Sex-specific associations of adiposity with cardiometabolic traits in the UK: A multi–life stage cohort study with repeat metabolomics

Background Sex differences in cardiometabolic disease risk are commonly observed across the life course but are poorly understood and may be due to different associations of adiposity with cardiometabolic risk in females and males. We examined whether adiposity is differently associated with cardiometabolic trait levels in females and males at 3 different life stages. Methods and findings Data were from 2 generations (offspring, Generation 1 [G1] born in 1991/1992 and their parents, Generation 0 [G0]) of a United Kingdom population-based birth cohort study, the Avon Longitudinal Study of Parents and Children (ALSPAC). Follow-up continues on the cohort; data up to 25 y after recruitment to the study are included in this analysis. Body mass index (BMI) and total fat mass from dual-energy X-ray absorptiometry (DXA) were measured at mean age 9 y, 15 y, and 18 y in G1. Waist circumference was measured at 9 y and 15 y in G1. Concentrations of 148 cardiometabolic traits quantified using nuclear magnetic resonance spectroscopy were measured at 15 y, 18 y, and 25 y in G1. In G0, all 3 adiposity measures and the same 148 traits were available at 50 y. Using linear regression models, sex-specific associations of adiposity measures at each time point (9 y, 15 y, and 18 y) with cardiometabolic traits 3 to 6 y later were examined in G1. In G0, sex-specific associations of adiposity measures and cardiometabolic traits were examined cross-sectionally at 50 y. A total of 3,081 G1 and 4,887 G0 participants contributed to analyses. BMI was more strongly associated with key atherogenic traits in males compared with females at younger ages (15 y to 25 y), and associations were more similar between the sexes or stronger in females at 50 y, particularly for apolipoprotein B–containing lipoprotein particles and lipid concentrations. For example, a 1 standard deviation (SD) (3.8 kg/m2) higher BMI at 18 y was associated with 0.36 SD (95% confidence interval [CI] = 0.20, 0.52) higher concentrations of extremely large very-low-density lipoprotein (VLDL) particles at 25 y in males compared with 0.15 SD (95% CI = 0.09, 0.21) in females, P value for sex difference = 0.02. By contrast, at 50 y, a 1 SD (4.8 kg/m2) higher BMI was associated with 0.33 SD (95% CI = 0.25, 0.42) and 0.30 SD (95% CI = 0.26, 0.33) higher concentrations of extremely large VLDL particles in males and females, respectively, P value for sex difference = 0.42. Sex-specific associations of DXA-measured fat mass and waist circumference with cardiometabolic traits were similar to findings for BMI and cardiometabolic traits at each age. The main limitation of this work is its observational nature, and replication in independent cohorts using methods that can infer causality is required. Conclusions The results of this study suggest that associations of adiposity with adverse cardiometabolic risk begin earlier in the life course among males compared with females and are stronger until midlife, particularly for key atherogenic lipids. Adolescent and young adult males may therefore be high priority targets for obesity prevention efforts.

Population study of the gut microbiome: associations with diet, lifestyle, and cardiometabolic disease

Background The human gut harbors trillions of microbes that play dynamic roles in health. While the microbiome contributes to many cardiometabolic traits by modulating host inflammation and metabolism, there is an incomplete understanding regarding the extent that and mechanisms by which individual microbes impact risk and development of cardiovascular disease (CVD). The Framingham Heart Study (FHS) is a multi-generational observational study following participants over decades to identify risk factors for CVD by correlating genetic and phenotypic factors with clinical outcomes. As a large-scale population-based cohort with extensive clinical phenotyping, FHS provides a rich landscape to explore the relationships between the gut microbiome and cardiometabolic traits. Methods We performed 16S rRNA gene sequencing on stool from 1423 participants of the FHS Generation 3, OMNI2, and New Offspring Spouse cohorts. Data processing and taxonomic assignment were performed with the 16S bioBakery workflow using the UPARSE pipeline. We conducted statistical analyses to investigate trends in overall microbiome composition and diversity in relation to disease states and systematically examined taxonomic associations with a variety of clinical traits, disease phenotypes, clinical blood markers, and medications. Results We demonstrate that overall microbial diversity decreases with increasing 10-year CVD risk and body mass index measures. We link lifestyle factors, especially diet and exercise, to microbial diversity. Our association analyses reveal both known and unreported microbial associations with CVD and diabetes, related prescription medications, as well as many anthropometric and blood test measurements. In particular, we observe a set of microbial species that demonstrate significant associations with CVD risk, metabolic syndrome, and type 2 diabetes as well as a number of shared associations between microbial species and cardiometabolic subphenotypes. Conclusions The identification of significant microbial taxa associated with prevalent CVD and diabetes, as well as risk for developing CVD, adds to increasing evidence that the microbiome may contribute to CVD pathogenesis. Our findings support new hypothesis generation around shared microbe-mediated mechanisms that influence metabolic syndrome, diabetes, and CVD risk. Further investigation of the gut microbiomes of CVD patients in a targeted manner may elucidate microbial mechanisms with diagnostic and therapeutic implications.

