Introduction Lipids of the endocannabinoid (eCB), N-acyl ethanolamine (NAE), and ceramide (CER) classes are potential novel biomarkers of coronary artery disease and type-2 diabetes. Major-gene effects have been discovered for certain lipid species, notably lipoprotein(a). We sought to establish the heritability of eCB, NAE, and CER species, and identify DNA variants influencing their concentrations in plasma.
Methods We undertook heritability (QTDT, GCTA) and GWAS analyses (FaST-LMM) of 11 eCBs and NAEs, and 37 CERs in 1,016 plasma samples from 196 British Caucasian families ascertained through a hypertensive proband, using targeted lipidomics by mass spectrometry, and Illumina 660W-Quad genotyping.
Results Anandamide (AEA), a potent eCB, was found heritable (h2AEA = 32–35%; P<5.80 × 10–11), however other less studied NAE species, presented higher estimates of heritability (h2NAEs = 41–79%; P<3.89 × 10–13). 24–46% of the variation in potential biomarker CER is due to genetic factors (P<1.00 × 10–7). GWAS identified associations with eQTLs of proteins involved in the metabolism of eCB and NAE (e.g. FAAH; PNAEDHEA <6.33 × 10–12,) as well as CER (SPTLC3; PCERN(24)S(18) <8.99 × 10–19) and novel loci implicated in cancer risk and non-alcoholic fatty liver disease (e.g. FBXO28; PCERN(24)S(19)ratio <1.95 × 10–8, SULT1C4; PCERN(24)S(19) <8.99 × 10–19).
Two-sample Mendelian randomisation suggests that a variant in FAAH (rs324420) influencing the level of plasma NAEs (e.g. PNAEDHEA<6.33 × 10–12) is causally associated with obesity, drug addiction, and anxiety. As an example, participants with the rs324420 AA genotype had a mean DHEA plasma concentration of 518 pg/ml, which decreased by 27% in those with the AC genotype, and a further 11% in those carrying the CC genotype.
Conclusion We demonstrate for the first time estimates of heritability for this extended array of bioactive lipids, identify GWAS-significant SNPs associating with their levels in circulation, and implicate the lipid species studied here in cardiovascular disease, cancer, and drug addition. The results shown here can be used for prioritisation of lipid mediators for large-scale Mendelian Randomisation studies, and in the identification of causal metabolic pathways, novel diagnostics and drug targets for disease intervention.
Conflict of Interest None
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.