Quality control and removal of technical variation of NMR metabolic biomarker data in ~120,000 UK Biobank participants

A Machine Learning Analysis of Big Metabolomics Data for Classifying Depression: Model Development and Validation

BACKGROUND

Many metabolomics studies of depression have been performed, but these have been limited by their scale. A comprehensive in silico analysis of global metabolite levels in large populations could provide robust insights into the pathological mechanisms underlying depression and candidate clinical biomarkers.

METHODS

Depression-associated metabolomics was studied in 2 datasets from the UK Biobank database: participants with lifetime depression (N = 123,459) and participants with current depression (N = 94,921). The Whitehall II cohort (N = 4744) was used for external validation. CatBoost machine learning was used for modeling, and Shapley additive explanations were used to interpret the model. Fivefold cross-validation was used to validate model performance, training the model on 3 of the 5 sets with the remaining 2 sets for validation and testing, respectively. Diagnostic performance was assessed using the area under the receiver operating characteristic curve.

RESULTS

In the lifetime depression and current depression datasets and sex-specific analyses, 24 significantly associated metabolic biomarkers were identified, 12 of which overlapped in the 2 datasets. The addition of metabolic features slightly improved the performance of a diagnostic model using traditional (nonmetabolomics) risk factors alone (lifetime depression: area under the curve 0.655 vs. 0.658 with metabolomics; current depression: area under the curve 0.711 vs. 0.716 with metabolomics).

CONCLUSIONS

The machine learning model identified 24 metabolic biomarkers associated with depression. If validated, metabolic biomarkers may have future clinical applications as supplementary information to guide early and population-based depression detection.

An interactive atlas of genomic, proteomic, and metabolomic biomarkers promotes the potential of proteins to predict complex diseases

Multiomics analyses have identified multiple potential biomarkers of the incidence and prevalence of complex diseases. However, it is not known which type of biomarker is optimal for clinical purposes. Here, we make a systematic comparison of 90 million genetic variants, 1453 proteins, and 325 metabolites from 500,000 individuals with complex diseases from the UK Biobank. A machine learning pipeline consisting of data cleaning, data imputation, feature selection, and model training using cross-validation and comparison of the results on holdout test sets showed that proteins were most predictive, followed by metabolites, and genetic variants. Only five proteins per disease resulted in median (min-max) areas under the receiver operating characteristic curves for incidence of 0.79 (0.65-0.86) and 0.84 (0.70-0.91) for prevalence. In summary, our work suggests the potential of predicting complex diseases based on a limited number of proteins. We provide an interactive atlas (macd.shinyapps.io/ShinyApp/) to find genomic, proteomic, or metabolomic biomarkers for different complex diseases.

Interplay Between Plasma Glycine and Branched-Chain Amino Acids Contributes to the Development of Hypertension and Coronary Heart Disease

BACKGROUND

Higher levels of plasma glycine are linked to a reduced risk, while increased levels of total branched-chain amino acids (tBCAAs) are associated with a higher risk of essential hypertension and coronary heart disease (CHD). As these metabolic components are interconnected, analyzing the tBCAAs/glycine ratio may help to understand their interplay in the pathogenesis of cardiovascular disease.

METHODS

The Cox regression approach was combined with the development of novel genetic tools for assessments of associations between plasma metabolomic data (glycine, tBCAAs, and tBCAAs/glycine ratio) from the UK Biobank and the development of hypertension and CHD. Genome-wide association study was performed on 186 523 White UK Biobank participants to identify new independent genetic instruments for the 2-sample Mendelian randomization analyses. P-gain statistic >10 identified instruments associated with tBCAAs/glycine ratio significantly stronger compared with individual amino acids. Outcomes of genome-wide association study on hypertension and CHD were derived from the UK Biobank (nonoverlapping sample), FinnGen, and CARDIoGRAMplusC4D.

RESULTS

The tBCAAs/glycine ratio was prospectively associated with a higher risk of developing hypertension and CHD (hazard ratio quintile Q5 versus Q1, 1.196 [95% CI, 1.109-1.289] and 1.226 [95% CI, 1.160-1.296], respectively). Mendelian randomization analysis demonstrated that tBCAAs/glycine ratio (P-gain >10) was a risk factor for hypertension (meta-analyzed inverse-variance weighted causal estimate 0.45 log odds ratio/SD (95% CI, 0.26-0.64) and CHD (0.48 [95% CI, 0.29-0.67]) with an absolute effect significantly larger compared with the effect of glycine (-0.06 [95% CI, -0.1 to -0.03] and -0.08 [95% CI, -0.11 to -0.05], respectively) or tBCAAs (0.22 [95% CI, 0.09-0.34] and 0.12 [95% CI, 0.01-0.24], respectively).

