[Skip to Content]
[Skip to Content Landing]
Figure 1.  Manhattan Plot of the Results From the Genome-Wide Association Study of Anxiety and Stress-Related Disorders
Manhattan Plot of the Results From the Genome-Wide Association Study of Anxiety and Stress-Related Disorders

This plot displays 12 665 individuals and 19 225 controls. Single-nucleotide polymorphisms in green are in linkage disequilibrium with the index single-nucleotide polymorphisms (diamonds) and have a P value less than .001. Index variants located with a distance less than 400 kilobase are considered as 1 locus.

Figure 2.  Significant Genetic Correlations Between Anxiety and Stress-Related Disorders and Other Heritable Traits
Significant Genetic Correlations Between Anxiety and Stress-Related Disorders and Other Heritable Traits

Bonferroni correction P < 2.19 × 10−4. In total, 228 traits were tested, including psychiatric traits, educational outcomes, obesity-related phenotypes, smoking, and reproductive success (eTable 4 in the Supplement). Error bars indicate 95% confidence limits. GPC indicates Genetics of Personality Consortium; PGC, Psychiatric Genomics Consortium; SSGAC, Social Science Genetic Association Consortium.

Figure 3.  Differential Pde4b Gene Expression Levels in Mice Exposed to Chronic Psychosocial Stress
Differential Pde4b Gene Expression Levels in Mice Exposed to Chronic Psychosocial Stress

Each sample is represented by a dot superimposed on the box plot. Number of mice per group: B6 strain controls = 6 (mPFC and vHPC), resilient = 6 (mPFC) and 8 (vHPC), susceptible = 6 (mPFC) and 3 (vHPC); D2 strain controls = 6 (mPFC and vHPC), resilient = 0, susceptible = 8 (mPFC) and 5 (vHPC). CPM indicates log counts per million; mPFC, medial prefrontal cortex; vHPC, ventral hippocampus.

Table 1.  Sample Characteristics
Sample Characteristics
Table 2.  Results for Index Variants in the Top 10 Loci Associated With Anxiety and Stress-Related Disordersa
Results for Index Variants in the Top 10 Loci Associated With Anxiety and Stress-Related Disordersa
Supplement.

eMethods

eReferences

eFigure 1. Q-Q plot of GWAS analysis

eFigure 2. Regional association plots for index SNPs in PDE4B

eFigure 3. Forest plots for the genome-wide significant index SNP in PDE4B

eFigure 4. Manhattan plot of the results from the GWAS of anxiety disorders

eFigure 5. Manhattan plot of the results from the GWAS of stress-related disorders

eFigure 6. Manhattan plot of the results from the GWAS excluding adjustment disorders

eFigure 7. Manhattan plot of the results from the GWAS for “population” Design

eFigure 8. Manhattan plot of the results from the GWAS for the “covariate” Design

eFigure 9. Manhattan plot of the results from the GWAS for “propensity” Design

eFigure 10. Manhattan plot of the results from the GWAS excluding schizophrenia

eFigure 11. Manhattan plot of the results from the GWAS including OCD

eFigure 12. Partitioning of heritability by functional annotations

eFigure 13. Partitioning of heritability by tissue-group annotations

eFigure 14. Partitioning of heritability by tissue-specific H3K4Me1 annotations

eFigure 15. Multivariate regression of normalized polygenic risk scores on diagnostic subtypes

eFigure 16. Multivariate regression of normalized polygenic risk scores on comorbid subgroups

eTable 1. Summary of results for PDE4B

eTable 2. Gene-based test results

eTable 3. Pathway analysis results

eTable 4. Genetic correlations for “naïve” Design

eTable 5. Genetic correlations for “naïve” Design (UKBB traits)

eTable 6. Genetic correlations for “prevalence” Design

eTable 7. Genetic correlations for “prevalence” Design (UKBB traits)

eTable 8. Genetic correlations for “covariate” Design

eTable 9. Genetic correlations for “covariate” Design (UKBB traits)

eTable 10. Genetic correlations for “propensity” Design

eTable 11. Genetic correlations for “propensity” Design (UKBB traits)

