one or more perceived defects or flaws in physical appearance that are not observable or appear slight to others”. It involves “repetitive behaviors or mental acts in response to the appearance concerns” and is the cause of “clinically significant distress or impairment in social, occupational, or other important areas of functioning” (APA, 2013).

1.3.3. Hoarding disorder

Prior to release of the DSM-5, hoarding was considered a symptom of OCD and patients were either diagnosed with OCD, anxiety disorder, or OCPD (APA, 2013). In the DSM-5, hoarding disorder is described by “persistent difficulty discarding or parting with possessions, regardless of their actual value” and the behavior leads to an “accumulation of possessions that congest or clutter active living areas and substantially compromise their intended use”.

Hoarding “causes clinically significant distress or impairment in social, occupational, or other important areas of functioning” (APA, 2013). Hoarding disorder is estimated to affect 2-5% of adults and 2% of adolescents. Little is known, however, about its prevalence in children

(Burton et al., 2015a). Its new designation is intended to motivate further research into hoarding as a distinct disorder (APA, 2013).

1.3.4. Trichotillomania (hair-pulling disorder)

Trichotillomania is described as the “recurrent pulling out of one’s hair resulting in hair loss” that is not described by another mental disorder or medical condition. The patient must have made “repeated attempts to decrease or stop pulling” and the hair pulling “causes clinically significant distress or impairment in social, occupational, or other important areas of functioning” (APA, 2013).

9 1.3.5. Excoriation (skin-picking) disorder

Based on similar ritualistic and compulsive components, excoriation (otherwise known as dermatillomania, or compulsive skin picking) has clinically been compared to OCD in the past. Given that it affects 2-4% of the population and that findings suggest distinct symptomatology and treatment, excoriation is now also a separate disorder in the DSM-5. The disorder is described as “recurrent skin picking resulting in skin lesions” that is not described by another mental disorder, medical condition, or substance use. The patient must have made

“repeated attempts to decrease or stop skin picking” and the skin picking causes “clinically significant distress or impairment in social, occupational, or other important areas of functioning” (APA, 2013).

1.4. Etiology of OCD: Differences in pediatric etiology

The etiology of OCD remains elusive. A number of lines of evidence point to a prominent role for genetic risk factors in OCD. First, OCD is familial. Relatives of individuals with

OCD are at higher risk of having the disorder than relatives of healthy individuals, with an

8.2% aggregate risk in first-degree relatives of subjects with OCD versus 2% in control relatives (Hettema et al., 2001; Pauls et al., 2014). Second, twin studies demonstrate that the familiality in OCD is largely due to genetic factors; i.e., the disorder is heritable (Pauls et al.,

2014). Genetic factors may play a particularly relevant role in childhood or early-onset OCD.

OCD is more common in first-degree relatives of probands with early-onset OCD, compared to first-degree relatives of probands with late-onset OCD (onset typically in late adolescence or early adulthood) (do Rosario-Campos et al., 2005; Hanna et al., 2005; Taylor, 2011; Pauls

10 et al., 2014). Symptom heritability is also higher in children with OCD (45 to 65%) versus adults with OCD (27 to 47%) (van Grootheest et al., 2005). There is substantial evidence showing that early-onset OCD is a neurodevelopmental disorder (Pauls et al., 2014).

Approximately one third to one half of adults with OCD have developed the disorder in childhood (Pauls et al., 2014). In children, onset in males ranges from 9 to 11 years of age and onset in females ranges from 11 to 13 years of age (MacMaster, 2010). Between 30 and 50% of cases have their onset in childhood and it has been postulated that this early-onset form may be neurodevelopmentally distinct (Pauls et al., 2014). To further support this concept, the aforementioned differences derived from child versus adult studies of symptom dimension breakdown may be attributed, in part, to developmental differences (Bloch et al., 2008a).

Intense research has endeavored to identify the specific risk variants associated with OCD, with much focus on the neurotransmitter, serotonin (5-hydroxytryptamine, or 5-HT) (Pauls et al., 2014).

1.5. OCD and the serotonin system

OCD is most commonly treated with drugs that act on the serotonin system, serotonin reuptake inhibitors (SRIs), which include selective serotonin reuptake inhibitors (SSRIs), combined serotonin-norepinephrine reuptake inhibitors (SNRIs), and clomipramine (a tricyclic compound) (Murphy et al., 2004; Millan et al., 2015). The discovery of SRI efficacy in treating OCD symptoms was the original catalyst for decades of investigation into the role of serotonin, including serotonin system genetics, in the pathobiology of OCD. Several neuroimaging studies have examined the serotonin transporter protein (SERT) and serotonin

2A (5-HT2A) receptor binding or availability in OCD with mixed results: some studies

11 showing increased (Pogarell et al., 2003; Adams et al., 2005), decreased (Stengler-Wenzke et al., 2004; Hesse et al., 2005; Hasselbalch et al., 2007; Reimold et al., 2007; Zitterl et al., 2007;

Perani et al., 2008; Matsumoto et al., 2010), or similar binding or availability within specific brain regions of interest (Simpson et al., 2003; van der Wee et al., 2004; Simpson et al., 2011).

A number of studies also reported neuroimaging differences related to age of OCD onset

(Pogarell et al., 2003; Simpson et al., 2011) or symptom severity (Hesse et al., 2005; Reimold et al., 2007; Zitterl et al., 2007; Perani et al., 2008; Hesse et al., 2011).

Serotonin is a major monoamine neurotransmitter in the central nervous system. The serotonin system originates in the midbrain raphe nuclei and projects to several cortical and sub-cortical brain regions. Serotonin is essential to a wide range of functions, including mood, behavior, eating patterns, cognition, sleep, reproduction, and motor functions (Vanhoutte,

1990; Murphy et al., 2004; Murphy et al., 2008; Murphy & Lesch, 2008). The serotonin system has also been implicated in the pathogenesis of various psychiatric traits, disorders, and medical conditions. These include OCD, depression, anxiety, neuroticism, autism,

ADHD, bipolar disorder, TS, sudden infant death syndrome (SIDS), irritable bowel syndrome, and pulmonary hypertension (Murphy et al., 2004; Nonnis Marzano et al., 2008).

Serotonin is released from the presynaptic neuron into the synapse where it can activate an array of downstream receptors in both the pre and postsynaptic neuronal cell membranes.

SERT is located in the presynaptic membrane and facilitates the reuptake of serotonin back into the presynaptic cell to be recycled after its use. As such, SERT plays a critical role in serotonin-based neuronal signalling (Stahl, 1998; Vanhoutte, 1990; Heils et al., 1996; Murphy et al., 2004; Murphy et al., 2008; Murphy & Lesch, 2008; Nichols & Nichols, 2008) (Figure

12 1). At least 14 serotonin receptor subtypes have been identified (Nichols & Nichols, 2008;

Palacios, 2016). Among the most commonly studied in OCD are the 5-HT2A receptor, the serotonin 1B (5-HT1B) receptor (or 5-HT1Dβ receptor in earlier literature), and the serotonin

2C (5-HT2C) receptor. The 5-HT2A receptor activates phosphoinositide hydrolysis to accommodate developmental and cell migration processes in response to serotonin (Dickel et al., 2007; Nichols & Nichols, 2008). The 5-HT1B receptor modulates serotonin release from neuronal axon terminals of serotonergic neurons in the raphe nuclei and also plays a role in the function of non-serotonergic neurons, such as γ-aminobutyric acid (GABA)-ergic and glutamatergic neurons (Dickel et al., 2007; Nichols & Nichols, 2008). Both the 5-HT2A receptor and the 5-HT2C receptor influence dopaminergic neurotransmission (Lappalainen et al., 1995; Nichols & Nichols, 2008).

Figure 1. Serotonin signaling: Diagram of a synapse between two neurons communicating via the neurotransmitter, serotonin, referred to here as 5-HT. 5-HT is released into the synaptic space where it can activate downstream receptors in the pre and postsynaptic membranes, such as the serotonin 2A receptor, the serotonin 1B receptor, the serotonin 2C receptor, and the serotonin 1A receptor (proteins respectively referred to here as 5-HT2A, 5-HT1B, 5-HT2C, and 5-HT1A). The serotonin transporter protein (SERT), located in the presynaptic membrane, is responsible for the reuptake of 5-HT back into the presynaptic neuron.

1.6. Discovery of serotonin

Early research into the effects of a substance (which would later come to be known as serotonin) began in 1868 (Mylecharane, 2013). In 1914, pathologist Pierre Masson discovered that specific intestinal cells are able to directly reduce silver nitrate (Masson, 1914; Sourkes,

2009). Also, studies in hypertension led to Irvine H. Page’s interest in a vasoconstrictor present in serum when the blood clots and in platelet extracts. Together, researchers Maurice

Rapport, Arda Green, and Irving H. Page later isolated and studied the nature and structure of

14 this substance, which they called “serotonin”, and identified it as 5-hydroxytryptamine

(Rapport et al., 1948a-c; Rapport, 1949; Sourkes, 2009; Mylecharane, 2013). In the 1930s,

Vittorio Erspamer also independently researched the substance that Masson had described and called it “enteramine”. He later obtained a sample of serotonin from Rapport and identified the substance as 5-hydroxytryptamine, which he noted was the same as “enteramine”

(Erspamer & Asero, 1952; Sourkes, 2009; Mylecharane, 2013).

George Reid and Michael Rand published a number of studies detailing the pharmacological and physiological actions of purified serotonin both in vivo and in vitro and noted such effects as an influence on blood pressure, broncho-, bladder, and pupil constriction in mammals, and the contraction of uterus and intestine preparations using mammalian tissue (Rand & Reid

1951; Reid, 1951; Reid & Rand 1951; Rand & Reid 1952; Reid & Rand 1952; Mylecharane,

2013). John Henry Welsh and his student Betty Mack Twarog, interested in neurotransmission, provided us with further evidence of serotonin behaving as a neurotransmitter. When they introduced serotonin to heart cells in Mercenaria and to muscle cells in Mytilus, they observed that it was excitatory in the heart and inhibitory in smooth muscle (Welsh, 1953; Welsh, 1957; Sourkes, 2009). Serotonin was later discovered in the mammalian brain (Twarog & Page, 1953) and in the nerve tracts originating from the raphe nuclei in the brain (Dahlstroem & Fuxe, 1964; Sourkes, 2009).

1.6.1. The function of serotonin as a neurotransmitter

Neurotransmitters act as chemical signals and mediate communication between neurons. The presynaptic neuron will release neurotransmitters into the synaptic cleft where they are

15 received by receptors on the postsynaptic membrane (and/or on the presynaptic membrane itself) to facilitate communication. Serotonin is classified as a small monoamine compound

(like norepinephrine) and can either excite, inhibit, or do both, depending on the receptors available on the neuron receiving the message. This in turn can activate a myriad of effects through a cascade of downstream signaling events. Serotonin, in particular, allows for a diverse range of outcomes given the large number of serotonin receptors. The raphe nuclei are a collection of neuron groups in the brainstem that use serotonin to communicate. Each group has distinct projections to different brain regions or to the spinal cord to mediate these numerous functions. (Martin, 2012)

1.6.2. Discovery of serotonergic antidepressants

In 1957, psychiatrist Roland Kuhn noted the antidepressant properties of a tricyclic agent previously thought to be an antipsychotic (Kuhn, 1957; Millan et al., 2015). Later re-coined imipramine, this tricyclic agent was the first to be approved by the Food and Drug

Administration (FDA) for the treatment of depression. Studies of the biological mechanisms of action only followed after the tricyclic agent was found to have antidepressant properties, as well as an effect on anxiety. The first mechanism of action proposed in depression was a resulting increase in norepinephrine levels in the brain (Schildkraut, 1965; Millan et al., 2015).

In the 1960s, examination of brain tissue led to a new hypothesis suggesting that there were decreased serotonin levels in the brains of depressed patients (Coppen, 1967; Millan et al.,

2015). The finding that some tricyclic agents blocked serotonin reuptake from the neuronal synapse followed this, resulting in the proposal that tricyclic agents improve mood by blocking serotonin reuptake (Carlsson et al., 1969; Millan et al., 2015). Clomipramine was

16 found to be a more effective serotonin reuptake inhibitor, compared to other tricyclic agents.

The FDA approved it in 1964 for its use in depression (Carlsson et al., 1969; Millan et al.,

2001; Millan et al., 2015).

The serotonin hypothesis led to the development of SSRIs throughout the 1970s and 1980s, with the advantage of having less adverse effects than tricyclic antidepressants (Millan et al.,

2015). Though known to block the reuptake of serotonin and therefore increase serotonin’s availability in the brain, the exact mechanisms by which SSRIs operate are still not completely understood today. SNRIs were later developed to block both serotonin and norepinephrine reuptake and may be used in patients who do not respond to SSRIs (Millan et al., 2015).

1.7. OCD treatment

1.7.1. Pharmacotherapy

Pharmacological treatment in OCD is chiefly based on the serotonin system. Clomipramine, initially used to treat depression, was the first drug found to effectively treat OCD as well

(Fenske & Petersen, 2015; Millan et al., 2015). SSRIs, also initially developed to treat depression, were found to be effective in OCD and are now the preferred first-line pharmacological treatment for OCD because they are believed to have less adverse effects.

SSRIs currently available include citalopram, escitalopram, fluoxetine, fluvoxamine, paroxetine, and sertraline. Higher SSRI dosages are normally required to treat OCD, compared to depression. Therapy should consist of a gradual increase in delivery of the medication up to a maximum tolerable dosage and should be administered for 8-12 weeks

17 before assessing its full effect. For patients who do not respond well to SSRIs, atypical antipsychotics are sometimes added to complement SSRI treatment. Alternatively, some OCD patients who do not respond to SSRIs are prescribed SNRIs, for which there is some evidence to support its effectiveness in OCD. Glutamatergic agents are currently being investigated for use in OCD as well (Fenske & Petersen, 2015).

1.7.2. Pharmacological mechanisms of SRIs

The exact mechanisms of action of SRIs are still largely unknown, with hypotheses largely derived from studies conducted in rodents (Blier & de Montigny, 1998; Stahl, 1998; Murphy et al., 2008). SRIs directly block the reuptake of serotonin via negative allosteric modulation of SERT, resulting in an immediate accumulation of serotonin in the synapse (Blier et al.,

1990; Billett et al., 1997; Stahl, 1998; Murphy et al., 2008). Chronic SRI treatment increases extracellular serotonin, which produces several downstream effects. These effects include desensitization of serotonin receptor subtypes, such as the serotonin 1A (5-HT1A) receptor and the 5-HT2A receptor, and ultimately enhanced serotonin neurotransmission (Blier et al., 1990;

Blier & de Montigny, 1998; Stahl, 1998). Overall, SRIs trigger a number of complex downstream actions that vary by SRI type, brain region, and neurodevelopmental stage. Acute effects are believed to mediate side effects that accompany SRI usage while effects of chronic treatment mediate the therapeutic effects of SRIs, as well as developed tolerance to side effects. SRIs take longer to yield a therapeutic response in patients with OCD rather than depression, suggesting that there are different mechanisms driving OCD pathogenesis and that

SRIs may take longer to affect these mechanisms (Blier et al., 1990; Blier & de Montigny,

1998; Stahl, 1998).

1.7.3. Psychotherapy and other treatment

The most effective form of psychotherapy for OCD treatment is cognitive behavioral therapy

(CBT). Like SSRIs, it a first-line treatment and may be offered alone or in conjunction with pharmacotherapy (Albert & Bogetto; 2015; Fenske & Petersen, 2015). In the case of severe

OCD that is resistant to any other form of treatment, deep brain stimulation may be effective, though not as well studied and more stringently regulated as of yet (Fenske & Petersen, 2015).

1.8. A review of the role of serotonin system genes in OCD

OCD is a debilitating neuropsychiatric disorder that causes the patient to experience intrusive thoughts and/or to carry out repetitive, ritualized behaviors that are time consuming and impairing. OCD is familial and heritable. The genetic factors responsible for pathogenesis, however, remain largely unknown despite the numerous candidate gene studies conducted.

Based on efficacy of SRIs in treating OCD, serotonin system genes have been a dominant focus in OCD candidate gene studies. We review the most commonly studied candidate serotonin system gene variants (specifically in SLC6A4, HTR2A, HTR1B, and HTR2C) and their association with OCD. Although findings to date are mixed, serotonin transporter polymorphism 5-HTTLPR and HTR2A polymorphism rs6311 (or rs6313) are most consistently associated with OCD. Mixed findings may be the result of genetic complexity and phenotypic heterogeneity that future studies should account for. Homogenous patient subgroups reflecting OCD symptom dimensions, OCD subtypes, and sex should be used for gene discovery.

1.9. Candidate gene studies of serotonin system genes in OCD

To date, over 100 association studies of OCD have been published. Candidate gene studies have focused mainly on genes from neurotransmitter systems including serotonin, dopamine, and glutamate (i.e., transporter genes SLC6A4, SLC6A3, and SLC1A1), as well as related genes

(i.e., catechol-O-methyltransferase (COMT) and monoamine oxidase A (MAOA)), genes involved in neurodevelopment (i.e., brain-derived neurotrophic factor (BDNF)), and genes involved in immune response (i.e., myelin oligodendrocyte glycoprotein (MOG)) (Zai et al.,

2004; Grados, 2010; Taylor, 2013). Serotonin system genes have been among the most extensively studied in OCD. In this article, we will review the most commonly examined serotonin gene variants in OCD, found in the genes that code for SERT and the 5-HT2A, 5-

HT1B, and 5-HT2C receptors. We adopted the approach used in the most comprehensive OCD genetic meta-analysis conducted to date (Taylor, 2013), which included 20 polymorphisms studied in at least 5 datasets. This comprised candidate gene variants from 179 potential studies, 113 of which were appropriate for meta-analysis, and with the number of OCD cases or probands ranging from 9 to 459 (Taylor, 2013). The effects for all 20 candidate polymorphisms included were small (Taylor, 2013), suggesting that OCD results from multiple genes of small effect. When combining this data with the meta-analysis conducted by the same group in a larger and more recent study replicating the significant findings in

OCD, mean odds ratios for the serotonin system gene polymorphisms ranged from 0.96 to

1.25 (Taylor, 2013; Taylor, 2016) (Table 2).

20 Table 2. Summary of serotonin gene findings: A summary compiling the most recent and comprehensive findings for serotonin system gene polymorphisms (using data from the main meta-analysis conducted by Taylor (2013) and from a newer meta-analysis conducted by Taylor (2016) replicating the results for 5-HTTLPR (coded as LA vs [LG+S]) and HTR2A). Significant findings are in bold.

Gene Polymorphism Number Mean OR P-value for of Studies Mean OR SLC6A4 5-HTTLPR (coded as L vs S) 34 1.05 0.266

SLC6A4 5-HTTLPR (coded as LA vs [LG+S]) 10 1.25 0.006 SLC6A4 STin2 VNTR (12 vs non-12 repeats) 11 1.05 0.554 HTR2A rs6311 (A vs G) or rs6313 (T vs C) 22 1.24 <0.001 HTR1B rs6296 (G vs C) 11 1.14 0.232 HTR2C rs6318 (Cys vs Ser) 6 0.96 0.843

Abbreviations: SLC6A4, solute carrier family 6 member 4 (serotonin transporter gene); 5-HTTLPR, serotonin transporter-linked polymorphic region (in the promoter of SLC6A4); L, long variant of 5-HTTLPR; S, short variant of 5-HTTLPR; LA, long variant of 5-HTTLPR with A allele at SNP rs25531; LG, long variant of 5- HTTLPR with G allele at SNP rs25531; STin2 VNTR, variable number of tandem repeats polymorphism in intron 2 (of SLC6A4); HTR2A, 5-HT2A receptor gene; HTR1B, 5-HT1B receptor gene; HTR2C, 5-HT2C receptor gene; OR, odds ratio.

In this article, we review the polymorphisms summarized in Table 2, as well as an additional rare variant in SERT that has also been the subject of multiple previous studies.

1.10. The serotonin transporter gene (SLC6A4)

In humans, SLC6A4 (solute carrier family 6 member 4) maps to chromosome 17q11.2, is approximately 40 kb in length, and is composed of 15 exons (Lesch et al., 1994; Murphy et al.,

2008). This gene gives rise to a 630 amino acid protein with 12 transmembrane domains

(Lesch et al., 1993; Murphy et al., 2008). The serotonin transporter gene is highly conserved throughout evolution, which underlines its significance. Homologs have been identified in over ten species, including non-human primates, rodents, Caenorhabditis elegans, and

Drosophila melanogaster (Murphy et al., 2004; Norton et al., 2008). Repetitive sequences

21 across the homologous serotonin transporter genes differ, but specific variants are conserved across different species (Soeby et al., 2005). Several variants affecting SERT expression and function have been identified in SLC6A4. Two of the most commonly studied polymorphisms, as well as additional variants examined, are described below.

1.10.1. The serotonin transporter-linked polymorphic region (5-HTTLPR)

SLC6A4 contains flanking regulatory regions composed of sequences of tandem repeats that serve to regulate transcription. Specifically, polymorphisms in the promoter region are of great interest because of their observed effect on gene expression (Heils et al., 1996; Lesch et al., 1996; Hu et al, 2006; Murphy et al., 2008). 5-HTTLPR is a frequently investigated polymorphic region within the promoter of SLC6A4. The major variants explored include the short (S) variant, composed of 14 sets of GC-rich 20-23 bp tandem repeats, and the long (L) variant, composed of 16 sets of these GC-rich 20-23 bp repeats. Both in vivo and in vitro studies have shown that the L variant (which has two additional sets of repeats) results in increased expression of SERT mRNA and more membrane-bound SERT, while the S variant results in decreased expression of SERT mRNA and less membrane-bound SERT (Heils et al.,

1996; Lesch et al., 1996, Little et al., 1998; Greenberg et al., 1999; Heinz et al., 2000; Hu et al., 2006).

The effect of the L variant on SERT expression varies based on an A/G single nucleotide polymorphism (SNP), rs25531, found within the first of the two additional sets of repeats in the L variant. The presence of this SNP means that there are 3 possible variations when considering L versus S variation and rs25531: L variant with an A allele (LA), L variant with a

22 G allele (LG), and S variant. The LG variant yields SERT expression that is nearly equivalent to the S variant, as it acquires a new AP2 transcription factor binding site which hinders transcription (Hu et al., 2006). Although 5-HTTLPR is triallelic (LA, LG, S), it is essentially biallelic in function because the LG and S variants result in similarly lower levels of SERT expression. Thus, LA results in higher mRNA expression, increased membrane-bound SERT, and increased SERT-mediated serotonin reuptake, while both LG and S result in lower mRNA expression, decreased membrane-bound SERT, and decreased SERT-mediated serotonin reuptake (Heils et al., 1996; Lesch et al., 1996; Hu et al., 2006; Bloch et al., 2008b). The patterns of SERT expression also suggest codominance of the above 5-HTTLPR variants

(whereby the alleles are additive in their effect). Genotype LA/LA yields greater SERT expression than genotype S/S, whereas LA, in combination with LG or S, yields an intermediate expression level (because LG expresses like the S variant) (Hu et al., 2006). Of note, two versions of S have also been identified, SA and SG, but SG is very rare in comparison to LA, LG, and SA. Most studies accounting for the added variance in 5-HTTLPR are referring to SA when they state “S” (Wendland et al., 2007; Voyiaziakis et al., 2011; Taylor, 2013). Another

5-HTTLPR SNP has been identified: intrinsic C/T SNP rs25532, located in the third-last repetitive element of the L and S variants (Wendland et al., 2008b). This SNP also contributes to SERT expression but is not as well characterized as rs25531.

The S and L variants, with 14 and 16 sets of repeats respectively, are not the only variants found in 5-HTTLPR. Additional 5-HTTLPR variants, including 15, 18, 19, 20, and 22 sets of repeats have been identified. These variants are less frequently observed (Heils et al., 1996;

Nakamura et al., 2000), and their exact SNP composition and effects on SERT expression warrant further exploration.

1.10.2. Variable number of tandem repeats polymorphism in intron 2 (STin2 VNTR) and other SLC6A4 variants

Another frequently investigated polymorphism is the STin2 VNTR residing in intron 2 of

SLC6A4 (Ogilvie et al., 1996). The most commonly identified variants of this polymorphism include those with 9, 10 and 12 repeats of a 16-17bp element. More repeats translate into increased SLC6A4 transcription, as shown by both in vitro and in vivo studies (Ogilvie et al.,

1996, Haddley et al., 2008; Murphy & Moya, 2011). Additionally, rare coding region variants in SLC6A4 have been reported including I425V, I425L, and G56A. These variants result in increased serotonin uptake (Kilic et al., 2003; Murphy et al., 2008; Murphy & Lesch, 2008).

Taken together, different combinations of variants identified within SLC6A4 may constitute up to a 20-fold difference in SERT expression and function (Murphy et al., 2008; Murphy &

Lesch, 2008). Data for the effect of serotonin receptor gene polymorphisms on gene expression is much more limited compared with the extensive studies of the effects of 5-

HTTLPR and other SLC6A4 variants.

1.11. SLC6A4: Genetic association with OCD

1.11.1. 5-HTTLPR

5-HTTLPR is a key candidate gene region in the study of OCD, though findings have been inconsistent. Several meta-analyses of the L/S polymorphism in OCD have been conducted.

Some report either mixed findings (Dickel et al., 2007; Lin, 2007; Taylor, 2013; Walitza et al.,

24 2014), or no association between the L/S polymorphism and OCD (Bloch et al., 2008b; Mak et al., 2015). A possible explanation for the inconclusive findings is the failure of some of the earlier studies to consider the rs25531 A/G SNP and to properly account for the altered expression induced by the LG variant in their analyses. Three meta-analyses divided the L/S polymorphism based on functionality, and therefore compared LA to [LG+S] in their analyses.

All three reported that the LA variant was significantly associated with OCD, one meta- analysis of which was conducted by the same author and replicated the same finding in a larger, more recent dataset (Taylor, 2013; Walitza et al., 2014; Taylor, 2016).

Another possible explanation for the mixed results is that the association between 5-HTTLPR and OCD may vary based on age, sex, ethnicity, and study design (i.e., case-control versus family-based study). For example, some of the meta-analyses only reported associations after further stratification. One reported that the L variant was significantly associated with OCD in children, in Caucasians, and in family-based studies (Bloch et al., 2008b). Another, which designated the LG variant as S only where provided, reported a trending association between the S variant and OCD in females (Mak et al., 2015). Though still largely inconclusive, studies indicate the importance of taking LG functionality into account, as well as stratifying participants into homogeneous subgroups when conducting analyses because different subgroups may have slightly different genetic underpinnings.

1.11.2. STin2 VNTR and other SLC6A4 variants

In a recent meta-analysis comparing 12 to non-12-repeat STin2 VNTR variants, the STin2

VNTR was not significantly associated with OCD (Taylor, 2013), despite individual studies

25 from the meta-analysis (conducted in varying ethnic groups) reporting a significantly higher proportion of the 12-repeat STin2 VNTR variant in OCD patients versus healthy controls

(Ohara et al., 1999; Baca-Garcia et al., 2007; Saiz et al., 2008). In a large, family-based study also included in the meta-analysis, genotype 9-repeat/10-repeat of STin2 VNTR was significantly associated with OCD only in females (Voyiaziakis et al., 2011), suggesting that this polymorphism may be sex-specific.

The rare I425V SLC6A4 variant has been reported in two unrelated families whereby six of seven family members with the mutation had OCD or obsessive-compulsive personality disorder (OCPD), many also with several comorbid disorders. The most severely affected individuals carried both the rare I425V mutation, which increases SERT transcription, and the

L variant of 5-HTTLPR, which was also understood to increase SERT transcription before the effects of LG were known (Ozaki et al, 2003). The I425V variant has also been found significantly more frequently in OCD patients versus controls, but only as ascertained in a study reviewing cumulative data from 5 unrelated families (Wendland et al., 2008a). A more recent, large-scale study also identified three individuals with I425V. Two of them were affected with either OCD or OCPD, and all three of them had offspring affected with OCD but did not pass the variant onto their affected offspring (Voyiaziakis et al., 2011). Evidence suggests that the I425V variant may be associated with a rare and familial OCD phenotype and may act in conjunction with other variants or risk factors to yield a more severe phenotype. It appears that I425V does not always co-segregate with OCD, however. More research is therefore required to understand the role of this variant in families and the combined effect of this variant with other candidate gene variants in OCD.

26 1.12. Serotonin 2A receptor gene (HTR2A): Genetic association with OCD

HTR2A encodes the 5-HT2A receptor and is located on chromosome 13q14 (Nichols &

Nichols, 2008). In OCD, the most studied HTR2A polymorphisms are -1438G/A (rs6311, promoter variant) and 102C/T (rs6313) (Taylor, 2013). These two polymorphisms are in strong linkage disequilibrium, whereby the 102C allele almost exclusively occurs with the -

1438G allele, as does the 102T allele with the -1438A allele (Spurlock et al., 1998; Arnold et al., 2004).

In a recent meta-analysis, OCD was significantly associated with the A allele of rs6311 or the linked T allele of rs6313 (Taylor, 2013). This finding was replicated in a more recent meta- analysis conducted by the same author in a larger dataset (Taylor, 2016). Results from older, individual studies testing the association between the rs6311/rs6313 polymorphisms and OCD vary in their findings (Enoch et al., 1998; Frisch et al., 2000; Enoch et al., 2001; Walitza et al.,

2002; Hemmings et al., 2003; Tot et al., 2003; Lochner et al., 2004; Meira-Lima et al., 2004;

Denys et al., 2006; Dickel et al., 2007; Liu et al, 2011). Similar to other variants, findings seem to indicate that rs6311/rs6313 is associated with specific OCD subgroups reflecting sex, age of onset, and presence of tic disorder comorbidity. For example, the A allele of rs6311 showed a nominally significant association with OCD with comorbid tic disorder (Dickel et al., 2007). As another example, a family-based study noted a significant association between the -1438G/A polymorphism and OCD. Upon stratifying further, they found that males with

OCD and individuals with late-onset OCD were driving the finding (Liu et al., 2011).

Similarly, we see evidence suggesting that rs6311/rs6313, or a closely linked variant, may play a role in the biology of specific behavioral traits that are shared across related disorders.

27 For example, the -1438A allele was more commonly found in patients with OCD relative to controls and in patients with anorexia relative to controls, suggesting the involvement of this variant in behavioral traits like perfectionism and obsessionality (Enoch et al., 1998). Overall, individual small studies have found interesting associations with one or another homogenous subgroup. Replication of these association results for variants of rs6311/rs6313 and homogenous subgroups or behavioral traits is required since results may be due to chance.

1.13. Serotonin 1B receptor gene (HTR1B): Genetic association with OCD

HTR1B encodes the 5-HT1B receptor and is located on chromosome 6q13 (Nichols & Nichols,

2008). The polymorphism in HTR1B most tested for association with OCD is 861G/C

(rs6296) (Taylor, 2013).

A recent meta-analysis found no significant association between the rs6296 polymorphism and

OCD, although there was a trend indicating that the effect of this polymorphism is larger in adults than in children and adolescents (Taylor, 2013). Additionally, a group of studies suggested that 861G/C may be associated specifically with severity of OCD (a quantitative trait) or with only a homogeneous subgroup of patients (Mundo et. al 2002; Camarena et al.,

2004; Liu et al., 2011). Two of these studies either revealed a trend or provided preliminary evidence that showed higher Y-BOCS obsession scores in G allele-carriers (Mundo et al.,

2002; Camarena et al., 2004), while another noted a significant association between 861G/C and early-onset OCD, as well as between 861G/C and males with OCD (Liu et al., 2011).

Although evidence from the meta-analysis suggests that the HTR1B polymorphism may not be associated with OCD as a whole, it did note an age-specific trend. Findings from the

28 individual studies listed here also suggest that the polymorphism may be more specifically associated with one or more homogenous subgroups reflective of age of onset, sex, and/or

OCD severity.

1.14. Serotonin 2C receptor gene (HTR2C): Genetic association with OCD

HTR2C encodes the 5-HT2C receptor and is located on chromosome Xq24 (Nichols & Nichols,

2008). The polymorphism (rs6318) yields an amino acid substitution that results in one of two variants: Cys23 or Ser23 (Lappalainen et al., 1995). The original suspected link between the

5-HT2C receptor and OCD stemmed from the ability for meta-chlorophenylpiperazine (mCPP), a 5-HT2C receptor agonist, to transiently exacerbate OCD symptoms in OCD patients (Zohar et al., 1987). In animal models, modulating 5-HT2C function induces behaviors that resemble aspects of OCD. For example, administering mCPP to rats increases grooming (Bagdy et al.,

1992) and 5-HT2C knockout mice display compulsive-like behaviors as measured through non- nutritive chewing and head dipping patterns (Chou-Green et al., 2003; Murphy et al., 2013).

A recent meta-analysis found no significant association between OCD and either variant of

Cys23Ser, but did report a trend indicating that the effect of this polymorphism was larger in children and adolescents than in adults (Taylor, 2013). Unlike for other serotonin genes, this polymorphism does not seem to be associated with subgroups of OCD based on sex, tic disorder comorbidity, or family history (Cavallini et al., 1998). Overall, there is very limited evidence to suggest the involvement of the 5-HT2C receptor in OCD. Although this receptor is located on the X chromosome, there are no sex-specific effects of this polymorphism in OCD.

29 1.15. Pharmacogenetics

Few studies have examined the relationship between the serotonin system gene polymorphisms most commonly studied in OCD and treatment response to SRIs. Most of these studies were reviewed in an article by Brandl and colleagues (2012).

In a study using trios to test for association between OCD and 5-HTTLPR L versus S variants, the L variant was overtransmitted to SRI drug nonresponders, while no difference in transmission of either variant was reported in SRI drug responders (McDougle et al., 1998).

Three studies found no association between the 5-HTTLPR polymorphism and treatment response to SRIs (Billett et al., 1997; Zhang et al., 2004; Miguita et al., 2011). One study of

OCD found no significant association between 5-HTTLPR genotypes L/L, L/S, and S/S and treatment response to the SSRI fluvoxamine, measured using Y-BOCS total scores. When considering Y-BOCS obsessions and compulsions subtotal scores separately, however, a significant interaction between fluvoxamine treatment time and genotype was found when using the compulsions subtotal. Also, when examining only OCD patients without comorbid tic disorder, a significant interaction between time and genotype was found using patterns of change in total Y-BOCS score over time, during treatment with fluvoxamine. Upon stratifying further, this effect remained significant only when using Y-BOCS compulsions subtotal score. These results should be interpreted with caution given the degree of stratification (Di Bella et al., 2002). Also, the majority of pharmacogenetics studies did not consider the LG variant and its altered functionality. The only study to examine the association of STin2 VNTR in SLC6A4 and treatment response to SRIs (clomipramine in this case) in OCD was negative (Miguita et al., 2011).

In a study of OCD patients treated with the SSRI paroxetine, or with the SNRI venlafaxine, a significantly higher frequency of heterozygous 5-HTTLPR genotype L/S was found in responders versus nonresponders, only in the group treated with venlafaxine. When examining the -1438G/A polymorphism in HTR2A, this same study reported a significantly higher frequency of the homozygous genotype G/G in responders versus nonresponders only in the group treated with paroxetine. These findings remained significant after correction for multiple testing and may be highlighting possible differences in SRIs and their mechanisms of action. Of particular interest, patients with the combined G/G genotype in -1438G/A and the

L/S genotype in 5-HTTLPR all responded to treatment with either SRI, indicative of a possible additive genetic effect (Denys et al., 2007). Another study identified an association between homozygosity of the -1438G/A polymorphism and SRI treatment response in OCD patients

(Zhang et al., 2004), however two other studies of the -1438G/A polymorphism and/or 102C/T linked variant found no association with SRI treatment response in OCD (Tot et al., 2003;

Miguita et al., 2011).

The few studies examining other candidate serotonin genes for association with treatment response to SRIs in OCD have been negative, including three studies of the 861G/C polymorphism in HTR1B (Denys et al., 2007; Miguita et al., 2011; Corregiari et al., 2012), and one study of the Cys23Ser variant in HTR2C and clomipramine response (Cavallini et al.,

1998). Overall, pharmacogenetic studies to date have been limited by small sample sizes and have been conducted in different ethnic groups, which should be taken into account. Subtypes of OCD (i.e., based on gender, age of onset, presence of comorbid tics) should also be examined in future pharmacogenetic studies. A recent genome-wide association study

31 (GWAS) showed no compelling evidence that any one variant in the serotonergic neurotransmission pathway was associated with SRI treatment response. Instead an enrichment analysis, conducted as part of the study, indicated that multiple genes in the serotoninergic neurotransmission pathway including SLC6A4 and HTR2A (as well as multiple genes in the glutamatergic neurotransmission pathway) may together contribute to SRI treatment response in OCD, which may explain some of the past null findings with respect to individual variants (Qin et al., 2016).

1.16. Intermediate phenotypes: Neuroimaging

The mixed findings in the genetic studies of OCD, as highlighted above, may be due to the heterogeneous nature of the disorder. The current consensus is that OCD represents a continuous spectrum of possibly overlapping symptom dimensions, which extend beyond traditional boundaries of the disorder, each with a wide range of severity (Mataix-Cols et al.,

2005; Miguel et al., 2005; Leckman et al., 2007). Factor analyses of the Y-BOCS of patients with OCD have identified between 3 and 5 symptom dimensions (Goodman et al., 1989a;

Goodman et al., 1989b; Cullen et al., 2007; Pinto et al., 2008). This dimensional approach is consistent with the Research Domain Criteria (RDoC) approach (National Institute of Mental

Health [NIMH], 2008) which aims to guide research in clinical neuroscience and genomics by integrating observable, quantitative, behavioral dimensions with scientific evidence of dysfunctional neurocircuitry based on biologically salient “domains” measured using a range of methods (i.e., neuroimaging, neuropsychological tasks) (Insel et al., 2010, Casey et al.,

2014). Moving away from using the traditional diagnostic case/control design and towards dimensions or intermediate phenotypes that are closer to the genetic basis of the disorder may

32 help cut through heterogeneity and in turn facilitate gene discovery in OCD (Murphy et al.,

2008).