Conventional and bi-directional genetic evidence on resting heart rate and cardiometabolic traits

Objectives To examine the direction, strength and causality of the associations of resting heart rate (RHR) with cardiometabolic traits. Methods We assessed the strength of associations between measured RHR and cardiometabolic traits in 506,211 and 372,452 participants from China Kadoorie Biobank (CKB) and UK Biobank (UKB). Mendelian randomization (MR) analyses were used to make causal inferences in 99,228 and 371,508 participants from CKB and UKB, respectively. Results We identified significant, directionally-concordant observational associations between RHR and higher total cholesterol, triglycerides (TG), low-density lipoprotein, C-reactive protein (CRP), glucose, body mass index, waist-hip ratio (WHR), systolic blood pressure (SBP) and diastolic blood pressure (DBP) after the Bonferroni correction. MR analyses showed that 10 beat/min higher genetically-predicted RHR were trans-ethnically associated with a higher DBP (beta 2.059 [95%CI 1.544, 2.574] mmHg in CKB; 2.037 [1.845, 2.229] mmHg in UKB), higher CRP (0.180 [0.057, 0.303] log mg/L in CKB; 0.154 [0.134, 0.174] log mg/L in UKB), higher TG (0.052 [-0.009, 0.113] log mmol/L in CKB; 0.020 [0.010, 0.030] log mmol/L in UKB) and higher WHR (0.218 [-0.033, 0.469] % in CKB; 0.225 [0.111, 0.339] % in UKB). In the opposite direction, higher genetically-predicted SBP, TG, glucose, WHR and lower high-density lipoprotein were associated with elevated RHR. Conclusion Our large-scale analyses provide causal evidence between RHR and cardiometabolic traits, highlighting the importance of monitoring heat rate as a means of alleviating the adverse effect of metabolic disorders.

Body Fat Distribution, Cardiometabolic Traits, and Risk of Major Lower-Extremity Arterial Disease in Postmenopausal Women

<b>OBJECTIVE </b> <p>To assess the relationship between body fat distribution and incident lower-extremity arterial disease (LEAD). </p> <p><b>RESEARCH DESIGN AND METHODS </b></p> <p>We included 155,925 postmenopausal women with anthropometric measures from the Women’s Health Initiative who had no known LEAD at recruitment. A subset of 10,894 participants had body composition data quantified by dual energy X-ray absorptiometry (DXA). Incident cases of symptomatic LEAD were ascertained and adjudicated via medical record review.</p> <p><b>RESULTS </b></p> <p>We identified 1152 incident LEAD cases during a median 18.8 years follow-up. After multivariable adjustment and mutual adjustment, waist and hip circumference was positively and inversely associated with risk of LEAD, respectively (both P-trend values <0.0001). In a subset (n = 22,561) where various cardiometabolic biomarkers were quantified, a similar positive association of waist circumference with risk of LEAD was eliminated after adjustment for diabetes and HOMA-IR (P-trend = 0.89), whereas hip circumference remained inversely associated with the risk after adjustment for major cardiometabolic traits (P-trend = 0.0031). In the DXA subset, higher trunk fat (P-trend = 0.0081) and higher leg fat (P-trend <0.0001) was associated with higher and lower risk of LEAD, respectively. Further adjustment for diabetes, dyslipidemia, and blood pressure diminished the association for trunk fat (P-trend = 0.49), yet the inverse association for leg fat persisted (P-trend = 0.0082).</p> <p><b>CONCLUSIONS</b></p> <p>Among US postmenopausal women, a positive association of upper-body fat with risk of LEAD appeared to be attributable to traditional risk factors especially insulin resistance. Lower-body fat was inversely associated with risk of LEAD beyond known risk factors. </p>

Body Fat Distribution, Cardiometabolic Traits, and Risk of Major Lower-Extremity Arterial Disease in Postmenopausal Women