eTable 12. SNP Heritability estimates across varying analytical designs

1.
Kessler  RC, Avenevoli  S, Costello  EJ,  et al.  Prevalence, persistence, and sociodemographic correlates of DSM-IV disorders in the National Comorbidity Survey Replication Adolescent Supplement.  Arch Gen Psychiatry. 2012;69(4):372-380. doi:10.1001/archgenpsychiatry.2011.160PubMedGoogle ScholarCrossref
2.
Wittchen  HU, Jacobi  F, Rehm  J,  et al.  The size and burden of mental disorders and other disorders of the brain in Europe 2010.  Eur Neuropsychopharmacol. 2011;21(9):655-679. doi:10.1016/j.euroneuro.2011.07.018PubMedGoogle ScholarCrossref
3.
Hettema  JM, Neale  MC, Kendler  KS.  A review and meta-analysis of the genetic epidemiology of anxiety disorders.  Am J Psychiatry. 2001;158(10):1568-1578. doi:10.1176/appi.ajp.158.10.1568PubMedGoogle ScholarCrossref
4.
Tambs  K, Czajkowsky  N, Røysamb  E,  et al.  Structure of genetic and environmental risk factors for dimensional representations of DSM-IV anxiety disorders.  Br J Psychiatry. 2009;195(4):301-307. doi:10.1192/bjp.bp.108.059485PubMedGoogle ScholarCrossref
5.
Erhardt  A, Czibere  L, Roeske  D,  et al.  TMEM132D, a new candidate for anxiety phenotypes: evidence from human and mouse studies.  Mol Psychiatry. 2011;16(6):647-663. doi:10.1038/mp.2010.41PubMedGoogle ScholarCrossref
6.
Otowa  T, Yoshida  E, Sugaya  N,  et al.  Genome-wide association study of panic disorder in the Japanese population.  J Hum Genet. 2009;54(2):122-126. doi:10.1038/jhg.2008.17PubMedGoogle ScholarCrossref
7.
Otowa  T, Tanii  H, Sugaya  N,  et al.  Replication of a genome-wide association study of panic disorder in a Japanese population.  J Hum Genet. 2010;55(2):91-96. doi:10.1038/jhg.2009.127PubMedGoogle ScholarCrossref
8.
Otowa  T, Kawamura  Y, Nishida  N,  et al.  Meta-analysis of genome-wide association studies for panic disorder in the Japanese population.  Transl Psychiatry. 2012;2:e186. doi:10.1038/tp.2012.89PubMedGoogle ScholarCrossref
9.
Logue  MW, Baldwin  C, Guffanti  G,  et al.  A genome-wide association study of post-traumatic stress disorder identifies the retinoid-related orphan receptor alpha (RORA) gene as a significant risk locus.  Mol Psychiatry. 2013;18(8):937-942. doi:10.1038/mp.2012.113PubMedGoogle ScholarCrossref
10.
Xie  P, Kranzler  HR, Yang  C, Zhao  H, Farrer  LA, Gelernter  J.  Genome-wide association study identifies new susceptibility loci for posttraumatic stress disorder.  Biol Psychiatry. 2013;74(9):656-663. doi:10.1016/j.biopsych.2013.04.013PubMedGoogle ScholarCrossref
11.
Guffanti  G, Galea  S, Yan  L,  et al.  Genome-wide association study implicates a novel RNA gene, the lincRNA AC068718.1, as a risk factor for post-traumatic stress disorder in women.  Psychoneuroendocrinology. 2013;38(12):3029-3038. doi:10.1016/j.psyneuen.2013.08.014PubMedGoogle ScholarCrossref
12.
Stein  MB, Chen  CY, Ursano  RJ,  et al; Army Study to Assess Risk and Resilience in Servicemembers (STARRS) Collaborators.  Genome-wide association studies of posttraumatic stress disorder in 2 cohorts of US Army soldiers.  JAMA Psychiatry. 2016;73(7):695-704. doi:10.1001/jamapsychiatry.2016.0350PubMedGoogle ScholarCrossref
13.
Ashley-Koch  AE, Garrett  ME, Gibson  J,  et al; Veterans Affairs Mid-Atlantic Mental Illness Research, Education, and Clinical Center Workgroup.  Genome-wide association study of posttraumatic stress disorder in a cohort of Iraq-Afghanistan era veterans.  J Affect Disord. 2015;184:225-234. doi:10.1016/j.jad.2015.03.049PubMedGoogle ScholarCrossref
14.
Nievergelt  CM, Maihofer  AX, Mustapic  M,  et al.  Genomic predictors of combat stress vulnerability and resilience in U.S. Marines: a genome-wide association study across multiple ancestries implicates PRTFDC1 as a potential PTSD gene.  Psychoneuroendocrinology. 2015;51:459-471. doi:10.1016/j.psyneuen.2014.10.017PubMedGoogle ScholarCrossref
15.
Kilaru  V, Iyer  SV, Almli  LM,  et al.  Genome-wide gene-based analysis suggests an association between Neuroligin 1 (NLGN1) and post-traumatic stress disorder.  Transl Psychiatry. 2016;6:e820. doi:10.1038/tp.2016.69PubMedGoogle ScholarCrossref
16.
Dunn  EC, Sofer  T, Gallo  LC,  et al.  Genome-wide association study of generalized anxiety symptoms in the Hispanic Community Health Study/Study of Latinos.  Am J Med Genet B Neuropsychiatr Genet. 2017;174(2):132-143. doi:10.1002/ajmg.b.32448PubMedGoogle ScholarCrossref
17.
Walter  S, Glymour  MM, Koenen  K,  et al.  Performance of polygenic scores for predicting phobic anxiety.  PLoS One. 2013;8(11):e80326. doi:10.1371/journal.pone.0080326PubMedGoogle ScholarCrossref
18.
Stein  MB, Chen  CY, Jain  S,  et al; Army STARRS Collaborators.  Genetic risk variants for social anxiety.  Am J Med Genet B Neuropsychiatr Genet. 2017;174(2):120-131. doi:10.1002/ajmg.b.32520PubMedGoogle ScholarCrossref
19.
Otowa  T, Hek  K, Lee  M,  et al.  Meta-analysis of genome-wide association studies of anxiety disorders.  Mol Psychiatry. 2016;21(10):1391-1399. doi:10.1038/mp.2015.197PubMedGoogle ScholarCrossref
20.
Purves  KL, Coleman  JRI, Meier  SM,  et al.  A major role for common genetic variation in anxiety disorders.  Bio Rxiv. https://www.biorxiv.org/content/10.1101/203844v2. Published April 11, 2019. Accessed April 15, 2019.Google Scholar
21.
Kessler  RC, Sampson  NA, Berglund  P,  et al.  Anxious and non-anxious major depressive disorder in the World Health Organization World Mental Health Surveys.  Epidemiol Psychiatr Sci. 2015;24(3):210-226. doi:10.1017/S2045796015000189PubMedGoogle ScholarCrossref
22.
Pedersen  CB, Bybjerg-Grauholm  J, Pedersen  MG,  et al.  The iPSYCH2012 case-cohort sample: new directions for unravelling genetic and environmental architectures of severe mental disorders.  Mol Psychiatry. 2018;23(1):6-14.PubMedGoogle ScholarCrossref
23.
Nørgaard-Pedersen  B, Hougaard  DM.  Storage policies and use of the Danish Newborn Screening Biobank.  J Inherit Metab Dis. 2007;30(4):530-536. doi:10.1007/s10545-007-0631-xPubMedGoogle ScholarCrossref
24.
Hollegaard  MV, Grove  J, Thorsen  P, Nørgaard-Pedersen  B, Hougaard  DM.  High-throughput genotyping on archived dried blood spot samples.  Genet Test Mol Biomarkers. 2009;13(2):173-179. doi:10.1089/gtmb.2008.0073PubMedGoogle ScholarCrossref
25.
Lynge  E, Sandegaard  JL, Rebolj  M.  The Danish national patient register.  Scand J Public Health. 2011;39(7)(suppl):30-33. doi:10.1177/1403494811401482PubMedGoogle ScholarCrossref
26.
Mors  O, Perto  GP, Mortensen  PB.  The Danish psychiatric central research register.  Scand J Public Health. 2011;39(7)(suppl):54-57. doi:10.1177/1403494810395825PubMedGoogle ScholarCrossref
27.
van Steensel  FJ, Bögels  SM, Perrin  S.  Anxiety disorders in children and adolescents with autistic spectrum disorders: a meta-analysis.  Clin Child Fam Psychol Rev. 2011;14(3):302-317. doi:10.1007/s10567-011-0097-0PubMedGoogle ScholarCrossref
28.
Grove  J, Ripke  S, Als  TD,  et al; Autism Spectrum Disorder Working Group of the Psychiatric Genomics Consortium; BUPGEN; Major Depressive Disorder Working Group of the Psychiatric Genomics Consortium; 23andMe Research Team.  Identification of common genetic risk variants for autism spectrum disorder.  Nat Genet. 2019;51(3):431-444. doi:10.1038/s41588-019-0344-8PubMedGoogle ScholarCrossref
29.
Ripkelab/ricopili. Main ricopili repo for public releases. https://github.com/Ripkelab/ricopili. Accessed April 15, 2019.
30.