Brain structure and function are intermediate phenotypes that could be particularly useful in the study of OCD genetics (Gottesman & Gould, 2003; Miguel et al., 2005; Murphy et al.,

2008, Arnold et al., 2009b; Nicolini et al., 2009). Neuroimaging findings consistently point to dysfunctional cortico-striato-thalamo-cortical (CSTC) circuitry in OCD, with the striatum

(caudate and putamen), thalamus, anterior cingulate cortex (ACC), and orbitofrontal cortex

(OFC) being the regions most consistently associated with OCD (MacMaster, 2010; Pauls et al., 2014). Very few studies have examined brain imaging and OCD in the context of serotonin system gene variants (Atmaca et al., 2011; Hesse et al., 2011). One study used magnetic resonance imaging (MRI) to compare OCD patients to healthy controls and showed a significant 5-HTTLPR genotype-diagnosis interaction on OFC volume. Furthermore, in

OCD patients (but not in healthy controls) individuals carrying the 5-HTTLPR S variant had significantly reduced OFC volume relative to individuals with genotype L/L (Atmaca et al.,

2011). The second study used positron emission tomography (PET) and a SERT-selective radiotracer. When comparing 5-HTTLPR in cases to controls, there was no significant effect of the interaction between presence of the disorder and 5-HTTLPR genotype on SERT availability, but when examining OCD patients alone, there was an overall trend showing increased SERT availability in patients with 5-HTTLPR genotype S/S. This trend was apparent in all examined brain regions, reaching significance in the midbrain. When examining only OCD patients, STin2 VNTR genotype had a significant effect on SERT availability in the ACC and putamen, and the interaction between 5-HTTLPR and STin2 VNTR genotype had a significant effect on SERT availability in the ACC, putamen, and

33 hypothalamus (Hesse et al., 2011). From these studies, we conclude that serotonin gene variants may be related to brain changes specifically in OCD, but more studies are required given the small number of studies conducted to date, given that the studies require replication, and given that the studies conducted were exploratory in nature and did not correct for multiple comparisons.

1.17. Discussion

Much research has focused on the role of serotonin system genes in the etiology of OCD. As described in this review, the evidence on the association of genetic variants in SLC6A4,

HTR2A, HTR1B, and HTR2C is mixed with many contradictory findings. These inconsistencies likely result from both phenotypic heterogeneity and genetic complexity.

1.17.1. Genetic complexity of OCD

As with most common psychiatric and medical disorders, OCD is hypothesized to be genetically complex and the result of multiple genes/variants of small effect (Plomin et al.,

2009; Taylor, 2013). Furthermore, there are likely a number of factors contributing to this complexity, including genetic and allelic heterogeneity (different genes or variants yielding similar disease characteristics), epistasis (interaction between genes), pleiotropy (one gene influencing multiple phenotypic traits), epigenetics (i.e., DNA methylation patterns), and gene-environment (GxE) interaction (Arnold & Richter, 2007; Grados, 2010). Larger samples will be critical to acquiring sufficient power to parse out these mechanisms driving genetic complexity, some of which are discussed in the following paragraphs.

The expression of genes can be modified by environmental factors, as has been demonstrated for variants within SLC6A4. In a seminal paper by Caspi et al., participants exposed to stressful life events in early adulthood that carried the S variant were more likely than those with the genotype L/L to experience depression later in life (Caspi et al., 2003). This was one of the first studies to provide evidence that stressful environmental events differentially affect individuals based on their genetics. To further explain this complexity, a number of investigators have proposed an alternative to the classic “stress-diathesis” hypothesis, in which the effect of certain genetic variants (notably including 5-HTTLPR) is to confer differential

“sensitivity” or “susceptibility” to environmental influences. This hypothesized model, known by different terms including “differential susceptibility to the environment” (Boyce & Ellis,

2005) or “phenotypic capacitance” (Conley et al., 2013) has been illustrated using an analogy to orchids versus dandelions. The “orchid” allele/variant makes its carrier more sensitive to the environment, whereby it confers an advantage under ideal and nurturing conditions and a disadvantage under detrimental conditions. On the other hand, the “dandelion” allele/variant makes its carrier less responsive to the environment, which allows the individual to adequately handle both ideal and detrimental environmental circumstances (Conley et al., 2013). A number of studies in depression show the S variant to be orchid-like, as S-carriers perform better than normal under supportive conditions and worse under unfavourable conditions

(Conley et al., 2013). Conversely, the L variant confers an effect that is more stable, irrespective of environmental conditions. Conley et al. then go on to detail the mechanism driving this effect, hypothesizing in the case of 5-HTTLPR that because the L variant yields higher SERT mRNA and protein levels, it functions as a capacitor to dull the effects of environmental variation. The low SERT expression level of the S variant would mean that it

35 does not have the capacity to confer such stability in the carrier in the face of environmental variation (Conley et al., 2013). Higher SERT expression is thought to mediate the influence of

5-HTTLPR on phenotypic variation by affecting neuronal functions like synaptic plasticity

(Niven, 2004). Although requiring further testing, the differential susceptibility model may prove to be a useful model for the study of GxE interactions in psychiatric genetics in general; with a recent meta-analysis reaffirming that the S variant is associated with increased risk of developing depression when combined with life stress (Karg et al., 2011). There is very limited data on GxE interactions in OCD and no published reports in OCD that focus on environmental interactions with serotonin gene variants.

Epigenetics is a mechanism by which environmental influences on the genome may be mediated, through effects on gene expression. Epigenetic mechanisms include DNA methylation and histone modification (chromatin remodeling), and RNA-based regulation

(Gibney & Nolan, 2010). There is evidence, for example, that pre and postnatal environmental circumstances can directly affect methylation status in the genome, influence neurodevelopment, and impact future psychiatric state (Kundakovic & Champagne, 2015).

Studies show that different 5-HTTLPR variants yield different levels of SLC6A4 gene expression in response to different types of stressors like early life stress or recent stress; however, to date, there are limited findings to suggest that DNA methylation is a mechanism responsible for modifying SERT expression (Wankerl et al., 2014; Duman & Canli, 2015). A study by Duman and Canli (2015) conducted in healthy males showed that S-carriers had increased SLC6A4 CpG island methylation in response to stress, with the site of methylation specific to the type of stressor, but it still remains unclear whether or not changes in

36 methylation patterns are driving changes in SLC6A4 gene expression (Wankerl et al., 2014;

Duman & Canli, 2015).

Two epigenetic studies have been conducted in OCD. One examined the DNA methylation profiles (using preselected CpG sites) of 14 candidate genes, including SLC6A4. These investigators looked for an association with OCD diagnosis, severity, and/or treatment response in 21 female children and adolescents and 12 female age- and sex-matched controls, using blood spot samples taken from these children at birth and again near the time of diagnosis. There were no significant findings pertaining to SLC6A4 and OCD (Nissen et al.,

2016). Another group conducted a genome-wide DNA methylation analysis in 65 patients with OCD and 96 healthy control subjects. Although not among their most strongly associated findings, HTR2C was noted to contain more than 3 probes that were differentially methylated in OCD patients relative to controls (Yue et al., 2016). Results for both of these studies should be interpreted with caution given their small sample sizes. Overall, further studies examining GxE effects and epigenetic mechanisms are warranted, including studies focused on serotonin system genes.

1.17.2. Animal models of OCD: Genetic mouse models of obsessive-compulsive-like behaviors

As reviewed by Albelda & Joel (2012), the two most promising genetic models of OCD are the Sapap3 and the Slitrk5 knockout mouse models. The Sapap3 gene in the mouse, referred to as DLGAP3 in humans, encodes a postsynaptic scaffolding protein present at excitatory cortico-striatal synapses within the brain. It is highly expressed in the striatum. The Sapap3

37 knockout mouse, developed by Welch and colleagues (2007), reveals structural defects in the formation of this postsynaptic complex, within the striatum, as well as abnormalities within the glutamatergic system. Respective mice show excessive self-grooming and increased anxiety-like behaviors which are reduced following pharmacological treatment with SSRI, fluoxetine (Welch et al., 2007). The Slitrk5 gene in the mouse encodes a neuron-specific transmembrane protein. The Slitrk5 knockout mouse, developed by Shmelkov and colleagues

(2010), exhibits increased expression of transcription factor FosB within the OFC, as well as structural abnormalities within the striatum. Respective mice show increased self-grooming and increased anxiety, as well as increased marble burying behavior. As is the case with the

Sapap3 knockout mouse, behaviors show improvement following pharmacological treatment with SSRI, fluoxetine (Shmelkov et al., 2010). Though an explicit “OCD mouse” does not exist, these models demonstrate behavioral aspects that resemble obsessive or compulsive traits seen in the disorder. Most importantly, the above knockout models reveal a resulting disruption in glutamatergic functioning and an amelioration of symptoms via SSRI treatment.

This supports the involvement of the glutamatergic and serotonergic systems in OCD behaviors and suggests a dynamic relationship between these two systems (Albelda & Joel,

2012; Murphy et al., 2013; Pauls et al., 2014).

Additional genetic mouse models have related mainly dopaminergic, serotonergic, and glutamatergic genes to such OCD-related behaviors as increased anxiety and excessive self- grooming. These genetic models include the dopamine transporter (DAT) knockdown mouse,

D1CT-7 transgenic mouse, 5-HT2C knockout mouse, Hoxb2 mutant mouse, aromatase knockout mouse, and SLC1A1 knockout mouse, though none of the studies implicating these models report response to conventional pharmacological treatments used in OCD (Albelda &

38 Joel, 2012; Murphy et al., 2013; Pauls et al., 2014). The study by Fox et al., however, reveals that serotonergic manipulation alters DAT knockout mouse behavior (Fox et al., 2013). This, again, supports the concept of a dynamic relationship between candidate gene systems in

OCD.

1.17.3. Phenotypic heterogeneity of OCD

OCD is also phenotypically heterogeneous. Several OCD subtypes have been proposed as a function of their accompanying demographic, clinical, and genetic features. These include

OCD with comorbid tic disorder (now reflected in the addition of a “tic-related” specifier for

OCD in the DSM-5), early-onset OCD, and acute-onset OCD (PANS or pediatric acute-onset neuropsychiatric syndrome) which may be associated with streptococcal infections (PANDAS or pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections)

(Swedo et al., 1998; Grados, 2010; Leckman et al., 2010; Taylor, 2011; Swedo et al., 2012;

APA, 2013; Williams et al., 2013). A recent cross-disorder GWAS examined the genetic relationship between tic disorders and OCD and suggested that OCD with comorbid TS or chronic tics has different and distinct genetic etiology compared to OCD alone (Yu et al.,

2015). As another example of the distinct nature of these subtypes, compared to late-onset

OCD, early-onset OCD is 1) more common in males than females, 2) more heritable, 3) likely to present more compulsions without obsessions, 4) more likely to present with comorbid tics and TS, and 5) associated with poorer response to SRIs (Lensi et al., 1996; do Rosario-

Campos et al., 2001; Miguel et al., 2005; Arnold & Richter, 2007).

39 As noted above, factor analyses of the Y-BOCS symptom checklist of patients with OCD have identified distinct symptom dimensions. The most recent meta-analysis of these factor analyses reported four symptom dimensions: 1) symmetry, 2) forbidden thoughts, 3) cleaning, and 4) hoarding (Bloch et al., 2008a). In addition to specific symptom dimensions within

OCD, it is now well established that several disorders share many phenotypic features with

OCD and are recognized under the heading “Obsessive-Compulsive and Related Disorders” in the DSM-5. These disorders include BDD, trichotillomania (hair-pulling disorder), excoriation (skin-picking) disorder, and hoarding disorder (APA, 2013). Most importantly, hoarding was reclassified from an OCD symptom to a distinct diagnosis because of consistent, biological and clinical evidence showing that hoarding is separate from, though related to,

OCD (Mataix-Cols & Pertusa, 2012). Hoarding differences include variations in heritability relative to other OCD symptom dimensions (Iervolino et al., 2011; Burton et al., 2018), and some genetic and neuroimaging differences between hoarding and OCD (van Ameringen et al., 2014) including evidence of disorder-specific genetic influences for hoarding disorder

(Monzani et al., 2014). Hoarding symptoms differ in their response to SRIs, whereby OCD patients with hoarding symptoms show a poorer response to SRIs than do OCD patients without hoarding (Murphy et al., 2004; Bloch et al., 2014).

One implication of the heterogeneous clinical presentation of OCD is that the different subgroups described, and those reflecting other contributing factors (age of onset, sex, etc.), may have somewhat different etiologies. Previous studies have already suggested that there may be different genetic mechanisms driving the various OC symptoms or behaviors (Grados,

2010). Results from OCD candidate gene meta-analyses also suggest that the effects of gene variants may vary by several phenotypic factors. Moderator analyses were conducted in the

40 Taylor (2013) meta-analysis to assess the effect of study design (case-control versus family- based), sex, race (Asian versus Caucasian), and age group (child and adolescent versus adult) on the results. There was no significant effect of any one of these variables for serotonin system gene variants, but there were trends suggesting that age may moderate the association between HTR1B and OCD and between HTR2C and OCD (Taylor, 2013). Despite the lack of significant subgroup effects in this meta-analysis, several individual studies showed significant gene associations when subjects were stratified into homogenous subgroups and analyzed separately. To reduce heterogeneity, future studies in larger sample sizes could focus on these individual symptom dimensions or separately study OCD subtypes or obsessive-compulsive related disorders (Figure 2). Potential differences due to key biological variables in OCD, such as sex, age of onset, and ethnicity, must also be taken into account in genetic analysis. Furthermore, as described above, the use of intermediate phenotypes may reduce heterogeneity, thereby potentially increasing power for gene discovery and also clarifying mechanisms by which specific genetic variants are involved in pathogenesis.

Figure 2. OCD as a complex disorder: Diagram summarizing some of the genetic complexity and phenotypic heterogeneity to consider when studying OCD and/or related disorders. OCD, obsessive-compulsive disorder; COMT, catechol-O-methyltransferase; MAOA, monoamine oxidase A; BDNF brain-derived neurotrophic factor; MOG, myelin oligodendrocyte glycoprotein; GxE, gene-environment; BDD, body dysmorphic disorder.

1.17.4. General conclusion

Overall, evidence generally supports that serotonin system genes play a role in OCD, but the exact genetic variants involved are still unclear. The strongest evidence from meta-analyses

42 suggest that the LA variant in 5-HTTLPR and the A allele of rs6311 or the linked T allele of rs6313 in HTR2A are the serotonin genes most relevant to OCD pathogenesis (Taylor, 2013;

Taylor, 2016). Only limited evidence supports the association between rs6296 in HTR1B and

OCD as a whole or between rs6318 in HTR2C and OCD (Taylor, 2013). Recent GWAS results from two groups showed no significant associations between SNPs in serotonin genes and OCD (Stewart et al., 2013; Mattheisen et al., 2015). In addition, there were no significant associations found between serotonin genes and OCD in gene-level analyses conducted

(Mattheisen et al., 2015), nor was there evidence of SNPs within previously identified OCD candidate genes reaching genome-wide significance or being among the top “hits” in the association analyses (Stewart et al., 2013). It is important to note, though, that variation in 5-

HTTLPR is not captured using GWAS arrays. Next generation sequencing may help reveal genome-wide variants not currently captured via GWAS that may be relevant to OCD.

Future research should stratify into homogenous subgroups reflecting symptom dimensions, subtypes, and demographics (where possible based on sample size) to help clarify if different genes are involved in different OCD subgroups and to help explain past inconsistencies or lack of significant results. When considering potential genetic associations with a trait, it is also important to be mindful of the possible influence of neurodevelopmental stage (Shonkoff &

Garner, 2012). Increasing availability of cost-effective, genome-wide technological approaches, including genome-wide association arrays and next generation sequencing, will enable us to advance beyond candidate gene analyses. This will help us better understand the full extent of genetic variation and functional effects of genes both within and outside of the serotonin system in OCD. Serotonin interacts closely with several neurotransmitter systems.

For example, evidence indicates that serotonin can influence dopaminergic activity directly via

43 different serotonin receptors on dopamine neurons or indirectly via the GABA-ergic and glutamatergic systems (Di Giovanni et al., 2008). As larger samples are accumulated on OCD patients with genome-wide data, it will become possible to examine complex interactions between genes using pathway analyses, moving beyond the effects of single genes or candidate gene systems such as the serotonin system. Elucidation of the genetic risk factors underlying OCD, including the specific role of serotonin genes, will inform future treatment studies and enable precision medicine in which treatments target an individual’s genotypic and phenotypic characteristics.

44 2. AIMS AND HYPOTHESES

2.1. Rationale and general aims

OCD is a complex disorder with multiple levels of phenotypic heterogeneity and genetic complexity that make it difficult to decipher its underlying causes. Phenotypic heterogeneity manifests itself through the various symptoms presented. Symptoms commonly group into dimensions reflecting symmetry, forbidden thoughts, cleaning, and hoarding. Furthermore, there are sex and age of onset-related subgroups within OCD that may differ in their etiology and which should be accounted for in genetic studies. Genetic complexity is seen through the multiple genes and gene systems thought to be implicated in the disorder. Some candidate gene systems include serotonergic genes, dopaminergic genes, and glutamatergic genes.

Moreover, these genes can interact differently with one another and with the environment to yield obsessive-compulsive symptoms.

Serotonin system genes have been most extensively studied in OCD because primary pharmacological treatment for the disorder, which was discovered serendipitously, mainly targets the serotonin system. Researchers have since strived to understand the respective mechanisms. To date, genetic studies in OCD have been inconsistent in their findings, possibly because of a failure to account for all of these potential factors driving OCD pathogenesis.

The thesis focuses specifically on candidate serotonin system genes and aims to address this multilayered complexity in OCD by reducing heterogeneity and investigating if specific gene

45 variants may be driving pathogenesis in different homogenous OCD subgroups. This approach can help account for previously inconclusive findings. The thesis also aims to address heterogeneity by studying putative intermediate phenotypes in OCD to help delineate potential mechanisms in the disorder, as they relate to serotonin system genes. Overall, the thesis will help us better understand the underlying complexity in OCD and serve to narrow the search for a complete understanding of its etiology, facilitate early identification of at-risk groups, and guide new and personalized therapeutics.

2.2. Specific aims

1.) To study the association between serotonergic genes and obsessive-compulsive (OC) trait dimensions in a general population of children and adolescents

2.) To study the association between serotonergic genes and hoarding, specifically, in a general population of children and adolescents

3.) To study the relationship between serotonergic genes and putative intermediate phenotypes in OCD in children and adolescents in the clinic

2.3. Hypotheses

1.) Based on previous evidence for association of serotoninergic genes with clinically diagnosed OCD, and on newer evidence that OCD is a heterogeneous entity best understood as several partially overlapping dimensions, we hypothesized that different serotonergic gene variants will be specifically associated with different OC trait dimensions in a general, pediatric population.

2.) Based on recent biological and clinical evidence suggesting that hoarding symptoms may be etiologically distinct from other OCD symptoms, and supporting hoarding’s new distinction as a separate obsessive-compulsive related disorder, we hypothesized that specific serotonin system gene variants would be uniquely associated with hoarding alone.

3.) Based on evidence that the serotonin system is implicated in OCD pathogenesis and based on evidence that CSTC circuitry is implicated in OCD pathogenesis, we hypothesized that

OCD diagnosis will have a significant effect on the relationship between specific serotonin system gene variants and distinct structural differences in CTSC-related brain regions.

2.4. Research study 1

2.4.1. Part A

The 5-HTTLPR polymorphism lies in the serotonin transporter gene (SLC6A4) and is not covered in GWAS array platforms. It has been implicated in previous studies of OCD and has been the most heavily studied polymorphism, to date, with respect to the disorder. We examined the association of 5-HTTLPR variants with OC traits overall, and with specific OC trait dimensions in a pediatric, population-based sample and in males and females separately.

5-HTTLPR contains a long/short (L/S) repetitive region and an intrinsic A/G SNP (rs25531) within 5-HTTLPR. This region was directly genotyped, in two consecutive steps, to identify variants LA, LG, and S.

47 2.4.2. Part B

Our interest in serotonergic candidate genes SLC6A4, HTR2A, and HTR1B was based on a priori evidence of their involvement in OCD. We examined the association of SNPs across these candidate genes with OC traits overall, and with specific OC trait dimensions in a pediatric, population-based sample and in males and females separately. SNP data for the genes was obtained using information from GWAS array platforms.

2.5. Research study 2

2.5.1. Part A

We examined the association between 5-HTTLPR variants and hoarding without other OC traits present, hoarding with other OC traits present, and OC traits without hoarding traits in a pediatric, population-based sample and in males and females separately.

2.5.2. Part B

We examined the association between SNPs across serotonergic candidate genes SLC6A4,

HTR2A, and HTR1B and hoarding without other OC traits present, hoarding with other OC traits present, and OC traits without hoarding traits in a pediatric, population-based sample and in males and females separately.

48 2.6. Research study 3

2.6.1. Part A

We examined the association of 5-HTTLPR variants and of additional SNPs across SLC6A4,

HTR2A, HTR1B, and HTR2C (added to our imaging study since the 5-HT2C receptor is widely distributed throughout the central nervous system and has been the subject of imaging and anatomic studies showing brain region-specific differences in expression, mRNA editing, and protein activity) with OCD in a clinical, pediatric sample and in males and females separately.

2.6.2. Part B

We assessed the effect of 5-HTTLPR variants or additional SNPs across SLC6A4, HTR2A,

HTR1B, and HTR2C on brain volume irrespective of OCD diagnosis, in a combined group of

OCD patients and healthy controls and in males and females separately. Our brain regions were selected based on a priori evidence of their involvement in OCD and were structures implicated in CSTC circuitry. Volumetric data was obtained via structural magnetic resonance imaging (sMRI) data collected in all cases and controls.

2.6.3. Part C

We assessed the effect of genotype-diagnosis interaction on brain volume. Using interaction model analyses, and in males and females separately, we examined whether the relationship between 5-HTTLPR variants or additional SNPs across SLC6A4, HTR2A, HTR1B, and HTR2C

49 and brain volume differed in OCD patients versus healthy controls. Therefore, we directly assessed whether the presence of an OCD diagnosis significantly changed the association between serotonin gene variant and brain region volume.

50 3. STUDY 1

Serotonin system genes and obsessive-compulsive trait dimensions in a population-based, pediatric sample: A genetic association study

This chapter is modified from the following article and contains the entire article contents, in full:

Copyright permission was obtained from journal publisher, Wiley, for full use of the article and its contents in this dissertation.

Authors

Vanessa M. Sinopolia,b, Lauren Erdmanb,c, Christie L. Burtonb,d, Laura S. Parka,b, Annie Dupuise,f, Janet Shand, Tara Goodaled, S-M Shaheeng, Jennifer Crosbied,h, Russell J. Schachara,d,h, Paul D. Arnoldb,g,i,*

Affiliations aInstitute of Medical Science, University of Toronto, Toronto, ON, Canada bProgram in Genetics & Genome Biology, The Hospital for Sick Children, Toronto, ON, Canada cDepartment of Computer Science, University of Toronto, Toronto, ON, Canada dProgram in Neurosciences and Mental Health, The Hospital for Sick Children, Toronto, ON, Canada eClinical Research Services, The Hospital for Sick Children, Toronto, ON, Canada fDalla Lana School of Public Health, University of Toronto, Toronto, ON, Canada gMathison Centre for Mental Health Research & Education, Hotchkiss Brain Institute, University of Calgary, Calgary, ON, Canada hDepartment of Psychiatry, University of Toronto, Toronto, ON, Canada iDepartments of Psychiatry and Medical Genetics, Cumming School of Medicine, University of Calgary, Calgary, ON, Canada * Corresponding Author: Paul Daniel Arnold, MD, PhD, FRCPC, Mathison Centre for Mental Health Research and Education, 4th floor, Teaching, Research and Wellness (TRW) Building, 3280 Hospital Dr NW, Calgary AB, T2N 4Z6, Canada; Email: [email protected]; Phone: (403) 210-6464

Statement of Contributions

The authors contributed to the work presented in the chapter as follows:

Lauren Erdman – statistical analysis

Christie L. Burton – coauthor, editing

Laura S. Park – wet lab assistance

Annie Dupuis – data management, statistical insight

Janet Shan – data management

Tara Goodale – data and sample/participant management

S-M Shaheen – lab technician

Jennifer Crosbie – supervision

Russell J. Schachar – editing, supervision

Paul D. Arnold – editing, supervision

Conflicts of Interest

Russell J. Schachar consults for Highland Therapeutics, Purdue Pharma, and ehave, and is the

Toronto Dominion Bank Financial Group Chair in Child and Adolescent Psychiatry. Vanessa

M. Sinopoli, Lauren Erdman, Christie L. Burton, Laura S. Park, Annie Dupuis, Janet Shan,

Tara Goodale, S-M Shaheen, Jennifer Crosbie, and Paul D. Arnold have no conflicts of interest to declare.

Acknowledgements

This work was supported by the Canadian Institutes of Health Research (P.D.A., MOP-106573 and R.J.S., MOP-93696). Dr. Paul D. Arnold receives funding from the Alberta Innovates

52 Health Solutions (AIHS) Translational Health Chair in Child and Youth Mental Health.

Vanessa Sinopoli received funding from the Canadian Institutes of Health Research (CIHR)

Master's Award: Frederick Banting and Charles Best Canada Graduate Scholarships, Ontario

Graduate Scholarship (OGS), and the Hospital for Sick Children Restracomp Studentship.

The authors also wish to acknowledge the contributions of Ms. Clara Tam in study coordination, Mr. Bingbin Li for laboratory technical assistance, and the Ontario Science

Centre for collaborating in the recruitment of participants.

Abstract

Background: Serotonin system genes are commonly studied in obsessive-compulsive disorder (OCD), but genetic studies to date have produced inconsistent results, possibly because phenotypic heterogeneity has not been adequately accounted for. In this paper, we studied candidate serotonergic genes and homogenous phenotypic subgroups as presented through obsessive-compulsive (OC) trait dimensions in a general population of children and adolescents. We hypothesized that different serotonergic gene variants are associated with different OC trait dimensions and, furthermore, that they vary by sex.

Methods: OC trait dimensions (Cleaning/Contamination, Counting/Checking,

Symmetry/Ordering, Superstition, Rumination, and Hoarding) were examined in a total of

5213 pediatric participants in the community using the Toronto Obsessive-Compulsive Scale

(TOCS). We genotyped candidate serotonin genes (directly genotyping the 5-HTTLPR polymorphism in SLC6A4 for 2018 individuals and using single nucleotide polymorphism

53 (SNP) array data for genes SLC6A4, HTR2A, and HTR1B for 4711 individuals). We assessed the association between variants across these genes and each of the OC trait dimensions, within males and females separately.

Results: The [LG + S] variant in 5-HTTLPR was significantly associated with hoarding in males (odds ratio, OR, of 1.35; P-value of 0.003). There were no significant findings for 5-

HTTLPR in females. Using SNP array data, there were no significant findings in males or females after correction for multiple comparisons, but our top SNP findings in females also suggested that serotonin system gene variation (HTR1B) may be implicated in hoarding.

Conclusions: This represents the first genetic association study of OC trait dimensions in a community-based pediatric sample. Our strongest results indicate that hoarding may be distinct in its underlying serotonin system genetics and that the associated serotonin system gene variation may be specific to sex. Future genetic association studies in OCD should properly account for heterogeneity, using homogenous subgroups stratified by symptom dimension, sex, and age group.

54 3.1. Introduction

Obsessive-compulsive disorder (OCD) is a debilitating psychiatric disorder characterized by recurring, disturbing thoughts and/or repetitive behaviors carried out in response to mounting anxiety (American Psychiatric Association [APA], 2013). The estimated worldwide prevalence of OCD is 2-3% (Angst et al., 2004, Kessler et al., 2005; Murphy et al., 2013) and

0.5-2% in children and adolescents, although OCD is often underdiagnosed in the pediatric population (Geller & March, 2012; Alvarenga et al., 2015). The most efficacious, pharmacological treatments for OCD are medications that act on the serotonin system, collectively referred to as serotonin reuptake inhibitors (SRIs), which include selective serotonin reuptake inhibitors (SSRIs), combined serotonin-norepinephrine reuptake inhibitors

(SNRIs), and clomipramine (a tricyclic compound) (Murphy et al., 2004).

OCD is familial and heritable with estimates ranging from 27-47% in adults and 45-65% in children (van Grootheest et al., 2005, Pauls et al., 2014). Genetic studies have focused on serotonin system gene variants because of the efficacy of SRIs for treating OCD (Millan et al.,

2015). SRIs block serotonin reuptake in the brain via binding and allosteric modulation of the serotonin transporter (SERT) (Blier et al., 1990; Billett et al., 1997). The most commonly studied variants are in the genes coding for SERT (SLC6A4) and the serotonin receptors 5-

HT2A (HTR2A) and 5-HT1B (HTR1B) (Taylor, 2013; Sinopoli et al., 2017).

One of the most researched serotonin gene polymorphisms is 5-HTTLPR (5-HTT-linked polymorphic region), located within the promoter region of SLC6A4 (Sinopoli et al., 2017). 5-

HTTLPR exists in two major forms: the long (L) or short (S) variant. The L variant is made up

55 of 16 sets of 20-23 base pair (bp) tandem repeats, while the S variant is made up of 14 sets of

20-23 bp tandem repeats. The L variant results in greater SERT expression than the S variant

(Heils et al., 1996; Lesch et al., 1996). Recently, an intrinsic single nucleotide polymorphism

(SNP), rs25531, was discovered within the additional repeats in the L variant. When the less frequent G allele of rs25531 is present in the L variant, SERT expression is similar to the S variant. This results in two functional groups: LA variant with high SERT expression and

LG/S variant with low SERT expression (Hu et al., 2006). The rs25531 SNP was not accounted for in early association studies, which may contribute to some of the discrepancies in the literature to date (Sinopoli et al., 2017). In a recent meta-analysis, the LA variant of 5-

HTTLPR and the A allele of SNP rs6311 or the linked T allele of SNP rs6313 in HTR2A were associated with OCD (Taylor, 2013; Taylor, 2016).

OCD is phenotypically complex and can present with various types of symptoms. A meta- analysis of previous factor analyses of the gold standard measure of OCD, the Yale-Brown

Obsessive-Compulsive Scale (Y-BOCS), in adults and children with OCD, reported four symptom dimensions: 1) Symmetry 2) Forbidden thoughts, 3) Cleaning, and 4) Hoarding.

Although similar factors were identified in children and adults, the specific items contributing to each dimension differed based on age group (Bloch et al., 2008a). Factor analysis also demonstrates that symptom dimensions exist for obsessive-compulsive (OC) traits in the general population (Burton et al., 2018). The prevalence and distribution of symptom dimensions in clinical and in general population samples also differ between males and females (Flament, 1990; Ruscio et al., 2010; Alvarenga et al., 2015). OCD symptom dimensions are heritable, with evidence of both shared and/or unique genetic influences between them in clinical and population-based samples (van Grootheest et al., 2008; Katerberg

56 et al., 2010; Burton et al., 2018), and they differ in terms of their neurobiology, comorbidities, and response to treatment (Pauls et al., 2014). For example, response to SRIs varies depending on prevailing OCD symptoms or symptom dimensions (Landeros-Weisenberger et al., 2010). Thus, OCD symptom dimensions may be mediated by different underlying biological mechanisms.

The genetic basis of OCD is unclear, likely because it is often treated as a unidimensional phenomenon. The purpose of this study was to examine if different candidate serotonin system gene variants are associated with different obsessive-compulsive trait dimensions.

Given that pediatric OCD is often underdiagnosed, and that children and adolescents with mild to moderate symptom severity may not present in the clinic (Geller & March, 2012), we used a population-based sample to capture symptoms that may otherwise remain undetected. To account for previously reported genetic differences in OCD between sexes (Mattina & Steiner,

2016), we ran our analyses in males and females separately. Since candidate serotonin genes have been implicated in OCD and given that OCD symptom dimensions differ in terms of their heritability, neurobiology, and clinical presentation, we hypothesized that we would find different serotonin gene variants associated with different OC trait dimensions and that these findings would vary based on sex.

57 3.2. Methods

3.2.1. Participants

17,263 children and adolescents, ages 6-18 years, were recruited and assessed at the Ontario

Science Centre in Toronto, Canada for the Thoughts, Actions and Genes (TAG) project, described in detail elsewhere (Crosbie et al., 2013; Burton et al., 2016; Park et al., 2016).

Informed consent, and assent where applicable, was obtained for all participants. The study was approved by the Hospital for Sick Children Research Ethics Board. 16,718 participants had complete demographic information and questionnaires from either parent-respondent

(N=13,680) or self-respondent (N=3,038). From this sample, we obtained two smaller, partially overlapping groups of participants who had 4 grandparents of European descent. The first group had 2100 participants with DNA samples readily available to directly genotype the

5-HTTLPR region of interest in SLC6A4. The second group had 4810 participants with genome-wide association study (GWAS) data available, from which candidate gene SNPs were derived.

3.2.2. Obsessive-compulsive features

Participants or their parents completed the 21-item Toronto Obsessive Compulsive Scale

(TOCS) that measures OC traits (Park et al., 2016). Each of the 21 items was scored on a scale of -3 to +3, where -3 signifies that the child performs the action of interest far less often than his or her peers, and +3 signifies that the child performs the action of interest far more often than his or her peers. A previous factor analysis of the TOCS identified 6 factors: 1)

58 Cleaning/Contamination, 2) Counting/Checking, 3) Symmetry/Ordering, 4) Superstition, 5)

Rumination, 6) Hoarding (Burton et al., 2018). Heritability estimates for each of these dimensions range from 30 to 77% (Burton et al., 2018). We dichotomized each item, whereby a score of ≥ 2 on an item indicates presence of the obsessive-compulsive trait and any other score indicates absence of that trait. Individuals were assigned “affected” status for the OC group if they had at least one reported obsessive-compulsive trait. They were also assigned

“affected” status for groups, based on the previous factor analysis, if they had at least one reported trait within the dimension. Note that individuals could be deemed “affected” for more than one dimension. Individuals who did not have any OC traits were deemed controls.

3.2.3. DNA collection and extraction

Saliva was collected from all participants using Oragene×DNA (OG-500) kits and DNA extracted using the protocol recommended by the manufacturer (DNA Genotek). We also centrifuged the samples at 10,000 RPM for another 10 min to remove any precipitated carbohydrates. DNA was quantified using the Quant-iT™ PicoGreen® dsDNA Assay Kit

(Invitrogen). DNA samples were run on agarose gels to ensure good quality before using microarrays. Samples with DNA concentration below 60ng/µl (if genotyped on the

HumanCoreExome-12 v1.0 microarray (Illumina)) and samples with poor DNA quality were excluded.

59 3.2.4. Selection of candidate genes

We selected serotonin candidate genes SLC6A4, HTR2A, and HTR1B based on their inclusion in a recent OCD meta-analysis (Taylor, 2013). Given our interest in the serotonin system, we included 3 serotonin genes in which the polymorphisms with the best evidence of association with OCD were located.

3.2.5. 5-HTTLPR

3.2.5.1. Direct genotyping

Variation in 5-HTTLPR is not captured via microarrays and, thus, was directly genotyped in a two-step process (Wendland et al., 2006), in conjunction with The Centre for Applied

Genomics (TCAG) at the Hospital for Sick Children in Toronto:

A.) Identifying L or S variant by polymerase chain reaction (PCR) amplification of 5-HTTLPR polymorphism:

5-HTTLPR was PCR amplified using forward primer 5'-

ATGCCAGCACCTAACCCCTAATGT-3’, which was 5’-labeled with HEX fluorescent dye for visualization, and reverse primer 5'-GGACCGCAAGGTGGGCGGGA-3' (Gelernter et al.,

1997). PCR involved 50 ng of our genomic template DNA, 10mM dNTP mix (Fermentas,

Life Technologies), the PCRx Enhancer System (Invitrogen, Life Technologies) inclusive of

10X PCRx Enhancer Solution, 50 mM MgSO4, 10X PCRx Amplification Buffer, and 5000

U/ml Taq DNA polymerase (New England BioLabs). PCR cycling conditions involved 30

60 cycles of 95ºC for 30 sec, 55ºC for 45 sec, and 68ºC for 60 sec. 1 µl volumes of the PCR products were suspended in a 10 µl mixture of 7 µl GeneScan 500 LIZ dye Size Standard in

993 µl Hi-Di formamide (Applied Biosystems). Samples were subsequently run on an

ABI3730XL genetic analyzer using the POP-7 polymer and Dye Set G5 (Applied

Biosystems). The Peak Scanner Software v1.0 (Thermo Fisher Scientific) was used to analyze results. The amplified 5-HTTLPR products were predicted to be 419 bp and 375 bp in length for the L and S variant, respectively. We found the mobility of the amplified fragments on the

ABI3730XL genetic analyzer to read as 412 bp and 370 bp for L and S, respectively.

B.) Determining rs25531 SNP by digestion of identified L or S variant:

Once the 5-HTTLPR was identified as L or S, the DNA was digested to determine an A or G

SNP at rs25531. A restriction enzyme digested the amplified PCR product to cut the sequence at rs25531 if a G was present. As per the manufacturer’s protocol we used 0.1-0.5 µg of PCR reaction DNA, 10X Buffer Tango, and 0.5-2 µl of restriction enzyme MspI (Fermentas, Life

Technologies). 1 µl volumes of the resulting DNA products were suspended in a 10 µl mixture of 7 µl GeneScan 500 LIZ dye Size Standard in 993 µl Hi-Di formamide (Applied

Biosystems). Samples were run on an ABI3730XL genetic analyzer using the POP-7 polymer and Dye Set G5 (Applied Biosystems) and the Peak Scanner Software v1.0 (Thermo Fisher

Scientific) was used to analyze results. We identified that if no G was present at rs25531, the digested L variant yielded a 321 bp fragment and the digested S variant yielded a 278 bp fragment. If a G, but not an A, was present at rs25531, the enzyme cut at rs25531 to yield a

149 bp fragment. It is documented that the G SNP is located in the additional stretch of DNA within the L variant (Hu et al., 2006), but there have been reports of a rare SG variant that carries a G SNP at a site resulting in the same length fragment as LG when digested (Wendland

61 et al., 2006; Voyiaziakis et al., 2011). Thus, the resulting digested fragments would be the same length for both LG and SG variants. It is for this reason that the L versus S variant data was obtained from the initial PCR amplification, prior to digestion. L versus S data was then combined with data from the rs25531 SNP restriction enzyme method to complete 5-HTTLPR genotyping and accurately identify LA, LG, and S variants (inclusive of an SA variant and the rare SG variant) (Figure 3). We then grouped our variants into two allelic groups, based on functionality, as follows: LA and [LG + S]. We also took into account that the alleles are additive in their effect (Hu et al., 2006).