CONCLUSIONS

The total BCAAs/glycine ratio is a key element of the metabolic signature contributing to hypertension and CHD, which may reflect biological pathways shared by glycine and tBCAAs.

A genome-first approach to variants in MLXIPL and their association with hepatic steatosis and plasma lipids

BACKGROUND

Common variants of the max-like protein X (MLX)-interacting protein-like (MLXIPL) gene, encoding the transcription factor carbohydrate-responsive element-binding protein, have been shown to be associated with plasma triglyceride levels. However, the role of these variants in steatotic liver disease (SLD) is unclear.

METHODS

We used a genome-first approach to analyze a variety of metabolic phenotypes and clinical outcomes associated with a common missense variant in MLXIPL, Gln241His, in 2 large biobanks: the UK Biobank and the Penn Medicine Biobank.

RESULTS

Carriers of MLXIPL Gln241His were associated with significantly lower serum levels of triglycerides, apolipoprotein-B, gamma-glutamyl transferase, and alkaline phosphatase. Additionally, MLXIPL Gln241His carriers were associated with significantly higher serum levels of HDL cholesterol and alanine aminotransferase. Carriers homozygous for MLXIPL Gln241His showed a higher risk of SLD in 2 unrelated cohorts. Carriers of MLXIPL Gln241His were especially more likely to be diagnosed with SLD if they were female, obese, and/or also carried the PNPLA3 I148M variant. Furthermore, the heterozygous carriage of MLXIPL Gln241His was associated with significantly higher all-cause, liver-related, and cardiovascular mortality rates. Nuclear magnetic resonance metabolomics data indicated that carriage of MLXIPL Gln241His was significantly associated with lower serum levels of VLDL and increased serum levels of HDL cholesterol.

CONCLUSIONS

Analyses of the MLXIPL Gln241His polymorphism showed a significant association with a higher risk of SLD diagnosis and elevated serum alanine aminotransferase as well as significantly lower serum triglycerides and apolipoprotein-B levels. MLXIPL might, therefore, be a potential pharmacological target for the treatment of SLD and hyperlipidemia, notably for patients at risk. More mechanistic studies are needed to better understand the role of MLXIPL Gln241His on lipid metabolism and steatosis development.

Remnant Cholesterol, Not LDL Cholesterol, Explains Peripheral Artery Disease Risk Conferred by apoB: A Cohort Study

BACKGROUND

Elevated apoB-containing lipoproteins (=remnants+LDLs [low-density lipoproteins]) are a major risk factor for atherosclerotic cardiovascular disease, including peripheral artery disease (PAD) and myocardial infarction. We tested the hypothesis that remnants and LDL both explain part of the increased risk of PAD conferred by elevated apoB-containing lipoproteins. For comparison, we also studied the risk of chronic limb-threatening ischemia and myocardial infarction.

METHODS

apoB, remnant cholesterol, and LDL cholesterol were measured in 93 461 individuals without statin use at baseline from the Copenhagen General Population Study (2003-2015). During up to 15 years of follow-up, 1207 had PAD, 552 had chronic limb-threatening ischemia, and 2022 had myocardial infarction in the Danish National Patient Registry. Remnant and LDL cholesterol were calculated from a standard lipid profile. Remnant and LDL particle counts were additionally measured with nuclear magnetic resonance spectroscopy in 25 347 of the individuals. Results were replicated in 302 167 individuals without statin use from the UK Biobank (2004-2010).

RESULTS

In the Copenhagen General Population Study, multivariable adjusted hazard ratios for risk of PAD per 1 mmol/L (39 mg/dL) increment in remnant and LDL cholesterol were 1.9 (95% CI, 1.5-2.4) and 1.1 (95% CI, 1.0-1.2), respectively; corresponding results in the UK Biobank were 1.7 (95% CI, 1.4-2.1) and 0.9 (95% CI, 0.9-1.0), respectively. In the association from elevated apoB to increased risk of PAD, remnant and LDL cholesterol explained 73% (32%-100%) and 8% (0%-46%), respectively; corresponding results were 63% (30%-100%) and 0% (0%-33%) for risk of chronic limb-threatening ischemia and 41% (27%-55%) and 54% (38%-70%) for risk of myocardial infarction; results for remnant and LDL particle counts corroborated these findings.

CONCLUSIONS

PAD risk conferred by elevated apoB-containing lipoproteins was explained mainly by elevated remnants, while myocardial infarction risk was explained by both elevated remnants and LDL.

