Fenglei Wang1, Megu Y Baden1, Marta Guasch-Ferré1,2, Clemens Wittenbecher1, Jun Li1, Yanping Li1, Yi Wan1, Shilpa N Bhupathiraju1,2, Deirdre K Tobias1,3, Clary B Clish4, Lorelei A Mucci5, A Heather Eliassen2,5, Karen H Costenbader6, Elizabeth W Karlson6, Alberto Ascherio1,2,5, Eric B Rimm1,2,5, JoAnn E Manson2,3,5,7, Liming Liang5,8, Frank B Hu9,10,11. 1. Department of Nutrition, Harvard T.H. Chan School of Public Health, Boston, MA, USA. 2. Channing Division of Network Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, MA, USA. 3. Division of Preventive Medicine, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, MA, USA. 4. Metabolomics Platform, Broad Institute of MIT and Harvard, Cambridge, MA, USA. 5. Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, MA, USA. 6. Division of Rheumatology, Inflammation and Immunity, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, MA, USA. 7. Mary Horrigan Connors Center for Women's Health and Gender Biology, Brigham and Women's Hospital and Harvard Medical School, Boston, MA, USA. 8. Department of Biostatistics, Harvard T.H. Chan School of Public Health, Boston, MA, USA. 9. Department of Nutrition, Harvard T.H. Chan School of Public Health, Boston, MA, USA. fhu@hsph.harvard.edu. 10. Channing Division of Network Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, MA, USA. fhu@hsph.harvard.edu. 11. Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, MA, USA. fhu@hsph.harvard.edu.
Abstract
AIMS/HYPOTHESIS: Plant-based diets, especially when rich in healthy plant foods, have been associated with a lower risk of type 2 diabetes. However, whether plasma metabolite profiles related to plant-based diets reflect this association was unknown. The aim of this study was to identify the plasma metabolite profiles related to plant-based diets, and to evaluate the associations between the identified metabolite profiles and the risk of type 2 diabetes. METHODS: Within three prospective cohorts (Nurses' Health Study, Nurses' Health Study II and Health Professionals Follow-up Study), we measured plasma metabolites from 10,684 participants using high-throughput LC MS. Adherence to plant-based diets was assessed by three indices derived from the food frequency questionnaire: an overall Plant-based Diet Index (PDI), a Healthy Plant-based Diet Index (hPDI), and an Unhealthy Plant-based Diet Index (uPDI). Multi-metabolite profiles related to plant-based diet were identified using elastic net regression with a training/testing approach. The prospective associations between metabolite profiles and incident type 2 diabetes were evaluated using multivariable Cox proportional hazards regression. Metabolites potentially mediating the association between plant-based diets and type 2 diabetes risk were further identified. RESULTS: We identified multi-metabolite profiles comprising 55 metabolites for PDI, 93 metabolites for hPDI and 75 metabolites for uPDI. Metabolite profile scores based on the identified metabolite profiles were correlated with the corresponding diet index (Pearson r = 0.33-0.35 for PDI, 0.41-0.45 for hPDI, and 0.37-0.38 for uPDI, all p<0.001). Metabolite profile scores of PDI (HR per 1 SD higher = 0.81 [95% CI 0.75, 0.88]) and hPDI (HR per 1 SD higher = 0.77 [95% CI 0.71, 0.84]) showed an inverse association with incident type 2 diabetes, whereas the metabolite profile score for uPDI was not associated with the risk. Mutual adjustment for metabolites selected in the metabolite profiles, including trigonelline, hippurate, isoleucine and a subset of triacylglycerols, attenuated the associations of diet indices PDI and hPDI with lower type 2 diabetes risk. The explainable proportion of PDI/hPDI-related diabetes risk by these metabolites ranged between 8.5% and 37.2% (all p<0.05). CONCLUSIONS/ INTERPRETATION: Plasma metabolite profiles related to plant-based diets, especially a healthy plant-based diet, were associated with a lower risk of type 2 diabetes among a generally healthy population. Our findings support the beneficial role of healthy plant-based diets in diabetes prevention and provide new insights for future investigation.