Figure 3. 5-HTTLPR; L and S variants and rs25531: Green circles represent fluorescent probes. Red arrows represent where restriction enzyme, MspI, cuts the DNA at its appropriate recognition sequence if present. Upon digestion, the following sizes of fluorescently probed fragments are identified: LA = 321bp, SA = 278bp, LG and rare SG = 149bp. This information is combined with the L versus S information from the original PCR to distinguish between LG and SG. 5-HTTLPR, 5-HTT-linked polymorphic region; L, long variant; S, short variant; LA, long variant with single nucleotide polymorphism A at rs25531; SA, short variant with single nucleotide polymorphism A at rs25531; LG, long variant with single nucleotide polymorphism G at rs25531; SG, short variant with single nucleotide polymorphism G at rs25531; PCR, polymerase chain reaction.

For the 5-HTTLPR sample, quality control (QC) was implemented when visualizing and analyzing genotypes, using the standard Peak Scanner Software v1.0 (Thermo Fisher

62 Scientific) program settings. Individuals were genotyped with a 97% completion rate.

Reported siblings and individuals of indeterminate ethnicity were removed. When siblings were identified, we kept the individual in the OC group if possible. We identified 12 individuals carrying the rare SG variant in our final, total 5-HTTLPR-genotyped sample. Given that we found no evidence of differential expression of the SG in the literature, we chose to include individuals with this variant in our analysis and considered both SA and the rare SG variants as “S”. After QC and removal of ensuing individuals, we had a final sample size of

2018.

3.2.5.2. Statistical analyses

Analyses were conducted using the program, R, version 3.0.1 (https://www.R-project.org).

The expected number of tested alleles (0, 1, or 2 [LG + S] alleles) was used in a logistic regression as a predictor of affected versus control status for each of the 7 trait groups. An additive model was used. Analyses were conducted in male and in female subjects separately, using age and questionnaire respondent type (parent or self) as covariates. To account for multiple comparisons, our threshold of significance was a P-value < 0.007 (0.05 / 7 analyses) for both males and females.

63 3.2.6. Candidate gene SNPs

3.2.6.1. Genotyping

Samples were genotyped on the HumanCoreExome-12 v1.0 microarray (Illumina) for collection of SNP data at TCAG. The microarray data was available from an ongoing GWAS study conducted by our group (Burton et al., 2015b). We analyzed SNPs across SLC6A4,

HTR2A, and HTR1B (UCSC Genome Browser, hg19). Standard QC removed SNPs with call rate < 0.97 and samples with sex misspecification and ambiguity. Only one member per family was used in the analyses using the PI_HAT cutoff = 0.12. European ethnicity was verified via principal component analysis (PCA) using EIGENSTRAT version 3.2.10 (Price et al., 2006) with HapMap populations CEU and TSI to ensure that our individuals’ genetic data clustered with Europeans, relative to other ethnicities, and then via another PCA to identify and remove outliers within our own population. An additional PCA was conducted without outliers, followed by formation of a scree plot, to determine that there were seven principal components (PCs) we needed to use to control for population structure in our analyses. After

QC and removal of ensuing individuals, we had a final sample size of 4711.

Genetic data was imputed over the genetic regions for SLC6A4, HTR2A, and HTR1B along with ±50 kb flanking regions for each gene using SHAPEIT2 version 2.790 (Delaneau et al.,

2014) and IMPUTE2 version 2.3.1 (Howie et al., 2009). 1000 Genomes Phase 3 reference data was used for both pre-phasing and imputation. After imputation, SNPs were removed for low quality imputation (information score < 0.8). SNPs with a minor allele frequency (MAF)

64 < 0.05 were excluded from the analysis. Data were kept as dosage calls for all analyses.

Table 3 shows final SNP counts across each of our 3 genes of interest.

Table 3. SNP counts across 3 candidate genes, including all genotyped and imputed SNPs: Common SNPs (MAF ≥ 0.05) are shown for each gene and its flanking regions included in the analyses for males and in the analyses for females.

Gene Number of SNPs in Males Number of SNPs in Females SLC6A4 121 142 HTR2A 411 361 HTR1B 242 242

Abbreviations: SNP, single nucleotide polymorphism; SLC6A4, solute carrier family 6 member 4 (serotonin transporter gene); HTR2A, 5-HT2A receptor gene; HTR1B, 5-HT1B receptor gene.

3.2.6.2. Statistical analyses

Using R, version 3.0.1 (https://www.R-project.org), logistic regressions for each SNP were conducted using the expected number of tested alleles (0, 1, or 2) to predict affected versus control status for each of the 7 trait groups. An additive model was used. Regressions were conducted for males and females separately, using age, respondent, and the identified PCs as covariates. To address multiple comparisons, we used the Genetic Type 1 error calculator

(GEC), developed to account for dependent SNPs (Li et al., 2012). Our threshold of significance was a P-value < 2.94e-5 (2.06e-4 / 7 analyses) for males and a P-value < 2.90e-5

(2.03e-4 / 7 analyses) for females.

65 3.2.7. Sample size considerations

The final sample size for 5-HTTLPR (N=2018) was smaller than the final sample size for our candidate gene SNPs (N=4711), with a 29% overlap between these two groups (502 participants genotyped in the 5-HTTLPR group only; 3195 participants genotyped in the candidate gene SNP group only; 1516 participants genotyped for and used in both the 5-

HTTLPR and candidate gene SNP groups). The smaller sample size for 5-HTTLPR was deemed more than sufficient given the less stringent statistical correction required for the single variant versus the set of candidate gene SNPs, and given the increased labor and cost required to directly genotype 5-HTTLPR.

3.3. Results

3.3.1. 5-HTTLPR analyses

Demographics for the 5-HTTLPR-genotyped individuals are reported in Table 4.

Using t-tests, we identified no significant differences for age between affected males and male controls, or between affected females and female controls. We identified a significant difference for respondent between affected males and male controls (absolute t-statistic = 3.09,

P-value = 0.002), but not between affected females and female controls. Table 5 shows the affected versus control counts for OC traits overall and for each of the OC trait dimensions in our 5-HTTLPR-genotyped individuals.

66 Table 4. 5-HTTLPR-genotyped individuals: Demographics and group characteristics are shown for total individuals combined, all affected individuals, and control individuals.

Total Individuals Male Female N 1050 968 Mean Age (SD) 10.7 (2.5) 11.3 (2.9) Age Range 6.1 - 17.9 6.3 - 17.9 Respondent: % Parent-report, % Self-report 90.1, 9.9 81.5, 18.5

All Affected Individuals Male Female N 590 571 Mean Age (SD) 10.6 (2.5) 11.2 (3.0) Age Range 6.3 - 17.9 6.3 - 17.9 OCD Diagnosis, N (%) 32 (5.4) 30 (5.3) Mood Disorder Diagnosis, N (%) 12 (2.0) 24 (4.2) ADHD Diagnosis, N (%) 106 (18.0) 39 (6.8) Anxiety Disorder Diagnosis, N (%) 74 (12.5) 66 (11.6) ASD Diagnosis, N (%) 59 (10.0) 12 (2.1) Tic Disorder Diagnosis, N (%) 28 (4.7) 16 (2.8) Taking SRI Medication, N (%) 20 (3.4) 19 (3.3) Respondent: % Parent-report, % Self-report 89.8, 10.2 81.3, 18.7 Allele Frequency: LA, [LG + S] 0.48, 0.52 0.52, 0.48

Control Individuals Male Female N 460 397 Mean Age (SD) 10.8 (2.5) 11.3 (2.9) Age Range 6.1 - 17.7 6.3 - 17.9 OCD Diagnosis, N (%) 0 (0.0) 0 (0.0) Mood Disorder Diagnosis, N (%) 3 (0.7) 1 (0.3) ADHD Diagnosis, N (%) 25 (5.4) 10 (2.5) Anxiety Disorder Diagnosis, N (%) 3 (0.7) 5 (1.3) ASD Diagnosis, N (%) 1 (0.2) 0 (0.0) Tic Disorder Diagnosis, N (%) 2 (0.4) 1 (0.3) Taking SRI Medication, N (%) 0 (0.0) 0 (0.0) Respondent: % Parent-report, % Self-report 90.4, 9.6 81.9, 18.1

Allele Frequency: LA, [LG + S] 0.52, 0.48 0.53, 0.47

Abbreviations: 5-HTTLPR, 5-HTT-linked polymorphic region; N, sample size; SD, standard deviation; OCD, obsessive-compulsive disorder; ADHD, attention-deficit/hyperactivity disorder; ASD, autism spectrum disorder; SRI, serotonin reuptake inhibitor; LA, long variant of 5-HTTLPR with A allele at SNP rs25531; LG, long variant of 5-HTTLPR with G allele at SNP rs25531; S, short variant of 5-HTTLPR.

67 Table 5. Trait dimension group counts for 5-HTTLPR-genotyped individuals: Affected versus control counts for OC traits overall and for each of the OC trait dimensions.

OC Male Female Affected Individuals 590 571 Control Individuals 460 397

Cleaning/Contamination Male Female Affected Individuals 238 238 Control Individuals 460 397

Counting/Checking Male Female Affected Individuals 191 165 Control Individuals 460 397

Symmetry/Ordering Male Female Affected Individuals 358 333 Control Individuals 460 397

Superstition Male Female Affected Individuals 116 162 Control Individuals 460 397

Rumination Male Female Affected Individuals 217 233 Control Individuals 460 397

Hoarding Male Female Affected Individuals 387 380 Control Individuals 460 397 Abbreviations: OC, obsessive-compulsive.

68 As shown in Figure 4, after correction for multiple comparisons, we found a significant association between the [LG + S] variant and hoarding in males (odds ratio, OR, of 1.35; P- value of 0.003). 5-HTTLPR was not significantly associated with OC traits or any specific OC trait dimension aside from hoarding in males, after correction for multiple comparisons. 5-

HTTLPR was not significantly associated with OC traits or any specific OC trait dimension in females.

Figure 4. Association between 5-HTTLPR and OC traits overall/OC trait dimensions: Forest plots for males (left) and females (right). Colored points show the odds ratios for the [LG + S] allele, representing the effect size estimate for the allele in each trait group relative to controls. The length of each horizontal line is the confidence interval for each effect size estimate. Significant finding is indicated by *. 5-HTTLPR, 5-HTT-linked polymorphic region; OC, obsessive-compulsive.

69 3.3.2. Candidate gene SNP analyses

Demographics for the candidate gene SNP individuals are reported in Table 6. Using t-tests, we identified no significant differences for age or for respondent between affected males and male controls, or between affected females and female controls. Table 7 shows the affected versus control counts for OC traits overall and for each of the OC trait dimensions in our individuals genotyped for candidate gene SNPs.

70 Table 6. Individuals genotyped for candidate gene SNPs: Demographics and group characteristics are shown for total individuals combined, all affected individuals, and control individuals.

Total Individuals Male Female N 2439 2272 Mean Age (SD) 10.7 (2.6) 11.1 (2.9) Age Range 6.1 - 17.9 6.1 - 17.9 Respondent: % Parent-report, % Self-report 89.7, 10.3 81.0, 19.0

All Affected Individuals Male Female N 1025 1043 Mean Age (SD) 11.1 (2.8) 11.8 (3.1) Age Range 6.1 - 17.9 6.4 - 17.9 OCD Diagnosis, N (%) 28 (2.7) 27 (2.6) Mood Disorder Diagnosis, N (%) 18 (1.8) 30 (2.9) ADHD Diagnosis, N (%) 157 (15.3) 44 (4.2) Anxiety Disorder Diagnosis, N (%) 95 (9.3) 82 (7.9) ASD Diagnosis, N (%) 65 (6.3) 11 (1.1) Tic Disorder Diagnosis, N (%) 30 (2.9) 15 (1.4) Taking SRI Medication, N (%) 30 (2.9) 20 (1.9) Respondent: % Parent-report, % Self-report 82.8, 17.2 71.9, 28.1

Control Individuals Male Female N 1414 1229 Mean Age (SD) 10.3 (2.3) 10.6 (2.6) Age Range 6.1 - 17.7 6.1 - 17.9 OCD Diagnosis, N (%) 0 (0) 0 (0) Mood Disorder Diagnosis, N (%) 8 (0.6) 2 (0.2) ADHD Diagnosis, N (%) 95 (6.7) 39 (3.2) Anxiety Disorder Diagnosis, N (%) 18 (1.3) 22 (1.8) ASD Diagnosis, N (%) 10 (0.7) 1 (0.1) Tic Disorder Diagnosis, N (%) 11 (0.8) 5 (0.4) Taking SRI Medication, N (%) 0 (0) 0 (0) Respondent: % Parent-report, % Self-report 94.8, 5.2 89.7, 10.3

Abbreviations: SNP, single nucleotide polymorphism; N, sample size; SD, standard deviation; OCD, obsessive- compulsive disorder; ADHD, attention-deficit/hyperactivity disorder; ASD, autism spectrum disorder; SRI, serotonin reuptake inhibitor.

71 Table 7. Trait dimension group counts for individuals genotyped for candidate gene SNPs: Affected versus control counts for OC traits overall and for each of the OC trait dimensions.

OC Male Female Affected Individuals 1025 1043 Control Individuals 1414 1229

Cleaning/Contamination Male Female Affected Individuals 331 319 Control Individuals 1414 1229

Counting/Checking Male Female Affected Individuals 221 222 Control Individuals 1414 1229

Symmetry/Ordering Male Female Affected Individuals 502 515 Control Individuals 1414 1229

Superstition Male Female Affected Individuals 128 209 Control Individuals 1414 1229

Rumination Male Female Affected Individuals 337 384 Control Individuals 1414 1229

Hoarding Male Female Affected Individuals 465 500 Control Individuals 1414 1229 Abbreviations: OC, obsessive-compulsive.

72 Most of the top SNPs reported in males corresponded with HTR2A and were associated with rumination (OR of 0.62 to 0.65; P-value of 1.34e-3 to 3.63e-3; Table 8). For females, all of the top reported SNPs corresponded with HTR1B and were associated with hoarding (OR of

1.33 to 1.42; P-value of 1.05e-3 to 4.60e-3; Table 9). None of these findings remained significant after correction for multiple comparisons.

Table 8. Association between serotonin gene SNPs and OC traits overall/OC trait dimensions in males: Top SNP findings with P-value < 1.00e-2.

Tested SNP Gene MAF OR P-value Trait Group Allele rs3742278 HTR2A G 0.132 0.64 1.34E-03 Rumination rs9567735 HTR2A G 0.132 0.64 1.47E-03 Rumination rs9562684 HTR2A C 0.131 0.64 1.60E-03 Rumination rs1923884 HTR2A T 0.131 0.64 1.65E-03 Rumination rs9562685 HTR2A A 0.131 0.64 1.65E-03 Rumination rs9567736 HTR2A A 0.132 0.64 1.81E-03 Rumination rs62416430 HTR1B G 0.066 0.51 2.38E-03 Rumination rs17068986 HTR2A T 0.133 0.65 2.90E-03 Rumination rs62416428 HTR1B G 0.067 0.52 3.06E-03 Rumination rs62416429 HTR1B T 0.067 0.52 3.06E-03 Rumination rs6505166 SLC6A4 A 0.052 1.93 3.50E-03 Counting/Checking rs9534505 HTR2A G 0.085 0.62 3.63E-03 Rumination rs2020932 SLC6A4 A 0.051 2.04 3.96E-03 Counting/Checking rs62416430 HTR1B G 0.068 0.37 5.83E-03 Superstition rs2770301 HTR2A C 0.231 0.63 8.68E-03 Superstition

Abbreviations: SNP, single nucleotide polymorphism; MAF, minor allele frequency; OR, odds ratio; SLC6A4, solute carrier family 6 member 4 (serotonin transporter gene); HTR2A, 5-HT2A receptor gene; HTR1B, 5-HT1B receptor gene.

73 Table 9. Association between serotonin gene SNPs and OC traits overall/OC trait dimensions in females: Top SNP findings with P-value < 1.00e-2.

Tested SNP Gene MAF OR P-value Trait Group Allele rs1777762 HTR1B T 0.154 1.39 1.05E-03 Hoarding rs1777763 HTR1B T 0.154 1.39 1.07E-03 Hoarding rs2798528 HTR1B T 0.154 1.39 1.09E-03 Hoarding rs1738506 HTR1B A 0.154 1.39 1.09E-03 Hoarding rs2223832 HTR1B C 0.154 1.39 1.11E-03 Hoarding rs1343334 HTR1B A 0.154 1.39 1.17E-03 Hoarding rs2207053 HTR1B C 0.153 1.38 1.32E-03 Hoarding rs1228806 HTR1B C 0.153 1.38 1.34E-03 Hoarding rs1228805 HTR1B G 0.154 1.38 1.48E-03 Hoarding rs2207056 HTR1B T 0.154 1.38 1.49E-03 Hoarding rs1228797 HTR1B T 0.154 1.38 1.49E-03 Hoarding rs1228798 HTR1B G 0.154 1.38 1.49E-03 Hoarding rs1228800 HTR1B G 0.154 1.38 1.49E-03 Hoarding rs1343336 HTR1B T 0.154 1.38 1.51E-03 Hoarding rs1228802 HTR1B A 0.153 1.37 1.65E-03 Hoarding rs1145827 HTR1B A 0.153 1.37 1.73E-03 Hoarding rs2798529 HTR1B G 0.151 1.37 1.81E-03 Hoarding rs1228804 HTR1B G 0.155 1.37 1.82E-03 Hoarding rs6453980 HTR1B G 0.153 1.37 1.90E-03 Hoarding rs58608035 HTR1B T 0.153 1.37 1.90E-03 Hoarding rs60174069 HTR1B C 0.153 1.37 1.90E-03 Hoarding rs59414600 HTR1B C 0.153 1.37 2.03E-03 Hoarding rs1228803 HTR1B T 0.154 1.36 2.03E-03 Hoarding rs61295513 HTR1B T 0.153 1.37 2.03E-03 Hoarding rs6938832 HTR1B T 0.152 1.36 2.24E-03 Hoarding rs6939163 HTR1B C 0.152 1.36 2.24E-03 Hoarding rs4708338 HTR1B C 0.152 1.36 2.26E-03 Hoarding rs6938765 HTR1B A 0.152 1.36 2.27E-03 Hoarding rs4708340 HTR1B C 0.152 1.36 2.28E-03 Hoarding rs1777767 HTR1B C 0.148 1.37 2.33E-03 Hoarding rs10806097 HTR1B T 0.121 1.42 2.33E-03 Hoarding rs17272333 HTR1B C 0.121 1.41 2.78E-03 Hoarding rs112390126 HTR1B C 0.121 1.41 2.79E-03 Hoarding rs11755194 HTR1B C 0.122 1.40 3.17E-03 Hoarding rs55636038 HTR1B T 0.122 1.40 3.17E-03 Hoarding rs11753559 HTR1B A 0.120 1.40 3.20E-03 Hoarding rs12110491 HTR1B T 0.123 1.40 3.21E-03 Hoarding rs17272829 HTR1B A 0.121 1.40 3.50E-03 Hoarding rs62416427 HTR1B A 0.121 1.40 3.59E-03 Hoarding rs11757592 HTR1B C 0.150 1.33 4.60E-03 Hoarding

74 Abbreviations: SNP, single nucleotide polymorphism; MAF, minor allele frequency; OR, odds ratio; HTR1B, 5- HT1B receptor gene.

3.4. Discussion

Previous studies have suggested that variants in serotonin system genes may be involved in

OCD, but findings have been mixed. One reason for the mixed findings may be that previous studies have not examined whether or not different serotonin gene variants are associated with different symptom dimensions (Taylor, 2013; Taylor, 2016, Sinopoli et al., 2017). Our study was the first to examine if serotonin genes were associated with OC trait dimensions in males and in females in a pediatric, population-based sample. Overall, our findings support our hypothesis that different OC trait dimensions may have different genetic underpinnings and that genetic influences may be sex-specific, particularly showing that 5-HTTLPR is significantly associated with hoarding traits in males.

Our findings suggest that hoarding traits are biologically distinct from other OC traits. When comparing the two functionally distinct variants of 5-HTTLPR (LA versus [LG + S]), the [LG +

S] variant, associated with lower SERT expression, was significantly associated with hoarding traits in males, but not in females. In our post hoc analysis combining males and females, the association between [LG + S] and hoarding was no longer statistically significant, suggesting that males are driving the association. Many studies have hypothesized an increase in SERT expression in OCD (Murphy et al., 2013; Sinopoli et al., 2017), but findings have been inconsistent overall. Our finding suggests that a decrease in SERT expression may be correlated with hoarding in males.

75 Our hoarding-driven findings are consistent with recent literature suggesting that hoarding is clinically and biologically distinct from OCD (Mataix-Cols & Pertusa, 2012; Snowdon et al.,

2012). The Fifth Edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-

5) now classifies hoarding as a distinct disorder in the “Obsessive-Compulsive and Related

Disorders” category (APA, 2013). Recent research has also identified phenotypic, biological/genetic, and sex differences in patients with primary hoarding symptoms versus

OCD patients without hoarding symptoms (Murphy et al., 2013) and shown that hoarding symptoms may have distinct neurological and genetic underpinnings (van Ameringen et al.,

2014). As of yet, there is no clear evidence linking serotonergic genes to hoarding specifically. Our data suggests that particular serotonin gene variants may be associated with hoarding traits and, furthermore, that this association varies by sex. 5-HTTLPR appears to be involved in hoarding symptomatology in males and there was a trend for variation downstream of HTR1B to be associated with hoarding symptomatology in females. Our data suggests that distinct biological underpinnings may also be driving other symptom dimensions within OCD and that these underpinnings vary by sex, as we observed a trend for variation across HTR2A to be association between the rumination trait dimension and in males.

Our findings show that when studying the genetic basis of OCD, and of psychiatric disorders in general, it is important to reduce heterogeneity wherever possible to increase the power of genetic analyses and to study if different trait dimensions have shared or distinct etiologies.

The National Institute of Mental Health (NIMH) launched the Research Domain Criteria

(RDoC) project to help researchers account for heterogeneity when studying mental health.

The framework suggests researching basic dimensional constructs (identified through molecular, genetic, neurocircuitry-related, and behavioral analyses) that underlie higher-level

76 behaviors and thus complex psychiatric disorders (NIMH, 2008). The major strength of our study includes our effort to address heterogeneity and use a dimensional approach consistent with RDoC.

Some inconsistencies in previous genetic studies could be attributed to failure to account for phenotypic heterogeneity in OCD, particularly stratifying data both by symptom dimensions and sex. For example, three studies used Y-BOCS-derived factors and analyzed association with 5-HTTLPR in adults with OCD. Results varied across all three with one identifying an association between L-carriers and a religious/somatic factor (Kim et al., 2005), another identifying an association between the S allele and the S/S genotype and symmetry obsessions/repeating counting and ordering/arranging compulsions (Hasler et al., 2006), and another reporting an associating between the L/L genotype (versus L/S or S/S) of 5-HTTLPR and counting and repeating rituals in individuals with comorbid tics (Cavallini et al., 2002).

Mixed results from these early studies may have been due to a failure to account for sex differences in addition to symptom dimensions. For example, another study examined SNPs in a glutamate receptor gene GRIN2B and found no association between GRIN2B SNPs and

OCD. After stratifying by symptom dimension and sex, rs1805476 was significantly associated with contamination obsessions and cleaning compulsions in male patients (Alonso et al., 2012). Our results, although requiring replication, highlight the need to account for key sources of heterogeneity (including symptom dimensions and sex) to ensure that important associations are identified and to gain a fuller understanding of the complex etiology of OCD and traits that contribute to the disorder.

77 One of the limitations of our study was our sample size. Although adequate for a candidate gene approach, the size of our sample precluded an alternative, more comprehensive approach such as performing a GWAS in which we could comprehensively analyze all common variation across the genome. A GWAS of multiple symptom dimensions would require very stringent correction for multiple comparisons (standard GWAS plus correction for multiple phenotypes) and our sample would be underpowered to detect associations. Therefore, we instead elected to perform a more focused study to examine whether candidate serotonergic gene variants were associated with OC trait dimensions in our pediatric, population-based sample. Another limitation was a difference in the initial sample collection approach used for our 5-HTTLPR-genotyped group versus our candidate gene SNP group. We initially began collecting individuals for the 5-HTTLPR-genotyped group using an extreme trait approach, which would result in higher TOCS scores in affected individuals and which likely explains why there are more individuals who reported an OCD diagnosis in this group (relative to our larger group of individuals genotyped for candidate gene SNPs). Though we cannot rule out whether or not our significant hoarding finding was being driven by a higher number of OCD- diagnosed individuals in the upper extreme, future work by our group is exploring the effect of non-hoarding OC traits on our main hoarding finding. Finally, in a study of the TOCS by our group, we identified significant differences between parent- and self-reported OC trait scores, with parent-reported scores lower for most items (Park et al., 2016). For this reason, we corrected for respondent in each of our analyses. It is important to recognize, however, that the sensitivity and specificity were higher in parent-respondents than self-respondents and that most of our gathered information was based on parent report (Park et al., 2016).

78 3.5. Conclusion

Our paper highlights the importance of addressing phenotypic heterogeneity based on clinical symptomatology and based on sex. More specifically, our findings suggest that hoarding is distinct in its underlying serotonin gene variation, relative to other OC dimensions. Future studies are necessary to replicate our finding of association between serotonin genetic variation and hoarding traits. Additional studies are also required to verify that these trends go beyond the general population and occur in clinical settings. Although our findings were strongest for hoarding, we predict that different gene variants (serotonin genes and genes in other gene systems) are implicated in different OCD symptom dimensions and we, therefore, encourage prospective studies to use our approach to heterogeneity to build on our findings in independent, large datasets. Future studies should also examine additional sources of heterogeneity including gene-gene interaction and gene-environment interaction. Improved understanding of the relationship between genetic variants and specific OC traits will ultimately lead to more refined, targeted treatments for patients with OCD and related disorders.

79 4. STUDY 2

Serotonin system genes and hoarding with and without other obsessive-compulsive traits in a population-based, pediatric sample: A genetic association study

This chapter is modified from the following article and contains the entire article contents, in full:

Authors

80 Statement of Contributions

The authors contributed to the work presented in the chapter as follows:

Lauren Erdman – statistical analysis

Christie L. Burton – coauthor, editing

Laura S. Park – wet lab assistance

Annie Dupuis – data management, statistical insight

Janet Shan – data management

Tara Goodale – data and sample/participant management

S-M Shaheen – lab technician

Jennifer Crosbie – supervision

Russell J. Schachar – editing, supervision

Paul D. Arnold – editing, supervision

Conflicts of Interest

Russell J. Schachar consults for Highland Therapeutics, Purdue Pharma, and ehave, and is the

Toronto Dominion Bank Financial Group Chair in Child and Adolescent Psychiatry. Vanessa

M. Sinopoli, Lauren Erdman, Christie L. Burton, Laura S. Park, Annie Dupuis, Janet Shan,

Tara Goodale, S-M Shaheen, Jennifer Crosbie, and Paul D. Arnold have no conflicts of interest to declare.

Acknowledgements

This work was supported by the Canadian Institutes of Health Research (P.D.A., MOP-106573 and R.J.S., MOP-93696). Dr. Paul D. Arnold receives funding from the Alberta Innovates

Health Solutions (AIHS) Translational Health Chair in Child and Youth Mental Health.

81 Vanessa Sinopoli received funding from the Canadian Institutes of Health Research (CIHR)

Master's Award: Frederick Banting and Charles Best Canada Graduate Scholarships, Ontario

Graduate Scholarship (OGS), and the Hospital for Sick Children Restracomp Studentship.

The authors also wish to acknowledge the contributions of Ms. Clara Tam in study coordination, Mr. Bingbin Li for laboratory technical assistance, and the Ontario Science

Centre for collaborating in the recruitment of participants.

Abstract

Background: Hoarding, originally only considered an OCD symptom, is now categorized as a separate disorder in the Fifth Edition of the Diagnostic and Statistical Manual of Mental

Disorders (DSM-5). Here, we aimed to examine the distinctness of hoarding by studying the association between serotonergic genes and hoarding with and without other obsessive- compulsive (OC) traits in a general, pediatric population. We hypothesized that unique serotonin gene variants would be associated with hoarding alone.

Methods: We examined OC traits, including hoarding, in a total of 5213 pediatric participants in the community. We genotyped candidate serotonin genes (5-HTTLPR polymorphism in SLC6A4 for 2018 individuals and single nucleotide polymorphisms (SNPs) across genes SLC6A4, HTR2A, and HTR1B for 4711 individuals). In a previous study conducted by our group in the same sample, we examined OC trait dimensions and showed a significant association between 5-HTTLPR and hoarding in males. In this study, we examined

82 hoarding more closely by testing the association between serotonin gene variants and hoarding traits with and without other accompanying OC traits, separately in males and females.

Results: The [LG + S] variant in 5-HTTLPR was significantly associated with hoarding alone in males (odds ratio, OR, of 1.43; P-value of 0.009). There were no significant findings for 5-

HTTLPR in females. There were no significant findings in males or females after correction for multiple comparisons using SNP array data, but our top SNP findings suggested that variation downstream of HTR1B may also be implicated in hoarding alone in females.

Conclusions: Our results propose that specific serotonin gene variants are associated with hoarding traits alone, differing between sexes. Our top findings are in line with our former study, further suggesting that individuals with hoarding alone (without other OC traits present) were driving our previous results. This supports hoarding disorder’s new designation in the

DSM-5.

83 4.1. Introduction

Obsessive-compulsive disorder (OCD) is a psychiatric disorder characterized by intrusive and repetitive thoughts and behaviors and is debilitating in nature (American Psychiatric

Association [APA], 2013), with an estimated lifetime prevalence of 2-3% worldwide (Angst et al., 2004; Kessler et al., 2005; Murphy et al., 2013).

OCD is phenotypically heterogeneous. A meta-analysis of factor analyses of the Yale-Brown

Obsessive-Compulsive Scale (Y-BOCS) in adults and/or children with OCD reported four symptom dimensions: 1) Symmetry, 2) Forbidden thoughts, 3) Cleaning, and 4) Hoarding

(Bloch et al., 2008a). Based on mounting scientific evidence (Mataix-Cols & Pertusa, 2012), hoarding is now classified as a separate disorder in the Fifth Edition of the Diagnostic and

Statistical Manual of Mental Disorders (DSM-5), under the new category “Obsessive-

Compulsive and Related Disorders” (APA, 2013).

Hoarding disorder is estimated to affect 2-5% of adults (Samuels et al., 2008) and 2% of adolescents (Ivanov et al., 2013), with little known about its prevalence in children (Burton et al., 2015a). It is characterized by difficulty discarding acquired possessions, so as to cause significant clutter and/or financial strain and negatively impact quality of life for patients and their families, and often presents with excessive acquisition of materials (APA, 2013; Burton et al., 2015a). Less than 20% of adults with hoarding disorder have OCD, meaning over 80% of adults present with hoarding independent of OCD (Frost et al., 2011). Between 19 and 31% of patients with pediatric OCD also have hoarding symptoms (Storch et al., 2007, Sheppard et al., 2010; Samuels et al., 2014).

OCD has been identified as familial and heritable, with estimated heritability ranging from 27-

47% in adults and 45-65% in children (van Grootheest et al., 2005, Pauls et al., 2014).

Hoarding is also familial (Frost & Gross, 1993; Samuels et al., 2002; Samuels et al., 2007a;

Pertusa et al., 2008; Steketee et al., 2015) and heritable, with estimates in adult twins ranging from 36-49% (Iervolino et al., 2009; Taylor et al., 2010; Mathews et al., 2014). In adolescence to young adulthood, heritability was shown to range from 29-41% (decreasing with age), with authors noting substantially higher heritability in males versus females in the youngest age group (Ivanov et al., 2017). Among the symptom dimensions of OCD, hoarding was most influenced by unique genetic effects (Iervolino et al., 2011; Burton et al., 2018).

A number of candidate genes have been studied in OCD including SLC6A4 (solute carrier family 6 member 4), HTR2A, and HTR1B, respectively corresponding to the serotonin transporter (SERT) and serotonin receptors 5-HT2A and 5-HT1B (Taylor et al., 2013). Within the promoter of SLC6A4 lies a heavily researched polymorphism in OCD, 5-HTTLPR (5-HTT- linked polymorphic region). It exists as a long (L) variant or a short (S) variant (Heils et al.,

1996; Lesch et al., 1996). Within the L variant lies single nucleotide polymorphism (SNP) rs25531 (A or G). Thus, there are two resulting functional groups: LA, which yields greater

SERT expression and LG/S, which yields lower SERT expression (Hu et al., 2006). Overall, findings as to which variants across these candidate genes are associated with OCD have been inconsistent (Sinopoli et al., 2017). This may result from failure to adequately address underlying heterogeneity in OCD, such as presence or absence of hoarding symptoms. For example, some genetic and neuroimaging studies show that the underlying biology of hoarding differs from OCD (van Ameringen et al., 2014). Hoarding symptoms may also differ

85 in their response to the most commonly prescribed medications for OCD, serotonin reuptake inhibitors (SRIs) (Murphy et al., 2004). Patients with OCD and hoarding symptoms respond more poorly to SRIs compared to OCD patients without hoarding symptoms (Bloch et al.,

2014), supporting the notion of unique underlying biology. One possibility for this difference in treatment response may be genetic differences between patients with or without hoarding symptoms.

In a recent study, our group examined serotonin candidate gene variants for their association with obsessive-compulsive (OC) trait dimensions in a community pediatric sample and tested whether these associations differed between males and females. Our main finding was that

[LG + S] (versus LA) was significantly associated with hoarding in males. We also identified a trend for variation downstream of HTR1B to be associated with hoarding in females (Sinopoli et al., 2019b). Given that recent comprehensive meta-analyses found an association between

LA, as opposed to [LG + S], and OCD (Taylor, 2013; Taylor 2016), our results suggest that genetic heterogeneity in the serotonin system may be one factor that underlies the phenotypic heterogeneity observed between OC dimensions and hoarding. What we do not yet know is if hoarding traits alone are driving our previous serotonergic findings. If so, this would suggest that hoarding has genetic correlates that are distinct from OCD and, more specifically, that there are unique serotonin system gene variants that are associated with hoarding alone and not with other OC traits.

In this study, we aimed to further characterize the relationship between hoarding and candidate serotonin gene variants in the same large, pediatric, population-based sample (Sinopoli et al.,

2019b). We aimed to extend the results of the first study (Sinopoli et al., 2019b) and identify

86 if different serotonin system genes variants are associated with each of the following more homogenous, trait-based subgroups: hoarding traits with other OC traits, hoarding traits without other OC traits, and OC traits without hoarding traits. Given that hoarding most often presents independently from OCD and studies suggest biological distinctness of hoarding

(Snowdon et al., 2012; van Ameringen et al., 2014), we hypothesized that our previously noted findings, whereby unique serotonergic gene variants were associated with hoarding (as one of the OC trait dimensions), were being driven by individuals with hoarding traits alone.

We therefore expected to identify the same genetic variants to be associated with the hoarding only group and not with the hoarding plus OC group or the OC only group. We also hypothesized that our findings would differ between sex groups as seen previously (Sinopoli et al., 2019b).

4.2. Methods

4.2.1. Participants

As explained in our previous study (Sinopoli et al., 2019b), a total of 17,263 children and adolescents, ranging in age from 6-18, were recruited from and evaluated at the Ontario

Science Centre in Toronto, Canada for the Thoughts, Actions and Genes (TAG) project further described elsewhere (Crosbie et al., 2013; Burton et al., 2016; Park et al., 2016). Informed consent (and assent where applicable) was acquired for participants and study approval was obtained through the Research Ethics Board at the Hospital for Sick Children. Complete demographic information and questionnaires were available for 16,718 participants, whereby information was provided via parent-respondent (N=13,680) or self-respondent (N=3,038).

87 Our current study used the same overall sample of individuals used in our previous study

(Sinopoli et al., 2019b) to further examine hoarding traits in the community. As was the case for our previous study, two smaller, partially overlapping groups of participants were gathered from our initial, overall, population-based sample. Selected individuals had 4 grandparents of

European descent. The first group included 2100 individuals who had DNA available to directly genotype 5-HTTLPR in SLC6A4. The second group included 4810 individuals who had genome-wide association study (GWAS) data available and from which SNPs in our candidate genes were obtained.

4.2.2. Hoarding/obsessive-compulsive features

As described previously (Sinopoli et al., 2019b), participants or their parents completed the

Toronto Obsessive Compulsive Scale (TOCS) which measures OC traits (Park et al., 2016).

Each of the 21 items was scored on a scale of -3 to +3, whereby -3 means the participant performs the behavior far less often than his or her peers and +3 means the participant performs the behavior far more often than his or her peers. The TOCS factors into 6 dimensions: Cleaning/Contamination, Counting/Checking, Symmetry/Ordering, Superstition,

Rumination, and Hoarding (Burton et al., 2018). Two items directly pertained to hoarding traits: one examining excessive acquisition of useless objects and another examining difficulty discarding things. We dichotomized each TOCS item, with a score of ≥ 2 indicating presence of the trait and with any other score indicating absence of the trait. We assigned “affected” participants into the following groups, separately for males and females: 1) Individuals with at least one hoarding trait and at least one other non-hoarding OC trait (Hoarding Plus OC group), 2) Individuals with at least one hoarding trait and without any other non-hoarding OC

88 traits (Hoarding Only group), 3) Individuals with no hoarding traits and at least one OC trait that is not a hoarding trait (OC Only group). Our control group included all individuals with no hoarding or non-hoarding OC traits.

4.2.3. DNA collection and extraction

Saliva was collected from each participant using Oragene×DNA (OG-500) kits and DNA was extracted using the recommended protocol (DNA Genotek). In addition to the protocol, we centrifuged the samples at 10,000 RPM for 10 min more to remove precipitated carbohydrates.

We quantified the DNA using the Quant-iT™ PicoGreen® dsDNA Assay Kit (Invitrogen).

To ensure adequate DNA quality, samples were run on agarose gels before being used for the microarrays. DNA samples that had a concentration below 60ng/µl (if genotyped on the

HumanCoreExome-12 v1.0 microarray (Illumina)) and DNA samples of poor quality were excluded.

4.2.4. Selection of candidate genes

As in our previous study (Sinopoli et al., 2019b), we examined candidate serotonin genes

SLC6A4, HTR2A, and HTR1B based on their inclusion in a recent OCD meta-analysis and within which lie polymorphisms that have the best evidence of association with OCD (Taylor,

2013).

