Aim 1: Identification of gut microbiota signatures that increase fracture risk
We have successfully established the network of population‑based cohorts required to address this aim. These cohorts provide comprehensive phenotypic data, metagenomic sequencing of the gut microbiota, and information on musculoskeletal traits and incident fractures.
As an initial step, we conducted observational association analyses to identify gut microbiota–related signatures associated with musculoskeletal parameters and fracture risk in human cohorts. In a first publication based on a single cohort, we demonstrated that gut microbiota diversity, as well as specific microbial taxa and functional pathways, are associated with fracture risk. This work was published in npj Biofilms and Microbiomes in 2024, with the applicant as senior author, providing a strong proof‑of‑concept for the project hypothesis. However, no replication phase in an independent cohort was included. Building on this foundation, we are currently establishing the logistical framework required to perform a large‑scale meta‑analysis across four independent cohorts with available metagenomic sequencing data and prospective information on incident fractures. This effort will substantially increase statistical power and robustness, allowing for more precise identification of reproducible gut microbiota signatures linked to fracture risk in ageing populations. In parallel, analyses in the HUNT cohort have revealed that associations between gut microbiota taxonomy and functionality are stronger for measures of muscle function than for bone mineral density. These findings suggest that the link between gut microbiota and fracture risk may be mediated to a greater extent through effects on muscle function, such as strength and performance, rather than through reduced bone mineral density alone. This insight represents an important conceptual advance and has informed the refinement of subsequent project aims, particularly with regard to mechanistic and causal analyses. Sex steroids are major regulators of musculoskeletal function and fracture risk. We recently demonstrated that the gut microbiota expresses enzymes capable of modulating androgen metabolism, thereby affecting circulating levels of the potent androgen dihydrotestosterone in women. This work, with the grantee as lead author, was published in EBioMedicine in 2025 and provides novel mechanistic insight directly relevant to the objectives of the ERC project.
In addition, we have made significant progress towards enabling causal analyses of the identified gut microbiota (GM) signatures in relation to musculoskeletal parameters and fracture risk using Mendelian randomization (MR). A key prerequisite for these analyses is the identification of robust human genetic determinants of gut microbiota composition and functionality at an overall level. To this end, we have conducted two large‑scale genome‑wide association studies (GWAS) focusing on human genetic influences on gut microbiota composition and functional pathways. These studies were performed using data from the FINRISK, HUNT, SCAPIS, and SIMPLER cohorts, with successful replication in independent populations. Together, these efforts have identified multiple reproducible human genetic signals associated with variation in gut microbiota features.
The results were published in two articles in Nature Genetics in 2026, one with the ERC grantee as senior author, and represent a substantial contribution to the field by providing well‑powered and carefully validated GWAS resources for gut microbiota traits.
The resulting GWAS summary statistics constitute a critical enabling resource for the planned causal analyses using MR. Importantly, we have demonstrated that these GWAS data can be successfully applied within an MR framework to investigate causal relationships involving gut microbiota as an outcome. Specifically, in one of the published studies we showed that the summary statistics could be used to assess the causal effects of host‑related factors, such as body mass index, on gut microbiota composition. This proof‑of‑principle supports the use of these GWAS resources for downstream causal analyses of factors influencing GM, while acknowledging that ongoing work is required to further refine and validate instruments for more granular microbiota traits.
Finally, in functional studies using faecal microbiota transplantation (FMT), we aimed to identify bacterial species that protect against or contribute to age‑dependent loss of bone and muscle mass and muscle strength. We demonstrated that transfer of gut microbiota from aged mice to young recipient mice reduced relative lean mass, whereas transplantation of microbiota from young mice to aged mice had no effect on lean mass (manuscript submitted).
Aim 2 focused on the development of novel second‑generation probiotics with musculoskeletal effects. We demonstrated that the novel Bifidobacterium longum strain DSM 32947 increases bone mineral density in female mice. This work was published in Calcified Tissue International in 2025, with the grantee as lead author. In addition, we showed that the beneficial effects of a defined probiotic mixture on bone mass and lean mass in female mice are diet‑dependent. This work was published in Scientific Reports in 2025, with the grantee as lead author.
Aim 3 aims to determine the clinical usefulness of identified gut microbiota (GM)‑related fracture markers. This part of the project has recently been initiated. We are currently establishing the logistical framework and access to relevant clinical cohorts, and are in the process of prioritising which GM‑derived markers should be evaluated. These markers will be assessed in relation to, and in combination with, existing fracture prediction tools to determine their potential added clinical value.