Genome-wide association study of depression phenotypes in UK Biobank identifies variants in excitatory synaptic pathways

David M. Howard, Mark J. Adams, Masoud Shirali, Toni‐Kim Clarke, Riccardo E. Marioni, Gail Davies, Jonathan R. I. Coleman, Clara Alloza, Xueyi Shen, Miruna C. Barbu, Eleanor M. Wigmore, Jude Gibson, Michelle Agee, Babak Alipanahi, Adam Auton, Robert K. Bell, Katarzyna Bryc, Sarah L. Elson, Pierre Fontanillas, Nicholas A. Furlotte, David A. Hinds, Karen E. Huber, Aaron Kleinman, Nadia K. Litterman, Jennifer C. McCreight, Matthew H. McIntyre, Joanna L. Mountain, Elizabeth S. Noblin, Carrie A. M. Northover, Steven J. Pitts, J. Fah Sathirapongsasuti, Olga V. Sazonova, Janie F. Shelton, Suyash Shringarpure, Chao Tian, Joyce Y. Tung, Vladimir Vacic, Catherine H. Wilson, Saskia P. Hagenaars, Cathryn M. Lewis, Joey Ward, Daniel J. Smıth, Patrick F. Sullivan, Chris Haley, Gerome Breen, Ian J. Deary, Andrew M. McIntoshNature Communications 2018 · DOI

TL;DR

This study leverages the massive UK Biobank dataset to conduct a genome-wide association study (GWAS) of depression, identifying 17 genetic loci associated with depression risk. These findings point towards the crucial role of excitatory synaptic pathways in the biological underpinnings of depression, suggesting that broader depression phenotypes may be more amenable to genetic discovery.

What Problem Does This Paper Address?

Depression is a complex, polygenic disorder with a significant heritability, yet identifying the specific genetic variants contributing to its risk has been challenging. Previous genetic studies have often relied on varying definitions of depression, making it difficult to compare findings and understand the underlying biological mechanisms. This paper aims to address this by using three distinct, well-defined depression phenotypes from the UK Biobank to conduct a large-scale GWAS, seeking to identify novel genetic associations and gain deeper insights into the biological pathways involved.

Methods

The researchers employed a genome-wide association study (GWAS) approach, a standard technique for identifying genetic variants associated with a trait. They utilized the UK Biobank, a large-scale biomedical database, and analyzed three depression phenotypes: "broad depression" (self-reported help-seeking), "probable major depressive disorder (MDD)" (based on self-reported symptoms and impairment), and "ICD-coded MDD" (diagnoses from hospital records). They performed a meta-analysis with an independent dataset (23andMe) for replication. Additionally, they used gene-set enrichment analysis (MAGMA) to identify biological pathways implicated by the associated genetic variants and tissue enrichment analysis (DEPICT) to explore where these genes are expressed.

Datasets

  • UK Biobank: A large population-based cohort with over 500,000 participants, providing extensive genetic and phenotypic data. For this study, up to 322,580 participants were used for the GWAS.
  • 23andMe Research Participants: An independent dataset of approximately 307,354 individuals used for replication of the identified genetic variants.

Key Insights and Ideas

  • Broad Depression as a Tractable Phenotype: The study highlights that the "broad depression" phenotype, while potentially capturing a wider range of conditions, proved to be the most successful for identifying genetic associations due to its larger sample size. This suggests that for genetic discovery, broader, more inclusive phenotypes might be more powerful, even if they are less specific.
  • Synaptic Pathways are Key: A major finding is the enrichment of associated genes in pathways related to excitatory neurotransmission, including postsynaptic functions, neuron spines, and dendrites. This strongly implicates the intricate workings of neuronal communication in the genetic architecture of depression.
  • Replication is Crucial: The consistent replication of effect directions in the 23andMe cohort, and the formal replication of seven variants, lends strong support to the findings and suggests these are genuine associations, not just statistical artifacts.
  • Novelty in Genetic Loci: A significant portion of the identified loci (14 out of 17) are likely novel, meaning they haven't been previously linked to depression in major GWAS. This expands our understanding of the genetic landscape of the disorder.

