Gut microbes related to the Dietary Approaches to Stop Hypertension index are associated with bone quantity but not with measures of bone quality in older Puerto Rican adult

Authors: MERRILL, LISA - University Of Massachusetts, Lowell; LOPEZ MARTINEZ, RAFAEL - University Of Massachusetts, Lowell; PALACIOS, NATALIA - University Of Massachusetts, Lowell; DAWSON-HUGHES, BESS - Jean Mayer Human Nutrition Research Center On Aging At Tufts University; NOEL, SABRINA - University Of Massachusetts, Lowell; WANG, YAN - University Of Massachusetts, Lowell; TUCKER, KATHERINE - University Of Massachusetts, Lowell; MANGANO, KELSEY - University Of Massachusetts, Lowell

Submitted to: The American Journal of Clinical Nutrition
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/24/2025
Publication Date: 12/17/2025
Citation: Merrill, L.C., Lopez Martinez, R., Palacios, N., Dawson-Hughes, B., Noel, S., Wang, Y., Tucker, K., Mangano, K. 2025. Gut microbes related to the Dietary Approaches to Stop Hypertension index are associated with bone quantity but not with measures of bone quality in older Puerto Rican adult. The American Journal of Clinical Nutrition. https://doi.org/10.1016/j.ajcnut.2025.101129.
DOI: https://doi.org/10.1016/j.ajcnut.2025.101129

Interpretive Summary: Limitations exist in understanding how lifestyle influences the strength and overall quality of bone. This research in the Boston Puerto Rican Osteoporosis Study examined the associations of diet quality with several bone assessments including bone material strength index, trabecular bone score, and bone mineral density. We also assessed the influence of the gut microbiome and inflammation markers on these associations. Diet quality was assessed as Dietary Approaches to Stop Hypertension (DASH) scores and the gut microbiome was assessed using deep shotgun metagenomics. We found that the DASH diet score was predictive of bone mineral density at the hip in women but not in men; it was not predictive of bone material strength or trabecular bone score in the total study population. The significant association with BMD was not influenced by systemic inflammation. One gut microbial species that was negatively associated with sodium was predictive of BMD. We conclude that diet predicted bone mineral density but did not predict measures of bone quality.

Technical Abstract: Background: Limitations exist in understanding the mechanistic underpinnings linking lifestyle to osteoporosis due, in part, to the inability of bone mineral density to fully explain incidence of fracture. Data are needed relating lifestyle factors to bone quality. Objective: This study examined the associations of diet quality with bone material strength index (BMSi), trabecular bone score (TBS), and bone mineral density (BMD); tested for the partial mediating influence of the gut microbiome; and tested for moderation by inflammatory cytokines in the Boston Puerto Rican Osteoporosis Study. Methods: This cross-sectional analysis included dietary intake assessed with a culturally tailored food-frequency questionnaire. Dietary Approaches to Stop Hypertension (DASH) scores were created to assess diet quality. BMSi was measured using micro indentation with the Osteoprobe, BMD (grams per centimeter squared) was measured using dual-energy X-ray absorptiometry (DXA), and TBS was calculated from DXA scans. The gut microbiome was assessed using deep shotgun metagenomics. Multivariable linear regression related DASH with each bone outcome. Microbiome Multivariate Association with Linear Models (MaAsLin2) identified diet-related gut microbial parameters. Mediation analysis was performed to evaluate direct and indirect associations between diet and bone. Three machine learning algorithms were tested to identify the important predictors of each bone outcome, and the Shapley Additive exPlanations (SHAP) framework was used to interpret these results. Results: The DASH score was predictive of BMD at the femoral neck and total hip in women (B = 0.005 + - 0.002, P < 0.01 and 0.006 + - 0.002, P < 0.01, respectively), but not of BMD in men, nor of BMSi or TBS in the total sample. The significant association with BMD was not moderated by systemic inflammation. Several gut microbial species were associated with DASH, including Lachnospira eligens which was positively related to BMD; of those that were negatively associated with sodium, only Enterocloster aldensis, was predictive of BMD. No microbial functional pathways were associated with DASH. Person-first features, including the inflammation score and serum 25(OH)D were related to BMD. Conclusions: Diet predicted bone quantity, but did not predict measures of bone quality. Select diet-related gut microbial species along with known bone covariates, an inflammation score, and laxative use were also predictive of BMD at the hip.