Genome-wide characterization of circulating metabolic biomarkers

Genome-wide association analyses using high-throughput metabolomics platforms have led to novel insights into the biology of human metabolism1-7. This detailed knowledge of the genetic determinants of systemic metabolism has been pivotal for uncovering how genetic pathways influence biological mechanisms and complex diseases8-11. Here we present a genome-wide association study for 233 circulating metabolic traits quantified by nuclear magnetic resonance spectroscopy in up to 136,016 participants from 33 cohorts. We identify more than 400 independent loci and assign probable causal genes at two-thirds of these using manual curation of plausible biological candidates. We highlight the importance of sample and participant characteristics that can have significant effects on genetic associations. We use detailed metabolic profiling of lipoprotein- and lipid-associated variants to better characterize how known lipid loci and novel loci affect lipoprotein metabolism at a granular level. We demonstrate the translational utility of comprehensively phenotyped molecular data, characterizing the metabolic associations of intrahepatic cholestasis of pregnancy. Finally, we observe substantial genetic pleiotropy for multiple metabolic pathways and illustrate the importance of careful instrument selection in Mendelian randomization analysis, revealing a putative causal relationship between acetone and hypertension. Our publicly available results provide a foundational resource for the community to examine the role of metabolism across diverse diseases.

SGLT2 inhibition, high-density lipoprotein, and kidney function: a mendelian randomization study

BACKGROUND

Sodium-glucose cotransporter 2 (SGLT2) inhibition is recognized for its evident renoprotective benefits in diabetic renal disease. Recent data suggest that SGLT2 inhibition also slows down kidney disease progression and reduces the risk of acute kidney injury, regardless of whether the patient has diabetes or not, but the mechanism behind these observed effects remains elusive. The objective of this study is to utilize a mendelian randomization (MR) methodology to comprehensively examine the influence of metabolites in circulation regarding the impact of SGLT2 inhibition on kidney function.

METHODS

We used a MR study to obtain associations between genetic proxies for SGLT2 inhibition and kidney function. We retrieved the most recent and comprehensive summary statistics from genome-wide association studies (GWAS) that have been previously published and involved kidney function parameters such as estimated glomerular filtration rate (eGFR), urine albumin-to-creatinine ratio (UACR), and albuminuria. Additionally, we included blood metabolite data from 249 biomarkers in the UK Biobank for a more comprehensive analysis. We performed MR analyses to explore the causal relationships between SGLT2 inhibition and kidney function and two-step MR to discover potential mediating metabolites.

RESULTS

The study found that a decrease in HbA1c levels by one standard deviation, which is genetically expected to result in SGLT2 inhibition, was linked to a decreased likelihood of developing type 2 diabetes mellitus (T2DM) (odds ratio [OR] = 0.55 [95% CI 0.35, 0.85], P = 0.007). Meanwhile, SGLT2 inhibition also protects eGFR (β = 0.05 [95% CI 0.03, 0.08], P = 2.45 × 10- 5) and decreased UACR (-0.18 [95% CI -0.33, -0.02], P = 0.025) and albuminuria (-1.07 [95% CI -1.58, -0.57], P = 3.60 × 10- 5). Furthermore, the study found that of the 249 metabolites present in the blood, only one metabolite, specifically the concentration of small high-density lipoprotein (HDL) particles, was significantly correlated with both SGLT2 inhibition and kidney function. This metabolite was found to play a crucial role in mediating the improvement of renal function through the use of SGLT2 inhibition (β = 0.01 [95% CI 0.005, 0.018], P = 0.001), with a mediated proportion of 13.33% (95% CI [5.71%, 26.67%], P = 0.020).

CONCLUSIONS

The findings of this investigation provide evidence in favor of a genetically anticipated biological linkage between the inhibition of SGLT2, the presence of circulating metabolites, and renal function. The findings demonstrate that the protective effect of SGLT2 inhibition on renal function is mostly mediated by HDL particle concentrations in circulating metabolites. These results offer significant theoretical support for both the preservation of renal function and a better comprehension of the mechanisms underlying SGLT2 inhibition.

An interactive atlas of genomic, proteomic, and metabolomic biomarkers promotes the potential of proteins to predict complex diseases

Multiomics analyses have identified multiple potential biomarkers of the incidence and prevalence of complex diseases. However, it is not known which type of biomarker is optimal for clinical purposes. Here, we make a systematic comparison of 90 million genetic variants, 1,453 proteins, and 325 metabolites from 500,000 individuals with complex diseases from the UK Biobank. A machine learning pipeline consisting of data cleaning, data imputation, feature selection, and model training using cross-validation and comparison of the results on holdout test sets showed that proteins were most predictive, followed by metabolites, and genetic variants. Only five proteins per disease resulted in median (min-max) areas under the receiver operating characteristic curves for incidence of 0.79 (0.65-0.86) and 0.84 (0.70-0.91) for prevalence. In summary, our work suggests the potential of predicting complex diseases based on a limited number of proteins. We provide an interactive atlas (macd.shinyapps.io/ShinyApp/) to find genomic, proteomic, or metabolomic biomarkers for different complex diseases.

