Leveraging a founder population to identify novel rare-population genetic determinants of lipidome

May E. Montasser, Stella Aslibekyan, Vinodh Srinivasasainagendra, Hemant K. Tiwari, Amit Patki, Minoo Bagheri, Tobias Kind, Dinesh Kumar Barupal, Sili Fan, James A. Perry, Kathleen A. Ryan, Donna K. Arnett, Amber L. Beitelshees, Marguerite R. Irvin, Jeffrey R. O’ConnellbioRxiv 2021 · DOI

TL;DR

This study leverages the unique genetic makeup of the Old Order Amish founder population to identify novel genetic variants influencing lipid levels. By analyzing a comprehensive set of 355 lipid species, researchers discovered three new rare genetic loci associated with sphingolipids and confirmed associations for two known rare variants, highlighting the power of founder populations for uncovering rare genetic influences on complex traits.

What Problem Does This Paper Address?

Understanding the genetic underpinnings of an individual's lipid profile (the "lipidome") is crucial for unraveling the mechanisms behind cardiovascular disease (CVD) risk. Traditional genetic studies, often conducted in large, diverse populations, are primarily powered to detect associations with common genetic variants. However, rare genetic variants, which can have larger effect sizes, are often missed in these studies. Founder populations, like the Old Order Amish, offer a unique advantage because genetic drift has led to an increased frequency of certain rare alleles compared to the general population. This study aims to harness this advantage to discover novel, rare genetic determinants of the lipidome that might be missed in broader population studies.

Methods

The researchers employed a Genome-Wide Association Study (GWAS) approach to identify genetic associations with lipid species. They analyzed 355 lipid species in 650 individuals from the Old Order Amish founder population. To identify potential causal variants within identified loci, they utilized fine-mapping analyses and functional annotation tools. Replication and validation of findings were performed using data from the Genetics of Lipid Lowering Drugs and Diet Network (GOLDN) study, which comprises outbred populations. They also leveraged publicly available association databases for further lookup.

Datasets

  • Discovery Cohort: 650 individuals from the Old Order Amish (OOA) population of Lancaster County, PA.
  • Replication Cohort: The Genetics of Lipid Lowering Drugs and Diet Network (GOLDN) study.
  • Public Databases: UK Biobank, FinnGen, BioBank Japan, dbSNP, gnomAD, GTEx, Open Targets Genetics, and UCSC Genome Browser were used for annotation and lookup.

Key Insights and Ideas

  • Founder Populations as Genetic Discovery Engines: The study strongly demonstrates that founder populations, due to increased allele frequencies of rare variants, are powerful resources for identifying novel genetic associations that would be difficult to detect in larger, outbred cohorts.
  • Lipidome Complexity Beyond Traditional Lipids: The analysis of 355 lipid species, including many not previously studied, reveals a more nuanced genetic architecture of lipid metabolism than previously understood by focusing solely on traditional lipids like cholesterol and triglycerides.
  • Rare Variants with Significant Impact: The identification of novel rare-population variants associated with specific lipid species underscores the importance of rare genetic variation in influencing complex metabolic traits.
  • Functional Annotation is Key: The researchers effectively used functional annotation to prioritize potential causal variants within identified genetic loci, linking them to plausible genes involved in lipid metabolism.

Key Results

  • The study identified three novel rare-population but Amish-enriched loci associated with lipid species, specifically sphingolipids. These include associations with:
  • GLTPD2_rs536055318 (associated with SM(d40:0))
    • CERS5_rs771033566 (associated with SM(d32:2) and SM(d30:1))
    • AKNA_rs531892793 (associated with glucosylceramide species)
  • They reported for the first time the association of lipid species with two previously known rare-population but Amish-enriched lipid variants:
  • APOB_rs5742904
    • APOC3_rs76353203
  • Seven previously known common loci were replicated, including novel associations for two sterols:
  • Androstenediol with a UGT locus on chromosome 2.
  • Estriol with an SLC22A8/A24 locus on chromosome 11.
  • The heritability of lipid species was found to be primarily additive, with sphingolipids showing higher heritability than glycerolipids.

Limitations and Future Work

  • Sample Size: While the Amish cohort is valuable, the authors acknowledge that a relatively small sample size (650 individuals) may limit the power to detect all rare variants. Larger sample sizes in the Amish and other founder populations are suggested for identifying more rare variants.
  • Replication Challenges: Replicating rare-population variants in outbred populations can be challenging due to their rarity. The study employed strategies to address this, but further validation in larger, diverse cohorts is always beneficial.
  • Functional Validation: While functional annotation was used to prioritize candidate variants, experimental validation of the identified variants and their proposed mechanisms is a crucial next step.
  • Comprehensive Lipidome Profiling: The study excluded lipid species with low-quality data. More complete profiling with higher quality data could lead to further discoveries.
  • Therapeutic Targets: The identified novel variants and their associated lipid species may represent potential therapeutic targets for CVD, warranting further investigation and personalized medicine approaches.

Summary updated

Abstract

ABSTRACT Identifying the genetic determinants of inter-individual variation in lipid species (lipidome) may provide deeper understanding and new insight into the mechanistic effect of complex lipidomic pathways in CVD risk and progression beyond simple traditional lipids. Previous studies have been largely population based and thus only powered to discover associations with common genetic variants. Founder populations represent a powerful resource to accelerate discovery of novel biology associated with rare population alleles that have risen to higher frequency due to genetic drift. We performed a GWAS of 355 lipid species in 650 individuals from the Old Order Amish founder population including 127 lipid species not previously tested. We report for the first time the lipid species associated with two rare-population but Amish-enriched lipid variants: APOB _rs5742904 and APOC3 _rs76353203. We also identified novel associations for 3 rare-population Amish-enriched loci with several sphingolipids and with proposed potential functional/causal variant in each locus including GLPTD2 _rs536055318, CERS5 _rs771033566, and AKNA _rs531892793. We replicated 7 previously known common loci including novel associations with two sterols: androstenediol with UGT locus on chromosome 2 and estriol with SLC22A8/A24 locus on chromosome 11. Our results show the power of founder populations to discover novel biology due to genetic drift that can increase the frequency of an allele from only few copies in large sample cohorts such as the UK Biobank to dozens of copies in sample size as small as 650.
BibTeX
@misc{montasser2021leveraging,
  author = {May E. Montasser and Stella Aslibekyan and Vinodh Srinivasasainagendra and Hemant K. Tiwari and Amit Patki and Minoo Bagheri and Tobias Kind and Dinesh Kumar Barupal and Sili Fan and James A. Perry and Kathleen A. Ryan and Donna K. Arnett and Amber L. Beitelshees and Marguerite R. Irvin and Jeffrey R. O’Connell},
  title = {{Leveraging a founder population to identify novel rare-population genetic determinants of lipidome}},
  year = {2021},
  doi = {10.1101/2021.05.21.445208},
  howpublished = {bioRxiv},
}
Figure