Schizophrenia Working Group of the Psychiatric Genomics Consortium.  Biological insights from 108 schizophrenia-associated genetic loci.  Nature. 2014;511(7510):421-427. doi:10.1038/nature13595PubMedGoogle ScholarCrossref
31.
Abecasis  GR, Altshuler  D, Auton  A,  et al; 1000 Genomes Project Consortium.  A map of human genome variation from population-scale sequencing.  Nature. 2010;467(7319):1061-1073. doi:10.1038/nature09534PubMedGoogle ScholarCrossref
32.
Delaneau  O, Marchini  J, Zagury  JF.  A linear complexity phasing method for thousands of genomes.  Nat Methods. 2011;9(2):179-181. doi:10.1038/nmeth.1785PubMedGoogle ScholarCrossref
33.
Howie  BN, Donnelly  P, Marchini  J.  A flexible and accurate genotype imputation method for the next generation of genome-wide association studies.  PLoS Genet. 2009;5(6):e1000529. doi:10.1371/journal.pgen.1000529PubMedGoogle ScholarCrossref
34.
Price  AL, Patterson  NJ, Plenge  RM, Weinblatt  ME, Shadick  NA, Reich  D.  Principal components analysis corrects for stratification in genome-wide association studies.  Nat Genet. 2006;38(8):904-909. doi:10.1038/ng1847PubMedGoogle ScholarCrossref
35.
Willer  CJ, Li  Y, Abecasis  GR.  METAL: fast and efficient meta-analysis of genomewide association scans.  Bioinformatics. 2010;26(17):2190-2191. doi:10.1093/bioinformatics/btq340PubMedGoogle ScholarCrossref
36.
Nagel  M, Watanabe  K, Stringer  S, Posthuma  D, van der Sluis  S.  Item-level analyses reveal genetic heterogeneity in neuroticism.  Nat Commun. 2018;9(1):905. doi:10.1038/s41467-018-03242-8PubMedGoogle ScholarCrossref
37.
de Leeuw  CA, Mooij  JM, Heskes  T, Posthuma  D.  MAGMA: generalized gene-set analysis of GWAS data.  PLoS Comput Biol. 2015;11(4):e1004219. doi:10.1371/journal.pcbi.1004219PubMedGoogle ScholarCrossref
38.
Bulik-Sullivan  BK, Loh  PR, Finucane  HK,  et al; Schizophrenia Working Group of the Psychiatric Genomics Consortium.  LD Score regression distinguishes confounding from polygenicity in genome-wide association studies.  Nat Genet. 2015;47(3):291-295. doi:10.1038/ng.3211PubMedGoogle ScholarCrossref
39.
Bulik-Sullivan  B, Finucane  HK, Anttila  V,  et al; ReproGen Consortium; Psychiatric Genomics Consortium; Genetic Consortium for Anorexia Nervosa of the Wellcome Trust Case Control Consortium 3.  An atlas of genetic correlations across human diseases and traits.  Nat Genet. 2015;47(11):1236-1241. doi:10.1038/ng.3406PubMedGoogle ScholarCrossref
40.
Pedersen  CB, Mors  O, Bertelsen  A,  et al.  A comprehensive nationwide study of the incidence rate and lifetime risk for treated mental disorders.  JAMA Psychiatry. 2014;71(5):573-581. doi:10.1001/jamapsychiatry.2014.16PubMedGoogle ScholarCrossref
41.
Kessler  RC, Berglund  P, Demler  O, Jin  R, Merikangas  KR, Walters  EE.  Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication.  Arch Gen Psychiatry. 2005;62(6):593-602. doi:10.1001/archpsyc.62.6.593PubMedGoogle ScholarCrossref
42.
Finucane  HK, Bulik-Sullivan  B, Gusev  A,  et al; ReproGen Consortium; Schizophrenia Working Group of the Psychiatric Genomics Consortium; RACI Consortium.  Partitioning heritability by functional annotation using genome-wide association summary statistics.  Nat Genet. 2015;47(11):1228-1235. doi:10.1038/ng.3404PubMedGoogle ScholarCrossref
43.
UK Biobank: Neale lab. http://www.nealelab.is/uk-biobank. Accessed April 15, 2019.
44.
Zheng  J, Erzurumluoglu  AM, Elsworth  BL,  et al; Early Genetics and Lifecourse Epidemiology (EAGLE) Eczema Consortium.  LD Hub: a centralized database and web interface to perform LD score regression that maximizes the potential of summary level GWAS data for SNP heritability and genetic correlation analysis.  Bioinformatics. 2017;33(2):272-279. doi:10.1093/bioinformatics/btw613PubMedGoogle ScholarCrossref
45.
Purcell  SM, Wray  NR, Stone  JL,  et al; International Schizophrenia Consortium.  Common polygenic variation contributes to risk of schizophrenia and bipolar disorder.  Nature. 2009;460(7256):748-752.PubMedGoogle ScholarCrossref
46.
Golden  SA, Covington  HE  III, Berton  O, Russo  SJ.  A standardized protocol for repeated social defeat stress in mice.  Nat Protoc. 2011;6(8):1183-1191. doi:10.1038/nprot.2011.361PubMedGoogle ScholarCrossref
47.
Hochberg  Y, Benjamini  Y.  More powerful procedures for multiple significance testing.  Stat Med. 1990;9(7):811-818. doi:10.1002/sim.4780090710PubMedGoogle ScholarCrossref
48.
Pardiñas  AF, Holmans  P, Pocklington  AJ,  et al; GERAD1 Consortium; CRESTAR Consortium.  Common schizophrenia alleles are enriched in mutation-intolerant genes and in regions under strong background selection.  Nat Genet. 2018;50(3):381-389. doi:10.1038/s41588-018-0059-2PubMedGoogle ScholarCrossref
49.
Li  Z, Chen  J, Yu  H,  et al.  Genome-wide association analysis identifies 30 new susceptibility loci for schizophrenia.  Nat Genet. 2017;49(11):1576-1583. doi:10.1038/ng.3973PubMedGoogle ScholarCrossref
50.
Pasman  JA, Verweij  KJH, Gerring  Z,  et al; 23andMe Research Team; Substance Use Disorders Working Group of the Psychiatric Genomics Consortium; International Cannabis Consortium.  GWAS of lifetime cannabis use reveals new risk loci, genetic overlap with psychiatric traits, and a causal influence of schizophrenia.  Nat Neurosci. 2018;21(9):1161-1170. doi:10.1038/s41593-018-0206-1PubMedGoogle ScholarCrossref
51.
Mühleisen  TW, Leber  M, Schulze  TG,  et al.  Genome-wide association study reveals two new risk loci for bipolar disorder.  Nat Commun. 2014;5:3339. doi:10.1038/ncomms4339PubMedGoogle ScholarCrossref
52.
is Duncan  LE, Ratanatharathorn  A, Aiello  AE,  et al.  Largest GWAS of PTSD (N=20 070) yields genetic overlap with schizophrenia and sex differences in heritability.  Mol Psychiatry. 2018;23(3):666-673. doi:10.1038/mp.2017.77PubMedGoogle ScholarCrossref
53.
Otowa  T, Kawamura  Y, Sugaya  N,  et al.  Association study of PDE4B with panic disorder in the Japanese population.  Prog Neuropsychopharmacol Biol Psychiatry. 2011;35(2):545-549. doi:10.1016/j.pnpbp.2010.12.013PubMedGoogle ScholarCrossref
54.
Zhang  C, Xu  Y, Zhang  HT, Gurney  ME, O’Donnell  JM.  Comparison of the pharmacological profiles of selective PDE4B and PDE4D inhibitors in the central nervous system.  Sci Rep. 2017;7:40115. doi:10.1038/srep40115PubMedGoogle ScholarCrossref
55.
Zhang  HT, Huang  Y, Masood  A,  et al.  Anxiogenic-like behavioral phenotype of mice deficient in phosphodiesterase 4B (PDE4B).  Neuropsychopharmacology. 2008;33(7):1611-1623. doi:10.1038/sj.npp.1301537PubMedGoogle ScholarCrossref
56.
Meier  SM, Petersen  L, Mattheisen  M, Mors  O, Mortensen  PB, Laursen  TM.  Secondary depression in severe anxiety disorders: a population-based cohort study in Denmark.  Lancet Psychiatry. 2015;2(6):515-523. doi:10.1016/S2215-0366(15)00092-9PubMedGoogle ScholarCrossref
57.
Cross-Disorder Group of the Psychiatric Genomics Consortium.  Identification of risk loci with shared effects on five major psychiatric disorders: a genome-wide analysis.  Lancet. 2013;381(9875):1371-1379. doi:10.1016/S0140-6736(12)62129-1PubMedGoogle ScholarCrossref
58.
de Lijster  JM, Dieleman  GC, Utens  EMWJ,  et al.  Social and academic functioning in adolescents with anxiety disorders: a systematic review.  J Affect Disord. 2018;230:108-117. doi:10.1016/j.jad.2018.01.008PubMedGoogle ScholarCrossref
59.
Hettema  JM, Prescott  CA, Myers  JM, Neale  MC, Kendler  KS.  The structure of genetic and environmental risk factors for anxiety disorders in men and women.  Arch Gen Psychiatry. 2005;62(2):182-189. doi:10.1001/archpsyc.62.2.182PubMedGoogle ScholarCrossref
Original Investigation
May 22, 2019