Sodium fluoride preserves blood metabolite integrity for biomarker discovery in large-scale, multi-site metabolomics investigations

Background: Metabolite profiling of blood by nuclear magnetic resonance (NMR) is invaluable to clinical biomarker discovery. To ensure robustness, biomarkers require validation in large cohorts and across multiple centres. However, collection procedures are known to impact on the stability of biofluids that may, in turn, degrade biomarker signals. We trialled three blood collection tubes with the aim of solving technical challenges due to preanalytical variation in blood metabolite levels that are common in cohort studies. Methods: We first investigated global NMR-based metabolite variability between biobanks, including the large-scale UK Biobank and TwinsUK biobank of the general UK population, and more targeted biobanks derived from multicentre clinical trials relating to inflammatory bowel disease. We then compared the blood metabolome of 12 healthy adult volunteers when collected into either sodium fluoride/potassium oxalate, lithium heparin, or serum blood tubes using different pre-processing parameters. Results: Preanalytical variation in the method of blood collection strongly influences metabolite composition within and between biobanks. This variability can largely be attributed to glucose and lactate. In the healthy control cohort, the fluoride oxalate collection tube prevented fluctuation in glucose and lactate levels for 24 hours at either 4 °C or room temperature (20 °C). Conclusions: Blood collection into a fluoride oxalate collection tube appears to preserve the blood metabolome with delayed processing up to 24 hours at 4 °C. This method may be considered as an alternative when rapid processing is not feasible.

Gut microbiota, circulating metabolites, and gallstone disease: a Mendelian randomization study

Background

The link between Gut microbiota (GM) and Gallstone disease (GSD) is well established, but it is not clear whether there is a causal relationship between the two associations.

Methods

We conducted bidirectional Mendelian randomization (MR) analyses, leveraging aggregated data from the Genome-Wide Association Study (GWAS) of GM and Circulating Metabolites. Our primary objective was to investigate the causal interplay between intestinal flora and GSD. Additionally, we performed mediational analyses, two-step MR, and multivariate MR to uncover the potential mediating effect of circulating metabolites in this relationship.

Result

Our study has revealed a causal relationship between GSD and six distinct bacterial groups. Genetically predicted Class Bacilli (Odds Ratio (OR): 0.901, 95% Confidence Interval (95% CI): 0.825-0.985; p = 0.021), Order Lactobacillales (OR: 0.895, 95% CI: 0.816-0.981; p = 0.017), and Genus Coprococcus 2 (OR: 0.884, 95% CI: 0.804-0.973; p = 0.011) were inversely associated with the risk of GSD. Conversely, the Genus Clostridiumsensustricto1 (OR: 1.158, 95% CI: 1.029-1.303; p = 0.015), Genus Coprococcus3 (OR: 1.166, 95% CI: 1.024-1.327; p = 0.020), and Genus Peptococcus (OR: 1.070, 95% CI: 1.017-1.125; p = 0.009) were positively associated with the risk of GSD. Moreover, our findings suggest that the positive influence of the Genus Peptococcus on GSD may be mediated through Omega-3 polyunsaturated fatty acids (PUFA).

Conclusion

This study reinforces the connection between the gut microbiome and the risk of GSD while also unveiling the mediating role of Omega-3 PUFA in the causal relationship between these factors.

Refinement of a published gene-physical activity interaction impacting HDL-cholesterol: role of sex and lipoprotein subfractions

Large-scale gene-environment interaction (GxE) discovery efforts often involve compromises in the definition of outcomes and choice of covariates for the sake of data harmonization and statistical power. Consequently, refinement of exposures, covariates, outcomes, and population subsets may be helpful to establish often-elusive replication and evaluate potential clinical utility. Here, we used additional datasets, an expanded set of statistical models, and interrogation of lipoprotein metabolism via nuclear magnetic resonance (NMR)-based lipoprotein subfractions to refine a previously discovered GxE modifying the relationship between physical activity (PA) and HDL-cholesterol (HDL-C). This GxE was originally identified by Kilpeläinen et al., with the strongest cohort-specific signal coming from the Women's Genome Health Study (WGHS). We thus explored this GxE further in the WGHS (N = 23,294), with follow-up in the UK Biobank (UKB; N = 281,380), and the Multi-Ethnic Study of Atherosclerosis (MESA; N = 4,587). Self-reported PA (MET-hrs/wk), genotypes at rs295849 (nearest gene: LHX1), and NMR metabolomics data were available in all three cohorts. As originally reported, minor allele carriers of rs295849 in WGHS had a stronger positive association between PA and HDL-C (pint = 0.002). When testing a range of NMR metabolites (primarily lipoprotein and lipid subfractions) to refine the HDL-C outcome, we found a stronger interaction effect on medium-sized HDL particle concentrations (M-HDL-P; pint = 1.0×10-4) than HDL-C. Meta-regression revealed a systematically larger interaction effect in cohorts from the original meta-analysis with a greater fraction of women (p = 0.018). In the UKB, GxE effects were stronger both in women and using M-HDL-P as the outcome. In MESA, the primary interaction for HDL-C showed nominal significance (pint = 0.013), but without clear differences by sex and with a greater magnitude using large, rather than medium, HDL-P as an outcome. Towards reconciling these observations, further exploration leveraging NMR platform-specific HDL subfraction diameter annotations revealed modest agreement across all cohorts in the interaction affecting medium-to-large particles. Taken together, our work provides additional insights into a specific known gene-PA interaction while illustrating the importance of phenotype and model refinement towards understanding and replicating GxEs.