89 4.2.5. 5-HTTLPR

4.2.5.1. Direct genotyping

5-HTTLPR variation is not captured using microarrays and so we directly genotyped this polymorphism as detailed in our previous study, carried out with The Centre for Applied

Genomics (TCAG) at the Hospital for Sick Children in Toronto (Sinopoli et al., 2019b). In brief, genotyping this region involved two steps (Wendland et al., 2006). First, to identify variants as L or S, 5-HTTLPR was PCR amplified using forward primer 5'-

ATGCCAGCACCTAACCCCTAATGT-3' (5’-labeled with HEX fluorescent dye) and reverse primer 5'-GGACCGCAAGGTGGGCGGGA-3' (Gelernter et al., 1997). Second, to identify A or G at rs25531, restriction enzyme MspI (Fermentas, Life Technologies) was used for digestion. Samples were run on an ABI3730XL genetic analyzer using the POP-7 polymer and Dye Set G5 (Applied Biosystems). The Peak Scanner Software v1.0 (Thermo Fisher

Scientific) was used to analyze results. 5-HTTLPR variants were identified as LA, LG, or S

(which includes the SA variant and the rare SG variant). Based on functionality, we sorted our alleles as follows: LA and [LG + S]. Alleles were treated as additive in their effect (Hu et al.,

2006). Standard quality control (QC) was implemented as previously described (Sinopoli et al., 2019b). Individuals were genotyped with a 97% completion rate and reported siblings and individuals of indeterminate ethnicity were removed. In our final, total 5-HTTLPR sample, we identified 12 individuals carrying the rare SG variant. We had a final N of 2018 after all QC.

90 4.2.5.2. Statistical analyses

Analyses were conducted (similar to what was done previously) using the program, R, version

3.0.1 (https://www.R-project.org). In separate analyses for male and for female subjects, and using an additive model, the expected number of tested alleles (0, 1, 2 [LG + S] alleles) was used in a logistic regression as a predictor of affected versus control status for each of the 3 trait groups. Age and questionnaire respondent type (parent or self) were used as covariates.

To account for multiple comparisons, the final significance threshold was a P-value < 0.017

(0.05 / 3 analyses) for both males and females.

4.2.6. Candidate gene SNPs

4.2.6.1. Genotyping

Candidate gene SNPs across SLC6A4, HTR2A, and HTR1B were obtained as previously described (Sinopoli et al., 2019b). In short, genotype data was collected using the

HumanCoreExome-12 v1.0 microarray (Illumina) and QC included removal of SNPs with a call rate < 0.97 and samples with sex misspecification and ambiguity. Only one member per family was kept for analyses (PI_HAT cutoff = 0.12) and European ethnicity was verified via principal component analysis (PCA) using EIGENSTRAT version 3.2.10 (Price et al., 2006) with HapMap populations CEU and TSI to ensure that our individuals’ genetic data clustered with Europeans. Another PCA was conducted to identify and remove outliers within our own population and an additional PCA was conducted without outliers, followed by formation of a

91 scree plot, to identify seven principal components (PCs) needed to control for population structure in our analyses. We had a final N of 4711 after QC.

Imputation was conducted across SLC6A4, HTR2A, and HTR1B along with ±50 kb flanking

DNA using SHAPEIT2 version 2.790 (Delaneau et al., 2014) and IMPUTE2 version 2.3.1

(Howie et al., 2009). 1000 Genomes Phase 3 reference data was used for pre-phasing and for imputation. SNPs were removed for low quality imputation (information score < 0.8) and

SNPs with a minor allele frequency (MAF) < 0.05 were excluded. Data were kept as dosage calls for the analyses. Final SNP counts across the 3 candidate genes are shown in Table 10 and are identical to those previously reported (Sinopoli et al., 2019b).

Table 10. SNP counts across 3 candidate genes, including all genotyped and imputed SNPs: Common SNPs (MAF ≥ 0.05) are shown for each gene and its flanking regions included in the analyses for males and in the analyses for females. Table obtained from Sinopoli et al. (2019b).

Gene Number of SNPs in Males Number of SNPs in Females SLC6A4 121 142 HTR2A 411 361 HTR1B 242 242

Abbreviations: SNP, single nucleotide polymorphism; SLC6A4, solute carrier family 6 member 4 (serotonin transporter gene); HTR2A, 5-HT2A receptor gene; HTR1B, 5-HT1B receptor gene.

4.2.6.2. Statistical analyses

R, version 3.0.1 (https://www.R-project.org), was used in a set of analyses similar to those previously conducted. Within male and female subjects separately, and using an additive model, logistic regressions were conducted for each SNP using the expected number of tested alleles (0, 1, 2) to predict affected versus control status for each of the 3 trait groups. Analyses were conducted using age, respondent, and the identified PCs (to control for population

92 structure) as covariates. The Genetic Type 1 error calculator (GEC) was used to account for dependent SNPs (Li et al., 2012) and to address multiple comparisons, resulting in a significance threshold P-value < 6.87e-5 (2.06e-4 / 3 analyses) for males and a significance threshold P-value < 6.77e-5 (2.03e-4 / 3 analyses) for females.

4.2.7. Sample size considerations

As in our previous study (Sinopoli et al., 2019b), the final sample size for 5-HTTLPR was smaller than the final sample size for our candidate gene SNPs (N of 2018 versus N of 4711).

The two groups share a 29% overlap (1516 individuals genotyped for and used in both the 5-

HTTLPR and candidate gene SNP groups). Given that less stringent statistical correction was required for the single variant versus the set of candidate gene SNPs and given that increased labor and cost was required to directly genotype 5-HTTLPR, the sample size for 5-HTTLPR was deemed sufficient even though smaller.

4.3. Results

4.3.1. 5-HTTLPR analyses

Demographics for 5-HTTLPR-genotyped individuals are in Table 11 and are identical to what was previously reported (Sinopoli et al., 2019b). As determined through the use of t-tests, there were no significant differences for age between affected males and male controls, or between affected females and female controls. There was a significant difference for respondent between affected males and male controls (absolute t-statistic = 3.09, P-value =

93 0.002), but not between affected females and female controls. Table 12 shows the affected versus control counts for each of the three trait groups in our 5-HTTLPR-genotyped individuals.

Table 11. 5-HTTLPR-genotyped individuals: Demographics and group characteristics are shown for total individuals combined, all affected individuals, and control individuals. Table obtained from Sinopoli et al. (2019b).

Total Individuals Male Female N 1050 968 Mean Age (SD) 10.7 (2.5) 11.3 (2.9) Age Range 6.1 - 17.9 6.3 - 17.9 Respondent: % Parent-report, % Self-report 90.1, 9.9 81.5, 18.5

Allele Frequency: LA, [LG + S] 0.52, 0.48 0.53, 0.47

94 Abbreviations: 5-HTTLPR, 5-HTT-linked polymorphic region; N, sample size; SD, standard deviation; OCD, obsessive-compulsive disorder; ADHD, attention-deficit/hyperactivity disorder; ASD, autism spectrum disorder; SRI, serotonin reuptake inhibitor; LA, long variant of 5-HTTLPR with A allele at SNP rs25531; LG, long variant of 5-HTTLPR with G allele at SNP rs25531; S, short variant of 5-HTTLPR.

Table 12. Trait group counts for 5-HTTLPR-genotyped individuals: Affected versus control counts for each of the three trait groups.

Hoarding Plus OC Male Female Affected Individuals 240 240 Control Individuals 460 397

Hoarding Only Male Female Affected Individuals 147 140 Control Individuals 460 397

OC Only Male Female Affected Individuals 203 191 Control Individuals 460 397

Abbreviations: OC, obsessive-compulsive.

As seen in Figure 5, in males, and after correction for multiple comparisons, the [LG + S] allele was significantly associated with hoarding only (odds ratio, OR, of 1.43; P-value of

0.009). There were no significant differences for females.

Figure 5. Association between 5-HTTLPR and trait groups: Forest plots (males on left, females on right). Odds ratios for the [LG + S] allele are shown as colored points, each representing the effect size estimate for the allele in that trait group relative to controls. The confidence interval for each effect size estimate is represented by the length of each horizontal line. Significant finding is indicated by *. 5-HTTLPR, 5-HTT-linked polymorphic region; OC, obsessive-compulsive.

4.3.2. Candidate gene SNP analyses

Demographics for the candidate gene SNP individuals are in Table 13 and are identical to what was previously reported (Sinopoli et al., 2019b). As determined through the use of t- tests, there were no significant differences for age or for respondent between affected males and male controls, or between affected females and female controls. Table 14 shows the affected versus control counts for each of the three trait groups in our individuals genotyped for candidate gene SNPs.

96 Table 13. Individuals genotyped for candidate gene SNPs: Demographics and group characteristics are shown for total individuals combined, all affected individuals, and control individuals. Table obtained from Sinopoli et al. (2019b).

Total Individuals Male Female N 2439 2272 Mean Age (SD) 10.7 (2.6) 11.1 (2.9) Age Range 6.1 - 17.9 6.1 - 17.9 Respondent: % Parent-report, % Self-report 89.7, 10.3 81.0, 19.0

97 Table 14. Trait group counts for individuals genotyped for candidate gene SNPs: Affected versus control counts for each of the three trait groups.

Hoarding Plus OC Male Female Affected Individuals 311 345 Control Individuals 1414 1229

Hoarding Only Male Female Affected Individuals 154 155 Control Individuals 1414 1229

OC Only Male Female Affected Individuals 560 543 Control Individuals 1414 1229

Abbreviations: OC, obsessive-compulsive.

The top 20 SNP findings across the three genes and trait groups, for males and females, are shown in Table 15 and Table 16 respectively. For males, the top SNPs reported were varied across genes and associated trait groups (OR of 0.63 to 1.33; P-value of 7.35e-3 to 2.56e-2) and were not significant after correction for multiple comparisons. For females, the top SNPs reported all corresponded with HTR1B and were associated with hoarding only (OR of 1.64 to

1.68; P-value of 5.78e-4 to 9.50e-4). Our findings did not remain significant after correction for multiple comparisons.

98 Table 15. Association between serotonin gene SNPs and trait groups in males: Top 20 SNP findings.

SNP Gene Tested Allele MAF OR P-value Trait Group rs62416430 HTR1B G 0.065 0.63 7.35E-03 OC Only rs62416429 HTR1B T 0.066 0.65 9.25E-03 OC Only rs62416428 HTR1B G 0.066 0.65 9.26E-03 OC Only rs2143824 HTR1B T 0.146 0.76 1.09E-02 OC Only rs7214991 SLC6A4 G 0.373 0.73 1.13E-02 Hoarding Only rs1487971 SLC6A4 T 0.372 0.74 1.49E-02 Hoarding Only rs7215330 SLC6A4 C 0.372 0.74 1.52E-02 Hoarding Only rs6505167 SLC6A4 T 0.108 1.33 1.54E-02 OC Only rs2770301 HTR2A C 0.228 0.81 1.74E-02 OC Only rs6505169 SLC6A4 A 0.436 0.77 2.32E-02 Hoarding Only rs7329652 HTR2A G 0.074 0.72 2.39E-02 OC Only rs9896548 SLC6A4 G 0.431 0.76 2.40E-02 Hoarding Only rs7335941 HTR2A C 0.074 0.72 2.40E-02 OC Only rs7214248 SLC6A4 A 0.328 0.75 2.42E-02 Hoarding Only rs2025296 HTR2A G 0.097 0.71 2.44E-02 Hoarding Plus OC rs12945042 SLC6A4 T 0.321 0.75 2.46E-02 Hoarding Only rs7342921 SLC6A4 C 0.326 0.75 2.53E-02 Hoarding Only rs6505165 SLC6A4 C 0.434 0.77 2.53E-02 Hoarding Only rs3794806 SLC6A4 G 0.329 0.75 2.55E-02 Hoarding Only rs7208052 SLC6A4 C 0.327 0.75 2.56E-02 Hoarding Only

99 Table 16. Association between serotonin gene SNPs and trait groups in females: Top 20 SNP findings.

SNP Gene Tested Allele MAF OR P-value Trait Group rs1145827 HTR1B A 0.149 1.68 5.78E-04 Hoarding Only rs2207053 HTR1B C 0.149 1.67 6.23E-04 Hoarding Only rs1228806 HTR1B C 0.149 1.67 6.25E-04 Hoarding Only rs1777763 HTR1B T 0.150 1.67 6.33E-04 Hoarding Only rs1228805 HTR1B G 0.150 1.67 6.34E-04 Hoarding Only rs1777762 HTR1B T 0.150 1.67 6.34E-04 Hoarding Only rs2207056 HTR1B T 0.150 1.67 6.35E-04 Hoarding Only rs1228797 HTR1B T 0.150 1.67 6.35E-04 Hoarding Only rs1228798 HTR1B G 0.150 1.67 6.35E-04 Hoarding Only rs1228800 HTR1B G 0.150 1.67 6.35E-04 Hoarding Only rs1343334 HTR1B A 0.150 1.67 6.36E-04 Hoarding Only rs1343336 HTR1B T 0.150 1.67 6.37E-04 Hoarding Only rs2798528 HTR1B T 0.150 1.67 6.39E-04 Hoarding Only rs1738506 HTR1B A 0.150 1.67 6.39E-04 Hoarding Only rs2223832 HTR1B C 0.150 1.67 6.46E-04 Hoarding Only rs1228804 HTR1B G 0.152 1.66 6.78E-04 Hoarding Only rs2798529 HTR1B G 0.148 1.66 8.12E-04 Hoarding Only rs1228802 HTR1B A 0.149 1.64 9.47E-04 Hoarding Only rs59414600 HTR1B C 0.149 1.64 9.50E-04 Hoarding Only rs61295513 HTR1B T 0.149 1.64 9.50E-04 Hoarding Only

Abbreviations: SNP, single nucleotide polymorphism; MAF, minor allele frequency; OR, odds ratio; HTR1B, 5- HT1B receptor gene.

4.4. Discussion

Hoarding is phenotypically and biologically distinct from OCD (Snowdon et al., 2012; van

Ameringen et al., 2014). With hoarding now classified as a separate, but OCD-related disorder (APA, 2013), we sought to specifically examine serotonin system candidate genes in hoarding, using a sample of children and adolescents from the community to better capture individuals with hoarding traits who may not also have other OC traits. We previously examined six OC trait dimensions (Cleaning/Contamination, Counting/Checking,

100 Symmetry/Ordering, Superstition, Rumination, Hoarding) and our strongest findings were a significant sex-specific association between the [LG + S] allele of 5-HTTLPR and hoarding in males and a sex-specific trend for association between variation downstream of HTR1B and hoarding in females (Sinopoli et al., 2019b). What remained unanswered in our initial study was if our genetic findings in hoarding were unique to hoarding alone or if they were shared with other OC traits. For this reason, our current study aimed to further examine hoarding in this group to see if hoarding traits independent of other OC traits were driving our previous findings of associations with serotonergic gene variants in hoarding. This would further support the biological distinctness of hoarding from OCD by identifying distinct genetic correlates in hoarding.

We argued that hoarding without other OC traits would be associated with serotonin gene variants that would not be identified when hoarding presents with other OC traits. This is something we were unable to gather from our earlier study because individuals with hoarding alone and individuals with both hoarding and other OC traits were grouped together.

Consistent with our hypothesis, we found evidence to support that hoarding in the absence of other OC traits is associated with distinct serotonin gene variants, findings which are not present when examining other OC traits alone or hoarding plus OC traits. Furthermore, as in our previous study, genetic associations differed between sexes, which was also consistent with our hypothesis. Specifically, in males, we found a statistically significant association between [LG + S] and hoarding only and, in females, we identified a trending association between SNPs downstream of HTR1B and hoarding only. Since these were the same associations identified in our former paper (Sinopoli et al., 2019b), it appears individuals with hoarding only were driving our previous findings as suspected.

101

While our study identifies a statistically significant serotonin gene variant association with hoarding symptoms in the absence of OC symptoms, previous studies have studied the genetics of hoarding within the context of OCD. Samuels and colleagues (2007b) identified suggestive linkage of hoarding obsessions and/or compulsions to a region of chromosome 14

(at marker D14S588) in families with OCD, and also found significant linkage of OCD to a region of chromosome 14 (at marker C14S1937, nearby the first finding) in families with two or more hoarding relatives. Another study conducted in adult OCD patients with and without hoarding did not identify a significant association with 5-HTTLPR as we did, though the added variance of rs25531 was not considered in this study, nor did the study seek out patients with hoarding symptoms in the absence of OCD (Lochner et al., 2005).

Population-based studies, like ours, allow us to measure the full phenotypic range of traits that are widely distributed in the population as opposed to focusing on the subset of children and adolescents in the clinic. Examining trait-based measures allows us to parse out the effects of specific traits in the presence and absence of other traits, and to address heterogeneity. The

Research Domain Criteria (RDoC) project was launched by the National Institute of Mental

Health (NIMH) to encourage researchers to study the biological basis of simpler, behavioral traits underling more complex psychiatric disorders. This helps to address heterogeneity and is proposed as a more powerful approach to identifying genetic risk factors for psychiatric traits than the study of categorical diagnoses (NIMH, 2008). Our research underlines the importance of accounting for precise symptoms and comorbidity and suggests that unique serotonergic mechanisms may be driving hoarding traits in the absence of OCD. Our research also supports previous suggestions that different biological or genetic mechanisms may be

102 implicated in the pathogenesis of OCD depending on sex (Mattina et al., 2016). Overall, a failure to account for underlying heterogeneity may explain some of the past inconsistencies in genetic findings in OCD and related disorders to date (Murphy et al., 2013; Sinopoli et al.,

2017).

Like other psychiatric traits and disorders, hoarding is heterogeneous. One important source of heterogeneity is whether hoarding symptoms co-occur with symptoms of OCD. This led us to postulate that our previously identified serotonergic genetic association with hoarding was driven by one of the two subgroups composing the hoarding group in our initial study

(hoarding with and hoarding without comorbid OC traits). According to the DSM-5, hoarding disorder should only be diagnosed if the hoarding is not better explained as symptoms of another mental disorder, which includes OCD (APA, 2013). Hoarding behavior may, for example, result from an obsession in someone with OCD (Pertusa et al., 2010; Snowdon et al.,

2012; van Ameringen et al., 2014). OCD and hoarding disorder can also be diagnosed as comorbid disorders in the same individual (APA, 2013). One possible explanation of our findings is that our hoarding only group shared phenotypic features and genetic risk factors with hoarding disorder, however this remains speculative given that our population-based questionnaire does not enable us to diagnose DSM-5 disorders (which would require direct interview).

One of the limitations of our study was that our effective sample size was reduced by stratifying into multiple homogenous subgroups. This strategy was of course implemented to better study phenotypic heterogeneity in hoarding. The second limitation of our paper was

103 that our findings were limited to a subset of variants within serotonin candidate genes, so we cannot say whether or not other genes are implicated in hoarding traits.

4.5. Conclusion

This paper highlights the importance of addressing phenotypic heterogeneity based on clinical symptoms and sex. Our findings, although requiring replication in larger samples, suggest that hoarding traits, in the absence of OCD traits, are associated with variation in serotonin genes and that this association is both distinct from OCD and dependent on sex. Future studies should be conducted that replicate our serotonin-specific findings in hoarding, including studies in clinical samples of individuals with hoarding disorder as it is currently defined in the DSM-5 (APA, 2013). Studying both hoarding disorder (independent of OCD), OCD

(independent of hoarding disorder), and hoarding disorder comorbid with OCD in a larger sample will allow for a more extensive, genome-wide approach to discovery and will help identify if there are unique or shared genetic variants across these groups. Evidence has suggested that OCD patients with hoarding symptoms have a poorer response to conventional

OCD treatment when compared to OCD patients without hoarding symptoms (Bloch et al.,

2014). Better defining the underlying etiology of hoarding will help inform the design of more effective medications and help appropriately tailor current treatments to individual patients.

104 5. STUDY 3

Serotonin system gene variants and regional brain volume differences in pediatric OCD

This chapter is modified from the following article and contains the entire article contents, in full:

Copyright permission was obtained from journal publisher, Springer Nature, for full use of the article and its contents in this dissertation.

Authors

Vanessa M. Sinopolia,b, Lauren Erdmanb,c, Christie L. Burtonb,d, Phillip Eastere, Rageen Rajendramf, Gregory Baldwine, Kelli Petermane, Julie Costeb, S-M Shaheenb,h, Gregory L. Hannag, David R. Rosenberge, Paul D. Arnoldb,h,i,*

Affiliations iInstitute of Medical Science, University of Toronto, Toronto, ON, Canada bProgram in Genetics & Genome Biology, The Hospital for Sick Children, Toronto, ON, Canada cDepartment of Computer Science, University of Toronto, Toronto, ON, Canada dProgram in Neurosciences and Mental Health, The Hospital for Sick Children, Toronto, ON, Canada eDepartment of Psychiatry, Wayne State University School of Medicine, Detroit, MI, USA fUniversity of Toronto, Faculty of Medicine, Toronto, ON, Canada gDepartment of Psychiatry, University of Michigan, Ann Arbor, MI, USA hMathison Centre for Mental Health Research & Education, Hotchkiss Brain Institute, University of Calgary, Calgary, ON, Canada iDepartments of Psychiatry and Medical Genetics, Cumming School of Medicine, University of Calgary, Calgary, ON, Canada * Corresponding author at: Mathison Centre for Mental Health Research and Education, 4th floor, Teaching, Research and Wellness (TRW) Building, 3280 Hospital Dr NW, Calgary AB, T2N 4Z6, Canada; E-mail address: [email protected] (P.D. Arnold, MD, PhD, FRCPC)

105 Statement of Contributions

The authors contributed to the work presented in the chapter as follows:

Lauren Erdman – statistical analysis

Christie L. Burton – coauthor, editing

Phillip Easter – collaborator, sample collection, brain imaging data

Rageen Rajendram – genetic insight

Gregory Baldwin – collaborator, brain imaging

Kelli Peterman – collaborator, brain imaging

Julie Coste – wet lab assistance, organization

S-M Shaheen – laboratory technician

Gregory L. Hanna – collaborator, supervision

David R. Rosenberg – collaborator, supervision

Paul D. Arnold – editing, supervision

Conflict of Interest

The authors have no conflicts of interest to disclose.

Acknowledgements

This work was supported by the National Institutes of Health (NIH) grants R01-MH101493,

R01-MH085300, R01-MH085321, and R01-MH59299. Dr. Arnold receives support from the

Alberta Innovates Health Solutions (AIHS) Translational Health Chair in Child and Youth

Mental Health. Dr. Rosenberg receives support from the Lycaki-Young Fund, State of

Michigan, Miriam L. Hamburger Endowed Chair of Child Psychiatry, Paul and Anita Strauss

Endowment, Children’s Hospital of Michigan Foundation, and Donald and Mary Kosch

106 Foundation. Vanessa Sinopoli received funding from the Canadian Institutes of Health

Research (CIHR) Master's Award: Frederick Banting and Charles Best Canada Graduate

Scholarships, Ontario Graduate Scholarship (OGS), and the Hospital for Sick Children

Restracomp Studentship.

Abstract

Background: Obsessive-compulsive disorder (OCD) is phenotypically heterogeneous and genetically complex. This study aimed to reduce heterogeneity using structural brain imaging to study putative intermediate phenotypes for OCD. We hypothesized that select serotonin gene variants would differ in their relationship with brain volume in specific regions of the cortico-striato-thalamo-cortical (CSTC) circuits between OCD patients and controls.

Methods: In a total of 200 pediatric subjects, we genotyped candidate serotonin genes

(SLC6A4, HTR2A, HTR1B, and HTR2C) and conducted structural magnetic resonance imaging (sMRI) to measure regional brain volumes within CSTC circuits. In males and females separately, we first tested the association between serotonin gene variants and OCD and the effect of serotonin gene variants on brain volume irrespective of diagnosis. We then carried out a series of analyses to assess the effect of genotype-diagnosis interaction on brain volume.

Results: In females, but not in males, we identified a statistically significant genotype- diagnosis interaction for two single nucleotide polymorphisms (SNPs) in HTR2C, rs12860460

107 (interaction term estimate of 5.45 cc and interaction P-value of 9.70e-8) and rs12854485

(interaction term estimate of 4.28 cc and interaction P-value of 2.07e-6). The tested allele in each SNP was associated with decreased anterior cingulate cortex (ACC) volume in controls and with increased ACC volume in OCD patients.

Conclusions: Our findings suggest that, in females, sequence variation in HTR2C influences

ACC volume in pediatric OCD. The variants may contribute to differences in ACC volume and to OCD in a sex-specific manner when acting together with other genetic, biological, and/or environmental factors.

108 5.1. Introduction

Obsessive-compulsive disorder (OCD) is a neuropsychiatric disorder characterized by intrusive, repetitive thoughts and behaviors that are typically distressing to the patient

(National Institutes of Health [NIH], 2008; American Psychiatric Association [APA], 2013).

The disorder is heritable, as evidenced by twin studies and direct estimation of heritability from genome-wide common variant data (Davis et al., 2013; Pauls et al., 2014). Furthermore,

OCD is genetically complex and phenotypically heterogeneous (Arnold & Richter 2007;

Grados, 2010). Various approaches have been used to identify genetic variants associated with OCD. These include studies of variants within candidate serotonin system genes. The best-studied serotonin candidate genes for OCD include SLC6A4 (solute carrier family 6 member 4), HTR2A, HTR1B, and HTR2C, which correspond with the serotonin transporter, or

SERT, and serotonin receptors 5-HT2A, 5-HT1B, and 5-HT2C respectively (Taylor et al., 2013;

Sinopoli et al., 2017). Within SLC6A4, the most extensively studied polymorphism, 5-

HTTLPR (serotonin transporter-linked polymorphic region), lies within the promoter region of the gene (Heils et al., 1996; Lesch et al., 1996). The polymorphism exists in two major forms, one long (L) variant which consists of 16 sets of 20-23 bp tandem repeats and one short (S) variant which consists of 14 sets of 20-23 bp tandem repeats. In addition, an intrinsic A/G single nucleotide polymorphism (SNP), rs25531, occurs within the L variant and influences

SERT expression. Taken together, the 5-HTTLPR and rs25531 is a triallelic system (LA, LG and S variants) with the LA variant resulting in greater SERT expression compared with the S and LG variants. These variants appear to be additive in their effect (Hu et al., 2006).

109 Strong evidence supports an association between OCD and this commonly studied polymorphic region of SLC6A4 (Taylor, 2013; Walitza et al., 2014; Taylor, 2016; Sinopoli et al., 2017); however, findings have been somewhat mixed with respect to which variant of 5-

HTTLPR is associated with OCD. These mixed results may be because of heterogeneity in the

OCD phenotype that has not been adequately addressed in previous studies. Sources of heterogeneity include multiple overlapping symptom dimensions in OCD (Bloch et al., 2008a;

Alvarenga et al., 2015), subtypes of OCD that depend on factors such as age of onset or comorbidity with tic disorders (Grados, 2010; Leckman et al., 2010; APA, 2013; Williams et al., 2013) and sex differences (Alvarenga et al., 2015; Mak et al., 2015).

Biologically salient, quantitative measures that mediate genotype and phenotype, often called

“intermediate phenotypes”, are arguably less heterogeneous than clinical diagnostic syndromes like OCD (Meyer-Lindenberg & Weinberger, 2006). For example, brain differences measured by neuroimaging techniques like magnetic resonance imaging (MRI) may provide intermediate phenotypes that can be tested for their association with a particular set of candidate genes or used in a genome-wide association study (GWAS). There are two putative advantages of studying intermediate phenotypes compared with psychiatric diagnoses like OCD: 1) Reduced heterogeneity of such measures may lead to increased power to detect genetic association due to increased effect sizes (Leite et al., 2015), although recent imaging genetics studies suggest such effect sizes may be smaller than originally assumed (Carter et al., 2017), and 2) Imaging and other biologically salient measures may help delineate underlying biological mechanisms mediating gene variation and clinical symptoms (Meyer-

Lindenberg & Weinberger, 2006; Leite et al., 2015). OCD has been repeatedly linked to dysfunction in cortico-striato-thalamo-cortical (CSTC) circuits. Evidence includes volumetric

110 and functional abnormalities in various regions of CSTC circuits including the striatum

(caudate and putamen), thalamus, anterior cingulate cortex (ACC), and orbitofrontal cortex

(OFC), with some studies reporting differences in the directionality/magnitude of the effects between pediatric and adult OCD patients (MacMaster et al., 2010; Brem et al., 2012; Pauls et al., 2014; Boedhoe et al., 2017). Both pharmacotherapy with a serotonin reuptake inhibitor

(SRI) and cognitive behavioral therapy for OCD also alter metabolic activity in CSTC-related brain regions (Pauls et al., 2014).

Only three studies have been published which focus on the association between serotonin system gene variants and brain structure or function as measured by neuroimaging in individuals (predominantly adults) with OCD (Atmaca et al., 2011; Hesse et al., 2011; Honda et al., 2017). Upon examining the ACC, OFC, and thalamus using MRI, Atmaca et al. (2011) reported a significant genotype-diagnosis interaction whereby the effect of the 5-HTTLPR polymorphism on OFC volume was dependent on OCD diagnosis. S variant carriers showed a significantly smaller OFC volume in individuals with OCD, but not in healthy controls. Hesse et al. (2011) used positron emission tomography (PET) with a SERT-selective radiotracer and examined both 5-HTTLPR and STin2 VNTR (variable number of tandem repeats polymorphism in intron 2), another commonly studied polymorphism in SLC6A4. There was no significant effect of the interaction between presence of OCD and 5-HTTLPR genotype on

SERT availability. When examining OCD patients alone, however, the authors reported an overall trend towards increased SERT availability in patients with the genotype S/S of 5-

HTTLPR (reaching significance in the midbrain), a significant effect of STin2 VNTR genotype on SERT availability in the ACC and putamen, and a significant effect of the interaction between 5-HTTLPR and the STin2 VNTR on SERT availability in the ACC, putamen, and

111 hypothalamus. Neither the study by Atmaca et al. (2011) nor the study by Hesse et al. (2011) corrected for multiple comparisons. The most recent study examined association between 5-

HTTLPR and regional brain volumes measured using MRI, while accounting for the differential functionality imparted by rs25531. Authors reported a nominally significant genotype-diagnosis interaction whereby the effect of the 5-HTTLPR polymorphism on the right frontal pole was dependent on OCD diagnosis. They also noted a trend for OCD patients carrying LA to have reduced gray matter volume in the right frontal pole relative to healthy controls carrying LA (Honda et al., 2017). Overall, serotonin system gene variants may be associated with brain changes in OCD but, given the limited number of studies (which used different imaging modalities and different genotyping methods), further and more extensive studies are required.

To date, no studies examining serotonin system gene variants and their relationship to potential neuroimaging phenotypes in OCD have been conducted in pediatric samples. Early- onset OCD (onset typically in childhood or early adolescence) is considered a distinct subtype that differs from late-onset OCD (onset typically in late adolescence or early adulthood) with regards to a number of demographic and clinical features (MacMaster, 2010). To reduce heterogeneity, we have focused on children and adolescents with OCD (therefore predominantly capturing individuals with early-onset OCD) (Mattina & Steiner, 2016). There is also evidence from a recent meta-analysis suggesting that serotonin system gene variants differ in their association with OCD in males versus in females. Specifically, the S variant may be more associated with OCD in females than in males (Mak et al., 2015). To further reduce heterogeneity in our study, we therefore also chose to stratify by sex (Mattina &

112 Steiner, 2016). This decision is further supported by the fact that one of our genes of interest

(HTR2C) is located on chromosome X and may therefore be associated with sex differences.

In our study, we examined selected genetic variants in four candidate serotonin system genes,

SLC6A4, HTR2A, HTR1B, and HTR2C. Using brain volume in regions previously implicated in OCD, we aimed to identify associations between gene variants and neuroimaging phenotypes, specifically volumes of the ACC and other brain regions that have been implicated in OCD. We focused on children and adolescents, and we examined males and females separately to reduce heterogeneity. If the serotonin system is involved in OCD pathogenesis and if dysfunctional CSTC circuitry is involved in OCD pathogenesis, we hypothesized that we would find serotonin gene variants that differ in their association with brain volume in OCD patients versus in controls. Furthermore, given previously reported sex differences in OCD (Yücel et al., 2008; Mattina & Steiner, 2016), we hypothesized that our findings would differ between males and females in our pediatric population.

5.2. Materials and methods

5.2.1. Subjects

A total of 411 participants 6-19 years of age were initially recruited at two sites, which included Wayne State University in Detroit, Michigan and the University of Michigan in Ann

Arbor, Michigan, as part of a continuing collaborative study (Arnold et al., 2009a; Wu et al.,

2013). Written and informed consent and assent was obtained for all participants and, when applicable, from their parents. Each site received study approval from its respective Human

113 Investigation Committee prior to recruitment of participants into the study. OCD participants were required to have a lifetime OCD diagnosis and a current, minimum total score of 10 on the Children’s Yale-Brown Obsessive-Compulsive Scale (CY-BOCS) (Scahill et al., 1997).

OCD cases and healthy controls were assessed for lifetime diagnosis of Axis I psychiatric disorders, using DSM-IV-TR criteria (APA, 2000) based on information obtained through two semi-structured interviews, the Schedule for Affective Disorder and Schizophrenia for School-

Age Children-Present and Lifetime Version (K-SADS-PL) (Kaufman et al., 1997) and the pediatric version of the Schedule for Obsessive-Compulsive and Other Behavioral Syndromes

(SOCOBS) (Hanna, 2010), administered to both the participant and/or both parents.

Composite diagnoses were based on agreement between the two interviews regarding 1) specific diagnoses and 2) level of diagnosis-related interference. OCD cases and healthy controls were also assessed using the Pubertal Development Scale (PDS) (adapted from

Petersen et al., 1988).

For both OCD cases and healthy controls, participants were excluded if neurological disorders, such as seizure disorder, were identified. Participants were also excluded if they had a history of head injury with sustained loss of consciousness, chronic medical illness, history of substance abuse/dependence, history of psychosis, bipolar disorder, primary or recurrent major depressive disorder (MDD) (unless onset of MDD was after the onset of OCD), history of conduct disorder, an IQ less than or equal to 80, lifetime diagnosis of autism spectrum disorder (ASD) (including a score of 15 or higher on the Social Communication

Questionnaire, lifetime version) (Rutter et al., 2003), lifetime diagnosis of eating disorder, if they were adopted, or if they were living away from both biological parents at the time of the study. In addition to the exclusions above, healthy controls were also screened for lifetime

114 diagnosis of Axis I psychiatric disorders of any kind (APA, 2000). For this study, we included only unrelated participants and only participants with both parents of European Caucasian descent to minimize possible population stratification.

5.2.2. Imaging

All participants had T1-weighted structural magnetic resonance imaging (sMRI) data collected at the Children’s Hospital of Michigan Imaging Center. Detailed methods are described elsewhere (Szeszko et al., 2004; Arnold et al., 2009a; Wu et al., 2013). Briefly, a Sigma 3.0-

Tesla unit (Horizon HDX software, General Electric Medical Systems, Milwaukee, WI) was used to gather four separate contiguous three-dimensional (3D) MRI volumes. A 3D magnetization-prepared 180 degrees radio-frequency pulses and rapid gradient-echo (MP

RAGE) sequence was utilized. Acquisition parameters included echo time of 1.6 ms, repetition time of 2,200 ms, acquisition matrix of 512x512x204, pixel dimensions of

0.5x0.5x0.8 mm3, field of view 256x256 mm2, 204 axial slices, flip angle of 13 degrees, and inversion times of 776 ms, 780 ms, 794 ms, and 808 ms for each respective 3D MRI volume.

All MRIs were reviewed by a pediatric neuroradiologist to rule out clinically significant abnormalities and magnetic field inhomogeneities. A manual tracing technique was then carried out by trained operators, blind to case-control status, using BRAINS2 image processing software (Magnotta et al., 2002) to measure striatum, caudate, putamen, and thalamus, and using Analyze Direct 11.0 (AnalyzeDirect, Inc., Overland Park, KS) for the

ACC and the OFC. This measured regional brain volumes in select structures with a priori evidence of their involvement in OCD, along with the total intracranial volume (ICV) for each subject, in cubic centimeters (cc). Overall, the brain regions we examined were the caudate,

115 the putamen, the striatum (caudate + putamen combined), the thalamus, the ACC, and the

OFC. As a result, a total of six regions of interest (ROIs) were examined, with respective parameters for the brain regions outlined further in our past studies (Szeszko et al., 2004;

Arnold et al., 2009a; Wu et al., 2013). In this study, we combined left and right hemisphere and used the total volume for each ROI and we report intraclass correlation coefficient (ICC) values ranging from 0.970-0.999.

5.2.3. DNA collection and extraction

Either saliva or blood was collected from each participant. Oragene•DNA collection kits were used to extract DNA from saliva using the accompanying, recommended protocol (DNA

Genotek). An Autopure LS automated DNA extractor (QIAGEN) was used to extract whole

EDTA blood using Puregene chemistry (Gentra). Saliva DNA was quantified using the

Quant-iT™ PicoGreen® dsDNA Assay Kit (Invitrogen) and blood DNA was quantified using an FLx800 Fluorescence Reader (Biotek) and Hoechst 33258 dye (Sigma). DNA quality was verified using gel electrophoresis. Samples with DNA concentrations below 60ng/µl (if genotyped on the HumanOmni2.5-4 v1.0 or HumanCoreExome-12 v1.0 microarrays

(Illumina)) or samples with poor DNA quality were excluded.

5.2.4. Selection of candidate genes

The serotonin system has been the focus of numerous candidate gene studies in OCD, given the efficacy of SRIs in treating the disorder (Sinopoli et al., 2017). We therefore examined variants across serotonin system genes most commonly studied in OCD including SLC6A4,

116 HTR2A, HTR1B, and HTR2C (Taylor, 2013) to further understand the relationship of these genes with neuroimaging and OCD. Evidence suggests that the most promising of these genes in OCD are SLC6A4 (and more specifically polymorphism 5-HTTLPR) and HTR2A. There is more limited evidence to suggest the involvement of HTR1B in OCD, with studies indicating that this gene may be specific to certain OCD subgroups (Taylor et al., 2013; Sinopoli et al.,

2017). Though there is the least amount of evidence supporting the role of HTR2C in OCD, this gene has been implicated in a number of psychiatric disorders and is of interest to our structural neuroimaging study given that it is almost exclusively expressed in the central nervous system (Werry et al., 2008; Fagerberg et al., 2014).

5.2.5. Genotyping

5.2.5.1. 5-HTTLPR

Given that variation in 5-HTTLPR is not captured using microarrays, we directly genotyped this region of SLC6A4 in a two-step process (Wendland et al., 2006) as previously outlined

(Sinopoli et al., 2019b). This was carried out with The Centre for Applied Genomics (TCAG) at the Hospital for Sick Children (Toronto, Canada). First, we amplified 5-HTTLPR using forward primer 5'-ATGCCAGCACCTAACCCCTAATGT-3' (5’-labeled with HEX fluorescent dye) and reverse primer 5'-GGACCGCAAGGTGGGCGGGA-3' (Gelernter et al.,

1997). Variants were identified as L or S based on variant length. Second, DNA was digested at the rs25531 SNP location using restriction enzyme MspI (Fermentas, Life Technologies), which only cuts when G is present but not when A is present. Peak Scanner Software v1.0

(Thermo Fisher Scientific) was used to analyze results. Depending on the outcomes of the two

117 steps, variants were identified as LA, LG, or S. Based on their known functionality, we further categorized variants into two allelic groups: LA and [LG + S] which yield increased or decreased SERT expression respectively (Hu et al., 2006).