Genetic Variants Associated With Anxiety and Stress-Related Disorders: A Genome-Wide Association Study and Mouse-Model Study

Author Affiliations
  • 1Psychosis Research Unit, Aarhus University Hospital, Risskov, Denmark
  • 2The Lundbeck Foundation Initiative for Integrative Psychiatric Research, iPSYCH, Copenhagen, Denmark
  • 3Department of Psychiatry, Psychosomatics and Psychotherapy, Center of Mental Health, University Hospital Würzburg, Würzburg, Germany
  • 4now with the Department of Psychiatry, Dalhousie University, Halifax, Nova Scotia, Canada
  • 5Research Program of Molecular and Integrative Biosciences, Faculty of Biological and Environmental Sciences, Department of Psychology and Logopedics, Medicum, and SleepWell Research Program, University of Helsinki, Helsinki, Finland
  • 6Social, Genetic and Developmental Psychiatry Centre, Institute of Psychiatry, Psychology & Neuroscience, King’s College London, London, United Kingdom
  • 7Department of Biomedicine, Aarhus University, Aarhus, Denmark
  • 8Centre for integrative Sequencing (iSEQ), Aarhus University, Aarhus, Denmark
  • 9National Centre for Register-Based Research, Aarhus University, Aarhus, Denmark
  • 10Danish Centre for Neonatal Screening, Department for Congenital Disorders, Statens Serum Institut, Copenhagen, Denmark
  • 11Institute of Biological Psychiatry, Mental Health Centre Sct Hans, Copenhagen University Hospital, Roskilde, Denmark
  • 12Department of Clinical Medicine, University of Copenhagen, Copenhagen, Denmark
  • 13Mental Health Centre Copenhagen, Faculty of Health Sciences, University of Copenhagen, Copenhagen, Denmark
  • 14National Institute for Health Research Biomedical Research Centre for Mental Health, South London and Maudsley National Health Service Trust, London, United Kingdom
  • 15Department of Clinical Neuroscience, Centre for Psychiatric Research, Karolinska Institutet, Stockholm, Sweden
JAMA Psychiatry. 2019;76(9):924-932. doi:10.1001/jamapsychiatry.2019.1119
Key Points

Question  Which genetic variants are associated with anxiety and stress-related disorders and do they correlate with other traits?

Findings  In this study of genome-wide association data, PDE4B variants were associated with anxiety and stress-related disorders, and their genetic signature overlapped with other psychiatric traits, educational outcomes, obesity-related phenotypes, smoking, and reproductive success.

Meaning  Large samples are needed to validly identify genetic variants associated with anxiety and stress-related disorders.

Abstract

Importance  Anxiety and stress-related disorders are among the most common mental disorders. Although family and twin studies indicate that both genetic and environmental factors play an important role underlying their etiology, the genetic underpinnings of anxiety and stress-related disorders are poorly understood.

Objectives  To estimate the single-nucleotide polymorphism–based heritability of anxiety and stress-related disorders; to identify novel genetic risk variants, genes, or biological pathways; to test for pleiotropic associations with other psychiatric traits; and to evaluate the association of psychiatric comorbidities with genetic findings.

Design, Setting, Participants  This genome-wide association study included individuals with various anxiety and stress-related diagnoses and controls derived from the population-based Lundbeck Foundation Initiative for Integrative Psychiatric Research (iPSYCH) study. Lifetime diagnoses of anxiety and stress-related disorders were obtained through the national Danish registers. Genes of interest were further evaluated in mice exposed to chronic social defeat. The study was conducted between June 2016 and November 2018.

Main Outcomes and Measures  Diagnoses of a relatively broad diagnostic spectrum of anxiety and stress-related disorders.