Mendelian Randomization of Blood Metabolites Suggests Circulating Glutamine Protects Against Late-Onset Alzheimer's Disease

Background

Late-onset Alzheimer's disease (LOAD) represents a growing health burden. Previous studies suggest that blood metabolite levels influence risk of LOAD.

Objective

We used a genetics-based study design which may overcome limitations of other epidemiological studies to assess the influence of metabolite levels on LOAD risk.

Methods

We applied Mendelian randomization (MR) to evaluate bi-directional causal effects using summary statistics from the largest genome-wide association studies (GWAS) of 249 blood metabolites (n = 115,082) and GWAS of LOAD (ncase = 21,982, ncontrol = 41,944).

Results

MR analysis of metabolites as exposures revealed a negative association of genetically-predicted glutamine levels with LOAD (Odds Ratio (OR) = 0.83, 95% CI = 0.73, 0.92) that was consistent in multiple sensitivity analyses. We also identified a positive association of genetically-predicted free cholesterol levels in small LDL (OR = 1.79, 95% CI = 1.36, 2.22) on LOAD. Using genetically-predicted LOAD as the exposure, we identified associations with phospholipids to total lipids ratio in large LDL (OR = 0.96, 95% CI = 0.94, 0.98), but not with glutamine, suggesting that the relationship between glutamine and LOAD is unidirectional.

Conclusions

Our findings support previous evidence that higher circulating levels of glutamine may be a target for protection against LOAD.

The relationship of fatty acids to ischaemic heart disease and lifespan in men and women using Mendelian randomization

BACKGROUND

Observationally, polyunsaturated fatty acids (PUFAs) have health benefits compared with saturated fatty acids (SFAs); randomized controlled trials suggest fewer benefits. We used uni- and multi-variable Mendelian randomization to assess the association of major fatty acids and their sub-species with ischaemic heart disease (IHD) overall and sex-specifically and with lifespan sex-specifically, given differing lifespan by sex.

METHODS

We obtained strong (P <5x10-8), independent (r2<0.001) genetic predictors of fatty acids from genome-wide association studies (GWAS) in a random subset of 114 999 UK Biobank participants. We applied these genetic predictors to the Cardiogram IHD GWAS (cases = 60 801, controls = 123 504) and to the Finngen consortium GWAS (cases = 31 640, controls = 187 152) for replication and to the UK Biobank for sex-specific IHD and for lifespan based on parental attained age (fathers = 415 311, mothers = 412 937). We used sensitivity analysis and assessed sex differences where applicable.

RESULTS

PUFAs were associated with IHD [odds ratio 1.23, 95% confidence interval (CI) 1.05 to 1.44] and lifespan in men (-0.76 years, 95% CI -1.34 to -0.17) but not women (0.20, 95% CI -0.32 to 0.70). Findings were similar for omega-6 fatty acids and linoleic acid. Independent associations of SFAs, mono-unsaturated fatty acids or omega-3 fatty acids with IHD overall or lifespan in men and women were limited.

CONCLUSIONS

PUFAs, via specific subspecies, may contribute to disparities in lifespan by sex. Sex-specific dietary advice might be a start towards personalized public health and addressing inequities.

Ancestry-specific polygenic risk scores are risk enhancers for clinical cardiovascular disease assessments

Clinical implementation of new prediction models requires evaluation of their utility in a broad range of intended use populations. Here we develop and validate ancestry-specific Polygenic Risk Scores (PRSs) for Coronary Artery Disease (CAD) using 29,389 individuals from diverse cohorts and genetic ancestry groups. The CAD PRSs outperform published scores with an average Odds Ratio per Standard Deviation of 1.57 (SD = 0.14) and identify between 12% and 24% of individuals with high genetic risk. Using this risk factor to reclassify borderline or intermediate 10 year Atherosclerotic Cardiovascular Disease (ASCVD) risk improves assessments for both CAD (Net Reclassification Improvement (NRI) = 13.14% (95% CI 9.23-17.06%)) and ASCVD (NRI = 10.70 (95% CI 7.35-14.05)) in an independent cohort of 9,691 individuals. Our analyses demonstrate that using PRSs as Risk Enhancers improves ASCVD risk assessments outlining an approach for guiding ASCVD prevention with genetic information.

