Utilising the genetics of oestrogen traits in premenopausal and postmenopausal women to assess health impacts
Asteljoki, Juho (2024-11-18)
Utilising the genetics of oestrogen traits in premenopausal and postmenopausal women to assess health impacts
(18.11.2024)
Julkaisu on tekijänoikeussäännösten alainen. Teosta voi lukea ja tulostaa henkilökohtaista käyttöä varten. Käyttö kaupallisiin tarkoituksiin on kielletty.
suljettu
Julkaisun pysyvä osoite on:
https://urn.fi/URN:NBN:fi-fe20241216103423
https://urn.fi/URN:NBN:fi-fe20241216103423
Tiivistelmä
Genome-wide association studies (GWAS) have overlooked oestradiol (E2) and oestradiol-to-testosterone ratio (E2/T) in women due to the analytical challenges. Here, our objective was to characterise oestradiol traits in the UK Biobank. We aimed to employ more sophisticated GWAS models for E2 and E2/T. We also aimed to identify the sex hormone-binding globulin (SHBG)-associated variants that act as proxies for oestrogen. Finally, we sought to evaluate the health impacts of the traits investigated.
We partitioned women based on their menopausal status and employed GWAS models to adjust traits for group-specific covariates: the stage of the menstrual cycle and time since menopause. We identified oestrogen variants for premenopausal women from SHBG-associated variants by excluding those that are directly causal for SHBG or act through androgens. Health impacts were evaluated using genetic correlation and Mendelian randomisation.
We identified an SHBG locus for E2 and FSHB and CYP3A4 loci for E2/T in premenopausal women. The genetic correlations between the oestradiol traits and health traits did not reveal significant correlations. Using the SHBG proxy, we identified 10 loci plausibly reflecting oestrogens, many of which annotate to genes that encode proteins relevant to sex hormone regulation or metabolism (e.g. ESR1 and AKR1C4). These loci were utilised in Mendelian randomisation, which presented no statistically significant health impacts.
In summary, we identified 10 oestrogen loci in premenopausal women. Physiological variation in oestrogens did not impact the selected health traits, likely due to limited power. More comprehensive data are required for a more in-depth characterisation of oestrogens in women.
We partitioned women based on their menopausal status and employed GWAS models to adjust traits for group-specific covariates: the stage of the menstrual cycle and time since menopause. We identified oestrogen variants for premenopausal women from SHBG-associated variants by excluding those that are directly causal for SHBG or act through androgens. Health impacts were evaluated using genetic correlation and Mendelian randomisation.
We identified an SHBG locus for E2 and FSHB and CYP3A4 loci for E2/T in premenopausal women. The genetic correlations between the oestradiol traits and health traits did not reveal significant correlations. Using the SHBG proxy, we identified 10 loci plausibly reflecting oestrogens, many of which annotate to genes that encode proteins relevant to sex hormone regulation or metabolism (e.g. ESR1 and AKR1C4). These loci were utilised in Mendelian randomisation, which presented no statistically significant health impacts.
In summary, we identified 10 oestrogen loci in premenopausal women. Physiological variation in oestrogens did not impact the selected health traits, likely due to limited power. More comprehensive data are required for a more in-depth characterisation of oestrogens in women.