<b>OBJECTIVE </b> <p>To assess the relationship between body fat distribution and incident lower-extremity arterial disease (LEAD). </p> <p><b>RESEARCH DESIGN AND METHODS </b></p> <p>We included 155,925 postmenopausal women with anthropometric measures from the Women’s Health Initiative who had no known LEAD at recruitment. A subset of 10,894 participants had body composition data quantified by dual energy X-ray absorptiometry (DXA). Incident cases of symptomatic LEAD were ascertained and adjudicated via medical record review.</p> <p><b>RESULTS </b></p> <p>We identified 1152 incident LEAD cases during a median 18.8 years follow-up. After multivariable adjustment and mutual adjustment, waist and hip circumference was positively and inversely associated with risk of LEAD, respectively (both P-trend values <0.0001). In a subset (n = 22,561) where various cardiometabolic biomarkers were quantified, a similar positive association of waist circumference with risk of LEAD was eliminated after adjustment for diabetes and HOMA-IR (P-trend = 0.89), whereas hip circumference remained inversely associated with the risk after adjustment for major cardiometabolic traits (P-trend = 0.0031). In the DXA subset, higher trunk fat (P-trend = 0.0081) and higher leg fat (P-trend <0.0001) was associated with higher and lower risk of LEAD, respectively. Further adjustment for diabetes, dyslipidemia, and blood pressure diminished the association for trunk fat (P-trend = 0.49), yet the inverse association for leg fat persisted (P-trend = 0.0082).</p> <p><b>CONCLUSIONS</b></p> <p>Among US postmenopausal women, a positive association of upper-body fat with risk of LEAD appeared to be attributable to traditional risk factors especially insulin resistance. Lower-body fat was inversely associated with risk of LEAD beyond known risk factors. </p>

Parent-of-origin effects in the life-course evolution of cardio-metabolic traits

Objective: Human traits are heritable, and some of these including metabolic and lipid phenotypes show preferential parental transmissions, or parent-of-origin effects. These have been mostly studied in populations comprising adults. We aimed to investigate heritability and parent-of-origin effects on cardiometabolic and anthropometric traits in a birth-cohort with serial measurements to assess if these effects manifested at an early age. Research design and methods: We investigated heritability and parent-of-origin effects on cardiometabolic and anthropometric traits in the Pune Maternal Nutrition Study (PMNS) wherein offspring and parents were studied from birth and followed-up for 18 years. Heritability was estimated by calculating association between mid-parental phenotypes and offspring. Maternal and paternal effects on offspring phenotype were modelled by regression after adjusting for age, sex and BMI. Parent-of-origin effects were calculated by the difference between maternal and paternal effects. Results: Anthropomorphic traits and cardiometabolic traits were robustly heritable. Parent-of-origin effects were observed for glycemic traits at both 6- and 12-years, with a paternal effect at 6-years which transitioned to a maternal effect at 12-years. For insulin and HOMA-S, a negative maternal effect transitioned to a positive one at 12-years. For HOMA-B, a paternal effect at 6- years transitioned to a maternal one at 12-years. Lipid traits consistently showed stronger maternal influence while anthropometric traits did not show any parental biases. Conclusions: Our study highlights that parental programming of cardiometabolic traits is evident from early childhood and can transition during puberty. Further studies are needed to determine the mechanisms of underlying such effects.

Genome-wide analyses of multiple obesity-related cytokines and hormones informs biology of cardiometabolic traits

Background A complex set of perturbations occur in cytokines and hormones in the etiopathogenesis of obesity and related cardiometabolic conditions such as type 2 diabetes (T2D). Evidence for the genetic regulation of these cytokines and hormones is limited, particularly in African-ancestry populations. In order to improve our understanding of the biology of cardiometabolic traits, we investigated the genetic architecture of a large panel of obesity- related cytokines and hormones among Africans with replication analyses in African Americans. Methods We performed genome-wide association studies (GWAS) in 4432 continental Africans, enrolled from Ghana, Kenya, and Nigeria as part of the Africa America Diabetes Mellitus (AADM) study, for 13 obesity-related cytokines and hormones, including adipsin, glucose-dependent insulinotropic peptide (GIP), glucagon-like peptide-1 (GLP-1), interleukin-1 receptor antagonist (IL1-RA), interleukin-6 (IL-6), interleukin-10 (IL-10), leptin, plasminogen activator inhibitor-1 (PAI-1), resistin, visfatin, insulin, glucagon, and ghrelin. Exact and local replication analyses were conducted in African Americans (n = 7990). The effects of sex, body mass index (BMI), and T2D on results were investigated through stratified analyses. Results GWAS identified 39 significant (P value < 5 × 10−8) loci across all 13 traits. Notably, 14 loci were African-ancestry specific. In this first GWAS for adipsin and ghrelin, we detected 13 and 4 genome-wide significant loci respectively. Stratified analyses by sex, BMI, and T2D showed a strong effect of these variables on detected loci. Eight novel loci were successfully replicated: adipsin (3), GIP (1), GLP-1 (1), and insulin (3). Annotation of these loci revealed promising links between these adipocytokines and cardiometabolic outcomes as illustrated by rs201751833 for adipsin and blood pressure and locus rs759790 for insulin level and T2D in lean individuals. Conclusions Our study identified genetic variants underlying variation in multiple adipocytokines, including the first loci for adipsin and ghrelin. We identified population differences in variants associated with adipocytokines and highlight the importance of stratification for discovery of loci. The high number of African-specific loci detected emphasizes the need for GWAS in African-ancestry populations, as these loci could not have been detected in other populations. Overall, our work contributes to the understanding of the biology linking adipocytokines to cardiometabolic traits.