When Genomics meets Metabolomics by Circulating Metabolites in Blood

Metabolic syndrome is pathologically featured by a broad range of diseases, such as obesity, type 2 diabetes (T2D), cardiovascular and non-acholic fatty liver diseases (NAFLD), resulting from various genetic and acquired disturbances on energy homeostasis.

During the last decades, the number of people who suffer from those metabolic diseases has been skyrocketing worldwide, becoming an unaffordable burden on both individual health and society. Word Health organization reported that 2.5 billion adults are overweight or obese. Notably, childhood and adolescent obesity have been increasing to an altering level, affecting 390 million population aged between age 5 and 19 years.  Moreover, there are approximately 462 million people who had type 2 diabetes reported in 2017 and this number is expected to be increased to 640 million by 2030. Furthermore, the death toll directly caused or associated with those metabolic disorders has been top ranked right after cancer and infectious diseases. In 2019, diabetes and its associated kidney disease caused an estimated 2 million deaths globally. Undoubtably, these alarming facts have been showing the necessity of new treatment development based on further understanding on the disease mechanisms, and more essentially to implement accurate prognosis methods for early intervention especially for those who are inheritably susceptible towards those metabolic diseases. 

In the last two decades, a large number of human genome-wide association studies (hGWAS) have been conducted with the significant lift by the advances in the sequencing-based technologies from microarray to whole-genome and paneled WES sequencing, to statistically establish the association of genetic loci copy number or sequence variations with a specific trait or disease. By great effort, many novel genetic loci and variants have been successfully identified as the potential risk factors for those prevailing metabolic diseases, such as FTO gene polymorphisms linked with physical activity and appetite in several ethnicities (Harbron, J et al., Nature, 2014), one SNP in APOA1 gene cluster involved in lipid metabolism (Kristiansson K., Circ Cardiovasc Genet, 2012), and the transcript factor TCF7L2  strongly couples with T2D in multiple populations (Sladek et al., Nature, 2007) by regulating beta-cell functions and insulin secretion. However, most of those loci were shown later to have ambiguous direct causal relationship with diseases by more detailed physiological studies, hampering the interpretation of the biological meaning of hGWAS results. Such results also clearly indicate a third factor is indispensably needed to reinforce and validate the association between genetics and eventual biological outcomes.

To fill the missing pieces in this puzzle, the utilization of high-throughput metabolomics platforms on hGWAS analyses has set up a new foundation to better understand and to establish the pathogenic architecture of how genetic variants and pathways influence biological mechanisms and complex diseases, associating genetic information with changes on metabolic profiling. By the light of this methodology, a comprehensive genome-wide characterization study using circulating metabolic biomarkers in blood has been published in Nature (Karjalainen M, et al, 2024) recently by an international collaboration of geneticists, in which more than 400 independent loci and assign putative causal genes were identified by analyzing 233 NMR-quantified circulating metabolic trats in 136,016 participants from a large cohort.

Besides those previously reported loci associated with  adiposity and T2D risk, a number of novel loci and probable causal genes in both lipid and glucose homeostasis have been newly discovered by the manual curation and clustering approaches in this study, including LDL-RAP1’s association with lipoprotein and fatty acids, amino acid-associated SLC2A4RG and KCNK16 loci having  a regulatory role in activation of insulin-dependent glucose transporter GLUT4 and is T2D susceptible gene, respectively.  Moreover, a detailed metabolic profiling of lipoprotein- and lipid-associated variants was performed to deepen the understanding of how known and novel loci affect lipoprotein metabolism at a granular level. The clustering produced seven major categories of loci related to the risks in T2D, obesity and lipoprotein particles regulation. Despite those known loci-lipid traits correlations, TRIM5 best known for its antiviral role was identified by its lead variant (rs11601507) to associate with 42 lipoprotein and lipid traits, and thus potentially regulate lipid accumulation and inflammation via LDL receptor pathway and mTORC1 signaling. This new finding could raise the possibility of using TRIM5 as a therapeutical target to reduce the risk of cardiovascular diseases like PCSK9 inhibition, especially for statin intolerant individuals. To further demonstrate the values of combining the metabolic association information with disease associations proposed by this study, the associations of intrahepatic cholestasis of pregnancy (ICP) loci with metabolic traits was dissected into detail. The results identified nine replicated and three novel loci (UGT8, NUP153 and HKDC1), not only providing more insights on the metabolic groundwork of inadequately understood diseases like ICP, but also the new findings could be extrapolated into new treatments. Notably, ten loci with only one reported previously have been identified to have robust association with acetone indicated by Mendelian randomization analysis in the study and the positive association with hypertension can be robustly revealed by using only four loci (HMGCS2, OXCT1, CYP2E1 and SLC2A4/Glut4), enriching the genetic basis of the causal relationship between acetone with hypertension. Last but not least, the study emphasized the sample and participant characteristics, such as serum or plasma, fasting or non-fasting state, can have profound effects on the outcomes of those genetic association analyses.

diseases by more detailed physiological studies, hampering the interpretation of the biological meaning of hGWAS results. Such results also clearly indicate a third factor is indispensably needed to reinforce and validate the association between genetics and eventual biological outcomes.

To fill the missing pieces in this puzzle, the utilization of high-throughput metabolomics platforms on hGWAS analyses has set up a new foundation to better understand and to establish the pathogenic architecture of how genetic variants and pathways influence biological mechanisms and complex diseases, associating genetic information with changes on metabolic profiling. By the light of this methodology, a comprehensive genome-wide characterization study using circulating metabolic biomarkers in blood has been published in Nature (Karjalainen M, et al, 2024) recently by an international collaboration of geneticists, in which more than 400 independent loci and assign putative causal genes were identified by analyzing 233 NMR-quantified circulating metabolic trats in 136,016 participants from a large cohort.

Besides those previously reported loci associated with  adiposity and T2D risk, a number of novel loci and probable causal genes in both lipid and glucose homeostasis have been newly discovered by the manual curation and clustering approaches in this study, including LDL-RAP1’s association with lipoprotein and fatty acids, amino acid-associated SLC2A4RG and KCNK16 loci

This pioneering study highlights the value of associating data on genetics, metabolic traits, and disease outcomes at large scale to identify high-confidence causal relationship between genes and diseases. The summary statistics is publicly available through the NHGRI-EBI GWAS catalogue (GCST90301941–GCST90302173), and thus providing a cornerstone to continuously unmask novel genes’ associations with metabolic processes in diverse diseases.