The Associations of Genetically Predicted Plasma Alanine with Coronary Artery Disease and its Risk Factors: A Mendelian Randomization Study

BACKGROUND

Alanine is an amino acid commonly used as a nutritional supplement and plays a key role in the glucose-alanine cycle. Plasma alanine has been associated in observational studies with a higher risk of coronary artery disease (CAD) and unhealthier lipid profiles. However, evidence from large randomized controlled trials is lacking.

OBJECTIVES

Using Mendelian randomization (MR), we assessed the unconfounded associations of plasma alanine with CAD and CAD risk factors.

METHODS

We applied single nucleotide polymorphisms that were strongly (P < 5 ×10-8) associated with plasma alanine as genetic instruments to large genome-wide association studies of CAD (63,108 cases; 296,901 controls), diabetes (90,612 cases; 583,493 controls), glucose (515,538 participants), lipids (low-density lipoprotein [LDL] cholesterol, high-density lipoprotein [HDL] cholesterol, triglycerides, total cholesterol, and apolipoprotein B) (>1.1 million participants), blood pressure (BP) (757,601 participants), and body mass index (682,137 participants). Given the potential sex disparity, we also conducted sex-specific analyses. MR estimates per standard deviation increase in alanine concentrations were obtained using inverse variance weighting followed by sensitivity analyses using weighted median, MR-Egger, MR-Pleiotropy RESidual Sum and Outlier, and MR-Robust Adjusted Profile Score.

RESULTS

Genetically predicted plasma alanine was not associated with CAD but with a higher risk of diabetes (odds ratio [OR]: 1.35; 95% confidence interval [CI]: 1.06, 1.72), higher glucose (β: 0.11; 95% CI: 0.02, 0.19), LDL cholesterol (β: 0.08; 95% CI: 0.04, 0.12), triglycerides (β: 0.25; 95% CI: 0.13, 0.38), total cholesterol (β: 0.14; 95% CI: 0.08, 0.20), apolipoprotein B (β: 0.12; 95% CI: 0.03, 0.21), and BP (β: 1.17; 95% CI: 0.31, 2.04 for systolic BP: β: 0.97; 95% CI: 0.49, 1.45 for diastolic BP) overall. The positive associations of serum alanine with LDL cholesterol and triglycerides were more notable in women than in men.

CONCLUSIONS

Alanine or factors affecting alanine may have causal effects on diabetes, blood glucose, lipid profiles, and BP but not on CAD. Further studies are needed to clarify possible mechanisms.

SGLT2 inhibition, circulating metabolites, and atrial fibrillation: a Mendelian randomization study

BACKGROUND

Sodium-glucose cotransporter 2 (SGLT2) inhibitors have shown promise in reducing the risk of atrial fibrillation (AF). However, the results are controversial and the underlying metabolic mechanism remains unclear. Emerging evidence implied that SGLT2 inhibitors have extra beneficial metabolic effects on circulating metabolites beyond glucose control, which might play a role in reducing the risk of AF. Hence, our study aimed to investigate the effect of circulating metabolites mediating SGLT2 inhibition in AF by Mendelian randomization (MR).

METHODS

A two-sample and two-step MR study was conducted to evaluate the association of SGLT2 inhibition with AF and the mediation effects of circulating metabolites linking SGLT2 inhibition with AF. Genetic instruments for SGLT2 inhibition were identified as genetic variants, which were both associated with the expression of SLC5A2 gene and glycated hemoglobin level (HbA1c). Positive control analysis on type 2 diabetes mellitus (T2DM) was conducted to validate the selection of genetic instruments.

RESULTS

Genetically predicted SGLT2 inhibition (per 1 SD decrement in HbA1c) was associated with reduced risk of T2DM (odds ratio [OR] = 0.63 [95% CI 0.45, 0.88], P = 0.006) and AF (0.51 [0.27, 0.97], P = 0.039). Among 168 circulating metabolites, two metabolites were both associated with SGLT2 inhibition and AF. The effect of SGLT2 inhibition on AF through the total concentration of lipoprotein particles (0.88 [0.81, 0.96], P = 0.004) and the concentration of HDL particles (0.89 [0.82, 0.97], P = 0.005), with a mediated proportion of 8.03% (95% CI [1.20%, 14.34%], P = 0.010) and 7.59% ([1.09%, 13.34%], P = 0.011) of the total effect, respectively.