Results  The study sample included 12 655 individuals with various anxiety and stress-related diagnoses and 19 225 controls. Overall, 17 740 study participants (55.6%) were women. A total of 7308 participants (22.9%) were born between 1981-1985, 8840 (27.7%) between 1986-1990, 8157 (25.6%) between 1991-1995, 5918 (18.6%) between 1996-2000, and 1657 (5.2%) between 2001-2005. Standard association analysis revealed variants in PDE4B to be associated with anxiety and stress-related disorder (rs7528604; P = 5.39 × 10−11; odds ratio = 0.89; 95% CI, 0.86-0.92). A framework of sensitivity analyses adjusting for mental comorbidity supported this result showing consistent association of PDE4B variants with anxiety and stress-related disorder across analytical scenarios. In mouse models, alterations in Pde4b expression were observed in those mice displaying anxiety-like behavior after exposure to chronic stress in the prefrontal cortex (P = .002; t = −3.33) and the hippocampus (P = .001; t = −3.72). We also found a single-nucleotide polymorphism heritability of 28% (standard error = 0.027) and that the genetic signature of anxiety and stress-related overlapped with psychiatric traits, educational outcomes, obesity-related phenotypes, smoking, and reproductive success.

Conclusions and Relevance  This study highlights anxiety and stress-related disorders as complex heritable phenotypes with intriguing genetic correlations not only with psychiatric traits, but also with educational outcomes and multiple obesity-related phenotypes. Furthermore, we highlight the candidate gene PDE4B as a robust risk locus pointing to the potential of PDE4B inhibitors in treatment of these disorders.

Introduction

Anxiety disorders are characterized by excessive and inappropriate fear and anxiety triggered by stimuli perceived as threatening. They are among the most common mental disorders with a lifetime prevalence of more than 20%.1 Given the prevalence and the immense social and economic burden of these disorders,2 it is of strong interest to identify their risk factors.

Family and twin studies indicate that both genetic and environmental factors are of relevance in the etiology of anxiety disorders, with levels of familial aggregation and heritability at 30% to 50%.3 Although stress-related disorders share many symptom characteristics with anxiety disorders and both conditions are highly comorbid, they have recently been moved to a separate diagnostic category. Interestingly, susceptibility factors common to different anxiety and stress-related disorders seem to account for a larger proportion in heritability than factors predisposing to individual disorders.4 This indicates the potential of combining these phenotypes to identify their shared genetic underpinnings. Genome-wide association studies (GWAS) have proven to be an effective tool for the identification of common genetic variants increasing the susceptibility to complex disorders. Recently, GWAS of panic disorder,5-8 posttraumatic stress disorder,9-15 generalized anxiety disorders,16 phobias,17,18 and a composite indicator of anxiety disorders19,20 have been published. However, besides the study by Purves et al,20 most of these efforts were limited in sample size resulting in low overall power to detect significant associations.

In this study, we conducted a GWAS aggregating individuals in the Lundbeck Foundation Initiative for Integrative Psychiatric Research (iPSYCH) study with varying diagnoses of anxiety and stress-related disorders to identify their common genetic factors, extending previous successful attempts.19,20 As most individuals with these disorders experience another comorbid mental disorder, especially depression,21 we explored the association of mental comorbidity with the genetics of anxiety and stress-related disorders. Genes identified through this effort were further followed up in a mouse model of chronic social defeat. Our effort represents the first genetic study of this magnitude to explicitly target comorbidity of anxiety and stress-related disorders, to our knowledge.

Methods
Participants

The GWAS sample under analysis included 12 655 individuals with anxiety and stress-related diagnoses as well as 19 225 controls from Denmark. All study participants were enrolled in iPSYCH, a study designed to unravel risk factors of severe mental disorders, including individuals with schizophrenia, autism, attention-deficit/hyperactivity disorder, anorexia nervosa, and affective disorders (referred to in this article as design phenotypes) as wells as population-based controls. More information on iPSYCH can be found elsewhere.22 Of these design individuals, 4584 individuals were diagnosed as having an anxiety disorder and 9831 as having a stress-related disorder, of which 1760 received both diagnoses. DNA samples for iPSYCH were taken from the Danish Neonatal Screening Biobank.23 Following protocols for DNA extraction and amplification (described elsewhere24), all samples were genotyped using Illumina’s PsychChip (Illumina). Through the national research registers,25,26 we identified individuals with an anxiety and stress-related diagnosis assigned by a psychiatrist during routine clinical care according to the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (F40.0-F41.9; F43.0-F43.9). Individuals with comorbid autism were excluded. Although patients with autism experience anxiety, their anxiety is often reflecting their core autistic symptomatology and lacks the social component central to many anxiety and stress-related diagnoses.27 Exclusion criteria for control individuals were International Statistical Classification of Diseases and Related Health Problems, Tenth Revision diagnoses of anxiety, stress-related disorders, and mood disorders. Characteristics of the sample are displayed in Table 1. The study was conducted between June 2016 and November 2018. This study was approved by the Danish Data Protection Agency. By Danish law, registry-based studies do not require informed consent.

Quality Control, GWAS, and Gene-Based Analysis

Quality control, imputation, and primary association analyses in iPSYCH have been described elsewhere.28 We used the bioinformatics pipeline Ricopili29 developed by the Psychiatric Genomics Consortium.30 To avoid potential study effects of the 23 genotyping batches within the iPSYCH cohort, each batch was processed separately. Standard procedures for stringent quality control included filters for call rate, Hardy-Weinberg equilibrium, and heterozygosity rates. Each batch was phased and imputed using the 1000 Genomes Project phase 3 imputation reference panel31 using SHAPEIT32 and IMPUTE2,33 respectively. Cryptic relatedness and population structure were assessed on high-quality single-nucleotide polymorphisms (SNPs) with low linkage disequilibrium (LD).

Genome-wide association studies for the 23 batches in iPSYCH were performed using logistic regression models with the imputed marker dosages including the first 4 principal components to control for remaining population stratification.34 Subsequently, the results were meta-analyzed using an inverse variance–weighted fixed-effects model, implemented in METAL.35 Additionally, the analyses were stratified for anxiety disorders and stress-related disorders and different subtypes. Most individuals were diagnosed as having an additional mental disorder owing to their comorbidity with 1 of the design phenotypes in iPSYCH. We therefore aimed to adjust for mental comorbidity in a framework of sensitivity analyses (eMethods in the Supplement). There is currently no other study available with the same phenotype definition, to our knowledge; however, we sought for replication of our index SNP and correlated genome-wide significant variants (0.2 < r2 < 0.5, rs7528604 and rs7539350) in related phenotypes, such as anxiety20 and neuroticism.36 Finally, as obsessive-compulsive disorder has in previous classifications systems been categorized under anxiety disorders, analyses were supplemented by also including individuals with obsessive-compulsive disorder.

Gene-based associations were calculated with MAGMA37 using the summary statistics from the main GWAS analyses. Association was tested using the SNP-wise mean model, in which the sum of –log (SNP P value) for SNPs located within the transcribed region was used as test statistic. MAGMA37 controls for gene size, number of SNPs in a gene, and LD between markers estimated from 1000 Genomes Project phase 3 samples.

SNP Heritability and Genetic Correlation With Other Traits

Linkage disequilibrium score regression was used to dissect the relative contribution of polygenic effects and confounders (eg, cryptic relatedness, sample overlap, and population stratification) to deviation from the null in the genome-wide distribution of GWAS χ2 statistics.38,39 Prevalence was specified at 20%.40,41 Using LD score regression, SNP heritability was also partitioned by functional category and tissue association.38,39 Partitioning was performed for 53 overlapping functional categories as well as 220 cell-type–specific annotations grouped into 10 cell-type categories.42 Genetic correlations were tested for 228 phenotypes with publicly available GWAS summary statistics and 596 traits from the UK Biobank study43 using LD Hub.44 In addition, polygenic risk scores45 were constructed in our sample to explore polygenic heterogeneity across diagnostic subtypes and comorbidities (eMethods in the Supplement).

Mouse Model of Chronic Psychosocial Stress

Given that stress is known to increase the risk of anxiety and stress-related disorders, we aimed to establish the association of chronic psychosocial stress with the brain gene expression levels of significant genes. We used the chronic social defeat stress (CSDS) model46 including 2 inbred strains: C57/BL6NCrl (B6, innately nonanxious strain) and DBA/2NCrl (D2, innately anxious strain). Twenty-four hours after the last CSDS session, we tested all mice for social aversion comparing their explorative behavior in the area around a cylinder with and without a Clr-CD1 mouse. One week after the end of CSDS, we dissected the medial prefrontal cortex (mPFC) and ventral hippocampus (vHPC) for RNA sequencing. All animal procedures were approved by the Regional State Administration Agency for Southern Finland (license numbers ESAVI-3801-041003-2011 and ESAVI/2766/04.10.07/2014) and carried out according to directive 2010/63/EU of the European Parliament and of the Council and the Finnish Act on the Protection of Animals Used for Science or Educational Purposes (497/2013).