Quality control (QC) was applied using the standard settings in the Peak Scanner Software v1.0 (Thermo Fisher Scientific). Individuals were genotyped for 5-HTTLPR with a 95% completion rate. Reported siblings were removed. In our final sample, 1 5-HTTLPR- genotyped individual was carrying the previously noted rare SG variant (Wendland et al.,

2006; Voyiaziakis et al., 2011). After all QC and removal of ensuing individuals, we had a final N of 160 individuals with the 5-HTTLPR polymorphism genotyped (77 OCD cases and

83 healthy controls).

5.2.5.2. SNPs across SLC6A4, HTR2A, HTR1B, and HTR2C

Samples were originally genotyped to gather GWAS data for our continuing collaborative study (Arnold et al., 2009a; Wu et al., 2013) on the HumanOmni2.5-4 v1.0 or

HumanCoreExome-12 v1.0 microarrays (Illumina). We focused on data from SNPs across

SLC6A4, HTR2A, HTR1B, and HTR2C (UCSC Genome Browser, hg19).

SNPs with call rate < 0.97 were excluded. Samples were excluded on the basis of sex misspecification and ambiguity and cryptic relatedness of half sibling or above. We verified

European ethnicity via principal components analysis (PCA) using EIGENSTRAT version

3.2.10 (Price et al., 2006) and using HapMap populations CEU and TSI to ensure that the genetic data of our participants clustered with that of Europeans, relative to other ethnicities,

118 and then via another PCA to identify and remove outliers within our own population. An additional PCA was run without outliers. A Tracy-Widom test of the resulting principal components (PCs) showed no significant PCs, therefore none were retained to control for population structure in our analyses. After all quality control and removal of ensuing individuals, we had a final N of 154 individuals with SNPs genotyped across our four genes of interest (74 OCD cases and 80 healthy controls).

Chromosomes 6, 13, 17 and X were pre-phased using SHAPEIT2 version 2.790 (Delaneau et al., 2014). SNP data was imputed using IMPUTE2 version 2.3.1 (Howie et al., 2009) for each of the candidate genes, including a 50kb flanking region on either side of each gene. 1000

Genomes Phase 3 reference data was used for both pre-phasing and imputation. The non- pseudoautosomal region from 1000 Genomes chromosome X was used for chromosome X pre-phasing and imputation. After imputation, SNPs were removed for low quality imputation

(information score < 0.8). SNPs with MAF ≥ 0.05 were kept in the sample. Data were kept as dosage calls for all analyses. Table 17 shows SNP counts across SLC6A4, HTR2A, HTR1B, and HTR2C.

Table 17. SNP counts across 4 candidate genes: All common genotyped and imputed SNPs (MAF ≥ 0.05) are shown for each gene, and its flanking regions, included in the analyses for males and in the analyses for females.

Gene Number of SNPs in Males Number of SNPs in Females SLC6A4 117 102 HTR2A 301 313 HTR1B 241 235 HTR2C 401 482

Abbreviations: SNP, single nucleotide polymorphism; SLC6A4, solute carrier family 6 member 4 (serotonin transporter gene); HTR2A, 5-HT2A receptor gene; HTR1B, 5-HT1B receptor gene; HTR2C, 5-HT2C receptor gene.

119 5.2.6. Statistical analyses

T-tests were used to analyze demographics for statistically significant differences in age, in total ICV, and in PDS for all male cases versus male controls and for all female cases versus female controls. The following analyses were conducted for 5-HTTLPR and for all SNPs across SLC6A4, HTR2A, HTR1B, and HTR2C, using the program, R, Version 3.0.1

(https://www.R-project.org).

5.2.6.1. Assessing the association of serotonin gene variants with OCD

Logistic regressions were conducted in males and in females separately, using the expected number of tested alleles for each SNP (designated [LG + S] for 5-HTTLPR) against OCD case versus control status, in an additive model with age included as a covariate. The Genetic Type

1 error calculator (GEC) developed to account for dependent SNPs (Li et al., 2012) was used to calculate the P-value thresholds for the male and female analyses in the SNPs. The resulting significance P-value threshold for SNP analyses was 2.23e-4 for males and 2.15e-4 for females.

5.2.6.2. Assessing the effect of serotonin gene variants on brain volume

A linear regression was fit for each SNP using the expected number of tested alleles

(designated [LG + S] for 5-HTTLPR) against volume for each ROI and across the combined set of cases and controls (i.e., assessing the effect of SNP on brain volume irrespective of OCD diagnosis), in males and in females separately. An additive model was used including age and

120 total ICV as covariates. Analyses involving sMRI data used the GEC-generated values that were then corrected for the number of brain regions under investigation. This resulted in a brain region-adjusted significance P-value threshold of 3.72e-5 (2.23e-4 to account for dependent SNPs / 6 ROIs) for SNP analyses in males and of 3.58e-5 (2.15e-4 to account for dependent SNPs / 6 ROIs) for SNP analyses in females.

5.2.6.3. Assessing the effect of genotype-diagnosis interaction on brain volume

An interaction model was then fit for all individuals (cases and controls) with a main effect for both SNP (tested allele) and OCD (though we only report on SNP main effect, given our gene- driven hypotheses), along with an interaction term of OCD*[tested allele]. This allowed us to assess whether OCD status significantly changed the association of the tested allele with brain region volume (having corrected for age and total ICV by including these variables as covariates in the model). Brain region-adjusted significance P-value thresholds were used to assess significance in this model.

5.3. Results

Demographic characteristics for all 200 individuals in the study are reported in Table 18.

There were no significant differences to report in age, in total ICV, or in PDS between total male cases and total male controls. Between total female cases and total female controls, there was a significant difference in age (absolute t-statistic = 2.3, P-value = 0.02) and in PDS

(absolute t-statistic = 2.2, P-value = 0.03).

121 Table 18. Patient and control demographics: Group characteristics for all individuals combined, for OCD cases, and for controls.

All Individuals Males Females Total N 97 103 N with 5-HTTLPR Genotyped 75 85 N with SNPs Genotyped Across SLC6A4, 75 79 HTR2A, HTR1B, and HTR2C

OCD Cases Males Females Total N 46 51 Mean Age (SD) 13.2 (3.1) 13.6 (3.8) Age Range 6.5 - < 20.0 6.2 - < 20.0 Mean Total ICV (cc) (SD) 1684.4 (152.7) 1534.8 (133.3) Mean PDS Score (SD) 11.6 (3.4) 12.3 (5.6) PDS Score Range 5 - 17 1 - 19 Mean CY-BOCS Score (SD) 27.0 (6.3) 28.4 (7.0) CY-BOCS Score Range 13 - 37 10 - 39 MDD, N (%) 2 (4.3) 6 (11.8) ADHD, N (%) 9 (19.6) 11 (21.6) Anxietyi, N (%) 20 (43.5) 26 (51.0) Ticsii, N (%) 16 (34.8) 8 (15.7)

Controls Males Females Total N 51 52 Mean Age (SD) 14.2 (3.4) 15.4 (3.8) Age Range 6.2 - 19.9 6.8 - 19.8 Mean Total ICV (cc) (SD) 1676.6 (135.1) 1542.7 (120.9) Mean PDS Score (SD) 12.0 (3.5) 13.9 (3.4) PDS Score Range 4 - 18 5 - 19 iIncludes Separation Anxiety Disorder, Panic Disorder (with or without Agoraphobia Specified), Specific Phobia, Social Phobia, Generalized Anxiety Disorder (GAD) iiIncludes Tourette’s Disorder, Chronic Motor Tic Disorder, Chronic Vocal Tic Disorder, Transient Tic Disorder (Motor and/or Vocal), Lifetime History of Motor/Vocal Tic

Abbreviations: N, sample size; 5-HTTLPR, serotonin transporter-linked polymorphic region; SNP, single nucleotide polymorphism; SLC6A4, solute carrier family 6 member 4 (serotonin transporter gene); HTR2A, 5-

122 HT2A receptor gene; HTR1B, 5-HT1B receptor gene; HTR2C, 5-HT2C receptor gene; OCD, obsessive-compulsive disorder; SD, standard deviation; ICV, intracranial volume; PDS, Pubertal Development Scale; cc, cubic centimeters; CY-BOCS, Children’s Yale-Brown Obsessive-Compulsive Scale; MDD, major depressive disorder; ADHD, attention-deficit/hyperactivity disorder.

5.3.1. Assessing the association of serotonin gene variants with OCD

There was no significant association between 5-HTTLPR and OCD in males or in females. In the SNP analyses, the top 38 SNPs for OCD in males all corresponded with gene HTR1B

(odds ratio, OR, ranging from 0.37 to 0.43; P-value ranging from 1.27e-2 to 1.98e-2) and 24 of the 25 top SNPs for OCD in females all corresponded with gene SLC6A4 (OR ranging from

0.37 to 3.35; P-value ranging from 1.37e-3 to 4.89e-3). The SNP findings did not remain significant after correction for multiple comparisons.

5.3.2. Assessing the effect of serotonin gene variants on brain volume

There were no significant associations between 5-HTTLPR and brain volume for any of the

ROIs in males or in females. For the candidate gene SNPs, in males, the top 20 SNPs were all associated with reduced caudate volume, where 17 of the SNPs corresponded with HTR2A

(SNP effect ranging from -0.17 to -0.15 cc; P-value ranging from 2.68e-4 to 6.53e-4) and 3 of the SNPs corresponded with SLC6A4 (SNP effect of -0.15 cc; P-value of 6.23e-4) (Table 19).

In females, the top 20 SNPs were all associated with reduced volume of the striatum, where 8 of the SNPs corresponded with HTR1B (SNP effect ranging from -0.18 to -0.15 cc; P-value ranging from 2.70e-4 to 6.47e-4) and 8 of the SNPs corresponded with HTR2C (SNP effect ranging from -0.17 to -0.15 cc; P-value ranging from 3.64e-4 to 6.49e-4) (Table 20). None of the SNP findings remained significant after correction for multiple comparisons.

123

Table 19. Effect of serotonin gene variants on brain volume in males: Top 20 SNP findings.

Brain Tested SNP Gene MAF SNP Effect (cc) P-value Region Allele Caudate rs7330368 HTR2A A 0.22 -0.16 2.68E-04 Caudate rs9316235 HTR2A A 0.23 -0.17 2.92E-04 Caudate rs2770297 HTR2A T 0.26 -0.16 3.87E-04 Caudate rs61948331 HTR2A A 0.13 -0.16 4.56E-04 Caudate rs731779 HTR2A C 0.11 -0.16 4.60E-04 Caudate rs9562687 HTR2A A 0.10 -0.16 4.60E-04 Caudate rs9562688 HTR2A A 0.10 -0.16 4.60E-04 Caudate rs61948356 HTR2A A 0.14 -0.15 5.11E-04 Caudate rs61948354 HTR2A C 0.14 -0.15 5.13E-04 Caudate rs17287961 HTR2A T 0.07 -0.16 5.49E-04 Caudate rs61948357 HTR2A C 0.14 -0.15 5.61E-04 Caudate rs73477746 HTR2A G 0.14 -0.15 5.62E-04 Caudate rs8078900 SLC6A4 A 0.25 -0.15 6.23E-04 Caudate rs33980254 SLC6A4 T 0.25 -0.15 6.23E-04 Caudate rs73268099 SLC6A4 G 0.25 -0.15 6.23E-04 Caudate rs61647933 HTR2A C 0.17 -0.15 6.38E-04 Caudate rs61948335 HTR2A G 0.11 -0.15 6.42E-04 Caudate rs61948318 HTR2A T 0.13 -0.16 6.44E-04 Caudate rs9534501 HTR2A T 0.13 -0.16 6.53E-04 Caudate rs9567743 HTR2A C 0.13 -0.16 6.53E-04

Abbreviations: SNP, single nucleotide polymorphism; MAF, minor allele frequency; cc, cubic centimeters; SLC6A4, solute carrier family 6 member 4 (serotonin transporter gene); HTR2A, 5-HT2A receptor gene.

124 Table 20. Effect of serotonin gene variants on brain volume in females: Top 20 SNP findings.

Brain Tested SNP Gene MAF SNP Effect (cc) P-value Region Allele Striatum rs9343614 HTR1B G 0.32 -0.18 2.70E-04 Striatum rs9361228 HTR1B A 0.30 -0.18 2.96E-04 Striatum rs540285 HTR2C T 0.27 -0.17 3.64E-04 Striatum rs12836749 HTR2C A 0.13 -0.16 4.07E-04 Striatum rs1777761 HTR1B G 0.25 -0.16 5.12E-04 Striatum rs4543330 HTR1B C 0.08 -0.15 5.18E-04 Striatum rs2798949 HTR1B G 0.24 -0.16 5.22E-04 Striatum rs1891752 HTR1B G 0.25 -0.16 5.22E-04 Striatum rs199602838 HTR2C T 0.11 -0.15 5.39E-04 Striatum rs184155834 HTR1B G 0.08 -0.15 5.48E-04 Striatum rs9902340 SLC6A4 G 0.44 -0.16 5.72E-04 Striatum rs12861270 HTR2C G 0.10 -0.16 6.05E-04 Striatum rs11797988 HTR2C A 0.10 -0.16 6.34E-04 Striatum rs9567745 HTR2A C 0.06 -0.15 6.41E-04 Striatum rs3945573 HTR2A A 0.06 -0.15 6.41E-04 Striatum rs2226183 HTR1B A 0.07 -0.15 6.47E-04 Striatum rs3764450 SLC6A4 A 0.26 -0.16 6.48E-04 Striatum rs12855533 HTR2C A 0.18 -0.16 6.49E-04 Striatum rs12847225 HTR2C C 0.18 -0.16 6.49E-04 Striatum rs12842363 HTR2C A 0.18 -0.16 6.49E-04

5.3.3. Assessing the effect of genotype-diagnosis interaction on brain volume

When we assessed whether OCD diagnosis significantly changed the relationship between 5-

HTTLPR and brain volume, there were no significant findings to report for males or for females. When assessing whether OCD diagnosis significantly changed the relationship between SNPs across our four candidate genes and brain volume in males (Table 21), there were no findings that remained significant after correction for multiple comparisons. In females (Table 22), OCD diagnosis significantly changed the relationship between two

125 HTR2C SNPs (rs12860460 and rs12854485) and ACC volume. Both rs12860460 and rs12854485 were associated with a significant SNP main effect in the interaction model (SNP main effect of -4.37 cc and -3.26 cc respectively; SNP P-value of 1.75e-7 and 5.45e-6 respectively). There was a significant genotype-diagnosis interaction for both SNPs

(rs12860460, interaction term estimate of 5.45 cc and interaction P-value of 9.70e-8; rs12854485, interaction term estimate of 4.28 cc and interaction P-value of 2.07e-6). Figure 1 shows the change in the direction/magnitude of effect of rs12860460 (Figure 6-a) and rs12854485 (Figure 6-b) on ACC volume in female OCD patients versus in controls. For both

SNPs, the tested allele (versus what we observed for the alternate allele) was associated with reduced ACC volume in female controls and with increased ACC volume in female OCD patients.

126

Table 21. Effect of genotype-diagnosis interaction on brain volume in males: Top 20 SNP findings.

SNP Interacti Brain Tested Main SNP on Term Interaction SNP Gene MAF Region Allele Effect P-value Estimate P-value (cc) (cc) OFC rs1777764 HTR1B G 0.31 -0.97 3.52E-01 5.78 1.58E-03 OFC rs28656158 HTR1B C 0.31 -0.71 4.81E-01 5.43 2.23E-03 Thalamus rs17069005 HTR2A G 0.20 -0.74 4.24E-02 1.47 2.32E-03 Caudate rs4942577 HTR2A T 0.38 -0.89 3.02E-04 1.06 3.06E-03 OFC rs2223832 HTR1B C 0.14 -2.51 1.17E-01 7.20 3.61E-03 OFC rs11757592 HTR1B C 0.14 -2.51 1.17E-01 7.20 3.61E-03 Caudate rs9534488 HTR2A T 0.46 -0.69 3.04E-03 0.97 4.44E-03 Caudate rs9526236 HTR2A T 0.46 -0.69 3.22E-03 0.98 4.73E-03 OFC rs9343614 HTR1B G 0.20 -1.93 1.11E-01 6.22 4.77E-03 Caudate rs12110491 HTR1B T 0.12 0.46 2.37E-01 -1.72 4.90E-03 Caudate rs55636038 HTR1B T 0.12 0.46 2.37E-01 -1.72 4.90E-03 Caudate rs11755194 HTR1B C 0.12 0.46 2.37E-01 -1.72 4.90E-03 Caudate rs1228798 HTR1B G 0.12 0.46 2.37E-01 -1.72 4.90E-03 Caudate rs1228797 HTR1B T 0.12 0.46 2.37E-01 -1.72 4.91E-03 Caudate rs2207056 HTR1B T 0.12 0.46 2.37E-01 -1.72 4.92E-03 Caudate rs10806097 HTR1B T 0.12 0.48 2.34E-01 -1.74 5.06E-03 Caudate rs1228806 HTR1B C 0.11 0.50 2.36E-01 -1.76 5.47E-03 Caudate rs17069005 HTR2A G 0.20 -0.86 7.41E-03 1.17 5.50E-03 Caudate rs1228805 HTR1B G 0.11 0.50 2.36E-01 -1.77 5.56E-03 Caudate rs1228804 HTR1B G 0.11 0.51 2.38E-01 -1.77 5.74E-03

Abbreviations: SNP, single nucleotide polymorphism; MAF, minor allele frequency; cc, cubic centimeters; OFC, orbitofrontal cortex; HTR2A, 5-HT2A receptor gene; HTR1B, 5-HT1B receptor gene.

127 Table 22. Effect of genotype-diagnosis interaction on brain volume in females: Top 20 SNP findings. (Findings significant after correction for multiple comparisons are in bold.)

SNP Interacti Brain Tested Main SNP on Term Interaction SNP Gene MAF Region Allele Effect P-value Estimate P-value (cc) (cc) ACC rs12860460 HTR2C C 0.18 -4.37 1.75E-07 5.45 9.70E-08 ACC rs12854485 HTR2C G 0.23 -3.26 5.45E-06 4.28 2.07E-06 Striatum rs9567737 HTR2A T 0.42 0.94 8.56E-04 -1.45 3.40E-04 Striatum rs6561333 HTR2A C 0.42 0.95 8.77E-04 -1.45 3.42E-04 Striatum rs1923886 HTR2A T 0.42 0.90 1.53E-03 -1.43 4.37E-04 Striatum rs1885884 HTR2A G 0.13 1.21 1.15E-03 -1.78 6.59E-04 ACC rs35031982 HTR2C G 0.21 -2.81 2.53E-04 3.28 6.64E-04 Thalamus rs1738504 HTR1B C 0.26 0.42 1.30E-01 -1.43 7.04E-04 Thalamus rs1777764 HTR1B G 0.34 0.50 5.10E-02 -1.34 8.82E-04 ACC rs12861270 HTR2C G 0.10 -5.46 2.19E-03 6.20 1.01E-03 ACC rs2192371 HTR2C G 0.26 -2.36 5.44E-04 2.86 1.06E-03 ACC rs12862598 HTR2C C 0.10 -5.40 2.57E-03 6.17 1.14E-03 Thalamus rs2207055 HTR1B A 0.33 0.46 7.12E-02 -1.30 1.16E-03 ACC rs12851998 HTR2C T 0.10 -5.49 1.85E-03 6.00 1.23E-03 ACC rs12854729 HTR2C G 0.10 -5.46 2.25E-03 6.10 1.23E-03 ACC rs3813928 HTR2C A 0.10 -5.46 2.25E-03 6.09 1.24E-03 ACC rs3813929 HTR2C T 0.10 -5.46 2.25E-03 6.09 1.24E-03 ACC rs11798441 HTR2C G 0.10 -5.46 2.25E-03 6.09 1.24E-03 ACC rs17260565 HTR2C G 0.10 -5.46 2.25E-03 6.09 1.24E-03 ACC rs17326429 HTR2C A 0.10 -5.46 2.25E-03 6.09 1.24E-03

Abbreviations: SNP, single nucleotide polymorphism; MAF, minor allele frequency; cc, cubic centimeters; ACC, anterior cingulate cortex; HTR2A, 5-HT2A receptor gene; HTR1B, 5-HT1B receptor gene; HTR2C, 5-HT2C receptor gene.

128

Figure 6. Brain volume versus genotype in females graphed: a) HTR2C SNP rs12860460, showing females with genotype and corresponding ACC volume (cc) for controls (green) and for OCD cases (purple); ACC volume decreases by 4.37 cc in controls (with 0, 1, or 2 tested C alleles) and ACC volume increases by 1.08 cc in OCD cases (with 0, 1, or 2 tested C alleles). b) HTR2C SNP rs12854485, showing females with genotype and corresponding ACC volume (cc) for controls (green) and for OCD cases (purple); ACC volume decreases by 3.26 cc in controls (with 0, 1, or 2 tested G alleles) and ACC volume increases by 1.02 cc in OCD cases (with 0, 1, or 2 tested G alleles). OCD, obsessive-compulsive disorder; ACC, anterior cingulate cortex; cc, cubic centimeters.

5.4. Discussion

Studies to date have indicated that serotonin system genes may play a role in OCD, but the associated variants reported are inconsistent between studies (Sinopoli et al., 2017). Like other complex disorders, OCD is thought to result from multiple genes of small effect interacting with environmental factors (Pauls et al., 2014). In this study, we adopted a candidate gene approach to examine common potential serotonin system gene polymorphisms hypothesized to be involved in OCD. We aimed to address some of the limitations of previous studies including failure to account for sex or age of onset. To attempt to reduce phenotypic heterogeneity in our study, we used an exclusively pediatric group of individuals and stratified by sex. We also attempted to reduce heterogeneity by using structural brain imaging in order

129 to study putative intermediate phenotypes for OCD, specifically using CSTC circuitry relevant to OCD (MacMaster et al., 2010; Pauls et al., 2014).

In our case-control analyses, the top SNPs nominally associated with OCD in males were located in a region 5’ of HTR1B. These findings did not remain significant after correction for multiple comparisons. In females, the majority of our top SNPs were nominally associated with OCD and corresponded with SLC6A4. Again, these findings did not remain significant after correction for multiple comparisons, but the difference in trends in males and females is in line with previous literature showing that risk variants for OCD may differ by sex (Mattina

& Steiner, 2016). We did not find any significant association in males or in females between

OCD and the most commonly studied candidate gene polymorphism, 5-HTTLPR, which has been associated with the disorder in previous literature (Taylor, 2013). Though our case- control analyses were underpowered, our findings suggest a need to analyze candidate serotonin system genes in males and in females separately.

For our brain region analyses, we first examined the effect of candidate serotonin gene variants on brain volume in cases and controls combined (irrespective of diagnosis). None of the candidate variants were significantly associated with regional brain volumes in males or in females after correction for multiple comparisons. The results from our analyses assessing the effect of genotype-diagnosis interaction on brain volume supported our hypothesis that the relationship between some serotonin system gene variants and brain volume differs between

OCD patients and controls and that these differences vary between males and females. Our findings in males did not survive correction for multiple comparisons. In females, we identified a significant genotype-diagnosis interaction for two SNPs in HTR2C, rs12860460

130 and rs12854485. We also note that both significant SNPs were genotyped and not imputed, further supporting the reliability of our findings. Specifically, these HTR2C SNPs were differentially associated with ACC volume in female children/adolescents in such a way that the tested allele (versus what we observed for the alternate allele) in each SNP was associated with decreased ACC volume in healthy individuals and with increased ACC volume in OCD patients. The effect of gene variant on ACC volume was also smaller in OCD cases than it was in controls. We cannot refer to one of the variants in either SNP as a genetic risk factor, however, since we did not identify any to be significantly associated with OCD in our first set of analyses (though underpowered).

Our study implicated the ACC, consistent with prior findings of increased ACC volume in pediatric OCD (Rosenberg & Keshavan, 1998; Szeszko et al., 2004) and findings of ACC abnormalities being more pronounced in female OCD patients than in male OCD patients

(Yücel et al., 2008). Our study highlights a possible mechanism in OCD where the allelic variants in the implicated HTR2C SNPs act differently in cases versus in controls with regards to the direction and magnitude of effect on ACC volume.

Our diagnosis-specific findings are similar to what has previously been reported. Two prior studies that examined 5-HTTLPR in OCD showed a nominally significant genotype-diagnosis interaction effect on brain volume (Atmaca et al., 2011; Honda et al., 2017). Atmaca et al.,

(2011) noted a stronger effect of 5-HTTLPR genotype on OFC in OCD patients versus in controls, with reduced volume in the OFC of S carriers with OCD (versus individuals of genotype L/L with OCD). Honda et al., (2017) noted a stronger effect of 5-HTTLPR genotype on the right frontal pole in OCD patients versus in controls, with reduced gray matter volume

131 in the right frontal pole in LA carriers with OCD versus LA carriers without OCD (Honda et al., 2017). We similarly identified significant genotype-diagnosis interaction effects on brain volume for serotonin system gene variants and our findings survived correction for multiple comparisons. Unlike these previous studies, though, we did not observe a significant genotype-diagnosis effect for 5-HTTLPR. Instead, our strongest findings were within HTR2C.

Furthermore, the effects of the significant HTR2C SNPs were more marked in controls as opposed to in OCD cases, opposite to what was observed for 5-HTTLPR in previous studies.

The 5-HT2C receptor is widely distributed throughout the central nervous system, plays a number of roles in cell signalling, and is important in mood, sex, and appetite regulation

(Molineaux et al., 1989). Studies in psychiatry have also suggested the receptor’s involvement in anxiety, schizophrenia, depression, and suicide (Werry et al., 2008), with a group of studies reporting significant differences in RNA editing for HTR2C in post-mortem brain tissue of suicide victims (Niswender et al., 2001; Gurevich et al., 2002; Iwamoto & Kato, 2003;

Dracheva et al., 2008; Lyddon et al., 2013; Di Narzo et al., 2014; Weissmann et al., 2016).

Rs12854485 and rs12860460 are adjacent to one another (~4.67kb apart) in intron 2 of

HTR2C. Nearby the SNPs, and overlapping the same region as HTR2C intron 2, are 4 microRNA (miRNA) genes (MIR764, MIR1912, MIR1264, MIR1298) which yield non-coding

RNAs involved in gene expression and post-transcriptional regulation, and an H/ACA box small nucleolar RNA (snoRNA) gene (SNORA35), part of a class yielding non-coding RNAs which facilitate rRNA or spliceosomal RNA modification (Lestrade & Weber, 2006; UCSC

Genome Browser, hg19). SNORA35 is predominantly expressed in the brain (UCSC Genome

Browser, hg19; GTEx, Release V6). It could be the case that our significant SNPs lie in

132 potential regulatory regions of HTR2C that affect gene expression, or more specifically RNA editing or alternative splicing, depending on additional risk factors (Werry et al., 2008; Lu et al., 2012; Wang et al., 2015). The SNPs may also be tagging variants in another region of the gene(s) that are implicated in HTR2C expression or functionality or in the expression or functionality of the small regulatory genes overlapping HTR2C. Our significant genotype- diagnosis interaction shows us that the effect of each SNP variant on female ACC volume differs in OCD cases versus in controls. HTR2C variation is likely to be only one contributing factor in the complex etiology of OCD, which interacts with additional genetic, epigenetic, biological, and/or environmental risk factors to produce disease-relevant changes in ACC volume in female children and adolescents.

Though our study provides evidence of the value of studying the genetics of neural mechanisms for OCD in homogenous subgroups reflective of sex and age

(children/adolescents versus adults), it had a number of limitations. First, subgrouping by sex resulted in reduced sample size for each analysis, which could have reduced our power to detect associations in the absence of strong sex effects. To address this issue, we performed post hoc analyses for all autosomal variants (excluding HTR2C which is on chromosome X) combining males and females. There were no significant findings for these combined analyses after correction for multiple comparisons. The second limitation of our study was that our sample was too small to consider genome-wide data and all potential brain regions, both of which would have conferred a stricter significance threshold to account for multiple comparisons. We therefore elected to focus on identifying associations of interest confined to select candidate serotonin system genes and a priori brain regions pertinent to OCD.

133 Future studies are warranted to replicate our findings in a larger sample and, ideally, using a more comprehensive, genome-wide approach. Should our findings be replicated, more extensive studies would be necessary to identify additional pathological factors driving the development of OCD and that are specific to the relationship between ACC and HTR2C in pediatric females. Future genetic studies of OCD should similarly stratify by sex and age group, given that we have evidence supporting genetic association in a pediatric sample that would not have been identified without stratifying based on sex. Age has been shown to have an effect on brain volume, particularly in subcortical brain regions, with puberty status describing such developmental changes even better than age (Wierenga et al., 2018). Given that the age range of our sample traverses adolescence, we included age as a covariate in our study to control for its effects on brain volume. Future studies in larger samples may benefit from more detailed analyses of the influence of age and pubertal development (i.e., by stratifying based on these variables) given the effect of puberty on brain structure. Our sample size was insufficient to account for symptom dimensions in OCD, but future studies with larger samples should test for genetic associations based on symptom dimensions. This will help us to better understand the phenotypic heterogeneity of OCD and the complexity of the disorder at the level of gene systems and brain morphology.

5.5. Conclusion

In summary, although limited by a small sample size and by its focus on single candidate genes, our study is important given the novelty of our approach. Ours is one of the first studies of its kind (and the largest to date) to look across several candidate serotonin genes in

OCD, using brain imaging as a potential mediating biological factor in OCD and sex-based

134 analyses to reduce phenotypic heterogeneity. This approach has not yet been implemented on a genome-wide scale in OCD. Our findings, therefore, help lay methodological groundwork and will help guide future genome-wide studies aimed at understanding the genetic and neural mechanisms driving OCD. Consistent with a precision medicine approach, identification of potential genetic and neurobiological mechanisms in OCD, combined with information regarding key demographic features (sex, age of onset), may allow us to target OCD treatment to specific subgroups of patients.

135 6. GENERAL DISCUSSION, CONCLUSIONS AND FUTURE DIRECTIONS

6.1. General Discussion

6.1.1. Summary of key findings

As with any psychiatric disorder, we see multiple layers of complexity in OCD.

Phenotypically, the disorder presents with multiple symptom dimensions, such as preoccupation with symmetry, forbidden thoughts, cleaning, and/or hoarding (Bloch et al.,

2008a). More so, several OCD subtypes have been noted which reflect accompanying comorbidities, age of onset, etc. Facets such as sex distribution, symptom severity, response to treatment, and heritability may also differ between symptom dimensions and subtypes

(Swedo et al., 1998; Swedo et al., 2012; Grados, 2010; Leckman et al., 2010; APA, 2013;

Williams et al., 2013). Multiple gene systems have been implicated in OCD, including serotonin system genes, glutamate system genes, and dopamine system genes (Grados, 2010;

Taylor, 2013). Genetic studies in OCD have been largely inconclusive to date (Sinopoli et al.,

2017). We believe that this is in part due to the heterogeneous nature of the disorder, which many past studies failed to adequately address.

Overall, the experiments conducted in this thesis studied the underlying genetic basis of OCD or OC traits while addressing its multifaceted nature. We accomplished this across all three studies by focusing on serotonin system genes, specifically, and by attempting to use phenotypically homogenous subgroups for each of the analyses. Specifically, we studied the association between serotonergic genes and OC trait dimensions, as well as more closely

136 examined hoarding, in a general population of children and adolescents to test if specific serotonin gene variants were associated with specific symptom-based subgroups. We also studied the relationship between serotonergic genes and brain volume, a putative intermediate phenotype for OCD in children and adolescents from the clinic.

5-HTTLPR remains one of the most heavily studied candidate gene regions in OCD and one of the most commonly associated with the disorder (Taylor, 2013). Findings remain inconsistent as to the exact variant implicated, however, with studies suggesting it is important to take LG functionality into account, in addition to stratifying participants into homogeneous subgroups, when conducting analyses (Sinopoli et al., 2017). We were careful to include genotyping of rs25531 in order to account for the impact this SNP has on the function of 5-HTTLPR.

137 Table 23. Summary of thesis results: Significant findings in each of the 3 studies, after correction for multiple comparisons.

STUDY GENE POLYMORPHISM FINDING VALUES

1 SLC6A4 5-HTTLPR [LG + S] significantly OR of 1.35 and associated with ‘Hoarding’ P-value of 0.003 in Males

2 SLC6A4 5-HTTLPR [LG + S] significantly OR of 1.43 and associated with ‘Hoarding P-value of 0.009 Only’ in Males

3 HTR2C rs12860460 Significant SNP main SNP Main Effect of effect; Significant -4.37 cc and SNP genotype-diagnosis P-value of 1.75e-7; interaction (whereby C Interaction Term allele is associated with Estimate of 5.45 cc decreased ACC volume in and Interaction female healthy controls P-value of 9.70e-08 and with increased ACC volume in female OCD patients)

rs12854485 Significant SNP main SNP Main Effect of effect; Significant -3.26 cc and SNP genotype-diagnosis P-value of 5.45e-6; interaction (whereby G Interaction Term allele is associated with Estimate of 4.28 cc decreased ACC volume in and Interaction female healthy controls P-value of 2.07e-6 and with increased ACC volume in female OCD patients)

Abbreviations: SLC6A4, solute carrier family 6 member 4 (serotonin transporter gene); 5-HTTLPR, serotonin transporter-linked polymorphic region (in the promoter of SLC6A4); LA, long variant of 5-HTTLPR with A allele at SNP rs25531; LG, long variant of 5-HTTLPR with G allele at SNP rs25531; S, short variant of 5-HTTLPR; HTR2C, 5-HT2C receptor gene; SNP, single nucleotide polymorphism; ACC, anterior cingulate cortex; OR, odds ratio; cc, cubic centimeters.

A summary of the main findings in this thesis are summarized in Table 23. In Study 1, we used OC behaviors collected from children and adolescents to analyze OC trait dimensions in the community reflective of symptom dimensions in the clinic. We hypothesized that if the

138 serotonin system is implicated in OCD and that if OCD symptom dimensions differ in their underlying genetic basis, then different serotonin gene variants would be associated with different OC trait dimensions in our population-based sample and also differ between sexes.

Our results partially supported our hypothesis to the extent that we found a significant association between one specific serotonin variant and hoarding. We found a statistically significant association between the [LG + S] variant of 5-HTTLPR and hoarding traits in males. Our findings support both the DSM-5 and recent literature suggesting that hoarding is biologically and phenotypically distinct from, but related to, OCD (Mataix-Cols & Pertusa,

2012; APA, 2013). Interestingly, for 5-HTTLPR, we see odds ratios in males that indicate a stronger trending association with 5-HTTLPR when comparing trends for each trait dimension

(and OC overall) to females. Although this sex-specific effect was only significant for hoarding in our study, trends may emerge as significant for other dimensions in a larger sample.

We used our hoarding findings to guide the design of Study 2, where we used the same community-based sample to further examine hoarding as a potentially distinct behavioral and biological group. We examined the same serotonin candidate genes from Study 1 and examined individuals with hoarding plus other OC traits, with hoarding traits alone, and with non-hoarding OC traits. We hypothesized that if genetic associations identified in Study 1 were specific to hoarding, then we would find that these associations were strongest in the

“hoarding only” group. Our findings supported this hypothesis, specifically showing that the

[LG + S] variant of 5-HTTLPR was significantly associated with hoarding traits alone in males and that variants in HTR1B had a nominally significant association with hoarding traits alone in females. This supports the literature that the biological underpinnings for hoarding disorder

139 differ from OCD (Mataix-Cols & Pertusa, 2012) and also reinforces the concept of sex differences in hoarding (Ivanov et al., 2017).

Study 3 aimed to reduce heterogeneity in clinically diagnosed OCD by focusing on regional brain volumes identified using neuroimaging as putative intermediate phenotypes. We hypothesized that if serotonin system gene variants are associated with OCD, and if CSTC dysfunction is implicated in OCD, then we would find a subset of serotonin gene variants that would significantly differ in their relationship with volumetric changes in select CSTC brain regions in OCD patients versus controls. More so, we hypothesized that these findings would differ between sexes. Our findings supported our hypothesis, in that we identified significant genotype-diagnosis interactions for two SNPs in HTR2C (rs12860460 and rs12854485). The tested allele for each SNP (C allele in rs12860460; G allele in rs12854485) was associated with increased ACC volume in female OCD patients and with reduced ACC volume in female controls. Our findings, if replicated, may help to decipher potential pathogenic mechanisms in

OCD. SNP allele alone was not associated with OCD (though not surprising given that we were underpowered for the case-control portion of our study), nor was it associated with increased or decreased ACC volume. The specific gene variant-brain volume combination appears to be relevant to OCD pathogenesis. Overall, our study suggests that the underlying biological mechanisms in OCD are complex. The HTR2C SNPs in our genotype-diagnosis interactions likely act together with other genetic, biological, and/or environmental factors to contribute to OCD in ways which remain to be explored.

Studies 1 and 2 reduced phenotypic heterogeneity by stratifying groups by trait dimension and by sex. Both studies used the same population-based sample and indicated that 5-HTTLPR is

140 likely relevant to hoarding disorder in male children and adolescents. In study 3, we used an alternative approach to addressing heterogeneity in OCD by using a putative intermediate phenotype based on imaging, and again stratifying by sex. All three studies explored the same set of serotonin genes, with the addition of HTR2C in our imaging study. The 5-HT2C receptor is widely distributed throughout the central nervous system with most of its expression restricted to the brain (Molineaux et al., 1989). It has been the subject of numerous studies mainly detailing its functional and biological relevance to psychiatry and it has been the subject of imaging and anatomic studies showing brain region-specific differences in expression, mRNA editing, and protein activity (Drago & Serretti 2009; Quilter et al., 2012;

Weissmann et al., 2016), thus emphasizing the pertinence of HTR2C to the study of brain structures. From here, future studies are required to replicate our findings from Study 1 and 2 in a clinical setting to ensure transferability in the context of diagnosed OCD. In Study 3, we used a clinical OCD sample, but one of the limitations was our inability to examine OCD symptom dimensions given the smaller sample size.

6.1.2. Candidate gene approach in homogenous subgroups

In OCD, heritability ranges from 27 to 65%, reflecting genetic contribution to the disorder

(van Grootheest et al., 2005). The role of genetics is particularly relevant to pediatric OCD.