CONCLUSIONS

This study supported the association of SGLT2 inhibition with a reduced risk of AF. The total concentration of lipoprotein particles and particularly the concentration of HDL particles might mediate this association. Further mechanistic and clinical studies research are needed to understand the mediation effects of circulating metabolites especially blood lipids in the association between SGLT2 inhibition and AF.

An epidemiological introduction to human metabolomic investigations

Metabolomics holds great promise for uncovering insights around biological processes impacting disease in human epidemiological studies. Metabolites can be measured across biological samples, including plasma, serum, saliva, urine, stool, and whole organs and tissues, offering a means to characterize metabolic processes relevant to disease etiology and traits of interest. Metabolomic epidemiology studies face unique challenges, such as identifying metabolites from targeted and untargeted assays, defining standards for quality control, harmonizing results across platforms that often capture different metabolites, and developing statistical methods for high-dimensional and correlated metabolomic data. In this review, we introduce metabolomic epidemiology to the broader scientific community, discuss opportunities and challenges presented by these studies, and highlight emerging innovations that hold promise to uncover new biological insights.

Divergent Survival Outcomes Associated with Elevated Branched-Chain Amino Acid Levels among Older Adults with or without Hypertension and Diabetes: A Validated, Prospective, Longitudinal Follow-Up Study

Branched-chain amino acids are critical metabolic intermediates that can indicate increased risk of cardiometabolic disease when levels are elevated or, alternatively, suggest sufficient mitochondrial energy metabolism and reserve in old age. The interpretation of BCAA levels can be context-dependent, and it remains unclear whether abnormal levels can inform prognosis. This prospective longitudinal study aimed to determine the interrelationship between mortality hazard and fasting serum BCAA levels among older men and women aged ≥65 years with or without hypertension and diabetes mellitus. At baseline (0Y), fasting serum BCAA concentration in 2997 community-living older men and women were measured. Approximately 14 years later (14Y), 860 study participants returned for repeat measurements. Deaths were analysed and classified into cardiovascular and non-cardiovascular causes using International Classification of Diseases codes. Survival analysis and multivariable Cox regression were performed. During a median follow-up of 17Y, 971 (78.6%) non-cardiovascular and 263 (21.4%) cardiovascular deaths occurred among 1235 (41.2%) deceased (median age, 85.8 years [IQR 81.7-89.7]). From 0Y to 14Y, BCAA levels declined in both sexes, whereas serum creatinine concentration increased (both p < 0.0001). In older adults without hypertension or diabetes mellitus, the relationship between mortality hazard and BCAA level was linear and above-median BCAA levels were associated with improved survival, whereas in the presence of cardiometabolic disease the relationship was U-shaped. Overall, adjusted Cox regression determined that each 10% increment in BCAA concentration was associated with a 7% (p = 0.0002) and 16% (p = 0.0057) reduction in mortality hazard estimated at 0Y and 14Y, respectively. Our findings suggested that abnormally high or low (dyshomeostatic) BCAA levels among older adults with hypertension and/or diabetes mellitus were associated with increased mortality, whereas in those with neither disease, increased BCAA levels was associated with improved survival, particularly in the oldest-old.

Association of Baseline Metabolomic Profiles With Incident Stroke and Dementia and With Imaging Markers of Cerebral Small Vessel Disease

BACKGROUND AND OBJECTIVES

Cerebral small vessel disease is a major cause of stroke and dementia. Metabolomics can help identify novel risk factors to better understand pathogenesis and predict disease progression and severity.

METHODS

We analyzed baseline metabolomic profiles from 118,021 UK Biobank participants. We examined cross-sectional associations of 325 metabolites with MRI markers of small vessel disease, evaluated longitudinal associations with incident stroke and dementia, and ascertained causal relationships using Mendelian randomization.