Gene Expression Profiling

Total RNA was extracted using TriReagent. For RNA sequencing, we carried out ribosomal RNA depletion using Ribo-Zero (Illumina) followed by fragmentation with an S2 ultrasonicator (Covaris Inc). Messenger RNA sequencing libraries were prepared with Nextera (Illumina; vHPC samples) or ScriptSeq version 2 (Epicenter; mPFC samples) kits. Whole transcriptome level multiple testing correction was done with the Benjamini-Hochberg method,47 after which the expression levels of significant genes were extracted from the data set (eMethods in the Supplement). A 2-sided P value less than .05 was considered statistically significant.

Results
GWAS and Gene-Based Analysis

At a single locus on chromosome 1, 68 genetic variants exceeded the threshold for genome-wide significance in this GWAS for anxiety and stress-related disorders (ie, without correction for the study design). The locus overlaps with 1 gene (PDE4B) with the lead SNP rs7528604 (P = 5.39 × 10−11; odds ratio [OR] = 0.89; 95% CI, 0.86-0.92) (Figure 1, Table 230,48-51; eFigures 1 and 2 in the Supplement). We found no evidence for significant heterogeneity between genotyping batches for this marker (eFigure 3 in the Supplement). Stratified analyses focusing on anxiety and stress-related disorder phenotype definitions separately supported the identified locus (eFigures 4 and 5 in the Supplement), permutation tests indicated that the weaker signals can be attributed to a loss in sample size. Permutation tests also showed the difference in significance levels across diagnostic subgroups could be attributed to their sample size (for exclusion of adjustment disorders see eFigure 6 in the Supplement). In addition, results from analyses mimicking the comorbidity pattern of population-based samples (rs7528604; P = 1.20 × 10−11; OR = 0.88; 95% CI, 0.85-0.91) and including psychiatric phenotypes as covariates (rs7528604; P = 2.32 × 10−8; OR = 0.90; 95% CI, 0.87-0.93) were in line with our initial GWAS results (eFigures 7 and 8 in the Supplement). Even in the stringent sensitivity analyses, associations within the PDE4B gene were among the top signals although with a different lead SNP (rs17128482; P = 1.43 × 10−5; OR = 1.12; 95% CI, 1.07-1.17; eFigure 9 in the Supplement). Resampling analyses by removing individuals with comorbid depression from our data did not show a significant association of depression with outcomes of our result at the PDE4B locus (P = .46). To assess whether our PDE4B signal might have been based on comorbid schizophrenia, we excluded individuals with schizophrenia and still observed genome-wide significance and the same OR for that locus (P = 3.58 × 10−10; OR = 0.89; 95% CI, 0.86-0.91; eFigure 10 in the Supplement). Analyses including individuals with obsessive-compulsive disorder are displayed in eFigure 11 in the Supplement. A summary of all analyses can be found in eTable 1 in the Supplement. Genome-wide analyses using MAGMA37 identified 7 genome-wide significant genes, whereas no pathways were significant after correction for multiple testing (eTables 2 and 3 in the Supplement).

SNP Heritability and Genetic Correlation With Other Traits

Linkage disequilibrium–score regression38,39 was used to calculate SNP heritability of our naive anxiety and stress-related disorders GWAS and the genetic correlation with other phenotypes. Assuming a population prevalence of 20% for anxiety and stress-related disorders, we estimate the liability-scale SNP heritability at 0.28 (standard error [SE] = 0.027). This estimate is comparable with those reported in previous studies for posttraumatic stress disorder52 and anxiety disorders.20 Partitioning heritability based on functional annotations revealed significant enrichment in the heritability for SNPs located in conserved regions (enrichment = 2.01; SE = 0.32; P < .001), supporting the general biological importance of conserved regions and their potential association with susceptibility of anxiety and stress-related disorders (eFigure 12 in the Supplement). Cell-type–specific analyses revealed a significant enrichment in the heritability by SNPs located in central nervous system specific enhancers and promoters (enrichment = 2.93; SE = 0.50; P = 4.32 × 10−4; eFigures 13 and 14 in the Supplement).

Using LDhub,44 31 phenotypes displayed significant genetic correlation with anxiety and stress-related disorders after Bonferroni correction (P = 2.19 × 10−4), including psychiatric traits, educational outcomes, obesity-related phenotypes, smoking, and reproductive success (Figure 2). An overview of all genetic correlations can be found in eTables 4 and 5 in the Supplement. Linkage disequilibrium–score regression also revealed a strong genetic correlation (rfor genetic correlation = 0.55; P = 6.79 × 10−17) with the largest anxiety GWAS20 to date. We performed the same LD score regression–based analyses also for the results from our framework of sensitivity analyses. The association of the analyses mimicking the comorbidity pattern of population-based samples on SNP heritability and the genetic correlations with other traits was marginal (eTables 6 and 7 in the Supplement). Owing to the overrepresentation of psychiatric phenotypes in the control sample, the design including psychiatric phenotypes as covariates and the propensity score matched design resulted in lower SNP heritability and genetic correlations (eTables 8 to 11 in the Supplement). An overview with regard to SNP heritability reflecting the increased prevalence of anxiety and stress-related disorders in psychiatric phenotypes can be found in eTable 12 in the Supplement. Using polygenic risk scores trained on different phenotypes, we only observed limited heterogeneity across diagnostic and comorbidity groups (eFigures 15 and 16 in the Supplement).

Study Results in Context

We sought for replication of our index SNP and correlated genome-wide significant variants (rs7528604; P = 5.39 × 10−11; OR = 0.89; 95% CI, 0.86-0.92 and rs7539350; P = 1.46 × 10−8; OR = 1.10; 95% CI, 1.06-1.14) in related phenotypes.20,36PDE4B was associated with anxiety disorders in the largest GWAS to date20 (rs7528604; P = .18; OR = 0.97; 95% CI, 0.93-1.02 and rs7539350; P = .02; OR = 1.05; 95% CI, 1.01-1.09) and neuroticism36 (rs7528604; P = 9.01 × 10−3; OR = 0.99; 95% CI, 0.99-0.99 and rs7539350; P = 2.59 × 10−4; OR = 1.01; 95% CI, 1.01-1.01), which is often used as a proxy for anxiety and stress-related disorders.

Lower Pde4b Expression Levels in Mice

We determined the gene expression levels of Pde4b in mice exposed to CSDS using RNA sequencing (Figure 3). Stress-susceptible mice from the B6 strain had lower expression levels of Pde4b in the mPFC compared with both control (P < .01; t = −3.33; adjusted P = .06) and stress-resilient mice (P = .01; t = −2.97; adjusted P = .10), as well as in the vHPC compared with control (P < .01; t = −3.72; adjusted P < .01) and stress-resilient mice (P = .003; t = −3.29; adjusted P = .01). D2 mice are highly susceptible to CSDS and therefore, we were only able to compare stress-susceptible mice with controls. There were no differences in Pde4b expression levels between these groups in either brain region. Interestingly, the innately highly anxious D2 control mice had lower Pde4b expression levels compared with the nonanxious B6 control mice in the vHPC (P = 4.6 × 10−5; t = −4.81; adjusted P = .00016) but not in the mPFC (P = .13; t = −1.57; adjusted P = .26), suggesting that especially in the vHPC Pde4b expression levels may also contribute to innate anxiety levels.

Discussion

To our knowledge, we conducted the largest GWAS on anxiety and stress-related disorders to date and extend previous findings about shared genetic associations with other phenotypes. Specifically, we aggregated 12 655 individuals and 19 225 control individuals with the aim of identifying common variants underlying the etiology of these disorders. We tried to capture the clinical complexity of anxiety and stress-related phenotypes in a sample enriched for individuals with comorbid mental illnesses and identified genetic variants that were associated with disease susceptibility.

The most consistent association signal across our different analyses was observed for genetic variants located within the PDE4B gene, which regulates intracellular cyclic adenosine monophosphate signaling and is strongly expressed in the human brain. PDE4B has been proposed as a candidate gene for anxiety, in particular, panic disorder53; however, replication has so far been lacking. Pharmacologic profiles of selective PDE4B inhibitors have demonstrated clear antidepressant and anxiolytic benefits.54 Furthermore, mice deficient in Pde4b exhibit behavioral changes in a range of tests sensitive to anxiolytic drugs.55 We found the expression of Pde4b to be altered in B6 mice susceptible to chronic psychosocial stress compared with controls and stress-resilient mice. Lower expression levels were observed in brain regions (mPFC and vHPC), which are known to regulate emotional and social behavior in mice and humans. In innately anxious (D2 inbred) mice, Pde4b expression did not differ from controls after chronic psychosocial stress exposure indicating a genetic background effect.