As reviewed earlier, early-onset OCD has been proposed as a distinct OCD subtype (Swedo et al., 1998; Grados, 2010; Leckman et al., 2010; Taylor, 2011; Swedo et al., 2012; APA, 2013;

Williams et al., 2013) with OCD more common in first-degree relatives of probands with early-onset OCD versus first-degree relatives of probands with late-onset OCD (do Rosario-

Campos et al., 2005; Hanna et al., 2005; Taylor, 2011; Pauls et al., 2014) and with higher

141 symptom heritability in children with OCD versus adults with OCD (van Grootheest et al.,

2005). Heritability estimates for OCD in children alone range from 45-65% (van Grootheest et al., 2005, Pauls et al., 2014) and there is a heritability of 74% for OC traits in children and adolescents in the community (measured using TOCS) (Burton et al., 2018).

Several candidate genes have been the focus of study in OCD, including those implicated in the serotonin, dopamine, and glutamate systems (Grados, 2010; Taylor, 2013). The role of the serotonin system in OCD has been subject to extensive study and, amongst serotonin system genes, the most comprehensive meta-analyses convey that the LA variant in 5-HTTLPR and the A allele of rs6311 or the linked T allele of rs6313 in HTR2A have the best evidence of association with OCD, with more limited evidence supporting the association between rs6296 in HTR1B and OCD overall or between rs6318 in HTR2C and OCD (Taylor, 2013; Taylor

2016). In contrast to candidate gene studies, recent GWAS results have provided no indication that serotonin genes are significantly associated with OCD (Stewart et al., 2013;

Mattheisen et al., 2015), although GWAS arrays contain SNPs and do not currently include insertion/deletion polymorphisms such as 5-HTTLPR. This thesis proposed that studying the genetic basis of more homogenous OCD-based subgroups would help to address some of the inconsistencies in the literature and identify associations between serotonin gene variants and specific subgroups.

Evidence suggests that OCD results from multiple genes of small effect (Taylor, 2013;

Sinopoli et al., 2017), similar to other common, complex psychiatric disorders (Geschwind &

Flint, 2015). It is likely that the effect of these genes is mediated by several different biological mechanisms in OCD pathogenesis. Our use of homogenous OCD subgroups

142 increases our chance of detecting genetic variants of small effect because the effects may be stronger in phenotypic subgroups that may each have a more homogenous genetic etiology

(Traylor et al., 2015). For the 5-HTTLPR analyses in our population-based studies, we see that the two functional alleles are common and almost equally prevalent in the population.

The hypothesized contribution of 5-HTTLPR to genetic risk is an example of the “common disease, common variant” hypothesis in which multiple common variants of small effect are proposed to contribute to disease. This has implications for required sample sizes, as larger numbers are required to detect meaningful associations for common variants given their small effect size. This contrasts with the “common disease-rare variant” hypothesis in which rare, highly penetrant variants of large effect are thought to contribute to the risk for common disorders, and in this scenario, large sizes are also needed given the small proportion of individuals who carry a given rare variant (Schork et al., 2009).

Across our 3 studies, we used a candidate gene approach (specifically focusing on serotonin genes within which variants were most commonly found to be associated with OCD), in combination with a strict definition of our phenotypic subgroups, to maximize our chances of detecting gene variants responsible for each subgroup. In our population-based studies, we divided our participants by sex and by symptom dimension. Given that we stratified our participants into multiple groups, however, our total N for each group decreased in size and therefore simultaneously decreased our power to detect gene differences associated with each group. We reconciled this issue by opting for a candidate gene approach across our studies, which demanded less stringent statistical correction than a genome-wide approach.

143 Furthermore, we hypothesized that specific candidate serotonin gene variants, which we believe differ between phenotypic subgroups, are each also associated with specific neurobiological mechanisms in OCD as measured through brain structure. We stratified our analyses based on sex in our clinical, pediatric sample. Given the smaller sample sizes after stratification by sex and given the added correction for multiple comparisons to account for the 6 brain ROIs studied, we again opted for a candidate gene approach in order to streamline our research and maximize our chances of finding serotonin gene variants associated with specific volume differences in each of these regions and in our OCD, sex-based subgroups.

In-depth phenotyping using homogenous subgroups is key to deciphering specific pathological mechanisms in OCD. We hypothesized that specific OCD subgroups each have their own set of associated gene variants in serotonin system genes, which we elected to study using a candidate gene approach.

6.1.3. Community versus clinic-based samples in the study of OCD and other complex disorders

There are multiple advantages to the population-based study design used for Studies 1 and 2.

Pediatric OCD is often underdiagnosed since symptoms in children can be easily missed or improperly communicated (Geller & March, 2012). Also, some patients (pediatric or otherwise), present with subthreshold OCD symptoms which might lead to them being misclassified as controls in a clinical case-control study. A population-based approach allowed us to correctly classify participants on a continuum of OC traits.

144 Our Ontario Science Centre population provided additional advantages conducive to rapid gene discovery. It allowed us to obtain participants in a cost-effective and efficient manner.

Compared to a clinic-based study, we were able to recruit a large number of children and adolescents over a shorter period of time than a clinic setting would have permitted. It also allowed us to recruit and evaluate participants at a substantially lower cost than assessment in a clinic would have afforded. Clinic-based studies offer the advantage of more in-depth diagnostic assessment (Jarbin et al., 2017) using semi-structured clinician-administered interviews, so it is necessary to replicate genetic findings in a clinical setting to ensure transferability and validity, but we argue that an efficient population-based approach to data collection will accelerate research and ultimately expedite gene discovery through cost- effective collection of very large samples.

6.1.4. Intermediate phenotypes

The goal of neuroimaging in psychiatry is to help us better understand brain structure and function as it relates to mental illness. Results of neuroimaging can be helpful in defining more homogenous phenotypes (and therefore potential intermediate phenotypes) to help reduce genetic complexity and heterogeneity in OCD. The intent is to a) increase power to detect gene differences associated with a psychiatric disorder using the rationale that quantitative and biologically tangible traits like brain structural differences are closer to their underlying gene(s) and b) to help identify neural mechanisms that each discretely contribute to pathogenesis via specific gene(s) (Meyer-Lindenberg & Weinberger, 2006; Leite et al., 2015).

In this way, Study 3 aimed to examine the relationship between serotonin gene variants and

145 brain volume differences which are putative intermediate phenotypes relevant to OCD pathogenesis and mediate between genotype and behavior.

The term “intermediate phenotype” is often used interchangeably with “endophenotype”, where a genetic risk confers a heritable and primarily state-independent intermediate neural manifestation (brain imaging pattern), which is in turn associated with disease (Gottesman &

Gould, 2003). More recent advances in neuropsychiatry, however, challenge this unidirectional view and suggest a more complex model of disease with interaction between genome, environmental factors/experience, body, brain, and behavior (Bogdan et al., 2017).

Our results in Study 3 also challenge the unidirectional view of neuroimaging intermediate phenotypes with a finding of significant genotype-diagnosis interaction in OCD that suggests a more complex relationship between serotonin gene variation, brain structure, and behavior which differs between cases and controls. The underlying genotype-diagnosis interaction highlights potential abnormal neurocircuitry in females with OCD, though the exact mechanism(s) linking gene mechanism to brain structural difference to behavior is complex.

Specifically, we found significant genotype-diagnosis interactions in pediatric females. We noted that each allele for the significant SNP findings acts in an opposing manner in OCD cases versus controls. The tested allele for each SNP (both tested alleles minor in this case) was associated with increased ACC volume in OCD patients and decreased volume in healthy controls (Figure 7). Moreover, the effect of these SNPs on brain volume were stronger in controls than in OCD patients, which could also lead us to speculate that the magnitude of the brain volume difference between cases and controls could be of biological relevance. It is difficult to comment on the impact of directionality and findings could be due to chance.

146 Should our results be replicated in future studies, more work would be needed to understand the biological significance of this finding. Overall, we have identified a gene-ACC volume relationship that is dependent on diagnosis and which supports the notion that intermediate phenotypes may lie along a path from gene to behavior that is complex, and which may implicate additional mechanisms like environmental influences and interaction with other genes outside of serotonin system candidates.

Figure 7. Study 3 SNP alleles explained: How the two alleles in each of the SNPs for our significant genotype- diagnosis interactions differentially associate with ACC volume in OCD cases versus in healthy controls. HTR2C, 5-HT2C receptor gene; SNP, single nucleotide polymorphism; ACC, anterior cingulate cortex; OCD, obsessive-compulsive disorder.

147 6.1.5. Limitations

The main limitation in our population-based studies was sample size. Though sufficient for a candidate gene study, the degree of stratification to form symptom and sex-based homogenous subgroups limited our ability to use a more comprehensive genome-wide approach to gene discovery. This is because the reduced sample size per subgroup after stratification, coupled with stringent correction for multiple comparisons required for both a GWAS and the multiple symptom dimensions, would have rendered us underpowered to detect genetic associations.

As already mentioned, our focused approach was strategically carried out to better target phenotypic heterogeneity in OCD, and therefore improve power to find serotonin gene variant associations with each subgroup. We found a significant serotonin gene variant association with hoarding alone in males, though we cannot speak to the involvement of genes outside of the serotonin system in hoarding or other OC dimensions.

The main limitation in our clinic-based study was also sample size. Our sample size was too small to use a genome-wide approach and to look across all brain structures, as both of these added parameters would have required more stringent correction for multiple comparisons.

We were again limited to a candidate gene approach as well as an examination of only a priori brain regions relevant to OCD. This meant we were limited by our prior knowledge of the biology of OCD (both genetic and neural). Another limitation of Study 3 was that we did not include the globus pallidus in our brain region analyses. This region is a component of CSTC circuitry believed to be dysfunctional in OCD. Based on a review of the manual tracing technique used to measure our brain ROIs, we determined that the globus pallidus could not be reliably measured using this approach.

148

6.1.6. Thesis findings and the literature

Overall, this thesis provides tangible evidence that cutting through heterogeneity in OCD improves gene discovery. When studying distinct symptom-based dimensions, our findings from Studies 1 and 2 found a significant association between 5-HTTLPR and hoarding that complements the literature which suggests that hoarding is phenotypically and biologically distinct from OCD (Snowdon et al., 2012; van Ameringen et al., 2014), with substantial unique genetic factors compared to other symptom dimensions (Burton et al., 2018).

Furthermore, our findings support its reclassification as a separate disorder in the DSM-5

(APA, 2013). Our findings are the first to show a clear serotonin gene link to hoarding, specifically, and suggest that the [LG + S] variant plays a specific role in male pathogenesis.

While numerous studies have hypothesized increased SERT expression in OCD (Murphy et al., 2013; Sinopoli et al., 2017), our findings suggest that symptom and sex-based heterogeneity have been impeding our ability to find clear associations. Our findings suggest decreased SERT expression in males with hoarding traits, but this requires replication in the clinic and encourages study of more clearly-defined and ultimately biologically informed homogenous phenotypes. More specifically, our second study shows us how more clearly defined phenotyping can better clarify significant findings. Study 2 showed us that our significant 5-HTTLPR finding in hoarding was being driven by males with hoarding alone and did not remain significant in males with hoarding accompanied by other OC traits. Our methodological approach was also novel to the field, given that most past studies focused on hoarding symptoms within the context of diagnosed OCD (Lochner et al., 2005; Samuels et

149 al., 2007s; Samuels et al., 2007b) and also failed to identify a significant association with 5-

HTTLPR (though rs25531 was not genotyped in most previous studies) (Lochner et al., 2005).

In Study 3, we showed that reducing heterogeneity in diagnosed OCD (using brain structure as an intermediate phenotype and stratifying patients into homogenous sex-based subgroups) can improve gene discovery. Like two previous studies which identified a nominally significant genotype-diagnosis interaction for 5-HTTLPR and OCD for brain volume in the OFC (Atmaca et al., 2011) and in the right frontal pole (Honda et al., 2017), we also found serotonin gene- based genotype-diagnosis interaction that affected brain volume. Our significant interactions, however, were statistically significant after correction for multiple comparisons in the female subset of participants. We did not find a significant interaction for 5-HTTLPR, however. We identified significant gene-diagnosis interactions for variants in HTR2C for ACC volume. The

ACC finding compliments prior findings implicating ACC volume in pediatric OCD

(Rosenberg & Keshavan, 1998; Szeszko et al., 2004) and more specifically a sex-dependent finding of more prominent ACC abnormalities in female OCD patients than in male OCD patients (Yücel et al., 2008). The effects of our gene-diagnosis interactions were stronger in controls than they were in cases, however, which is opposite to what was seen with the previous 5-HTTLPR gene-diagnosis findings (Atmaca et al., 2011; Honda et al., 2017). Our study was also the largest study to date to look across multiple serotonin genes while using brain volume as an intermediate phenotype and the first study of its kind to focus primarily on pediatric OCD. Future studies are needed to replicate our findings which use a similar, careful approach to heterogeneity to increase chances of gene discovery.

150 6.2. Conclusions

6.2.1. Summary

In conclusion, the nature of OCD is multifaceted as evidenced by its phenotypic heterogeneity and genetic complexity previously discussed at length (Sinopoli et al., 2017). This thesis aimed to address some of this complexity, hypothesizing that specific OC subgroups each have their own set of associated serotonin system gene variants and that there are unique relationships between serotonin system gene variants and brain imaging patterns within CSTC circuity in OCD. Overall, our findings specifically showed that 5-HTTLPR is associated with hoarding in pediatric males, independent of other OC symptoms, which supports its newest reclassification in the DSM-5 as a separate but OCD related disorder (APA, 2013). When studying clinically diagnosed OCD, we showed significant genotype-diagnosis interaction for

HTR2C and the ACC in females.

6.2.2. Integrating information across genetic, functional, and behavioral levels

The RDoC framework proposes studying basic dimensional constructs within a mental disorder that together lead to diagnosable disease, and it suggests that these constructs can be examined using genetic/molecular, neural, or behavioral analysis (NIMH, 2008). This thesis is in line with this approach as it studies homogenous dimensions via use of symptom-based subgroups and the use of brain structure as a potentially less heterogeneous measure in OCD.

By integrating genetic analysis, neuroimaging, and behavioral analysis in a homogenous subgroup that reflects a simpler behavioral construct or domain within OCD, we are in effect

151 looking across this dimension to identify unique pathways or mechanisms that lead from gene to brain change to behavioral subgroup. In this way, this thesis set the groundwork for a multi-dimensional approach to studying the pathogenesis of OCD and obsessive-compulsive related disorders like hoarding (Figure 8). Looking across these biological levels can also help us examine developmental trajectories associated with pediatric OCD and the potential effects of environment/life experience (NIMH, 2008).

Future research using a multi-dimensional approach should include larger samples including genetic and structural and functional imaging data in children and adolescents who are well characterized for OCD and its symptom dimensions, as well as for hoarding. Our approach to phenotypic heterogeneity can be extended to a clinical setting. Although there are precedents for “deep phenotyping” in population-based samples of children and adolescents (i.e., the

Adolescent Brain Cognitive Development (ABCD) study conducted through the NIH), it is typically easier to carry out deep phenotyping in a clinic sample that includes imaging. A clinical sample also allows researchers to examine predictors of clinically meaningful outcomes, such as response to treatment.

Using two separate clinical cohorts, we can 1) study OCD and its symptom dimensions, and 2) study hoarding. This is important because some of our remaining questions would be better answered after collecting a sufficiently large sample of children and adolescents to then stratify based on clinically-derived symptom dimensions, as well as to specifically collect individuals with DSM-5 hoarding disorder (APA, 2013) in a similar manner to the OCD clinic sample of the thesis. Using 1) clinically diagnosed OCD and its symptom dimensions, and 2) clinically diagnosed hoarding disorder, and upon stratifying based on sex and age of onset, we

152 could then use an expansive genome-wide approach with the potential to identify all gene variants associated with each homogenous subgroup. Given the large clinical samples required, such an approach would only be possible through large, multi-site consortia, such as the Psychiatric Genomics Consortium in which our group already participates.

Finally, brain structure and/or function could help further address the issue of heterogeneity in clinically diagnosed OCD subgroups or in subgroups within hoarding, potentially serving as biologically concrete and quantitative measures in an otherwise heterogeneous phenotype.

Differences in structure/function across the entire brain could be assessed for their relationship to gene variants within different homogenous subgroups and these differences could then act as potential biological indicators of disease pathogenesis, possibly highlighting biological pathways that are characteristic of specific patient subgroups. Homogenous OCD or related subgroups may have distinct gene to behavior pathway that could be further examined in the context of environmental effects. Though there is literature to suggest that there are genetic variants more easily identified in specific homogenous groups, it is important to note that there is also literature on genes which confer risk to multiple disorders and traits, as identified through large cross-disorder samples like those used in the Province of Ontario

Neurodevelopmental Disorders (POND) Network.

153

Figure 8. A multi-dimensional approach to OCD and related disorders: Examples of areas of study at the genetic, functional, and behavioral level of psychiatric disorders. OCD, obsessive-compulsive disorder.

154 6.2.3. Translation to clinic

Further advancement in OCD treatment will be informed by a better understanding of its biological basis, which in turn requires integration across disciplines in neuroscience, from genetic and molecular studies to the study of neurocircuitry to careful examination of symptoms using standardized measures (Gordon, 2003; Dougherty et al., 2018). OCD likely involves multiple genes and biological pathways that interact with one another but that may also differ in how they each contribute to OCD diagnosis (Sinopoli et al., 2017). A recent review by Dougherty et al. (2018) highlights strategies that will help lead to neuroscientifically informed treatment planning for OCD.

First, Dougherty et al. (2018) acknowledge that OCD is phenotypically heterogeneous in its underlying symptomatology, and future studies are necessary to understand how the underlying biology differs between symptom subtypes/subgroups. Authors suggest deep phenotyping, an RDoC approach, and use of advanced neuroimaging techniques. Our work is highly consistent with this suggestion. Our studies were limited to candidate serotonin genes and to brain regions with a priori evidence of their involvement in OCD, but our findings inform future studies to use a similar approach to heterogeneity. While doing so, we encourage the use of more comprehensive genome-wide analyses in larger, clinical samples to better understand the involvement of all potential genes and gene systems in OCD and hoarding. We also encourage a broader look at brain structures that extend beyond CSTC circuitry as well as a study of brain function to better understand underlying neurocircuitry, perhaps even within individual OCD subgroups.

155 Second, Dougherty et al. (2018) review evidence supporting the emergent role of glutamate system dysfunction in OCD and suggest investigating glutamate-based treatment. We agree that it is necessary for future work to extend beyond serotonin-based treatment, but we also acknowledge that serotonergic medications still form the basis of most pharmacological treatment in OCD. Our study suggests that further study into how serotonin genetic variants predict OCD treatment response is warranted, building on previously discussed pharmacogenetics findings pertaining to serotonin genes, and with the intent to personalize treatment based on underlying, predictive genetics (Brandl et al., 2012; Sinopoli et al, 2017).

Lastly, Dougherty et al. (2018) explain the need to refine behavioral (i.e., exposure and response prevention and CBT) and neurostimulation approaches (i.e., transcranial stimulation and deep brain stimulation) for OCD. Combining these treatment approaches with genetic knowledge provides the field with tremendous potential. For example, our findings may have implications for refining these treatments using genetic variants to predict outcome of behavioral or neurostimulation approaches. Together, each of the above elements in neuroscience act together to paint a clearer picture of OCD (and related disorders like hoarding) and its working components, with the ultimate goal of informing patient treatment and individualizing it based on underlying biology, as emphasized by Dougherty et al. (2018).

6.3. Future Directions

My thesis supports the notion that OCD is both phenotypically heterogeneous and genetically complex. More specifically my results suggest that, even within the serotonin system itself, different variants are likely responsible for different OCD subgroups and intermediate

156 phenotypes. In this thesis, I used a candidate gene approach and selected a small number of candidate genes. I either directly genotyped or used genome-wide data specific to my genes of interest. This allowed me to create more phenotypically discrete subgroups without sacrificing too much statistical power, given that a full GWAS requires much more power.

Though a candidate gene approach is limited in its genetic scope and is dependent on what we know of the biology of OCD (Kwon & Goate, 2000), it allowed me to explore if different serotonin system genes are involved in different OCD subgroups, and if they differ between the sexes, within a pediatric population.

My findings indicate that 5-HTTLPR likely plays a role in hoarding behaviors in males, and that HTR2C may be implicated in the relationship between ACC volume and OCD in females.

Studies are required to replicate these findings and extend the community-based findings to the clinic. Future studies could also independently examine one or more of the identified subgroups using a more expansive genetic approach. My findings suggest that genome-wide association or next generation sequencing studies should consider symptom-based subgroups

(and more specifically hoarding disorder), sex-based subgroups, and age of onset related subgroups in studies to identify a wider range of gene variants that extend beyond candidate genes. Furthermore, functional genomics can be used to study the effects of replicated gene variants, and prospective biological pathways or systems, within specific homogenous subgroups.

Another necessary future direction is the study of GxE interactions in OCD. Amongst serotonin system genes, there is extensive literature to show that 5-HTTLPR is susceptible to environmental influences (Caspi et al, 2003; Karg et al., 2011; Conley et al., 2013; Sinopoli et

157 al., 2017). Though there is evidence showing that different variants yield different levels of serotonin transporter gene expression in response to different life stressors, evidence to suggest that epigenetics (i.e., DNA methylation) is responsible for the change in gene expression is minimal (Wankerl et al., 2014; Duman & Canli, 2015). More extensive studies are therefore required to better understand the mechanism driving GxE interactions for serotonin system genes like SLC6A4, but more specifically, studies are direly required in

OCD.

One putative risk factor for OCD that has received a lot of attention is streptococcal infections, which may result in the acute-onset subtype of OCD known as PANDAS. Although there is some emerging evidence pointing to the involvement of immune-related genes in acute-onset

OCD and genetic evidence suggesting PANDAS is related to autoimmune disfunction, there is no evidence to date to support the role of serotoninergic genes specifically in acute-onset OCD

(Chiarello et al., 2017). Further research is necessary to determine if serotonin system genes, or any other genes implicated in OCD, are also relevant to the acute-onset subtype.

6.3.1. Genetics and environmental risk in neurodevelopment

6.3.1.1. Environmental risk factors in OCD

Proposed risk factors in OCD include genetic influences (more commonly studied to date) and environmental influences (less commonly studied). Proposed environmental risk factors in

OCD include stressors like infection and/or trauma (Fernandez & Leckman, 2016). As an example of environmental risk in OCD, a recent large-scale study using a population-based

158 birth cohort of about 2.4 million children examined unfavourable perinatal events and determined that there was an increased risk of the child developing OCD if the mother smoked

10 or more cigarettes a day, or if the child presented as breech, was delivered by cesarean section (CS), was born preterm, had a low birth weight, was large for its gestational age, or had an Apgar score indicating distress at 5 minutes after birth. All these risks were deemed independent of shared environmental and genetic familial confounding factors, given the use of a family-based study design. The study also found dose-response relationships. It showed an association between increased number of perinatal risk factors and increased risk of OCD development in the child, as well a relationship between lower birth weight and increased risk of OCD development and between shorter gestational age and increased risk of OCD development (Brander et al., 2016). Another study, conducted in a large Danish birth cohort, looked at history of treated infections since birth and association of infection with the risk of developing a mental disorder in childhood or adolescence. The study showed associations for all mental health disorder, but the risk was most elevated for OCD and tic disorders (Köhler-

Forsberg et al., 2018). In other words, infection and immunity seems to play a role in OCD, with treated infection as a potential postnatal risk factor.

6.3.1.2. Effect of environmental risk on the genome

Further studies are necessary to merge our understanding of the genetic and environmental risk factors in OCD and to identify converging biological mechanisms (Fernandez &

Leckman, 2016). This includes looking across gene, epigenetic, and gene expression profiles while considering various environmental factors. For example, researchers have found that the prenatal epigenome is highly susceptible to environmentally-induced changes (Perera &

159 Herbstman, 2011) and that, when studying the brain, changes in the transcriptome were most substantial between brain regions and across time before birth (Kang et al., 2011). These studies seem to suggest prenatal risk factors are particularly important in development but, to date, there is limited evidence for the potential importance of epigenomic or transcriptomic changes in OCD specifically. We suggest that future studies are necessary to extend beyond our knowledge of environmental risk factors in OCD and to identify how potential risk factors act on the genome, particularly during neurodevelopment.

6.3.1.3. Epigenomic changes and OCD

We discussed earlier that epigenetics is a mechanism by which environmental influences on gene expression may be mediated, with evidence suggesting that pre and postnatal factors directly affect methylation status in the genome and affect neurodevelopment and future psychiatric state (Kundakovic & Champagne, 2015). As with most genes, such evidence to date is minimal for serotonin system genes in OCD (Nissen et al., 2016; Yue et al., 2016), with one of the studies showing differential methylation in HTR2C in OCD subjects (Yue et al.,

2016).

6.3.1.4. Conclusion

Overall, these studies suggest that environmental risk factors throughout neurodevelopment have a pertinent effect on the development of OCD, that the genome is particularly susceptible to environmentally-induced epigenetic changes and to transcriptomic changes during neurodevelopment, and that there may be epigenetic changes that impact OCD development

160 (though this still remains largely unknown). Our knowledge of environmental risk factors is still limited, however, and so is our understanding of how specific environmental elements like perinatal, prenatal, and postnatal environmental risk factors act on the genome to exert effects on gene expression that directly contribute to the neurodevelopmental changes in OCD.

Future studies are required to help us better understand the link between risk factor, gene expression, and ultimately OCD pathogenesis (Figure 9).

Figure 9. Potential neurodevelopmental changes in OCD: Examples of environmental risk and gene expression changes, as they may act together throughout neurodevelopment to lead to OCD behavior. OCD, obsessive- compulsive disorder.

6.3.2. Proposal for Future Experiment

Genes appear to be particularly susceptible to influence during neurodevelopment, and evidence suggests that risk factors during neurodevelopment (i.e., perinatal, pre/postnatal) can

161 increase risk for OCD development (Brander et al., 2016; Köhler-Forsberg et al., 2018).

Though there is evidence connecting early environmental risk to genomic influence (i.e., genomic methylation) in psychiatry (Kundakovic & Champagne, 2015), evidence specific to

OCD is only in its infancy. There is substantial evidence, however, to show that 5-HTTLPR is susceptibility to environmental influences (Caspi et al, 2003; Karg et al., 2011; Conley et al.,

2013; Sinopoli et al., 2017). I therefore propose to study this susceptibility to environmental risk factors in a GxE interaction study of OCD that considers the 5-HTTLPR polymorphism studied in our thesis and which also looks at DNA methylation differences in and around

SLC6A4 to better understand the disorder’s underlying biology as it pertains to neurodevelopment.

I aim to study failure to breastfeed, CS, and episodes of infection (from infancy to early adolescence) as potential risk factors for OCD development in a population-based, longitudinal study of pregnant mothers and their infants, following these infants into early adolescence and examining epigenetics. More specifically, I will look at DNA methylation in and around SLC6A4 and I will directly genotype the LA versus [LG + S] variants in 5-HTTLPR to study how different 5-HTTLPR genotypes interact with environmental risk factors and, more specifically, if different 5-HTTLPR variants are associated with differential DNA methylation in and around SLC6A4 in a way that may be mediating the interaction. Given that

CS has been associated with increased risk for OCD development (Brander et al., 2016) and has been associated with increased immune function impairment (Kristensen & Henriksen,

2016), given that breastfeeding has been shown to positively impact immunity and disease susceptibility (Dzidic et al., 2018), and given the proposed tie between infection and OCD

(Köhler-Forsberg et al., 2018), I hypothesize that the proposed risk factors will be associated

162 with increased risk of OCD development in the infants at childhood/early adolescence, that I will find a significant GxE interaction between the 5-HTTLPR polymorphism and the proposed risk factors in the development of OCD, and that different 5-HTTLPR variants will be associated with differential SLC6A4 DNA methylation between groups when comparing feeding, delivery mode, and/or number of acquired infections.

The ultimate goal of my future study is to steer OCD research away from strict genetic association studies to allow for the study of potential environmental risk factors and how they alter gene expression to yield OCD. Longitudinal approaches that consider development from an early stage would be ideal and would allow us to better understand the mechanisms implicated in such a complex neuropsychiatric disorder. This approach will essentially facilitate a change in how we examine OCD research by studying environmental factors that may be implicated in OCD and looking at how such risk factors act together to lead to OCD development.

163 REFERENCES

Adams, K. H., Hansen, E. S., Pinborg, L. H., Hasselbalch, S. G., Svarer, C., Holm, S., Bolwig, T. G., & Knudsen, G. M. (2005). Patients with obsessive-compulsive disorder have increased 5-HT2A receptor binding in the caudate nuclei. The International Journal of Neuropsychopharmacology / Official Scientific Journal of the Collegium Internationale Neuropsychopharmacologicum (CINP), 8(3), 391-401.

Albelda, N., & Joel, D. (2012). Current animal models of obsessive compulsive disorder: An update. Neuroscience, 211, 83-106.

Albert, U. & Bogetto, F. (2015). Treatment of obsessive-compulsive disorder: drugs, psychotherapy or combined treatments? Rivista di Psichiatria, 50(4), 153-4.

Alonso, P., Gratacós, M., Segalàs, C., Escaramís, G., Real, E., Bayés, M., Labad, J., López- Solà, C., Estivill, X., & Menchón, J. M. (2012). Association between the NMDA glutamate receptor GRIN2B gene and obsessive-compulsive disorder. Journal of Psychiatry & Neuroscience : JPN, 37(4), 273-281.

Alvarenga, P. G., Cesar, R. C., Leckman, J. F., Moriyama, T. S., Torres, A. R., Bloch, M. H., Coughlin, C. G., Hoexter, M. Q., Manfro, G. G., Polanczyk, G. V., Miguel, E. C., & do Rosario, M. C. (2015). Obsessive-compulsive symptom dimensions in a population-based, cross-sectional sample of school-aged children. Journal of Psychiatric Research, 62, 108- 114.

American Psychiatric Association (1994). Diagnostic and Statistical Manual of Mental Disorders (4th ed.). Washington, DC.

American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed., text rev.). Washington, DC.

American Psychiatric Association (2013). Diagnostic and Statistical Manual of Mental Disorders (5th ed.). Arlington, VA.

Angst, J., Gamma, A., Endrass, J., Goodwin, R., Ajdacic, V., Eich, D., & Rossler, W. (2004). Obsessive-compulsive severity spectrum in the community: Prevalence, comorbidity, and course. European Archives of Psychiatry and Clinical Neuroscience, 254(3), 156-164.

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. (2009a). Glutamate system genes associated with ventral prefrontal and thalamic volume in pediatric obsessive-compulsive disorder. Brain Imaging and Behavior, 3(1), 64-76.

Arnold, P. D., Macmaster, F. P., Richter, M. A., Hanna, G. L., Sicard, T., Burroughs, E., Mirza, Y., Easter, P. C., Rose, M., Kennedy, J. L., & Rosenberg, D. R. (2009b). Glutamate receptor gene (GRIN2B) associated with reduced anterior cingulate

164 glutamatergic concentration in pediatric obsessive-compulsive disorder. Psychiatry Research, 172(2), 136-139.

Arnold P. & Richter, M. A. (2007). Genetics of Obsessive-Compulsive Disorder: Evidence from Pediatric and Adult Studies. In: Storch, E. A. and Greffken, G. R., editors. Handbook of Child and Adolescent Obsessive-Compulsive Disorder. New Jersey: Lawrence Erlbaum Associates Inc.

Arnold, P. D., Zai, G., & Richter, M. A. (2004). Genetics of anxiety disorders. Current Psychiatry Reports, 6(4), 243-254.

Atmaca, M., Onalan, E., Yildirim, H., Yuce, H., Koc, M., Korkmaz, S., & Mermi, O. (2011). Serotonin transporter gene polymorphism implicates reduced orbito-frontal cortex in obsessive-compulsive disorder. Journal of Anxiety Disorders, 25(5), 680-685.

Baca-Garcia, E., Vaquero-Lorenzo, C., Diaz-Hernandez, M., Rodriguez-Salgado, B., Dolengevich-Segal, H., Arrojo-Romero, M., Botillo-Martin, C., Ceverino, A., Piqueras, J. F., Perez-Rodriguez, M. M., & Saiz-Ruiz, J. (2007). Association between obsessive- compulsive disorder and a variable number of tandem repeats polymorphism in intron 2 of the serotonin transporter gene. Progress in Neuro-Psychopharmacology & Biological Psychiatry, 31(2), 416-420.

Bagdy, G., Kalogeras, K. T., & Szemeredi, K. (1992). Effect of 5-HT1C and 5-HT2 receptor stimulation on excessive grooming, penile erection and plasma oxytocin concentrations. European Journal of Pharmacology, 229(1), 9-14.

Berrios, G. E. (1989). Obsessive-compulsive disorder: Its conceptual history in france during the 19th century. Comprehensive Psychiatry, 30(4), 283-295.

Billett, E. A., Richter, M. A., King, N., Heils, A., Lesch, K. P., & Kennedy, J. L. (1997). Obsessive compulsive disorder, response to serotonin reuptake inhibitors and the serotonin transporter gene. Molecular Psychiatry, 2(5), 403-406.

Blier, P. & de Montigny, C. (1998). Possible serotonergic mechanisms underlying the antidepressant and anti-obsessive-compulsive disorder responses. Biological Psychiatry, 44(5), 313-323.

Blier, P., de Montigny, C., & Chaput, Y. (1990). A role for the serotonin system in the mechanism of action of antidepressant treatments: Preclinical evidence. The Journal of Clinical Psychiatry, 51 Suppl, 14-20; discussion 21.

Bloch, M. H., Bartley, C. A., Zipperer, L., Jakubovski, E., Landeros-Weisenberger, A., Pittenger, C., & Leckman, J. F. (2014). Meta-analysis: Hoarding symptoms associated with poor treatment outcome in obsessive-compulsive disorder. Molecular Psychiatry, 19(9), 1025-1030.

165 Bloch, M. H., Landeros-Weisenberger, A., Rosario, M. C., Pittenger, C., & Leckman, J. F. (2008a). Meta-analysis of the symptom structure of obsessive-compulsive disorder. The American Journal of Psychiatry, 165(12), 1532-1542.

Bloch, M. H., Landeros-Weisenberger, A., Sen, S., Dombrowski, P., Kelmendi, B., Coric, V., Pittenger, C., & Leckman, J. F. (2008b). Association of the serotonin transporter polymorphism and obsessive-compulsive disorder: Systematic review. American Journal of Medical Genetics.Part B, Neuropsychiatric Genetics : The Official Publication of the International Society of Psychiatric Genetics, 147B(6), 850-858.

Boedhoe, P.S., Schmaal, L., Abe, Y., Ameis, S. H., Arnold, P. D., Batistuzzo, M. C., Benedetti, F., Beucke, J. C., Bollettini, I., Bose, A., Brem, S., Calvo, A., Cheng, Y., Cho, K. I., Dallaspezia, S., Denys, D., Fitzgerald, K. D., Fouche, J. P., Giménez, M., Gruner, P., Hanna, G. L., Hibar, D. P., Hoexter, M. Q., Hu, H., Huyser, C., Ikari, K., Jahanshad, N., Kathmann, N., Kaufmann, C., Koch, K., Kwon, J. S., Lazaro, L., Liu, Y., Lochner, C., Marsh, R., Martínez-Zalacaín, I., Mataix-Cols, D., Menchón, J. M., Minuzzi, L., Nakamae, T., Nakao, T., Narayanaswamy, J. C., Piras, F., Piras, F., Pittenger, C., Reddy, Y. C., Sato, J. R., Simpson, H. B., Soreni, N., Soriano-Mas, C., Spalletta, G., Stevens, M. C., Szeszko, P. R., Tolin, D. F., Venkatasubramanian, G., Walitza, S., Wang, Z., van Wingen, G. A., Xu, J., Xu, X., Yun, J. Y., Zhao, Q., ENIGMA OCD Working Group, Thompson, P. M., Stein, D. J., & van den Heuvel, O. A. (2017). Distinct Subcortical Volume Alterations in Pediatric and Adult OCD: A Worldwide Meta- and Mega-Analysis. American Journal of Psychiatry, 174(1), 60-69.

Bogdan, R., Salmeron, B. J., Carey, C. E., Agrawal, A., Calhoun, V. D., Garavan, H., Hariri, A. R., Heinz, A., Hill, M. N., Holmes, A., Kalin, N. H., & Goldman, D. (2017). Imaging genetics and genomics in psychiatry: A critical review of progress and potential. Biological Psychiatry, 82(3), 165-175.

Boyce, W. T. & Ellis, B. J. (2005). Biological sensitivity to context: I. an evolutionary- developmental theory of the origins and functions of stress reactivity. Development and Psychopathology, 17(2), 271-301.

Brander, G., Rydell, M., Kuja-Halkola, R., Fernandez de la Cruz, L., Lichtenstein, P., Serlachius, E., Rück, C., Almqvist, C., D'Onofrio, B. M., Larsson, H., & Mataix-Cols, D. (2016). Association of perinatal risk factors with obsessive-compulsive disorder: A population-based birth cohort, sibling control study. JAMA Psychiatry, 73(11), 1135- 1144.

Brandl, E. J., Muller, D. J., & Richter, M. A. (2012). Pharmacogenetics of obsessive- compulsive disorders. Pharmacogenomics, 13(1), 71-81.

Brem, S., Hauser, T. U., Iannaccone, R., Brandeis, D., Drechsler, R., & Walitza, S. (2012). Neuroimaging of cognitive brain function in paediatric obsessive compulsive disorder: A review of literature and preliminary meta-analysis. Journal of Neural Transmission (Vienna), 119(11), 1425-1448.

166 Burton, C. L., Arnold, P. D., & Soreni, N. (2015a). Three reasons why studying hoarding in children and adolescents is important. Journal of the Canadian Academy of Child and Adolescent Psychiatry, 24(2), 128-130.

Burton, C. L., Crosbie, J., Dupuis, A., Mathews, C. A., Soreni, N., Schachar, R., & Arnold, P. D. (2016). Clinical correlates of hoarding with and without comorbid obsessive- compulsive symptoms in a community pediatric sample. Journal of the American Academy of Child and Adolescent Psychiatry, 55(2), 114-21.e2.

Burton, C. L., Crosbie, J., Erdman, L., Dupuis, A., Park, L. S., Sinopoli, V., Paterson, A., Strug, L., Schachar, R., & Arnold, P. D. A hypothesis-driven genome-wide association study of an obsessive-compulsive quantitative trait in a community-based sample of children and adolescents. Oral talk at the 23nd World Congress of Psychiatric Genetics Meeting October 2015b, Toronto, ON.