RESULTS

In cross-sectional analyses, lower levels of apolipoproteins, free cholesterol, cholesteryl esters, fatty acids, lipoprotein particle concentrations, phospholipids, and triglycerides were associated with increased white matter microstructural damage on diffusion tensor MRI. In longitudinal analyses, lipoprotein subclasses of very large high-density lipoprotein cholesterol (HDL) were associated with an increased risk of stroke, and acetate and 3-hydroxybutyrate were associated with an increased risk of dementia. Mendelian randomization analyses identified strong evidence supporting causal relationships for many findings. A few metabolites had consistent associations across multiple analysis types. Increased total lipids in very large HDL and increased HDL particle size were associated with increased white matter damage (lower fractional anisotropy: OR: 1.44, 95% CI 1.07-1.95, and OR: 1.19, 95% CI 1.06-1.34, respectively; mean diffusivity: OR: 1.49, 95% CI 1.11-2.01, and OR: 1.24, 95% CI 1.11-1.40, respectively) and an increased risk of incident all stroke (HR: 4.04, 95% CI 2.13-7.64, and HR: 1.54, 95% CI 1.20-1.98, respectively) and ischemic stroke (HR: 3.12, 95% CI 1.53-6.38; HR: 1.37, 95% CI 1.04-1.81). Valine was associated with decreased mean diffusivity (OR: 0.51, 95% CI 0.30-0.88) and had a protective association with all-cause dementia (HR: 0.008, 95% CI 0.002-0.035). Increased levels of cholesterol in small HDL were associated with a decreased risk of incident all stroke (HR: 0.17, 95% CI 0.08-0.39) and ischemic stroke (HR: 0.19, 95% CI 0.08-0.46) and were supported by evidence of a causal association with MRI-confirmed lacunar stroke (OR: 0.96, 95% CI 0.93-0.99).

DISCUSSION

In this large-scale metabolomics study, we found multiple metabolites associated with stroke, dementia, and MRI markers of small vessel disease. Further studies may help inform the development of personalized prediction models and provide insights into mechanistic pathways and future treatment approaches.

Large-scale metabolomic profiling and incident non-alcoholic fatty liver disease

Non-alcoholic fatty liver disease (NAFLD) is a highly prevalent disease with no specific drug therapy. High-throughput metabolomics present an unprecedented opportunity to identify biomarkers and potentially causal risk factors for NAFLD. Here, we determined the impact of 21 circulating metabolites, 17 lipids, and 132 lipoprotein particle characteristics on NAFLD combining prospective observational and two-sample Mendelian randomization (MR) analyses in 121,032 UK Biobank participants. We identified several metabolic factors associated with NAFLD risk in observational and MR analyses including triglyceride-rich and high-density lipoprotein particles composition, as well as the ratio of polyunsaturated fatty acids to total fatty acids. This study, is one of the largest to investigate incident NAFLD, provides concordant observational and genetic evidence that therapies aimed at reducing circulating triglycerides and increasing large HDL particles, as well as interventions aimed at increasing polyunsaturated fatty acid content may warrant further investigation into NAFLD prevention and treatment.

An atlas of genetic scores to predict multi-omic traits

The use of omic modalities to dissect the molecular underpinnings of common diseases and traits is becoming increasingly common. But multi-omic traits can be genetically predicted, which enables highly cost-effective and powerful analyses for studies that do not have multi-omics1. Here we examine a large cohort (the INTERVAL study2; n = 50,000 participants) with extensive multi-omic data for plasma proteomics (SomaScan, n = 3,175; Olink, n = 4,822), plasma metabolomics (Metabolon HD4, n = 8,153), serum metabolomics (Nightingale, n = 37,359) and whole-blood Illumina RNA sequencing (n = 4,136), and use machine learning to train genetic scores for 17,227 molecular traits, including 10,521 that reach Bonferroni-adjusted significance. We evaluate the performance of genetic scores through external validation across cohorts of individuals of European, Asian and African American ancestries. In addition, we show the utility of these multi-omic genetic scores by quantifying the genetic control of biological pathways and by generating a synthetic multi-omic dataset of the UK Biobank3 to identify disease associations using a phenome-wide scan. We highlight a series of biological insights with regard to genetic mechanisms in metabolism and canonical pathway associations with disease; for example, JAK-STAT signalling and coronary atherosclerosis. Finally, we develop a portal ( https://www.omicspred.org/ ) to facilitate public access to all genetic scores and validation results, as well as to serve as a platform for future extensions and enhancements of multi-omic genetic scores.

Atlas of plasma NMR biomarkers for health and disease in 118,461 individuals from the UK Biobank

Blood lipids and metabolites are markers of current health and future disease risk. Here, we describe plasma nuclear magnetic resonance (NMR) biomarker data for 118,461 participants in the UK Biobank. The biomarkers cover 249 measures of lipoprotein lipids, fatty acids, and small molecules such as amino acids, ketones, and glycolysis metabolites. We provide an atlas of associations of these biomarkers to prevalence, incidence, and mortality of over 700 common diseases ( nightingalehealth.com/atlas ). The results reveal a plethora of biomarker associations, including susceptibility to infectious diseases and risk of various cancers, joint disorders, and mental health outcomes, indicating that abundant circulating lipids and metabolites are risk markers beyond cardiometabolic diseases. Clustering analyses indicate similar biomarker association patterns across different disease types, suggesting latent systemic connectivity in the susceptibility to a diverse set of diseases. This work highlights the value of NMR based metabolic biomarker profiling in large biobanks for public health research and translation.