In line with other GWAS,19,52 our SNP heritability estimate of 28% for anxiety and stress-related disorders indicates a substantial role for common genetic variation, accounting for a sizable portion of twin-based heritability.3 Conserved regions and regions containing enhancers and promoters of expression in the central nervous system tissues were found to be enriched for associations with anxiety disorders, consistent with findings for schizophrenia, bipolar disorder, and depression.42 More work is needed to unravel the nature of the genetic correlations described in this article and how different designs in our analytical framework affected these findings. Nevertheless, the range of genetic correlations with psychiatric traits, educational outcomes, obesity-related phenotypes, and smoking helps to broaden our conceptualization of anxiety and stress-related disorders. First, the strong positive genetic correlations of these disorders with depression and neuroticism in our naive GWAS reinforce clinical and epidemiologic observations. Anxiety and stress-related disorders are commonly comorbid with depression, often precede depression, and even affect the course of the depression.56 Second, the positive genetic correlations with schizophrenia and the cross–psychiatric disorder phenotype firmly anchor anxiety and stress-related disorders with other psychiatric traits and reflect the substantial evidence for partially shared genetic susceptibility across many psychiatric disorders.57 Third, negative associations between these disorders and educational attainment have been reported,58 and our results suggest that genetic factors may partially account for these reported associations.

Strengths and Limitations

A major strength of this study is the aim to identify genetic variants that play a central but nonspecific role in the susceptibility of anxiety and stress-related disorders. This contrasts with the approach taken in most psychiatric genetic studies, which generally focus on specific clinical diagnoses. However, it has long been recognized that clinical diagnoses poorly reflect etiological mechanisms, as both genetic and environmental factors have been found to have nonspecific effects across a wide range of diagnoses.59 Given how critical fear and anxiety are for human survival, it is very likely that conserved genes common to a range of anxiety and stress-related disorder regulate these basic biological processes.

Making use of registry-based diagnoses as a proxy for mental disorders in a research study that was primarily ascertained for closely related traits constitute the major limitation, but it is also a strength of our study enabling the characterization of anxiety and stress-related disorders susceptibility in the context of mental comorbidity. Similar to previous studies, the generalizability to milder forms of anxiety and stress-related disorders or truly population-based samples is difficult to assess.

Through different sensitivity analyses, we aimed to address some of the limitations (to the extent possible with the data at hand) and gained new insights that probably would not have been possible otherwise. Importantly, despite this lack in generalizability, these limitations are unlikely to lead to false positive associations in a narrow sense. However, they do ask for a reflection on the tested hypothesis in the sensitivity analyses under consideration. However, we cannot exclude the possibility that our results rather reflect the genetic underpinnings of stress-related disorders than anxiety or specific diagnostic subgroups (ie, adjustment disorders) given their differences in sample sizes, although our permutation analyses did not suggest so. As a final point, we would like to stress that our replication efforts were limited by the fact that our study is including a wider range of both anxiety and stress-related disorders than previous efforts.19,20,52

Conclusions

In summary, our results highlight anxiety and stress-related disorders to be a complex heritable phenotype. We highlight the candidate gene PDE4B as a robust risk locus (through studies in mice and humans), pointing to the potential of PDE4B inhibitors in treatment of these disorders. Future studies are needed to confirm these findings via independent replication and to detect additional loci, not only identifying potential pleiotropic effects across the full spectrum of anxiety and stress-related disorders, but also loci associated specifically with individual disorders.

Back to top
Article Information

Corresponding Author: Sandra M. Meier, PhD, Department of Psychiatry, Dalhousie University, 5850/5980 University Ave, PO Box 9700, Halifax, NS B3K 6R8, Canada (sandra.meier@iwk.nshealth.ca).

Accepted for Publication: March 20, 2019.

Published Online: May 22, 2019. doi:10.1001/jamapsychiatry.2019.1119

Author Contributions: Drs Meier and Mattheisen had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Drs Hovatta, Mattheisen, and Mors contributed equally.

Concept and design: Meier, Hougaard, Werge, Nordentoft, Breen, Børglum, Mattheisen, Mors.

Acquisition, analysis, or interpretation of data: Meier, Trontti, Purves, Als, Grove, Laine, Pedersen, Bybjerg-Grauholm, Bækved-Hansen, Sokolowska, Mortensen, Hougaard, Werge, Breen, Børglum, Eley, Hovatta, Mattheisen, Mors.

Drafting of the manuscript: Meier, Trontti, Laine, Breen, Hovatta, Mattheisen.

Critical revision of the manuscript for important intellectual content: Purves, Als, Grove, Pedersen, Bybjerg-Grauholm, Bækved-Hansen, Sokolowska, Mortensen, Hougaard, Werge, Nordentoft, Breen, Børglum, Eley, Mors.

Statistical analysis: Meier, Trontti, Purves, Als, Grove, Bybjerg-Grauholm, Bækved-Hansen, Breen, Mattheisen.

Obtained funding: Mortensen, Werge, Nordentoft, Breen, Børglum, Eley, Hovatta, Mors.

Administrative, technical, or material support: Bybjerg-Grauholm, Bækved-Hansen, Mortensen, Hougaard, Werge, Nordentoft, Breen, Børglum, Mattheisen, Mors.

Supervision: Sokolowska, Mortensen, Werge, Nordentoft, Breen, Børglum, Eley, Hovatta, Mattheisen, Mors.

Conflict of Interest Disclosures: Dr Mortensen reports grants from the Lundbeck Foundation during the conduct of the study and outside the submitted work. Dr Hougaard reports grants from the Lundbeck Foundation during the conduct of the study and personal fees from the Statens Serum Institut outside the submitted work. Dr Nordentoft reports grants from the Lundbeck Foundation during the conduct of the study. Dr Breen reports grants from the UK National Institute for Health Research during the conduct of the study. Dr Børglum reports grants from the Lundbeck Foundation during the conduct of the study and outside the submitted work. Dr Eley reports grants from the UK Medical Research Council, UK National Institute for Health Research, and Fondation Peters during the conduct of the study. Dr Hovatta reports grants from European Research Council, Sigrid Juselius Foundation, and University of Helsinki during the conduct of the study. Dr Mattheisen reports grants from the Lundbeck Foundation during the conduct of the study. Dr Mors reports grants from the Lundbeck Foundation during the conduct of the study. No other disclosures were reported.

Funding/Support: This article was funded by the Lundbeck Foundation (grants R102-A9118 and R155-2014-1724); the Novo Nordisk Foundation, who provided support for the Danish National Biobank resource; the European Research Council Starting Grant GenAnx; and the Sigrid Jusélius Foundation. The research in this study was partly funded by the National Institute for Health Research Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and King’s College London. Ms Purves’s work was funded by the Alexander von Humboldt Foundation. Dr Eley is partly funded by a program grant from the UK Medical Research Council (MR/M021475/1).

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The views expressed are those of the authors and not necessarily those of the National Health Service, the National Institute for Health Research, or the Department of Health.

Additional Contributions: We thank Naomi Wray, PhD (University of Queensland, Brisbane), and Jack Hettema, MD, PhD (Virginia Commonwealth University), for helpful comments and suggestions during the preparation of the manuscript. We also thank Ingrid Balcells, PhD; Natalia Kulesskaya, PhD; Zuzanna Misiewicz, PhD; Suvi Saarnio, MSc; Jenni Lahtinen, MSc; and Sanna Kängsep, PhD, from the Hovatta Lab for technical help and discussions on mouse experiments. None of these individuals received compensation for their work.