Burton, C. L., Park, L. S., Corfield, E. C., Forget-Dubois, N., Dupuis, A., Sinopoli, V. M., Shan, J., Goodale, T., Shaheen, S-M, Crosbie, J., Schachar, R. J., & Arnold, P. D. (2018). Heritability of obsessive-compulsive trait dimensions in youth from the general population. Translational Psychiatry, 8(1), 191.

Camarena, B., Aguilar, A., Loyzaga, C., & Nicolini, H. (2004). A family-based association study of the 5-HT-1Dbeta receptor gene in obsessive-compulsive disorder. The International Journal of Neuropsychopharmacology / Official Scientific Journal of the Collegium Internationale Neuropsychopharmacologicum (CINP), 7(1), 49-53.

Carlsson, A., Corrodi, H., Fuxe, K., & Hökfelt, T. (1969). Effect of antidepressant drugs on the depletion of intraneuronal brain 5-hydroxytryptamine stores caused by 4-methyl- alpha-ethyl-meta-tyramine. European Journal of Pharmacology, 5(4), 357-366.

Carter, C. S., Bearden, C. E., Bullmore, E. T., Geschwind, D. H., Glahn, D. C., Gur, R. E., Meyer-Lindenberg, A., & Weinberger, D. R. (2017). Enhancing the informativeness and replicability of imaging genomics studies. Biological Psychiatry, 82(3), 157-164.

Casey, B. J., Oliveri, M. E., & Insel, T. (2014). A neurodevelopmental perspective on the research domain criteria (RDoC) framework. Biological Psychiatry, 76(5), 350-353.

Caspi, A., Sugden, K., Moffitt, T. E., Taylor, A., Craig, I. W., Harrington, H., McClay, J., Mill, J., Martin, J., Braithwaite, A., & Poulton, R. (2003). Influence of life stress on depression: Moderation by a polymorphism in the 5-HTT gene. Science (New York, N.Y.), 301(5631), 386-389.

Cavallini, M. C., Di Bella, D., Pasquale, L., Henin, M., & Bellodi, L. (1998). 5HT2C CYS23/SER23 polymorphism is not associated with obsessive-compulsive disorder. Psychiatry Research, 77(2), 97-104.

167 Cavallini, M. C., Di Bella, D., Siliprandi, F., Malchiodi, F., & Bellodi, L. (2002). Exploratory factor analysis of obsessive-compulsive patients and association with 5-HTTLPR polymorphism. American Journal of Medical Genetics, 114(3), 347-353.

Chiarello, F., Spitoni, S., Hollander, E., Matucci Cerinic, M., & Pallanti, S. (2017). An expert opinion on PANDAS/PANS: Highlights and controversies. International Journal of Psychiatry in Clinical Practice, 21(2), 91-98. doi:10.1080/13651501.2017.1285941 [doi]

Chou-Green, J. M., Holscher, T. D., Dallman, M. F., & Akana, S. F. (2003). Compulsive behavior in the 5-HT2C receptor knockout mouse. Physiology & Behavior, 78(4-5), 641- 649.

Conley, D., Rauscher, E., & Siegal, M. L. (2013). Beyond orchids and dandelions: Testing the 5-HTT "risky" allele for evidence of phenotypic capacitance and frequency-dependent selection. Biodemography and Social Biology, 59(1), 37-56.

Coppen, A. (1967). The biochemistry of affective disorders. The British Journal of Psychiatry, 113(504),1236-1264.

Corregiari, F. M., Bernik, M., Cordeiro, Q., & Vallada, H. (2012). Endophenotypes and serotonergic polymorphisms associated with treatment response in obsessive-compulsive disorder. Clinics (Sao Paulo, Brazil), 67(4), 335-340.

Crosbie, J., Arnold, P., Paterson, A., Swanson, J., Dupuis, A., Li, X., Shan, J., Goodale, T., Tam, C., Strug, L. J., & Schachar, R. J. (2013). Response inhibition and ADHD traits: Correlates and heritability in a community sample. Journal of Abnormal Child Psychology, 41(3), 497-507.

Cullen, B., Brown, C. H., Riddle, M. A., Grados, M., Bienvenu, O. J., Hoehn-Saric, R., Shugart, Y. Y., Liang, K. Y., Samuels, J., & Nestadt, G. (2007). Factor analysis of the yale-brown obsessive compulsive scale in a family study of obsessive-compulsive disorder. Depression and Anxiety, 24(2), 130-138.

Dahlstroem, A. & Fuxe, K. (1964). Evidence for the existence of monoamine-containing neurons in the central nervous system. I. Demonstration of monoamine neurons in the cell bodies of brainstem neurons. Acta Physiologica Scandinavica, SUPPL 232, 1-55.

Davis, L. K., Yu, D., Keenan, C. L., Gamazon, E. R., Konkashbaev, A. I., Derks, E. M., Neale, B. M., Yang, J., Lee, S. H., Evans, P., Barr, C. L., Bellodi, L., Benarroch, F., Berrio, G. B., Bienvenu, O. J., Bloch, M. H., Blom, R. M., Bruun, R. D., Budman, C. L., Camarena, B., Campbell, D., Cappi, C., Cardona, Silgado, J. C., Cath, D. C., Cavallini, M. C., Chavira, D. A., Chouinard, S., Conti, D. V., Cook, E. H., Coric, V., Cullen, B. A., Deforce, D., Delorme, R., Dion, Y., Edlund, C. K., Egberts, K., Falkai, P., Fernandez, T. V., Gallagher, P. J., Garrido, H., Geller, D., Girard, S. L., Grabe, H. J., Grados, M. A., Greenberg, B. D., Gross-Tsur, V., Haddad, S., Heiman, G. A., Hemmings, S. M., Hounie, A. G., Illmann, C., Jankovic, J., Jenike, M. A., Kennedy, J. L., King, R. A., Kremeyer, B., Kurlan, R., Lanzagorta, N., Leboyer, M., Leckman, J. F., Lennertz, L., Liu, C., Lochner,

168 C., Lowe, T. L., Macciardi, F., McCracken, J. T., McGrath, L. M., Mesa Restrepo, S. C., Moessner, R., Morgan, J., Muller, H., Murphy, D. L., Naarden, A. L., Ochoa, W. C., Ophoff, R. A., Osiecki, L., Pakstis, A. J., Pato, M. T., Pato, C. N., Piacentini, J., Pittenger, C., Pollak, Y., Rauch, S. L., Renner, T. J., Reus, V. I., Richter, M. A., Riddle, M. A., Robertson, M. M., Romero, R., Rosàrio, M. C., Rosenberg, D., Rouleau, G. A., Ruhrmann, S., Ruiz-Linares, A., Sampaio, A. S., Samuels, J., Sandor, P., Sheppard, B., Singer, H. S., Smit, J. H., Stein, D. J., Strengman, E., Tischfield, J. A., Valencia Duarte, A. V., Vallada, H., Van Nieuwerburgh, F., Veenstra-Vanderweele, J., Walitza, S., Wang, Y., Wendland, J. R., Westenberg, H. G., Shugart, Y. Y., Miguel, E. C., McMahon, W., Wagner, M., Nicolini, H., Posthuma, D., Hanna, G. L., Heutink, P., Denys, D., Arnold, P. D., Oostra, B. A., Nestadt, G., Freimer, N. B., Pauls, D. L., Wray, N. R., Stewart, S. E., Mathews, C. A., Knowles, J. A., Cox, N. J., & Scharf, J. M. (2013). Partitioning the heritability of Tourette syndrome and obsessive compulsive disorder reveals differences in genetic architecture. PLoS Genetics, 9(10), e1003864.

Delaneau, O., Marchini, J., & 1000 Genomes Project Consortium. (2014). Integrating sequence and array data to create an improved 1000 genomes project haplotype reference panel. Nature Communications, 5, 3934.

Denys, D., Van Nieuwerburgh, F., Deforce, D., & Westenberg, H. G. (2006). Association between serotonergic candidate genes and specific phenotypes of obsessive compulsive disorder. Journal of Affective Disorders, 91(1), 39-44.

Denys, D., Van Nieuwerburgh, F., Deforce, D., & Westenberg, H. G. (2007). Prediction of response to paroxetine and venlafaxine by serotonin-related genes in obsessive- compulsive disorder in a randomized, double-blind trial. The Journal of Clinical Psychiatry, 68(5), 747-753.

Di Bella, D., Erzegovesi, S., Cavallini, M. C., & Bellodi, L. (2002). Obsessive-compulsive disorder, 5-HTTLPR polymorphism and treatment response. The Pharmacogenomics Journal, 2(3), 176-181.

Di Giovanni, G., Di Matteo, V., Pierucci, M., & Esposito, E. (2008). Serotonin-dopamine interaction: electrophysiological evidence. Progress in Brain Research, 172, 45-71.

Di Narzo, A. F., Kozlenkov, A., Roussos, P., Hao, K., Hurd, Y., Lewis, D. A., Sibille, E., Siever, L. J., Koonin, E., & Dracheva, S. (2014). A unique gene expression signature associated with serotonin 2C receptor RNA editing in the prefrontal cortex and altered in suicide. Human Molecular Genetics, 23(18), 4801-4813.

Dickel, D. E., Veenstra-VanderWeele, J., Bivens, N. C., Wu, X., Fischer, D. J., Van Etten-Lee, M., Himle, J. A., Leventhal, B. L., Cook, E. H. Jr., & Hanna, G. L. (2007). Association studies of serotonin system candidate genes in early-onset obsessive-compulsive disorder. Biological Psychiatry, 61(3), 322-329.

Do Rosario-Campos, M. C., Leckman, J. F., Curi, M., Quatrano, S., Katsovitch, L., Miguel, E. C., & Pauls, D. L. (2005). A family study of early‐onset obsessive‐compulsive disorder.

169 American Journal of Medical Genetics Part B: Neuropsychiatric Genetics, 136B(1), 92- 97.

Do Rosario-Campos, M. C., Leckman, J. F., Mercadante, M. T., Shavitt, R. G., Prado, H. S., Sada, P., Zamignani, D., & Miguel, E. C. (2001). Adults with early-onset obsessive- compulsive disorder. The American Journal of Psychiatry, 158(11), 1899-1903.

Dougherty, D. D., Brennan, B. P., Stewart, S. E., Wilhelm, S., Widge, A. S., & Rauch, S. L. (2018). Neuroscientifically informed formulation and treatment planning for patients with obsessive-compulsive disorder: A review. JAMA Psychiatry, 75(10), 1081-1087.

Dracheva, S., Patel, N., Woo, D. A., Marcus, S. M., Siever, L. J., & Haroutunian, V. (2008). Increased serotonin 2C receptor mRNA editing: A possible risk factor for suicide. Molecular Psychiatry, 13(11), 1001-1010.

Drago, A. & Serretti, A. (2009). Focus on HTR2C: A possible suggestion for genetic studies of complex disorders.American Journal of Medical Genetics.Part B, Neuropsychiatric Genetics : The Official Publication of the International Society of Psychiatric Genetics, 150B(5), 601-637.

Duman, E. A. & Canli, T. (2015). Influence of life stress, 5-HTTLPR genotype, and SLC6A4 methylation on gene expression and stress response in healthy caucasian males. Biology of Mood & Anxiety Disorders, 5, 2-015-0017-x. eCollection 2015.

Dzidic, M., Boix-Amoros, A., Selma-Royo, M., Mira, A., & Collado, M. C. (2018). Gut microbiota and mucosal immunity in the neonate. Medical Sciences (Basel, Switzerland), 6(3), 10.3390/medsci6030056.

Enoch, M. A., Greenberg, B. D., Murphy, D. L., & Goldman, D. (2001). Sexually dimorphic relationship of a 5-HT2A promoter polymorphism with obsessive-compulsive disorder. Biological Psychiatry, 49(4), 385-388.

Enoch, M. A., Kaye, W. H., Rotondo, A., Greenberg, B. D., Murphy, D. L., & Goldman, D. (1998). 5-HT2A promoter polymorphism -1438G/A, anorexia nervosa, and obsessive- compulsive disorder. Lancet (London, England), 351(9118), 1785-1786.

Erspamer, V. & Asero, B. (1952). Identification of enteramine, the specific hormone of the enterochromaffin cell system, as 5-hydroxytryptamine. Nature, 169(1952), 800-801.

Fagerberg, L., Hallstrom, B. M., Oksvold, P., Kampf, C., Djureinovic, D., Odeberg, J., Habuka M., Tahmasebpoor, S., Danielsson, A., Edlund, K., Asplund, A., Sjöstedt, E., Lundberg, E., Szigyarto, C. A., Skogs, M., Takanen, J. O., Berling, H., Tegel, H., Mulder, J., Nilsson, P., Schwenk, J. M., Lindskog, C., Danielsson, F., Mardinoglu, A., Sivertsson, A., von Feilitzen, K., Forsberg, M., Zwahlen, M., Olsson, I., Navani, S., Huss, M., Nielsen, J., Ponten, F., & Uhlen, M. (2014). Analysis of the human tissue-specific expression by genome-wide integration of transcriptomics and antibody-based proteomics. Molecular & Cellular Proteomics : MCP, 13(2), 397-406.

170

Fenske, J. N. & Petersen, K. (2015). Obsessive-Compulsive Disorder: Diagnosis and Management. American Family Physician, 92(10), 896-903.

Fernandez, T. V. & Leckman, J. F. (2016). Prenatal and perinatal risk factors and the promise of birth cohort studies: Origins of obsessive-compulsive disorder. JAMA Psychiatry, 73(11), 1117-1118.

Flament, M. (1990). Epidemiology of obsessive-compulsive disorder in children and adolescents. [Epidemiologie du trouble obsessionnel-compulsif chez l'enfant et l'adolescent] L'Encephale, 16 Spec No, 311-316.

Fox, M. A., Panessiti, M. G., Hall, F. S., Uhl, G. R., & Murphy, D. L. (2013). An evaluation of the serotonin system and perseverative, compulsive, stereotypical, and hyperactive behaviors in dopamine transporter (DAT) knockout mice. Psychopharmacology, 227(4), 685-695.

Frisch, A., Michaelovsky, E., Rockah, R., Amir, I., Hermesh, H., Laor, N., Fuchs, C., Zohar, J., Lerer, B., Buniak, S. F., Landa, S., Poyurovsky, M., Shapira, B., & Weizman, R. (2000). Association between obsessive-compulsive disorder and polymorphisms of genes encoding components of the serotonergic and dopaminergic pathways. European Neuropsychopharmacology : The Journal of the European College of Neuropsychopharmacology, 10(3), 205-209.

Frost, R. O., & Gross, R. C. (1993). The hoarding of possessions. Behaviour Research and Therapy, 31(4), 367-381.

Frost, R. O., Steketee, G., & Tolin, D. F. (2011). Comorbidity in hoarding disorder. Depression and Anxiety, 28(10), 876-884.

Gelernter, J., Kranzler, H., & Cubells, J. F. (1997). Serotonin transporter protein (SLC6A4) allele and haplotype frequencies and linkage disequilibria in african- and european- american and japanese populations and in alcohol-dependent subjects. Human Genetics, 101(2), 243-246.

Geller, D. A. & March, J. The AACAP Committee on Quality Issues (CQI). (2012). Practice parameter for the assessment and treatment of children and adolescents with obsessive- compulsive disorder. Journal of the American Academy of Child and Adolescent Psychiatry, 51(1), 98-113.

Geschwind, D. H. & Flint, J. (2015). Genetics and genomics of psychiatric disease. Science (New York, N.Y.), 349(6255), 1489-1494.

Gibney, E. R., & Nolan, C. M. (2010). Epigenetics and gene expression. Heredity, 105(1), 4- 13.

171 Goodman, W. K., Price, L. H., Rasmussen, S. A., Mazure, C., Delgado, P., Heninger, G. R., & Charney, D. S. (1989a). The Yale-Brown obsessive compulsive scale. II. validity. Archives of General Psychiatry, 46(11), 1012-1016.

Goodman, W. K., Price, L. H., Rasmussen, S. A., Mazure, C., Fleischmann, R. L., Hill, C. L., Heninger, G. R., & Charney, D. S. (1989b). The Yale-Brown obsessive compulsive scale. I. development, use, and reliability. Archives of General Psychiatry, 46(11), 1006-1011.

Gordon, E. (2003). Integrative neuroscience. Neuropsychopharmacology : Official Publication of the American College of Neuropsychopharmacology, 28 Suppl 1, S2-8.

Gottesman, I. I. & Gould, T. D. (2003). The endophenotype concept in psychiatry: Etymology and strategic intentions. The American Journal of Psychiatry, 160(4), 636-645.

Grados, M. A. (2010). The genetics of obsessive-compulsive disorder and tourette syndrome: An epidemiological and pathway-based approach for gene discovery. Journal of the American Academy of Child and Adolescent Psychiatry, 49(8), 810-9, 819.e1-2.

Greenberg, B. D., Tolliver, T. J., Huang, S. J., Li, Q., Bengel, D., & Murphy, D. L. (1999). Genetic variation in the serotonin transporter promoter region affects serotonin uptake in human blood platelets. American Journal of Medical Genetics, 88(1), 83-87.

Gurevich, I., Tamir, H., Arango, V., Dwork, A. J., Mann, J. J., & Schmauss, C. (2002). Altered editing of serotonin 2C receptor pre-mRNA in the prefrontal cortex of depressed suicide victims. Neuron, 34(3), 349-356.

Haddley, K., Vasiliou, A. S., Ali, F. R., Paredes, U. M., Bubb, V. J., & Quinn, J. P. (2008). Molecular genetics of monoamine transporters: Relevance to brain disorders. Neurochemical Research, 33(4), 652-667.

Hanna, G. L. (2010). Schedule for Obsessive-Compulsive and Other Behavioral Syndromes (SOCOBS). University of Michigan, Ann Arbor, MI.

Hanna, G. L., Himle, J. A., Curtis, G. C., & Gillespie, B. W. (2005). A family study of obsessive‐compulsive disorder with pediatric probands. American Journal of Medical Genetics Part B: Neuropsychiatric Genetics, 134B(1), 13-19.

Hasler, G., Kazuba, D., & Murphy, D. L. (2006). Factor analysis of obsessive-compulsive disorder YBOCS-SC symptoms and association with 5-HTTLPR SERT polymorphism. American Journal of Medical Genetics.Part B, Neuropsychiatric Genetics : The Official Publication of the International Society of Psychiatric Genetics, 141B(4), 403-408.

Hasselbalch, S. G., Hansen, E. S., Jakobsen, T. B., Pinborg, L. H., Lonborg, J. H., & Bolwig, T. G. (2007). Reduced midbrain-pons serotonin transporter binding in patients with obsessive-compulsive disorder. Acta Psychiatrica Scandinavica, 115(5), 388-394.

172 Heils, A., Teufel, A., Petri, S., Stober, G., Riederer, P., Bengel, D., & Lesch, K. P. (1996). Allelic variation of human serotonin transporter gene expression. Journal of Neurochemistry, 66(6), 2621-2624.

Heinz, A., Jones, D. W., Mazzanti, C., Goldman, D., Ragan, P., Hommer, D., Linnoila, M., & Weinberger, D. R. (2000). A relationship between serotonin transporter genotype and in vivo protein expression and alcohol neurotoxicity. Biological Psychiatry, 47(7), 643-649.

Hemmings, S. M., Kinnear, C. J., Niehaus, D. J., Moolman-Smook, J. C., Lochner, C., Knowles, J. A., Corfield, V.A., & Stein, D. J. (2003). Investigating the role of dopaminergic and serotonergic candidate genes in obsessive-compulsive disorder. European Neuropsychopharmacology : The Journal of the European College of Neuropsychopharmacology, 13(2), 93-98.

Hesse, S., Muller, U., Lincke, T., Barthel, H., Villmann, T., Angermeyer, M. C., Sabri, O., & Stengler-Wenzke, K. (2005). Serotonin and dopamine transporter imaging in patients with obsessive-compulsive disorder. Psychiatry Research, 140(1), 63-72.

Hesse, S., Stengler, K., Regenthal, R., Patt, M., Becker, G. A., Franke, A., Knüpfer, H., Meyer, P. M., Luthardt, J., Jahn, I., Lobsien, D., Heinke, W., Brust, P., Hegerl, U., & Sabri, O. (2011). The serotonin transporter availability in untreated early-onset and late-onset patients with obsessive-compulsive disorder. The International Journal of Neuropsychopharmacology / Official Scientific Journal of the Collegium Internationale Neuropsychopharmacologicum (CINP), 14(5), 606-617.

Hettema, J. M., Neale, M. C., & Kendler, K. S. (2001). A review and meta-analysis of the genetic epidemiology of anxiety disorders. The American Journal of Psychiatry, 158(10), 1568-1578.

Honda, S., Nakao, T., Mitsuyasu, H., Okada, K., Gotoh, L., Tomita, M., Sanematsu, H., Murayama, K., Ikari, K., Kuwano, M., Yoshiura, T., Kawasaki, H., & Kanba, S. (2017). A pilot study exploring the association of morphological changes with 5-HTTLPR polymorphism in OCD patients. Annals of General Psychiatry, 16, 2-017-0126-6. eCollection 2017.

Howie, B. N., Donnelly, P., & Marchini, J. (2009). A flexible and accurate genotype imputation method for the next generation of genome-wide association studies. PLoS Genetics, 5(6), e1000529.

Hu, X. Z., Lipsky, R. H., Zhu, G., Akhtar, L. A., Taubman, J., Greenberg, B. D., Xu, K., Arnold, P. D., Richter, M. A., Kennedy, J. L., Murphy, D. L., & Goldman, D. (2006). Serotonin transporter promoter gain-of-function genotypes are linked to obsessive- compulsive disorder. American Journal of Human Genetics, 78(5), 815-826.

Iervolino, A. C., Perroud, N., Fullana, M. A., Guipponi, M., Cherkas, L., Collier, D. A., & Mataix-Cols, D. (2009). Prevalence and heritability of compulsive hoarding: A twin study. The American Journal of Psychiatry, 166(10), 1156-1161.

173

Iervolino, A. C., Rijsdijk, F. V., Cherkas, L., Fullana, M. A., & Mataix-Cols, D. (2011). A multivariate twin study of obsessive-compulsive symptom dimensions. Archives of General Psychiatry, 68(6), 637-644.

Insel, T., Cuthbert, B., Garvey, M., Heinssen, R., Pine, D. S., Quinn, K., Sanislow, C., & Wang, P. (2010). Research domain criteria (RDoC): Toward a new classification framework for research on mental disorders. The American Journal of Psychiatry, 167(7), 748-751.

Ivanov, V. Z., Mataix-Cols, D., Serlachius, E., Lichtenstein, P., Anckarsater, H., Chang, Z., Gumpert, C. H., Lundström, S., Långström, N., & Ruck, C. (2013). Prevalence, comorbidity and heritability of hoarding symptoms in adolescence: A population based twin study in 15-year olds. PloS One, 8(7), e69140.

Ivanov, V. Z., Nordsletten, A., Mataix-Cols, D., Serlachius, E., Lichtenstein, P., Lundstrom, S., Magnusson, P. K. E., Kuja-Halkola, R., & Ruck, C. (2017). Heritability of hoarding symptoms across adolescence and young adulthood: A longitudinal twin study. PloS One, 12(6), e0179541.

Iwamoto, K. & Kato, T. (2003). RNA editing of serotonin 2C receptor in human postmortem brains of major mental disorders. Neuroscience Letters, 346(3), 169-172.

Jarbin, H., Andersson, M., Rastam, M., & Ivarsson, T. (2017). Predictive validity of the K- SADS-PL 2009 version in school-aged and adolescent outpatients. Nordic Journal of Psychiatry, 71(4), 270-276.

Jenike, M. A. (1983). Obsessive compulsive disorder. Comprehensive Psychiatry, 24(2), 99- 115.

Kang, H. J., Kawasawa, Y. I., Cheng, F., Zhu, Y., Xu, X., Li, M., Sousa, A. M., Pletikos, M., Meyer, K. A., Sedmak, G., Guennel, T., Shin, Y., Johnson, M. B., Krsnik, Z., Mayer, S., Fertuzinhos, S., Umlauf, S., Lisgo, S. N., Vortmeyer, A., Weinberger, D. R., Mane, S., Hyde, T. M., Huttner, A., Reimers, M., Kleinman, J. E., & Sestan, N. (2011). Spatio- temporal transcriptome of the human brain. Nature, 478(7370), 483-489.

Karg, K., Burmeister, M., Shedden, K., & Sen, S. (2011). The serotonin transporter promoter variant (5-HTTLPR), stress, and depression meta-analysis revisited: Evidence of genetic moderation. Archives of General Psychiatry, 68(5), 444-454.

Katerberg, H., Delucchi, K. L., Stewart, S. E., Lochner, C., Denys, D. A., Stack, D. E., Andresen, J. M., Grant, J. E., Kim, S. W., Williams, K. A., den Boer, J. A., van Balkom, A. J., Smit, J. H., van Oppen, P., Polman, A., Jenike, M. A., Stein, D. J., Mathews, C. A., & Cath, D. C. (2010). Symptom dimensions in OCD: Item-level factor analysis and heritability estimates. Behavior Genetics, 40(4), 505-517.

Kaufman, J., Birmaher, B., Brent, D., Rao, U., Flynn, C., Moreci, P., Williamson, D., & Ryan,

174 N. (1997). Schedule for Affective Disorders and Schizophrenia for School-Age Children- Present and Lifetime Version (K-SADS-PL): initial reliability and validity data. Journal of the American Academy of Child and Adolescent Psychiatry, 36(7):980-8.

Kessler, R. C., Berglund, P., Demler, O., Jin, R., Merikangas, K. R., & Walters, E. E. (2005). Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the national comorbidity survey replication. Archives of General Psychiatry, 62(6), 593-602.

Kilic, F., Murphy, D. L., & Rudnick, G. (2003). A human serotonin transporter mutation causes constitutive activation of transport activity. Molecular Pharmacology, 64(2), 440- 446.

Kim, S. J., Lee, H. S., & Kim, C. H. (2005). Obsessive-compulsive disorder, factor-analyzed symptom dimensions and serotonin transporter polymorphism. Neuropsychobiology, 52(4), 176-182.

Köhler-Forsberg, O., Petersen, L., Gasse, C., Mortensen, P. B., Dalsgaard, S., Yolken, R. H., Mors, O., & Benros, M. E. (2018). A nationwide study in denmark of the association between treated infections and the subsequent risk of treated mental disorders in children and adolescents. JAMA Psychiatry, 76(3), 271-279.

Kristensen, K. & Henriksen, L. (2016). Cesarean section and disease associated with immune function. The Journal of Allergy and Clinical Immunology, 137(2), 587-590.

Kuhn, R. (1957). Treatment of depressive states with an iminodibenzyl derivative (G 22355). [Uber die Behandlung depressiver Zustande mit einem Iminodibenzylderivat (G 22355)] Schweizerische Medizinische Wochenschrift, 87(35-36), 1135-1140.

Kundakovic, M. & Champagne, F. A. (2015). Early-life experience, epigenetics, and the developing brain. Neuropsychopharmacology : Official Publication of the American College of Neuropsychopharmacology, 40(1), 141-153.

Kwon, J. M. & Goate, A. M. (2000). The candidate gene approach. Alcohol Research & Health : The Journal of the National Institute on Alcohol Abuse and Alcoholism, 24(3), 164-168.

Landeros-Weisenberger, A., Bloch, M. H., Kelmendi, B., Wegner, R., Nudel, J., Dombrowski, P., Pittenger, C., Krystal, J. H., Goodman, W. K., Leckman, J. F., & Coric, V. (2010). Dimensional predictors of response to SRI pharmacotherapy in obsessive-compulsive disorder. Journal of Affective Disorders, 121(1-2), 175-179.

Lappalainen, J., Zhang, L., Dean, M., Oz, M., Ozaki, N., Yu, D. H., Virkkunen, M., Weight, F., Linnoila, M., & Goldman, D. (1995). Identification, expression, and pharmacology of a Cys23-Ser23 substitution in the human 5-HT2c receptor gene (HTR2C). Genomics, 27(2), 274-279.

175 Leckman, J. F., Denys, D., Simpson, H. B., Mataix-Cols, D., Hollander, E., Saxena, S., Miguel, E. C., Rauch, S. L., Goodman, W. K., Phillips, K. A., & Stein, D. J. (2010). Obsessive- compulsive disorder: A review of the diagnostic criteria and possible subtypes and dimensional specifiers for DSM-V. Depression and Anxiety, 27(6), 507-527.

Leckman, J. F., Rauch, S. L., & Mataix-Cols, D. (2007). Symptom dimensions in obsessive- compulsive disorder: Implications for the DSM-V. CNS Spectrums, 12(5), 376-87, 400.

Leite, J., Hespanhol, R., & Buchpiguel, C. A. (2015). Molecular imaging in genetics. Neuroimaging Clinics of North America, 25(1), 17-29.

Lensi, P., Cassano, G. B., Correddu, G., Ravagli, S., Kunovac, J. L., & Akiskal, H. S. (1996). Obsessive-compulsive disorder. familial-developmental history, symptomatology, comorbidity and course with special reference to gender-related differences. The British Journal of Psychiatry : The Journal of Mental Science, 169(1), 101-107.

Lesch, K. P., Balling, U., Gross, J., Strauss, K., Wolozin, B. L., Murphy, D. L., & Riederer, P. (1994). Organization of the human serotonin transporter gene. Journal of Neural Transmission.General Section, 95(2), 157-162.

Lesch, K. P., Bengel, D., Heils, A., Sabol, S. Z., Greenberg, B. D., Petri, S., Benjamin, J., Müller, C. R., Hamer, D. H., & Murphy, D. L. (1996). Association of anxiety-related traits with a polymorphism in the serotonin transporter gene regulatory region. Science (New York, N.Y.), 274(5292), 1527-1531.

Lesch, K. P., Wolozin, B. L., Murphy, D. L., & Reiderer, P. (1993). Primary structure of the human platelet serotonin uptake site: Identity with the brain serotonin transporter. Journal of Neurochemistry, 60(6), 2319-2322.

Lestrade, L. & Weber, M. J. (2006). snoRNA-LBME-db, a comprehensive database of human H/ACA and C/D box snoRNAs. Nucleic Acids Research, 34(Database issue), D158-62.

Li, M. X., Yeung, J. M., Cherny, S. S., & Sham, P. C. (2012). Evaluating the effective numbers of independent tests and significant p-value thresholds in commercial genotyping arrays and public imputation reference datasets. Human Genetics, 131(5), 747-756.

Lin, P. Y. (2007). Meta-analysis of the association of serotonin transporter gene polymorphism with obsessive-compulsive disorder. Progress in Neuro- Psychopharmacology & Biological Psychiatry, 31(3), 683-689.

Little, K. Y., McLaughlin, D. P., Zhang, L., Livermore, C. S., Dalack, G. W., McFinton, P. R., DelProposto, Z. S., Hill, E., Cassin, B. J., Watson, S. J., & Cook, E. H. (1998). Cocaine, ethanol, and genotype effects on human midbrain serotonin transporter binding sites and mRNA levels. The American Journal of Psychiatry, 155(2), 207-213.

176 Liu, W., Zhao, N., Xiong, J., Shi, M., & Hu, J. (2011). Association analysis of serotonin and catecholamine system candidate genes in obsessive-compulsive disorder in the chinese population. Psychiatry Research, 188(1), 170-172.

Lochner, C., Hemmings, S. M., Kinnear, C. J., Moolman-Smook, J. C., Corfield, V. A., Knowles, J. A., Niehaus D. J., & Stein, D. J. (2004). Gender in obsessive-compulsive disorder: Clinical and genetic findings. European Neuropsychopharmacology : The Journal of the European College of Neuropsychopharmacology, 14(2), 105-113.

Lochner, C., Kinnear, C. J., Hemmings, S. M., Seller, C., Niehaus, D. J., Knowles, J. A., Daniels, W., Moolman-Smook, J. C., Seedat, S., & Stein, D. J. (2005). Hoarding in obsessive-compulsive disorder: Clinical and genetic correlates. The Journal of Clinical Psychiatry, 66(9), 1155-1160.

Lu, Z. X., Jiang, P., & Xing, Y. (2012). Genetic variation of pre-mRNA alternative splicing in human populations. Wiley Interdisciplinary Reviews.RNA, 3(4), 581-592.

Lyddon, R., Dwork, A. J., Keddache, M., Siever, L. J., & Dracheva, S. (2013). Serotonin 2c receptor RNA editing in major depression and suicide. The World Journal of Biological Psychiatry : The Official Journal of the World Federation of Societies of Biological Psychiatry, 14(8), 590-601.

MacMaster, F. P. (2010). Translational neuroimaging research in pediatric obsessive- compulsive disorder. Dialogues in Clinical Neuroscience, 12(2), 165-174.

Magnotta, V. A., Harris, G., Andreasen, N. C., O'Leary, D. S., Yuh, W. T., & Heckel, D. (2002). Structural MR image processing using the BRAINS2 toolbox. Computerized Medical Imaging and Graphics : The Official Journal of the Computerized Medical Imaging Society, 26(4), 251-264.

Mak, L., Streiner, D. L., & Steiner, M. (2015). Is serotonin transporter polymorphism (5- HTTLPR) allele status a predictor for obsessive-compulsive disorder? A meta-analysis. Archives of Women's Mental Health, 18(3), 435-445.

Martin, J. H. (Ed.). (2012). Neuroanatomy Text and Atlas, Fourth Edition. New York, New York: McGraw-Hill Education.

Masson, P. (1914). La glande endocrine de l'intestin chez l'homme. Comptes Rendus de l'Académie des Sciences, Paris, 158, 59-61.

Mathews, C. A., Delucchi, K., Cath, D. C., Willemsen, G., & Boomsma, D. I. (2014). Partitioning the etiology of hoarding and obsessive-compulsive symptoms. Psychological Medicine, 44(13), 2867-2876.

Mataix-Cols, D., do Rosario-Campos, M. C., & Leckman, J. F. (2005). A multidimensional model of obsessive-compulsive disorder. The American Journal of Psychiatry, 162(2), 228-238.

177

Mataix-Cols, D. & Pertusa, A. (2012). Annual research review: Hoarding disorder: Potential benefits and pitfalls of a new mental disorder. Journal of Child Psychology and Psychiatry, and Allied Disciplines, 53(5), 608-618.

Matsumoto, R., Ichise, M., Ito, H., Ando, T., Takahashi, H., Ikoma, Y., Kosaka, J., Arakawa, R., Fujimura, Y., Ota, M., Takano, A., Fukui, K., Nakayama, K., & Suhara, T. (2010). Reduced serotonin transporter binding in the insular cortex in patients with obsessive- compulsive disorder: A [11C]DASB PET study. Neuroimage, 49(1), 121-126.

Mattheisen, M., Samuels, J. F., Wang, Y., Greenberg, B. D., Fyer, A. J., McCracken, J. T., Geller, D. A., Murphy, D. L., Knowles, J. A., Grados, M. A., Riddle, M. A., Rasmussen S. A., McLaughlin, N. C., Nurmi, E. L., Askland, K. D., Qin, H. D., Cullen, B. A., Piacentini, J., Pauls, D. L., Bienvenu, O. J., Stewart, S. E., Liang, K. Y., Goes, F. S., Maher, B., Pulver, A. E., Shugart, Y. Y., Valle, D., Lange C., & Nestadt, G. (2015). Genome-wide association study in obsessive-compulsive disorder: Results from the OCGAS. Molecular Psychiatry, 20(3), 337-344.

Mattina, G. F. & Steiner, M. (2016). The need for inclusion of sex and age of onset variables in genetic association studies of obsessive-compulsive disorder: Overview. Progress in Neuro-Psychopharmacology & Biological Psychiatry, 67, 107-116.

McDougle, C. J., Epperson, C. N., Price, L. H., & Gelernter, J. (1998). Evidence for linkage disequilibrium between serotonin transporter protein gene (SLC6A4) and obsessive compulsive disorder. Molecular Psychiatry, 3(3), 270-273.

Meira-Lima, I., Shavitt, R. G., Miguita, K., Ikenaga, E., Miguel, E. C., & Vallada, H. (2004). Association analysis of the catechol-o-methyltransferase (COMT), serotonin transporter (5-HTT) and serotonin 2A receptor (5HT2A) gene polymorphisms with obsessive- compulsive disorder. Genes, Brain, and Behavior, 3(2), 75-79.

Meyer-Lindenberg, A. & Weinberger, D. R. (2006). Intermediate phenotypes and genetic mechanisms of psychiatric disorders. Nature Reviews.Neuroscience, 7(10), 818-827.

Miguel, E. C., Leckman, J. F., Rauch, S., do Rosario-Campos, M. C., Hounie, A. G., Mercadante, M. T., Chacon, P., & Pauls, D. L. (2005). Obsessive-compulsive disorder phenotypes: Implications for genetic studies. Molecular Psychiatry, 10(3), 258-275.

Miguita, K., Cordeiro, Q., Shavitt, R. G., Miguel, E. C., & Vallada, H. (2011). Association study between genetic monoaminergic polymorphisms and OCD response to clomipramine treatment. Arquivos De Neuro-Psiquiatria, 69(2B), 283-287.

Millan, M. J., Dekeyne, A., Papp, M., La Rochell, C. D., MacSweeney, C., Peglion, J.L., Brocco, M. (2001). S33005, a novel ligand at both serotonin and norepinephrine transporters: II. Behavioral profile in comparison with venlafaxine, reboxetine, citalopram, and clomipramine. Journal of Pharmacology and Experimental Therapeutics, 298(2), 581-591.

178

Millan, M. J., Goodwin, G. M., Meyer-Lindenberg, A., & Ove Ogren, S. (2015). Learning from the past and looking to the future: Emerging perspectives for improving the treatment of psychiatric disorders. European Neuropsychopharmacology : The Journal of the European College of Neuropsychopharmacology, 25(5), 599-656.

Molineaux, S. M., Jessell, T. M., Axel, R., & Julius, D. (1989). 5-HT1c receptor is a prominent serotonin receptor subtype in the central nervous system. Proceedings of the National Academy of Sciences of the United States of America, 86(17), 6793-6797.

Monzani, B., Rijsdijk, F., Harris, J., & Mataix-Cols, D. (2014). The structure of genetic and environmental risk factors for dimensional representations of DSM-5 obsessive- compulsive spectrum disorders. JAMA Psychiatry, 71(2), 182-189.

Mundo, E., Richter, M. A., Zai, G., Sam, F., McBride, J., Macciardi, F., & Kennedy, J. L. (2002). 5HT1Dbeta receptor gene implicated in the pathogenesis of obsessive-compulsive disorder: Further evidence from a family-based association study. Molecular Psychiatry, 7(7), 805-809.

