Missense variants in FRS3 affect body mass index in populations of diverse ancestries

Abstract

Obesity is associated with adverse effects on health and quality of life. Improved understanding of its underlying pathophysiology is essential for developing counteractive measures. To search for sequence variants with large effects on BMI, we perform a multi-ancestry meta-analysis of 13 genome-wide association studies on BMI, including data derived from 1,534,555 individuals of European ancestry, 339,657 of Asian ancestry, and 130,968 of African ancestry. We identify an intergenic 262,760 base pair deletion at the MC4R locus that associates with 4.11 kg/m² higher BMI per allele, likely through downregulation of MC4R. Moreover, a rare FRS3 missense variant, p.Glu115Lys, only found in individuals from Finland, associates with 1.09 kg/m² lower BMI per allele. We also detect three other low-frequency FRS3 missense variants that associate with BMI with smaller effects and are enriched in different ancestries. We characterize FRS3 as a BMI-associated gene, encoding an adaptor protein known to act downstream of BDNF and TrkB, which regulate appetite, food intake, and energy expenditure through unknown signaling pathways. The work presented here contributes to the biological foundation of obesity by providing a convincing downstream component of the BDNF-TrkB pathway, which could potentially be targeted for obesity treatment.

Fig. 1. Forest plots showing the pattern of BMI effects of the four FRS3 missense variants identified in the study.

The plots illustrate the associations between the four FRS3 missense variants identified in the study and BMI: a p.Glu115Lys (rs773053137-T), b p.Pro137Arg (rs146730626-C), c p.Pro172Leu (rs74687105-A), and d p.Arg316Gln (rs35744673-T). Data are presented as effect sizes in standard deviation (SD) units with 95% confidence intervals, combined (from the meta-analysis) and for individual studies. Effect sizes were determined using a linear mixed model implemented in BOLT-LMM, assuming an additive genetic model, and two-sided P values were calculated. The P values presented in the figure have not been adjusted for multiple comparisons. The vertical dashed line indicates effect size equal to zero and horizontal dashed lines separate effect estimates by ancestry. Effects are not shown for allele frequencies below 0.1%. Source data are provided as a Source Data file. AF allele frequency, EUR European, SAS South Asian, EAS East Asian, AFR African, UKB UK Biobank, CHB/DBDS Copenhagen Hospital Biobank/Danish Blood Donor Study, GIANT Genetic Investigation of Anthropometric Traits, NashBio Nashville Biosciences, MVP Million Veteran Program, AAAGC African Ancestry Anthropometry Genetics Consortium.

Fig. 2. World maps showing the allele frequency distribution of the four FRS3 missense variants identified in the study.

Low allele frequencies are shown in light color, while higher frequencies are depicted in dark color. Gray indicates that zero participants reported being born in that country or region. Frequencies shown are for UK Biobank participants with origins in that region, except for Finland, for which the frequency in FinnGen was used. a p.Glu115Lys (rs773053137-T) is only found in Finland. b p.Pro137Arg (rs146730626-C) is predominantly found in East Asia but is present at lower frequencies in South Asia and the Indian Ocean. c p.Pro172Leu (rs74687105-A) is predominantly found in populations of African ancestry and is most common in Southern Africa. d p.Arg316Gln (rs35744673-T) is broadly distributed, but most common in populations of European ancestry. This figure was created with Natural Earth.

Fig. 3. Distribution of BMI categories for carriers and non-carriers of the FRS3 missense variant p.Glu115Lys among participants in the FinnGen study.

The BMI groups presented in the figure were defined according to the World Health Organization’s recommendations. Source data are provided as a Source Data file.

Fig. 4. A schematic overview of the proposed BDNF-TrkB-FRS3 signaling pathway, its role in energy balance regulation, and its connection to the leptin-melanocortin pathway.

Key anorexigenic components are illustrated in dark blue for the leptin-melanocortin pathway and lighter blue for the BDNF-TrkB-FRS3 pathway. Pointed arrows between the components represent activation of downstream proteins, where whole pointed arrows indicate that the linking mechanisms are well-established and dotted pointed arrows indicate that mechanisms remain to be clarified. The leptin-melanocortin pathway is activated in response to increased leptin levels and other hormones in the feeding state, resulting in decreased appetite and food intake, increased energy expenditure, and lower BMI. It also has an inhibiting mechanism, illustrated in light blue to the left, where AGRP acts as an antagonist of MC4R in response to low leptin levels with food deprivation, resulting in MC4R inhibition, increased appetite, decreased energy expenditure, and higher BMI. The BDNF-TrkB pathway is thought to be activated downstream of the leptin-melanocortin pathway to mediate its effect on energy balance. We postulate that FRS3 is a link in the undiscovered downstream signaling cascade of BDNF and TrkB. Components of these pathways are associated with both monogenic and polygenic forms of obesity, with the supporting evidence summarized to the right in the figure. Created in BioRender. Aegisdottir, H. (2025) https://BioRender.com/ a25a422. LEPR leptin receptor, AGRP agouti-related protein, POMC pro-opiomelanocortin, α-MSH α-melanocyte stimulating hormone, PCSK1 proprotein convertase subtilisin/kexin type 1, pLoF predicted loss-of-function, MC4R melanocortin 4 receptor, BDNF, brain derived neurotrophic factor, TrkB tropomyosin receptor kinase B, FRS3 fibroblast growth factor receptor substrate 3, EAF effect allele frequency.