References
1.
Kessler  RC, Avenevoli  S, Costello  EJ,  et al.  Prevalence, persistence, and sociodemographic correlates of DSM-IV disorders in the National Comorbidity Survey Replication Adolescent Supplement.  Arch Gen Psychiatry. 2012;69(4):372-380. doi:10.1001/archgenpsychiatry.2011.160PubMedGoogle ScholarCrossref
2.
Wittchen  HU, Jacobi  F, Rehm  J,  et al.  The size and burden of mental disorders and other disorders of the brain in Europe 2010.  Eur Neuropsychopharmacol. 2011;21(9):655-679. doi:10.1016/j.euroneuro.2011.07.018PubMedGoogle ScholarCrossref
3.
Hettema  JM, Neale  MC, Kendler  KS.  A review and meta-analysis of the genetic epidemiology of anxiety disorders.  Am J Psychiatry. 2001;158(10):1568-1578. doi:10.1176/appi.ajp.158.10.1568PubMedGoogle ScholarCrossref
4.
Tambs  K, Czajkowsky  N, Røysamb  E,  et al.  Structure of genetic and environmental risk factors for dimensional representations of DSM-IV anxiety disorders.  Br J Psychiatry. 2009;195(4):301-307. doi:10.1192/bjp.bp.108.059485PubMedGoogle ScholarCrossref
5.
Erhardt  A, Czibere  L, Roeske  D,  et al.  TMEM132D, a new candidate for anxiety phenotypes: evidence from human and mouse studies.  Mol Psychiatry. 2011;16(6):647-663. doi:10.1038/mp.2010.41PubMedGoogle ScholarCrossref
6.
Otowa  T, Yoshida  E, Sugaya  N,  et al.  Genome-wide association study of panic disorder in the Japanese population.  J Hum Genet. 2009;54(2):122-126. doi:10.1038/jhg.2008.17PubMedGoogle ScholarCrossref
7.
Otowa  T, Tanii  H, Sugaya  N,  et al.  Replication of a genome-wide association study of panic disorder in a Japanese population.  J Hum Genet. 2010;55(2):91-96. doi:10.1038/jhg.2009.127PubMedGoogle ScholarCrossref
8.
Otowa  T, Kawamura  Y, Nishida  N,  et al.  Meta-analysis of genome-wide association studies for panic disorder in the Japanese population.  Transl Psychiatry. 2012;2:e186. doi:10.1038/tp.2012.89PubMedGoogle ScholarCrossref
9.
Logue  MW, Baldwin  C, Guffanti  G,  et al.  A genome-wide association study of post-traumatic stress disorder identifies the retinoid-related orphan receptor alpha (RORA) gene as a significant risk locus.  Mol Psychiatry. 2013;18(8):937-942. doi:10.1038/mp.2012.113PubMedGoogle ScholarCrossref
10.
Xie  P, Kranzler  HR, Yang  C, Zhao  H, Farrer  LA, Gelernter  J.  Genome-wide association study identifies new susceptibility loci for posttraumatic stress disorder.  Biol Psychiatry. 2013;74(9):656-663. doi:10.1016/j.biopsych.2013.04.013PubMedGoogle ScholarCrossref
11.
Guffanti  G, Galea  S, Yan  L,  et al.  Genome-wide association study implicates a novel RNA gene, the lincRNA AC068718.1, as a risk factor for post-traumatic stress disorder in women.  Psychoneuroendocrinology. 2013;38(12):3029-3038. doi:10.1016/j.psyneuen.2013.08.014PubMedGoogle ScholarCrossref
12.
Stein  MB, Chen  CY, Ursano  RJ,  et al; Army Study to Assess Risk and Resilience in Servicemembers (STARRS) Collaborators.  Genome-wide association studies of posttraumatic stress disorder in 2 cohorts of US Army soldiers.  JAMA Psychiatry. 2016;73(7):695-704. doi:10.1001/jamapsychiatry.2016.0350PubMedGoogle ScholarCrossref
13.
Ashley-Koch  AE, Garrett  ME, Gibson  J,  et al; Veterans Affairs Mid-Atlantic Mental Illness Research, Education, and Clinical Center Workgroup.  Genome-wide association study of posttraumatic stress disorder in a cohort of Iraq-Afghanistan era veterans.  J Affect Disord. 2015;184:225-234. doi:10.1016/j.jad.2015.03.049PubMedGoogle ScholarCrossref
14.
Nievergelt  CM, Maihofer  AX, Mustapic  M,  et al.  Genomic predictors of combat stress vulnerability and resilience in U.S. Marines: a genome-wide association study across multiple ancestries implicates PRTFDC1 as a potential PTSD gene.  Psychoneuroendocrinology. 2015;51:459-471. doi:10.1016/j.psyneuen.2014.10.017PubMedGoogle ScholarCrossref
15.
Kilaru  V, Iyer  SV, Almli  LM,  et al.  Genome-wide gene-based analysis suggests an association between Neuroligin 1 (NLGN1) and post-traumatic stress disorder.  Transl Psychiatry. 2016;6:e820. doi:10.1038/tp.2016.69PubMedGoogle ScholarCrossref
16.
Dunn  EC, Sofer  T, Gallo  LC,  et al.  Genome-wide association study of generalized anxiety symptoms in the Hispanic Community Health Study/Study of Latinos.  Am J Med Genet B Neuropsychiatr Genet. 2017;174(2):132-143. doi:10.1002/ajmg.b.32448PubMedGoogle ScholarCrossref
17.
Walter  S, Glymour  MM, Koenen  K,  et al.  Performance of polygenic scores for predicting phobic anxiety.  PLoS One. 2013;8(11):e80326. doi:10.1371/journal.pone.0080326PubMedGoogle ScholarCrossref
18.
Stein  MB, Chen  CY, Jain  S,  et al; Army STARRS Collaborators.  Genetic risk variants for social anxiety.  Am J Med Genet B Neuropsychiatr Genet. 2017;174(2):120-131. doi:10.1002/ajmg.b.32520PubMedGoogle ScholarCrossref
19.
Otowa  T, Hek  K, Lee  M,  et al.  Meta-analysis of genome-wide association studies of anxiety disorders.  Mol Psychiatry. 2016;21(10):1391-1399. doi:10.1038/mp.2015.197PubMedGoogle ScholarCrossref
20.
Purves  KL, Coleman  JRI, Meier  SM,  et al.  A major role for common genetic variation in anxiety disorders.  Bio Rxiv. https://www.biorxiv.org/content/10.1101/203844v2. Published April 11, 2019. Accessed April 15, 2019.Google Scholar
21.
Kessler  RC, Sampson  NA, Berglund  P,  et al.  Anxious and non-anxious major depressive disorder in the World Health Organization World Mental Health Surveys.  Epidemiol Psychiatr Sci. 2015;24(3):210-226. doi:10.1017/S2045796015000189PubMedGoogle ScholarCrossref
22.
Pedersen  CB, Bybjerg-Grauholm  J, Pedersen  MG,  et al.  The iPSYCH2012 case-cohort sample: new directions for unravelling genetic and environmental architectures of severe mental disorders.  Mol Psychiatry. 2018;23(1):6-14.PubMedGoogle ScholarCrossref
23.
Nørgaard-Pedersen  B, Hougaard  DM.  Storage policies and use of the Danish Newborn Screening Biobank.  J Inherit Metab Dis. 2007;30(4):530-536. doi:10.1007/s10545-007-0631-xPubMedGoogle ScholarCrossref
24.
Hollegaard  MV, Grove  J, Thorsen  P, Nørgaard-Pedersen  B, Hougaard  DM.  High-throughput genotyping on archived dried blood spot samples.  Genet Test Mol Biomarkers. 2009;13(2):173-179. doi:10.1089/gtmb.2008.0073PubMedGoogle ScholarCrossref
25.
Lynge  E, Sandegaard  JL, Rebolj  M.  The Danish national patient register.  Scand J Public Health. 2011;39(7)(suppl):30-33. doi:10.1177/1403494811401482PubMedGoogle ScholarCrossref
26.
Mors  O, Perto  GP, Mortensen  PB.  The Danish psychiatric central research register.  Scand J Public Health. 2011;39(7)(suppl):54-57. doi:10.1177/1403494810395825PubMedGoogle ScholarCrossref
27.
van Steensel  FJ, Bögels  SM, Perrin  S.  Anxiety disorders in children and adolescents with autistic spectrum disorders: a meta-analysis.  Clin Child Fam Psychol Rev. 2011;14(3):302-317. doi:10.1007/s10567-011-0097-0PubMedGoogle ScholarCrossref
28.
Grove  J, Ripke  S, Als  TD,  et al; Autism Spectrum Disorder Working Group of the Psychiatric Genomics Consortium; BUPGEN; Major Depressive Disorder Working Group of the Psychiatric Genomics Consortium; 23andMe Research Team.  Identification of common genetic risk variants for autism spectrum disorder.  Nat Genet. 2019;51(3):431-444. doi:10.1038/s41588-019-0344-8PubMedGoogle ScholarCrossref
29.
Ripkelab/ricopili. Main ricopili repo for public releases. https://github.com/Ripkelab/ricopili. Accessed April 15, 2019.
30.
Schizophrenia Working Group of the Psychiatric Genomics Consortium.  Biological insights from 108 schizophrenia-associated genetic loci.  Nature. 2014;511(7510):421-427. doi:10.1038/nature13595PubMedGoogle ScholarCrossref
31.
Abecasis  GR, Altshuler  D, Auton  A,  et al; 1000 Genomes Project Consortium.  A map of human genome variation from population-scale sequencing.  Nature. 2010;467(7319):1061-1073. doi:10.1038/nature09534PubMedGoogle ScholarCrossref
32.
Delaneau  O, Marchini  J, Zagury  JF.  A linear complexity phasing method for thousands of genomes.  Nat Methods. 2011;9(2):179-181. doi:10.1038/nmeth.1785PubMedGoogle ScholarCrossref
33.
Howie  BN, Donnelly  P, Marchini  J.  A flexible and accurate genotype imputation method for the next generation of genome-wide association studies.  PLoS Genet. 2009;5(6):e1000529. doi:10.1371/journal.pgen.1000529PubMedGoogle ScholarCrossref
34.
Price  AL, Patterson  NJ, Plenge  RM, Weinblatt  ME, Shadick  NA, Reich  D.  Principal components analysis corrects for stratification in genome-wide association studies.  Nat Genet. 2006;38(8):904-909. doi:10.1038/ng1847PubMedGoogle ScholarCrossref
35.
Willer  CJ, Li  Y, Abecasis  GR.  METAL: fast and efficient meta-analysis of genomewide association scans.  Bioinformatics. 2010;26(17):2190-2191. doi:10.1093/bioinformatics/btq340PubMedGoogle ScholarCrossref
36.
Nagel  M, Watanabe  K, Stringer  S, Posthuma  D, van der Sluis  S.  Item-level analyses reveal genetic heterogeneity in neuroticism.  Nat Commun. 2018;9(1):905. doi:10.1038/s41467-018-03242-8PubMedGoogle ScholarCrossref
37.
de Leeuw  CA, Mooij  JM, Heskes  T, Posthuma  D.  MAGMA: generalized gene-set analysis of GWAS data.  PLoS Comput Biol. 2015;11(4):e1004219. doi:10.1371/journal.pcbi.1004219PubMedGoogle ScholarCrossref
38.
Bulik-Sullivan  BK, Loh  PR, Finucane  HK,  et al; Schizophrenia Working Group of the Psychiatric Genomics Consortium.  LD Score regression distinguishes confounding from polygenicity in genome-wide association studies.  Nat Genet. 2015;47(3):291-295. doi:10.1038/ng.3211PubMedGoogle ScholarCrossref
39.
Bulik-Sullivan  B, Finucane  HK, Anttila  V,  et al; ReproGen Consortium; Psychiatric Genomics Consortium; Genetic Consortium for Anorexia Nervosa of the Wellcome Trust Case Control Consortium 3.  An atlas of genetic correlations across human diseases and traits.  Nat Genet. 2015;47(11):1236-1241. doi:10.1038/ng.3406PubMedGoogle ScholarCrossref
40.
Pedersen  CB, Mors  O, Bertelsen  A,  et al.  A comprehensive nationwide study of the incidence rate and lifetime risk for treated mental disorders.  JAMA Psychiatry. 2014;71(5):573-581. doi:10.1001/jamapsychiatry.2014.16PubMedGoogle ScholarCrossref
41.
Kessler  RC, Berglund  P, Demler  O, Jin  R, Merikangas  KR, Walters  EE.  Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication.  Arch Gen Psychiatry. 2005;62(6):593-602. doi:10.1001/archpsyc.62.6.593PubMedGoogle ScholarCrossref
42.
Finucane  HK, Bulik-Sullivan  B, Gusev  A,  et al; ReproGen Consortium; Schizophrenia Working Group of the Psychiatric Genomics Consortium; RACI Consortium.  Partitioning heritability by functional annotation using genome-wide association summary statistics.  Nat Genet. 2015;47(11):1228-1235. doi:10.1038/ng.3404PubMedGoogle ScholarCrossref
43.
UK Biobank: Neale lab. http://www.nealelab.is/uk-biobank. Accessed April 15, 2019.
44.
Zheng  J, Erzurumluoglu  AM, Elsworth  BL,  et al; Early Genetics and Lifecourse Epidemiology (EAGLE) Eczema Consortium.  LD Hub: a centralized database and web interface to perform LD score regression that maximizes the potential of summary level GWAS data for SNP heritability and genetic correlation analysis.  Bioinformatics. 2017;33(2):272-279. doi:10.1093/bioinformatics/btw613PubMedGoogle ScholarCrossref
45.
Purcell  SM, Wray  NR, Stone  JL,  et al; International Schizophrenia Consortium.  Common polygenic variation contributes to risk of schizophrenia and bipolar disorder.  Nature. 2009;460(7256):748-752.PubMedGoogle ScholarCrossref
46.
Golden  SA, Covington  HE  III, Berton  O, Russo  SJ.  A standardized protocol for repeated social defeat stress in mice.  Nat Protoc. 2011;6(8):1183-1191. doi:10.1038/nprot.2011.361PubMedGoogle ScholarCrossref
47.
Hochberg  Y, Benjamini  Y.  More powerful procedures for multiple significance testing.  Stat Med. 1990;9(7):811-818. doi:10.1002/sim.4780090710PubMedGoogle ScholarCrossref
48.
Pardiñas  AF, Holmans  P, Pocklington  AJ,  et al; GERAD1 Consortium; CRESTAR Consortium.  Common schizophrenia alleles are enriched in mutation-intolerant genes and in regions under strong background selection.  Nat Genet. 2018;50(3):381-389. doi:10.1038/s41588-018-0059-2PubMedGoogle ScholarCrossref
49.
Li  Z, Chen  J, Yu  H,  et al.  Genome-wide association analysis identifies 30 new susceptibility loci for schizophrenia.  Nat Genet. 2017;49(11):1576-1583. doi:10.1038/ng.3973PubMedGoogle ScholarCrossref
50.
Pasman  JA, Verweij  KJH, Gerring  Z,  et al; 23andMe Research Team; Substance Use Disorders Working Group of the Psychiatric Genomics Consortium; International Cannabis Consortium.  GWAS of lifetime cannabis use reveals new risk loci, genetic overlap with psychiatric traits, and a causal influence of schizophrenia.  Nat Neurosci. 2018;21(9):1161-1170. doi:10.1038/s41593-018-0206-1PubMedGoogle ScholarCrossref
51.
Mühleisen  TW, Leber  M, Schulze  TG,  et al.  Genome-wide association study reveals two new risk loci for bipolar disorder.  Nat Commun. 2014;5:3339. doi:10.1038/ncomms4339PubMedGoogle ScholarCrossref
52.
is Duncan  LE, Ratanatharathorn  A, Aiello  AE,  et al.  Largest GWAS of PTSD (N=20 070) yields genetic overlap with schizophrenia and sex differences in heritability.  Mol Psychiatry. 2018;23(3):666-673. doi:10.1038/mp.2017.77PubMedGoogle ScholarCrossref
53.
Otowa  T, Kawamura  Y, Sugaya  N,  et al.  Association study of PDE4B with panic disorder in the Japanese population.  Prog Neuropsychopharmacol Biol Psychiatry. 2011;35(2):545-549. doi:10.1016/j.pnpbp.2010.12.013PubMedGoogle ScholarCrossref
54.
Zhang  C, Xu  Y, Zhang  HT, Gurney  ME, O’Donnell  JM.  Comparison of the pharmacological profiles of selective PDE4B and PDE4D inhibitors in the central nervous system.  Sci Rep. 2017;7:40115. doi:10.1038/srep40115PubMedGoogle ScholarCrossref
55.
Zhang  HT, Huang  Y, Masood  A,  et al.  Anxiogenic-like behavioral phenotype of mice deficient in phosphodiesterase 4B (PDE4B).  Neuropsychopharmacology. 2008;33(7):1611-1623. doi:10.1038/sj.npp.1301537PubMedGoogle ScholarCrossref
56.
Meier  SM, Petersen  L, Mattheisen  M, Mors  O, Mortensen  PB, Laursen  TM.  Secondary depression in severe anxiety disorders: a population-based cohort study in Denmark.  Lancet Psychiatry. 2015;2(6):515-523. doi:10.1016/S2215-0366(15)00092-9PubMedGoogle ScholarCrossref
57.
Cross-Disorder Group of the Psychiatric Genomics Consortium.  Identification of risk loci with shared effects on five major psychiatric disorders: a genome-wide analysis.  Lancet. 2013;381(9875):1371-1379. doi:10.1016/S0140-6736(12)62129-1PubMedGoogle ScholarCrossref
58.
de Lijster  JM, Dieleman  GC, Utens  EMWJ,  et al.  Social and academic functioning in adolescents with anxiety disorders: a systematic review.  J Affect Disord. 2018;230:108-117. doi:10.1016/j.jad.2018.01.008PubMedGoogle ScholarCrossref
59.
Hettema  JM, Prescott  CA, Myers  JM, Neale  MC, Kendler  KS.  The structure of genetic and environmental risk factors for anxiety disorders in men and women.  Arch Gen Psychiatry. 2005;62(2):182-189. doi:10.1001/archpsyc.62.2.182PubMedGoogle ScholarCrossref
×