AIMS/HYPOTHESIS: Plant-based diets, especially when rich in healthy plant foods, have been associated with a lower risk of type 2 diabetes. However, whether plasma metabolite profiles related to plant-based diets reflect this association was unknown. The aim of this study was to identify the plasma metabolite profiles related to plant-based diets, and to evaluate the associations between the identified metabolite profiles and the risk of type 2 diabetes. METHODS: Within three prospective cohorts (Nurses' Health Study, Nurses' Health Study II and Health Professionals Follow-up Study), we measured plasma metabolites from 10,684 participants using high-throughput LC MS. Adherence to plant-based diets was assessed by three indices derived from the food frequency questionnaire: an overall Plant-based Diet Index (PDI), a Healthy Plant-based Diet Index (hPDI), and an Unhealthy Plant-based Diet Index (uPDI). Multi-metabolite profiles related to plant-based diet were identified using elastic net regression with a training/testing approach. The prospective associations between metabolite profiles and incident type 2 diabetes were evaluated using multivariable Cox proportional hazards regression. Metabolites potentially mediating the association between plant-based diets and type 2 diabetes risk were further identified. RESULTS: We identified multi-metabolite profiles comprising 55 metabolites for PDI, 93 metabolites for hPDI and 75 metabolites for uPDI. Metabolite profile scores based on the identified metabolite profiles were correlated with the corresponding diet index (Pearson r = 0.33-0.35 for PDI, 0.41-0.45 for hPDI, and 0.37-0.38 for uPDI, all p<0.001). Metabolite profile scores of PDI (HR per 1 SD higher = 0.81 [95% CI 0.75, 0.88]) and hPDI (HR per 1 SD higher = 0.77 [95% CI 0.71, 0.84]) showed an inverse association with incident type 2 diabetes, whereas the metabolite profile score for uPDI was not associated with the risk. Mutual adjustment for metabolites selected in the metabolite profiles, including trigonelline, hippurate, isoleucine and a subset of triacylglycerols, attenuated the associations of diet indices PDI and hPDI with lower type 2 diabetes risk. The explainable proportion of PDI/hPDI-related diabetes risk by these metabolites ranged between 8.5% and 37.2% (all p<0.05). CONCLUSIONS/ INTERPRETATION: Plasma metabolite profiles related to plant-based diets, especially a healthy plant-based diet, were associated with a lower risk of type 2 diabetes among a generally healthy population. Our findings support the beneficial role of healthy plant-based diets in diabetes prevention and provide new insights for future investigation.
Authors: Paulette D Chandler; Raji Balasubramanian; Nina Paynter; Franco Giulianini; Teresa Fung; Lesley F Tinker; Linda Snetselaar; Simin Liu; Charles Eaton; Deirdre K Tobias; Fred K Tabung; JoAnn E Manson; Edward L Giovannucci; Clary Clish; Kathryn M Rexrode Journal: Am J Clin Nutr Date: 2020-08-01 Impact factor: 7.045
Authors: Mary C Playdon; Steven C Moore; Andriy Derkach; Jill Reedy; Amy F Subar; Joshua N Sampson; Demetrius Albanes; Fangyi Gu; Jukka Kontto; Camille Lassale; Linda M Liao; Satu Männistö; Alison M Mondul; Stephanie J Weinstein; Melinda L Irwin; Susan T Mayne; Rachael Stolzenberg-Solomon Journal: Am J Clin Nutr Date: 2016-12-28 Impact factor: 7.045
Authors: L Shi; C Brunius; I Johansson; I A Bergdahl; O Rolandsson; B van Guelpen; A Winkvist; K Hanhineva; R Landberg Journal: J Intern Med Date: 2019-12-09 Impact factor: 8.989
Authors: Casey M Rebholz; Alice H Lichtenstein; Zihe Zheng; Lawrence J Appel; Josef Coresh Journal: Am J Clin Nutr Date: 2018-08-01 Impact factor: 7.045
Authors: Ambika Satija; Shilpa N Bhupathiraju; Eric B Rimm; Donna Spiegelman; Stephanie E Chiuve; Lea Borgi; Walter C Willett; JoAnn E Manson; Qi Sun; Frank B Hu Journal: PLoS Med Date: 2016-06-14 Impact factor: 11.069