Key Results

  • 17 Independent Genetic Loci: The study identified 17 independent genetic loci that reached genome-wide significance (P < 5 × 10⁻⁸) across the three depression phenotypes.
  • Phenotype-Specific Associations: Of these 17 loci, 14 were associated with broad depression, 2 with probable MDD, and 1 with ICD-coded MDD.
  • Replication Success: 16 of the 17 variants showed the same direction of effect in the 23andMe replication cohort, with 7 variants achieving statistical significance after correction for multiple testing.
  • Enriched Pathways: Gene-set enrichment analysis revealed significant associations with pathways like GO_EXCITATORY_SYNAPSE, GO_MECHANOSENSORY_BEHAVIOR, GO_POSTSYNAPSE, GO_NEURON_SPINE, and GO_DENDRITE for broad depression.
  • Heritability Estimates: SNP-based heritability estimates for depression ranged from approximately 4.7% to 10.2% on the liability scale, consistent with previous research.

Limitations and Future Work

  • Phenotype Definitions: The authors acknowledge that none of the depression phenotypes used are based on a formal structured diagnostic interview (like the Structured Clinical Interview for DSM Axis I Disorders). Self-reported data can be subject to recall bias, and the broad depression phenotype might capture individuals with other personality or psychiatric disorders.
  • Sample Size for Stricter Phenotypes: The smaller sample sizes for probable MDD and ICD-coded MDD limited the power to detect significant genetic effects for these more specific phenotypes.
  • Heterogeneity of Depression: Depression is a heterogeneous condition, and while this study provides valuable insights, further research with larger sample sizes and more refined phenotyping will be necessary to fully understand its subtypes.
  • Future Directions: The paper suggests that broader depression phenotypes are more tractable for genetic discovery and that future research should continue to explore these pathways. Investigating the specific genes within the identified pathways and their functional roles in excitatory neurotransmission is a key next step. Further studies could also explore how these genetic variants interact with environmental factors to influence depression risk.

Summary updated

Abstract

Depression is a polygenic trait that causes extensive periods of disability. Previous genetic studies have identified common risk variants which have progressively increased in number with increasing sample sizes of the respective studies. Here, we conduct a genome-wide association study in 322,580 UK Biobank participants for three depression-related phenotypes: broad depression, probable major depressive disorder (MDD), and International Classification of Diseases (ICD, version 9 or 10)-coded MDD. We identify 17 independent loci that are significantly associated (P < 5 × 10<sup>-8</sup>) across the three phenotypes. The direction of effect of these loci is consistently replicated in an independent sample, with 14 loci likely representing novel findings. Gene sets are enriched in excitatory neurotransmission, mechanosensory behaviour, post synapse, neuron spine and dendrite functions. Our findings suggest that broad depression is the most tractable UK Biobank phenotype for discovering genes and gene sets that further our understanding of the biological pathways underlying depression.
BibTeX
@inproceedings{howard2018genome,
  author = {David M. Howard and Mark J. Adams and Masoud Shirali and Toni‐Kim Clarke and Riccardo E. Marioni and Gail Davies and Jonathan R. I. Coleman and Clara Alloza and Xueyi Shen and Miruna C. Barbu and Eleanor M. Wigmore and Jude Gibson and Michelle Agee and Babak Alipanahi and Adam Auton and Robert K. Bell and Katarzyna Bryc and Sarah L. Elson and Pierre Fontanillas and Nicholas A. Furlotte and David A. Hinds and Karen E. Huber and Aaron Kleinman and Nadia K. Litterman and Jennifer C. McCreight and Matthew H. McIntyre and Joanna L. Mountain and Elizabeth S. Noblin and Carrie A. M. Northover and Steven J. Pitts and J. Fah Sathirapongsasuti and Olga V. Sazonova and Janie F. Shelton and Suyash Shringarpure and Chao Tian and Joyce Y. Tung and Vladimir Vacic and Catherine H. Wilson and Saskia P. Hagenaars and Cathryn M. Lewis and Joey Ward and Daniel J. Smıth and Patrick F. Sullivan and Chris Haley and Gerome Breen and Ian J. Deary and Andrew M. McIntosh},
  title = {{Genome-wide association study of depression phenotypes in UK Biobank identifies variants in excitatory synaptic pathways}},
  booktitle = {Nature Communications},
  year = {2018},
  doi = {10.1038/s41467-018-03819-3},
}
Figure