Murphy, D. L., Fox, M. A., Timpano, K. R., Moya, P. R., Ren-Patterson, R., Andrews, A. M., Holmes, A., Lesch, K. P., & Wendland, J. R. (2008). How the serotonin story is being rewritten by new gene-based discoveries principally related to SLC6A4, the serotonin transporter gene, which functions to influence all cellular serotonin systems. Neuropharmacology, 55(6), 932-960.

Murphy, D. L., Lerner, A., Rudnick, G., & Lesch, K. P. (2004). Serotonin transporter: Gene, genetic disorders, and pharmacogenetics. Molecular Interventions, 4(2), 109-123.

Murphy, D. L. & Lesch, K. P. (2008). Targeting the murine serotonin transporter: Insights into human neurobiology. Nature Reviews.Neuroscience, 9(2), 85-96.

Murphy, D. L. & Moya, P. R. (2011). Human serotonin transporter gene (SLC6A4) variants: Their contributions to understanding pharmacogenomic and other functional GxG and GxE differences in health and disease. Current Opinion in Pharmacology, 11(1), 3-10.

Murphy, D. L., Moya, P. R., Fox, M. A., Rubenstein, L. M., Wendland, J. R., & Timpano, K. R. (2013). Anxiety and affective disorder comorbidity related to serotonin and other neurotransmitter systems: Obsessive-compulsive disorder as an example of overlapping clinical and genetic heterogeneity. Philosophical Transactions of the Royal Society of London.Series B, Biological Sciences, 368(1615), 20120435.

Murray, C. J. & Lopez, A. D. (1996). Evidence-based health policy--lessons from the global burden of disease study. Science (New York, N.Y.), 274(5288), 740-743.

Mylecharane, E. J. (2013). Serotonin and the australian connection: The science and the people. ACS Chemical Neuroscience, 4(1), 28-32.

179 Nakamura, M., Ueno, S., Sano, A., & Tanabe, H. (2000). The human serotonin transporter gene linked polymorphism (5-HTTLPR) shows ten novel allelic variants. Molecular Psychiatry, 5(1), 32-38.

National Institute of Mental Health. (2008). National Institute of Mental Health strategic plan. Retrieved from http://www.nimh.nih.gov/about/strategic-planning-reports/nimh-strategic- plan-2008.pdf

Nichols, D. E. & Nichols, C. D. (2008). Serotonin receptors. Chemical Reviews, 108(5), 1614- 1641.

Nicolini, H., Arnold, P., Nestadt, G., Lanzagorta, N., & Kennedy, J. L. (2009). Overview of genetics and obsessive-compulsive disorder. Psychiatry Research, 170(1), 7-14.

Nissen, J. B., Hansen, C. S., Starnawska, A., Mattheisen, M., Borglum, A. D., Buttenschon, H. N., & Hollegaard, M. (2016). DNA methylation at the neonatal state and at the time of diagnosis: Preliminary support for an association with the estrogen receptor 1, gamma- aminobutyric acid B receptor 1, and myelin oligodendrocyte glycoprotein in female adolescent patients with OCD. Frontiers in Psychiatry, 7, 35.

Niswender, C. M., Herrick-Davis, K., Dilley, G. E., Meltzer, H. Y., Overholser, J. C., Stockmeier, C. A., Emeson, R. B., & Sanders-Bush, E. (2001). RNA editing of the human serotonin 5-HT2C receptor. Alterations in suicide and implications for serotonergic pharmacotherapy. Neuropsychopharmacology : Official Publication of the American College of Neuropsychopharmacology, 24(5), 478-491.

Niven, J. E. (2004). Channelling evolution: canalization and the nervous system. PLoS Biology, 2(1), E19.

Nonnis Marzano, F., Maldini, M., Filonzi, L., Lavezzi, A. M., Parmigiani, S., Magnani, C., Bevilacqua, G., & Matturri, L. (2008). Genes regulating the serotonin metabolic pathway in the brain stem and their role in the etiopathogenesis of the sudden infant death syndrome. Genomics, 91(6), 485-491.

Norton, W. H., Folchert, A., & Bally-Cuif, L. (2008). Comparative analysis of serotonin receptor (HTR1A/HTR1B families) and transporter (slc6a4a/b) gene expression in the zebrafish brain. The Journal of Comparative Neurology, 511(4), 521-542.

Ogilvie, A. D., Battersby, S., Bubb, V. J., Fink, G., Harmar, A. J., Goodwim, G. M., & Smith, C. A. (1996). Polymorphism in serotonin transporter gene associated with susceptibility to major depression. Lancet (London, England), 347(9003), 731-733.

Ohara, K., Suzuki, Y., Ochiai, M., Tsukamoto, T., Tani, K., & Ohara, K. (1999). A variable- number-tandem-repeat of the serotonin transporter gene and anxiety disorders. Progress in Neuro-Psychopharmacology & Biological Psychiatry, 23(1), 55-65.

180 Ozaki, N., Goldman, D., Kaye, W. H., Plotnicov, K., Greenberg, B. D., Lappalainen, J., Rudnick, G., & Murphy, D. L. (2003). Serotonin transporter missense mutation associated with a complex neuropsychiatric phenotype. Molecular Psychiatry, 8(11), 933-936.

Palacios, J. M. (2016). Serotonin receptors in brain revisited. Brain Research, 1645, 46-49.

Park, L. S., Burton, C. L., Dupuis, A., Shan, J., Storch, E. A., Crosbie, J., Schachar, R. J., & Arnold, P. D. (2016). The Toronto Obsessive-Compulsive Scale: Psychometrics of a dimensional measure of obsessive-compulsive traits. Journal of the American Academy of Child and Adolescent Psychiatry, 55(4), 310-318.e4.

Pauls, D. L., Abramovitch, A., Rauch, S. L., & Geller, D. A. (2014). Obsessive-compulsive disorder: An integrative genetic and neurobiological perspective. Nature Reviews.Neuroscience, 15(6), 410-424.

Perani, D., Garibotto, V., Gorini, A., Moresco, R. M., Henin, M., Panzacchi, A., Matarrese, M., Carpinelli, A., Bellodi, L., & Fazio, F. (2008). In vivo PET study of 5HT(2A) serotonin and D(2) dopamine dysfunction in drug-naive obsessive-compulsive disorder. Neuroimage, 42(1), 306-314.

Perera, F. & Herbstman, J. (2011). Prenatal environmental exposures, epigenetics, and disease. Reproductive Toxicology (Elmsford, N.Y.), 31(3), 363-373.

Pertusa, A., Frost, R. O., & Mataix-Cols, D. (2010). When hoarding is a symptom of OCD: A case series and implications for DSM-V. Behaviour Research and Therapy, 48(10), 1012- 1020.

Pertusa, A., Fullana, M. A., Singh, S., Alonso, P., Menchon, J. M., & Mataix-Cols, D. (2008). Compulsive hoarding: OCD symptom, distinct clinical syndrome, or both? The American Journal of Psychiatry, 165(10), 1289-1298.

Petersen, A. C., Crockett, L., Richards, M., & Boxer, A. (1988). A self-report measure of pubertal status: Reliability, validity and initial norms. Journal of Youth and Adolescence, 7(2), 117-133.

Phillips, K. A., Stein, D. J., Rauch, S. L., Hollander, E., Fallon, B. A., Barsky, A., Fineberg, N., Mataix-Cols, D., Ferrão, Y. A., Saxena, S., Wilhelm, S., Kelly, M. M., Clark, L. A., Pinto, A., Bienvenu, O. J., Farrow, J., & Leckman, J. (2010). Should an obsessive- compulsive spectrum grouping of disorders be included in DSM-V? Depression and Anxiety, 27(6), 528-555.

Pinto, A., Greenberg, B. D., Grados, M. A., Bienvenu, O. J.,3rd, Samuels, J. F., Murphy, D. L., Hasler, G., Stout, R. L., Rauch, S. L., Shugart, Y. Y., Pauls, D. L., Knowles, J. A., Fyer, A. J., McCracken, J. T., Piacentini, J., Wang, Y., Willour, V. L., Cullen, B., Liang, K. Y., Hoehn-Saric, R., Riddle, M. A., Rasmussen, S. A., & Nestadt, G. (2008). Further development of YBOCS dimensions in the OCD collaborative genetics study: Symptoms vs. categories. Psychiatry Research, 160(1), 83-93.

181

Plomin, R., Haworth, C. M., & Davis, O. S. (2009). Common disorders are quantitative traits. Nature Reviews.Genetics, 10(12), 872-878.

Pogarell, O., Hamann, C., Popperl, G., Juckel, G., Chouker, M., Zaudig, M., Riedel, M., Möller, H. J., Hegerl, U., & Tatsch, K. (2003). Elevated brain serotonin transporter availability in patients with obsessive-compulsive disorder. Biological Psychiatry, 54(12), 1406-1413.

Price, A. L., Patterson, N. J., Plenge, R. M., Weinblatt, M. E., Shadick, N. A., & Reich, D. (2006). Principal components analysis corrects for stratification in genome-wide association studies. Nature Genetics, 38(8), 904-9.

Qin, H., Samuels, J. F., Wang, Y., Zhu, Y., Grados, M. A., Riddle, M. A., Greenberg, B. D., Knowles, J. A., Fyer, A. J., McCracken, J. T., Murphy, D. L., Rasmussen, S. A., Cullen, B. A., Piacentini, J., Geller, D., Stewart, S. E., Pauls, D., Bienvenu, O. J., Goes, F. S., Maher, B., Pulver, A. E., Valle, D., Lange, C., Mattheisen, M., McLaughlin, N. C., Liang, K. Y., Nurmi, E. L., Askland, K. D., Nestadt, G., & Shugart, Y. Y. (2016). Whole-genome association analysis of treatment response in obsessive-compulsive disorder. Molecular Psychiatry, 21(2), 270-276.

Quilter, C. R., Bagga, M., Moinie, A., Junaid, F., & Sargent, C. A. (2012). Gene structure and expression of serotonin receptor HTR2C in hypothalamic samples from infanticidal and control sows. BMC Neuroscience, 13, 37-2202-13-37.

Rand, M. & Reid, G. (1951). Source of ‘serotonin’ in serum. Nature, 168, 385.

Rand, M. & Reid, G. (1952). On the presence in rabbit serum of thrombotonin; (thrombocytin or serotonin). The Australian Journal of Experimental Biology and Medical Science, 30(2), 153-161.

Rapport, M. M. (1949). Serum vasoconstrictor (serotonin) the presence of creatinine in the complex; a proposed structure of the vasoconstrictor principle. The Journal of Biological Chemistry, 180(3), 961-969.

Rapport, M. M., Green, A. A., & Page, I. H. (1948a). Crystalline serotonin. Science, 108(2804), 329-330.

Rapport, M. M., Green, A. A., & Page, I. H. (1948b). Serum vasoconstrictor, serotonin; chemical inactivation. The Journal of Biological Chemistry, 176(3), 1237-1241.

Rapport, M. M., Green, A. A., & Page, I. H. (1948c). Serum vasoconstrictor, serotonin; isolation and characterization. The Journal of Biological Chemistry, 176(3), 1243-1251.

Reid, G. (1951). The pharmacology of tryptamine. The Australian Journal of Experimental Biology and Medical Science, 29(2), 102-116.

182 Reid, G. & Rand, M. (1951). Physiological actions of the partially purified serum vasoconstrictor (serotonin). The Australian Journal of Experimental Biology and Medical Science, 29(6), 401-415.

Reid, G. & Rand, M. (1952). Pharmacological actions of synthetic 5-hydroxytryptamine (serotonin, thrombocytin). Nature, 169, 801-802.

Reimold, M., Smolka, M. N., Zimmer, A., Batra, A., Knobel, A., Solbach, C., Mundt, A., Smoltczyk, H. U., Goldman, D., Mann, K., Reischl, G., Machulla, H. J., Bares, R., & Heinz, A. (2007). Reduced availability of serotonin transporters in obsessive-compulsive disorder correlates with symptom severity - a [11C]DASB PET study. Journal of Neural Transmission (Vienna, Austria : 1996), 114(12), 1603-1609.

Rosenberg, D. R. & Keshavan, M. S. (1998). A.E. Bennett research award. Toward a neurodevelopmental model of obsessive-compulsive disorder. Biological Psychiatry, 43(9), 623-640.

Ruscio, A.M., Stein, D. J., Chiu, W. T., & Kessler, R. C. (2010). The epidemiology of obsessive- compulsive disorder in the National Comorbidity Survey Replication. Molecular Psychiatry, 15(1), 53e63.

Rutter, M., Bailey, A., Berument, S. K., Lord, C., & Pickles, A. (2003): Social Communication Questionnaire. Los Angeles, CA: Western Psychological Services.

Saiz, P. A., Garcia-Portilla, M. P., Arango, C., Morales, B., Bascaran, M. T., Martinez- Barrondo, S., Florez, G., Sotomayor, E., Paredes, B., Alvarez, C., San Narciso, G., Carreño, E., Bombin, I., Alvarez, V., Coto, E., Fernandez, J. M., Bousoño, M., & Bobes, J. (2008). Association study between obsessive-compulsive disorder and serotonergic candidate genes. Progress in Neuro-Psychopharmacology & Biological Psychiatry, 32(3), 765-770.

Samuels, J. F., Bienvenu, O. J., Grados, M. A., Cullen, B., Riddle, M. A., Liang, K. Y., Eaton, W. W., & Nestadt, G. (2008). Prevalence and correlates of hoarding behavior in a community-based sample. Behaviour Research and Therapy, 46(7), 836-844.

Samuels, J. F., Bienvenu, O. J., 3rd, Pinto, A., Fyer, A. J., McCracken, J. T., Rauch, S. L., Murphy, D. L., Grados, M. A., Greenberg, B. D., Knowles, J. A., Piacentini, J., Cannistraro, P. A., Cullen, B., Riddle, M. A., Rasmussen, S. A., Pauls, D. L., Willour, V. L., Shugart, Y. Y., Liang, K. Y., Hoehn-Saric, R., & Nestadt, G. (2007a). Hoarding in obsessive-compulsive disorder: Results from the OCD collaborative genetics study.Behaviour Research and Therapy, 45(4), 673-686.

Samuels, J. F., Bienvenu, O. J., 3rd, Riddle, M. A., Cullen, B. A., Grados, M. A., Liang, K. Y., Hoehn-Saric, R., & Nestadt, G. (2002). Hoarding in obsessive compulsive disorder: Results from a case-control study. Behaviour Research and Therapy, 40(5), 517-528.

183 Samuels, J. F., Grados, M. A., Riddle, M. A., Bienvenu, O. J., Goes, F. S., Cullen, B., Wang, Y., Greenberg, B. D., Fyer, A. J., McCracken, J. T., Geller, D., Murphy, D. L., Knowles, J. A., Rasmussen, S. A., McLaughlin, N. C., Piacentini, J., Pauls, D. L., Stewart, S. E., Shugart, Y. Y., Maher, B., Pulver, A. E., & Nestadt, G. (2014). Hoarding in children and adolescents with obsessive-compulsive disorder. Journal of Obsessive-Compulsive and Related Disorders, 3(4), 325-331.

Samuels, J. F., Shugart, Y. Y., Grados, M. A., Willour, V. L., Bienvenu, O. J., Greenberg, B. D., Knowles, J. A., McCracken, J. T., Rauch, S. L., Murphy, D. L., Wang, Y., Pinto, A., Fyer, A. J., Piacentini, J., Pauls, D. L., Cullen, B., Rasmussen, S. A., Hoehn-Saric, R., Valle, D., Liang, K. Y., Riddle, M. A., & Nestadt, G. (2007b). Significant linkage to compulsive hoarding on chromosome 14 in families with obsessive-compulsive disorder: Results from the OCD collaborative genetics study. The American Journal of Psychiatry, 164(3), 493-499.

Scahill, L., Riddle, M. A., McSwiggin-Hardin, M., Ort, S. I., King, R. A., Goodman, W. K., Cicchetti, D., & Leckman, J. F. (1997). Children's Yale-Brown obsessive compulsive scale: Reliability and validity. Journal of the American Academy of Child and Adolescent Psychiatry, 36(6), 844-852.

Schildkraut, J. J. (1965). The catecholamine hypothesis of affective disorders: a review of supporting evidence. American Journal of Psychiatry, 122(5), 509-522.

Schork, N. J., Murray, S. S., Frazer, K. A., Topol, E. J. (2009). Common vs. rare allele hypotheses for complex diseases. Current Opinion in Genetics & Development, 19(3), 212-219.

Sheppard, B., Chavira, D., Azzam, A., Grados, M. A., Umana, P., Garrido, H., & Mathews, C. A. (2010). ADHD prevalence and association with hoarding behaviors in childhood-onset OCD. Depression and Anxiety, 27(7), 667-674.

Shmelkov, S. V., Hormigo, A., Jing, D., Proenca, C. C., Bath, K. G., Milde, T., Shmelkov, E., Kushner, J. S., Baljevic, M., Dincheva, I., Murphy, A. J., Valenzuela, D. M., Gale, N. W., Yancopoulos, G. D., Ninan, I., Lee, F. S., & Rafii, S. (2010). Slitrk5 deficiency impairs corticostriatal circuitry and leads to obsessive-compulsive-like behaviors in mice. Nature Medicine, 16(5), 598-602, 1p following 602.

Shonkoff, J. P., Garner, A. S., Committee on Psychosocial Aspects of Child and Family Health, Committee on Early Childhood, Adoption, and Dependent Care, & Section on Developmental and Behavioral Pediatrics. (2012). The lifelong effects of early childhood adversity and toxic stress. Pediatrics, 129(1), e232-46.

Simpson, H. B., Lombardo, I., Slifstein, M., Huang, H. Y., Hwang, D. R., Abi-Dargham, A., Liebowitz, M. R., & Laruelle, M. (2003). Serotonin transporters in obsessive-compulsive disorder: A positron emission tomography study with [(11)C]McN 5652. Biological Psychiatry, 54(12), 1414-1421.

184 Simpson, H. B., Slifstein, M., Bender, J. Jr., Xu, X., Hackett, E., Maher, M. J., & Abi- Dargham, A. (2011). Serotonin 2A receptors in obsessive-compulsive disorder: A positron emission tomography study with [11C]MDL 100907. Biological Psychiatry, 70(9), 897-904.

Sinopoli, V. M., Erdman, L., Burton, C. L., Easter, P., Rajendram, R., Baldwin, G., Peterman, K., Coste, J., Shaheen, S-M, Hanna, G. L., Rosenberg, D. R., Arnold, P. D. (2019a). Serotonin system gene variants and regional brain volume differences in pediatric OCD. Brain Imaging and Behavior, doi: 10.1007/s11682-019-00092-w. [Epub ahead of print]

Sinopoli, V. M., Erdman, L., Burton, C. L., Park, L. S., Dupuis, A., Shan, J., Goodale, T., Shaheen, S-M, Crosbie, J., Schachar, R. J., Arnold, P. D. (2019b). Serotonin system genes and obsessive-compulsive trait dimensions in a population-based, pediatric sample: A genetic association study. Journal of Child Psychology and Psychiatry, doi: 10.1111/jcpp.13079. [Epub ahead of print]

Snowdon, J., Pertusa, A., & Mataix-Cols, D. (2012). On hoarding and squalor: A few considerations for DSM-5. Depression and Anxiety, 29(5), 417-424.

Soeby, K., Larsen, S. A., Olsen, L., Rasmussen, H. B., & Werge, T. (2005). Serotonin transporter: Evolution and impact of polymorphic transcriptional regulation. American Journal of Medical Genetics.Part B, Neuropsychiatric Genetics : The Official Publication of the International Society of Psychiatric Genetics, 136B(1), 53-57.

Sourkes, T. L. (2009). Chapter 54: the discovery of neurotransmitters, and applications to neurology. Handbook of Clinical Neurology, 95, 869-883.

Spurlock, G., Heils, A., Holmans, P., Williams, J., D'Souza, U. M., Cardno, A., Murphy, K. C., Jones, L., Buckland, P. R., McGuffin, P., Lesch, K. P., & Owen, M. J. (1998). A family based association study of T102C polymorphism in 5HT2A and schizophrenia plus identification of new polymorphisms in the promoter. Molecular Psychiatry, 3(1), 42-49.

Stahl, S. M. (1998). Mechanism of action of serotonin selective reuptake inhibitors. serotonin receptors and pathways mediate therapeutic effects and side effects. Journal of Affective Disorders, 51(3), 215-235.

Stein, D. J. & Hollander, E. The spectrum of obsessive‐compulsive related disorders. In:

185 Hollander, E. (Ed.). (1993). Obsessive‐Compulsive Related Disorders. Washington, DC: American Psychiatric Press.

Stein, D. J., Hollander, E., & Rothbaum, B. O. (Eds.). (2009). Textbook of Anxiety Disorders. Washington, DC; London, England: American Psychiatric Publishing.

Steketee, G., Kelley, A. A., Wernick, J. A., Muroff, J., Frost, R. O., & Tolin, D. F. (2015). Familial patterns of hoarding symptoms. Depression and Anxiety, 32(10), 728-736.

Stengler-Wenzke, K., Muller, U., Angermeyer, M. C., Sabri, O., & Hesse, S. (2004). Reduced serotonin transporter-availability in obsessive-compulsive disorder (OCD). European Archives of Psychiatry and Clinical Neuroscience, 254(4), 252-255.

Stewart, S. E., Yu, D., Scharf, J. M., Neale, B. M., Fagerness, J. A., Mathews, C. A., Arnold, P. D., Evans, P. D., Gamazon, E. R., Davis, L. K., Osiecki, L., McGrath, L., Haddad, S., Crane, J., Hezel, D., Illman, C., Mayerfeld, C., Konkashbaev, A., Liu, C., Pluzhnikov, A., Tikhomirov, A., Edlund, C. K., Rauch, S. L., Moessner, R., Falkai, P., Maier, W., Ruhrmann, S., Grabe, H. J., Lennertz, L., Wagner, M., Bellodi, L., Cavallini, M. C., Richter, M. A., Cook, E. H. Jr., Kennedy, J. L., Rosenberg, D., Stein, D. J., Hemmings, S. M., Lochner, C., Azzam, A., Chavira, D. A., Fournier, E., Garrido, H., Sheppard, B., Umaña, P., Murphy, D. L., Wendland, J. R., Veenstra-VanderWeele, J., Denys, D., Blom, R., Deforce, D., Van Nieuwerburgh, F., Westenberg, H. G., Walitza, S., Egberts, K., Renner, T., Miguel, E. C., Cappi, C., Hounie, A. G., Conceição do Rosário, M., Sampaio, A. S., Vallada, H., Nicolini, H., Lanzagorta, N., Camarena, B., Delorme, R., Leboyer, M., Pato, C. N., Pato, M. T., Voyiaziakis, E., Heutink, P., Cath, D. C., Posthuma, D., Smit, J. H., Samuels, J., Bienvenu, O. J., Cullen, B., Fyer, A. J., Grados, M. A., Greenberg, B. D., McCracken, J. T., Riddle, M. A., Wang, Y., Coric, V., Leckman, J. F., Bloch, M., Pittenger, C., Eapen, V., Black, D. W., Ophoff, R. A., Strengman, E., Cusi, D., Turiel, M., Frau, F., Macciardi, F., Gibbs, J. R., Cookson, M. R., Singleton, A., North American Brain Expression Consortium, Hardy, J., UK Brain Expression Database, Crenshaw, A. T., Parkin, M. A., Mirel, D. B., Conti, D. V., Purcell, S., Nestadt, G., Hanna, G. L., Jenike, M. A., Knowles, J. A., Cox, N., & Pauls, D. L. (2013). Genome-wide association study of obsessive-compulsive disorder. Molecular Psychiatry, 18(7), 788-798.

Storch, E. A., Lack, C. W., Merlo, L. J., Geffken, G. R., Jacob, M. L., Murphy, T. K., & Goodman, W. K. (2007). Clinical features of children and adolescents with obsessive- compulsive disorder and hoarding symptoms. Comprehensive Psychiatry, 48(4), 313-318.

Swedo, S. E., Leckman, J. F., & Rose, N. R. (2012). From research subgroup to clinical syndrome: Modifying the PANDAS criteria to describe PANS (pediatric acute-onset neuropsychiatric syndrome). Pediatrics and Therapeutics, 2(2),1-8.

Swedo, S. E., Leonard, H. L., Garvey, M., Mittleman, B., Allen, A. J., Perlmutter, S., Lougee, L., Dow, S., Zamkoff, J., & Dubbert, B. K. (1998). Pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections: Clinical description of the first 50 cases. The American Journal of Psychiatry, 155(2), 264-271.

186 Szeszko, P. R., MacMillan, S., McMeniman, M., Chen, S., Baribault, K., Lim, K. O., Ivey, J., Rose, M., Banerjee, S. P., Bhandari, R., Moore, G. J., & Rosenberg, D. R. (2004). Brain structural abnormalities in psychotropic drug-naive pediatric patients with obsessive- compulsive disorder. The American Journal of Psychiatry, 161(6), 1049-1056.

Taylor, S. (2011). Early versus late onset obsessive-compulsive disorder: Evidence for distinct subtypes. Clinical Psychology Review, 31(7), 1083-1100.

Taylor, S. (2013). Molecular genetics of obsessive-compulsive disorder: A comprehensive meta-analysis of genetic association studies. Molecular Psychiatry, 18(7), 799-805.

Taylor, S. (2016). Disorder-specific genetic factors in obsessive-compulsive disorder: A comprehensive meta-analysis. American Journal of Medical Genetics.Part B, Neuropsychiatric Genetics : The Official Publication of the International Society of Psychiatric Genetics, 171B(3), 325-332.

Taylor, S., Jang, K. L., & Asmundson, G. J. (2010). Etiology of obsessions and compulsions: A behavioral-genetic analysis. Journal of Abnormal Psychology, 119(4), 672-682.

Tot, S., Erdal, M. E., Yazici, K., Yazici, A. E., & Metin, O. (2003). T102C and -1438 G/A polymorphisms of the 5-HT2A receptor gene in Turkish patients with obsessive- compulsive disorder. European Psychiatry : The Journal of the Association of European Psychiatrists, 18(5), 249-254.

Traylor, M., Markus, H., & Lewis, C. M. (2015). Homogeneous case subgroups increase power in genetic association studies. European Journal of Human Genetics : EJHG, 23(6), 863-869.

Twarog B. M. & Page, I. H. (1953). Serotonin content of some mammalian tissues and urine and a method for its determination. American Journal of Physiology, 175(1), 157-161.

Van Ameringen, M., Patterson, B., & Simpson, W. (2014). DSM-5 obsessive-compulsive and related disorders: Clinical implications of new criteria. Depression and Anxiety, 31(6), 487-493.

Van der Wee, N. J., Stevens, H., Hardeman, J. A., Mandl, R. C., Denys, D. A., van Megen, H. J., Kahn, R. S., & Westenberg, H. M. (2004). Enhanced dopamine transporter density in psychotropic-naive patients with obsessive-compulsive disorder shown by [123I]{beta}- CIT SPECT. The American Journal of Psychiatry, 161(12), 2201-2206.

Van Grootheest, D. S., Boomsma, D. I., Hettema, J. M., & Kendler, K. S. (2008). Heritability of obsessive-compulsive symptom dimensions. American Journal of Medical Genetics.Part B, Neuropsychiatric Genetics : The Official Publication of the International Society of Psychiatric Genetics, 147B(4), 473-478.

187 Van Grootheest, D. S., Cath, D. C., Beekman, A. T., & Boomsma, D. I. (2005). Twin studies on obsessive-compulsive disorder: A review. Twin Research and Human Genetics : The Official Journal of the International Society for Twin Studies, 8(5), 450-458.

Vanhoutte, P. M. (1990). Serotonergic antagonists and vascular disease. Cardiovascular Drugs and Therapy / Sponsored by the International Society of Cardiovascular Pharmacotherapy, 4 Suppl 1, 7-12.

Voyiaziakis, E., Evgrafov, O., Li, D., Yoon, H. J., Tabares, P., Samuels, J., Wang, Y., Riddle, M. A., Grados, M. A., Bienvenu, O. J., Shugart, Y. Y., Liang, K. Y., Greenberg, B. D., Rasmussen, S. A., Murphy, D. L., Wendland, J. R., McCracken, J. T., Piacentini, J., Rauch, S. L., Pauls, D. L., Nestadt, G., Fyer, A. J., & Knowles, J. A. (2011). Association of SLC6A4 variants with obsessive-compulsive disorder in a large multicenter US family study. Molecular Psychiatry, 16(1), 108-120.

Walitza, S., Marinova, Z., Grunblatt, E., Lazic, S. E., Remschmidt, H., Vloet, T. D., & Wendland, J. R. (2014). Trio study and meta-analysis support the association of genetic variation at the serotonin transporter with early-onset obsessive-compulsive disorder. Neuroscience Letters, 580, 100-103.

Walitza, S., Wewetzer, C., Warnke, A., Gerlach, M., Geller, F., Gerber, G., Görg, T., Herpertz-Dahlmann, B., Schulz, E., Remschmidt, H., Hebebrand, J., & Hinney, A. (2002). 5-HT2A promoter polymorphism -1438G/A in children and adolescents with obsessive- compulsive disorders. Molecular Psychiatry, 7(10), 1054-1057.

Wang, Y., Liu, J., Huang, B. O., Xu, Y. M., Li, J., Huang, L. F., Lin, J., Zhang, J., Min, Q. H., Yang, W. M., & Wang, X. Z. (2015). Mechanism of alternative splicing and its regulation. Biomedical Reports, 3(2), 152-158.

Wankerl, M., Miller, R., Kirschbaum, C., Hennig, J., Stalder, T., & Alexander, N. (2014). Effects of genetic and early environmental risk factors for depression on serotonin transporter expression and methylation profiles. Translational Psychiatry, 4, e402.

Weissmann, D., van der Laan, S., Underwood, M. D., Salvetat, N., Cavarec, L., Vincent, L., Molina, F., Mann, J. J., Arango, V., & Pujol, J. F. (2016). Region-specific alterations of A-to-I RNA editing of serotonin 2c receptor in the cortex of suicides with major depression. Translational Psychiatry, 6(8), e878.

Welsh, J. H. (1953). Excitation of the heart of Venus mercenaria. Naunyn-Schmiedebergs Archiv für experimentelle Pathologie und Pharmakologie, 219(1-2), 23-29.

Welsh, J. H. (1957). Serotonin as possible neurohumoral agent: evidence in lower animals. Annals of the New York Academy of Sciences, 66(32), 618-630.

Welch, J. M., Lu, J., Rodriguiz, R. M., Trotta, N. C., Peca, J., Ding, J. D., Feliciano, C., Chen, M., Adams, J. P., Luo, J., Dudek, S. M., Weinberg, R. J., Calakos, N., Wetsel, W. C., & Feng, G. (2007). Cortico-striatal synaptic defects and OCD-like behaviours in Sapap3-

188 mutant mice. Nature, 448(7156), 894-900.

Wendland, J. R., DeGuzman, T. B., McMahon, F., Rudnick, G., Detera-Wadleigh, S. D., & Murphy, D. L. (2008a). SERT Ileu425Val in autism, asperger syndrome and obsessive- compulsive disorder. Psychiatric Genetics, 18(1), 31-39.

Wendland, J. R., Kruse, M. R., Cromer, K. R., & Murphy, D. L. (2007). A large case-control study of common functional SLC6A4 and BDNF variants in obsessive-compulsive disorder. Neuropsychopharmacology : Official Publication of the American College of Neuropsychopharmacology, 32(12), 2543-2551.

Wendland, J. R., Martin, B. J., Kruse, M. R., Lesch, K. P., & Murphy, D. L. (2006). Simultaneous genotyping of four functional loci of human SLC6A4, with a reappraisal of 5-HTTLPR and rs25531. Molecular Psychiatry, 11(3), 224-226.

Wendland, J. R., Moya, P. R., Kruse, M. R., Ren-Patterson, R. F., Jensen, C. L., Timpano, K. R., & Murphy, D. L. (2008b). A novel, putative gain-of-function haplotype at SLC6A4 associates with obsessive-compulsive disorder. Human Molecular Genetics, 17(5), 717- 723.

Werry, T. D., Loiacono, R., Sexton, P. M., & Christopoulos, A. (2008). RNA editing of the serotonin 5HT2C receptor and its effects on cell signalling, pharmacology and brain function. Pharmacology & Therapeutics, 119(1), 7-23.

Wierenga, L. M., Bos, M. G. N., Schreuders, E., Kamp, F. vd, Peper, J. S., C. K. Tamnes, & Crone, E. A. (2018). Unraveling age, puberty and testosterone effects on subcortical brain development across adolescence. Psychoneuroendocrinology, 91, 105-114.

Williams, M. T., Mugno, B., Franklin, M., & Faber, S. (2013). Symptom dimensions in obsessive-compulsive disorder: Phenomenology and treatment outcomes with exposure and ritual prevention. Psychopathology, 46(6), 365-376.

Wu, K., Hanna, G. L., Easter, P., Kennedy, J. L., Rosenberg, D. R., & Arnold, P. D. (2013). Glutamate system genes and brain volume alterations in pediatric obsessive-compulsive disorder: A preliminary study. Psychiatry Research, 211(3), 214-220.

Yu, D., Mathews, C. A., Scharf, J. M., Neale, B. M., Davis, L. K., Gamazon, E. R., Derks, E. M., Evans, P., Edlund, C. K., Crane, J., Fagerness, J. A., Osiecki, L., Gallagher, P., Gerber, G., Haddad, S., Illmann, C., McGrath, L. M., Mayerfeld, C., Arepalli, S., Barlassina, C., Barr, C. L., Bellodi, L., Benarroch, F., Berrió, G. B., Bienvenu, O. J., Black, D. W., Bloch, M. H., Brentani, H., Bruun, R. D., Budman, C. L., Camarena, B., Campbell, D. D., Cappi, C., Silgado, J. C., Cavallini, M. C., Chavira, D. A., Chouinard, S., Cook, E. H., Cookson, M. R., Coric, V., Cullen, B., Cusi, D., Delorme, R., Denys, D., Dion, Y., Eapen, V., Egberts, K., Falkai, P., Fernandez, T., Fournier, E., Garrido, H., Geller, D., Gilbert, D. L., Girard, S. L., Grabe, H. J., Grados, M. A., Greenberg, B. D., Gross-Tsur, V., Grünblatt, E., Hardy, J., Heiman, G. A., Hemmings, S. M., Herrera, L. D., Hezel, D. M., Hoekstra, P. J., Jankovic, J., Kennedy, J. L., King, R. A., Konkashbaev, A.

189 I., Kremeyer, B., Kurlan, R., Lanzagorta, N., Leboyer, M., Leckman, J. F., Lennertz, L., Liu, C., Lochner, C., Lowe, T. L., Lupoli, S., Macciardi, F., Maier, W., Manunta, P., Marconi, M., McCracken, J. T., Mesa Restrepo, S. C., Moessner, R., Moorjani, P., Morgan, J., Muller, H., Murphy, D. L., Naarden, A. L., Nurmi, E., Ochoa, W. C., Ophoff, R. A., Pakstis, A. J., Pato, M. T., Pato, C. N., Piacentini, J., Pittenger, C., Pollak, Y., Rauch, S. L., Renner, T., Reus, V. I., Richter, M. A., Riddle, M. A., Robertson, M. M., Romero, R., Rosário, M. C., Rosenberg, D., Ruhrmann, S., Sabatti, C., Salvi, E., Sampaio, A. S., Samuels, J., Sandor, P., Service, S. K., Sheppard, B., Singer, H. S., Smit, J. H., Stein, D. J., Strengman, E., Tischfield, J. A., Turiel, M., Valencia Duarte, A. V., Vallada, H., Veenstra-VanderWeele, J., Walitza, S., Wang, Y., Weale, M., Weiss, R., Wendland, J. R., Westenberg, H. G., Shugart, Y. Y., Hounie, A. G., Miguel, E. C., Nicolini, H., Wagner, M., Ruiz-Linares, A., Cath, D. C., McMahon, W., Posthuma, D., Oostra, B. A., Nestadt, G., Rouleau, G. A., Purcell, S., Jenike, M. A., Heutink, P., Hanna, G. L., Conti, D. V., Arnold, P. D., Freimer, N. B., Stewart, S. E., Knowles, J. A., Cox, N. J., & Pauls, D. L. (2015). Cross-disorder genome-wide analyses suggest a complex genetic relationship between tourette's syndrome and OCD. The American Journal of Psychiatry, 172(1), 82- 93.

Yücel, M., Wood, S. J., Wellard, R. M., Harrison, B. J., Fornito, A., Pujol, J., Velakoulis, D., & Pantelis, C. (2008). Anterior cingulate glutamate-glutamine levels predict symptom severity in women with obsessive-compulsive disorder. The Australian and New Zealand Journal of Psychiatry, 42(6), 467-477.

Yue, W., Cheng, W., Liu, Z., Tang, Y., Lu, T., Zhang, D., Tang, M., & Huang, Y. (2016). Genome-wide DNA methylation analysis in obsessive-compulsive disorder patients. Scientific Reports, 6, 31333.

Zai, G., Bezchlibnyk, Y. B., Richter, M. A., Arnold, P., Burroughs, E., Barr, C. L., & Kennedy, J.L. (2004). Myelin oligodendrocyte glycoprotein (MOG) gene is associated with obsessive-compulsive disorder. American Journal of Medical Genetics Part B: Neuropsychiatric Genetics, 129B(1), 64-8.

Zhang, L., Liu, X., Li, T., Yang, Y., Hu, X., & Collier, D. (2004). Molecular pharmacogenetic studies of drug responses to obsessive-compulsive disorder and six functional genes [Article in Chinese]. Zhonghua Yi Xue Yi Chuan Xue Za Zhi = Zhonghua Yixue Yichuanxue Zazhi = Chinese Journal of Medical Genetics, 21(5), 479-481.

Zitterl, W., Aigner, M., Stompe, T., Zitterl-Eglseer, K., Gutierrez-Lobos, K., Schmidl-Mohl, B., Wenzel, T., Demal, U., Zettinig, G., Hornik, K., & Thau, K. (2007). 123I]-beta-CIT SPECT imaging shows reduced thalamus-hypothalamus serotonin transporter availability in 24 drug-free obsessive-compulsive checkers. Neuropsychopharmacology : Official Publication of the American College of Neuropsychopharmacology, 32(8), 1661-1668.

Zohar, J., Mueller, E. A., Insel, T. R., Zohar-Kadouch, R. C., & Murphy, D. L. (1987). Serotonergic responsivity in obsessive-compulsive disorder. comparison of patients and healthy controls. Archives of General Psychiatry, 44(11), 946-951.

190