Heritability refers to the difference in phenotypes, or expressed outward characteristics, due to variation in genotypes.
Image Credit: Lyubov Levitskaya/Shutterstock.com
It is estimated, from studies in twins and families, that about half of the differences in the levels of various metabolites, is due to genetic variance. However, the amount of variation is dependent upon the metabolite class. This can affect the individual’s chances of developing diabetes, for instance.
Genetic variance and metabolism
The effect of genetic regulation of metabolism is most obvious in the conditions called inborn errors of metabolism. Here, a single rare variant of a gene causes complete or marked disruption of the normal function of an enzyme, causing extreme and lethal accumulation of the concerned metabolite.
However, less toxic gene variants can be found more commonly to impact human metabolism. Their effects can be found using genome-wide association studies with metabolomics (mGWAS).
In a study with 284 participants, over 350 metabolites were scanned for a relationship between the presence of common single-nucleotide polymorphisms (SNPs) and metabolite levels.
Even in this small sample, there were 4 common SNPs which led to marked changes in the related metabolic process.
These associations between the SNPs and the metabolic traits are found to occur in groups of SNPs, coding for the same enzyme or another metabolic regulator, with large variance and strong effect sizes.
For instance, a single homozygous variant can show as much as 50% less enzyme activity than the major allele homozygote. These clustered SNPs encode the same genetically influenced metabotype (GIM).
“For example, differences in hair color are apparent to the observer at first glance. However, in the case of metabolism, it takes much more effort to identify the role which the respective gene variant plays in the metabolism of the affected person”, Karsten Suhre said.
Measuring heritability for metabolites
Even though the overall heritability of metabolite levels is 50%, many researchers have found that different types of lipids and lipoproteins have different heritability estimates. Amino acids have higher estimates than lipids.
Measured SNPs are strong contributors to trait heritability. The SNP-associated, together with the family pedigree-associated genetic effects, determine heritability. Different gene variants cause different metabolic capacities in different people.
In the case of family members, however, the shared environment and other genetic factors that contribute independently to the same trait can cause the variance to increase still more.
Using more than 10 years of genome-wide association and exome sequencing studies, it is possible to estimate the heritability for almost 370 metabolites. The inter-relationship between metabolites is accounted for by the presence of two genetic components that contribute to the metabolite levels. One is the known metabolite loci of a specific superclass, while the other is due to metabolite loci other than the first class.
There are over 1800 different metabolites or metabolite ratios, with about 950 unique known metabolites belonging to 12 superclasses, 43 classes, or 77 subclasses.
Most of the metabolites belong to the superclass lipids or the organic acids one. There are 8 lipid classes and 9 organic acid classes. Furthermore, lipids can be divided into 32 subclasses and organic acids in 17 subclasses.
Heritability differences among different metabolite classes
The median total heritability (h2total) for lipids is 0.47, ranging from 0.11 to 0.66. For organic acids, h2total is 0.41, ranging from 0.14 to 0.72. The contribution to lipid levels due to known genetic loci is 0.06, and 0.01 for organic acids. Most of this genetic variance behind the known loci comes from the loci linked to lipids or organic acids, respectively.
Some of these genetic loci, such as glucokinase (hexokinase 4) regulator (GCKR), are linked to short-chain fatty acids, but others like FADS1-3 are more strongly associated with longer fatty acids and with less saturation.
While total heritability estimates are similar among the different classes of metabolites, there are marked differences between the lipid and organic acid classes when it comes to the contribution of known genetic loci, and of the loci linked to the corresponding superclass.
With age, the genetic contribution wanes somewhat, with the increasing influence of the environment, including medications.
More carbons, more difference
Interestingly, within lipids, phosphatidylcholines and triglycerides have increasing heritability, with longer carbon chains and with increasing numbers of double bonds in the fatty acid side chains.
Forty-six triacylglycerols are inherited differently depending on these parameters. A possible hypothesis is that the number of metabolic conversions could underlie the increasing heritability with longer carbon backbones or increasing desaturation of the side chains.
There was no significant heritability difference between essential and non-essential amino acids, but there were significant, though small, differences in the mean heritability among different lipid classes.
Fatty acids, steroids, and lipoproteins had class-specific loci heritability estimates that were significantly greater. Among the lipid classes, total heritability was higher for sphingolipids and glycerolipids than for phospholipids.
Varying organic acid metabolism
The median heritability varies among different classes of organic acids, namely carboxylic, keto and hydroxy acids. The highest median total heritability is for keto acids > carboxylic acids > hydroxy acids. Keto acids also had the highest heritability due to loci linked to the corresponding class.
The hydroxy acids showed the highest genetic variance due to non-linked loci and due to known loci, both linked and non-linked, they had the lowest total heritability and the lowest contribution from linked loci.
The carboxylic acids had higher heritability variations due to known genetic loci compared to the keto acids.
There is no significant gap in the total heritability of all organic acid classes, nor in the metabolite levels attributable to loci linked to the corresponding class. However, the variance due to known genetic loci does show significant differences between individual classes.
We now know, through the efforts of many researchers, that both class-specific and non-class-specific loci for metabolites, contribute to different heritability estimates among different classes of metabolites and lipid classes.
The importance of this knowledge is in developing personalized interventions for the prevention and treatment of complex metabolic conditions or to predict how individuals will react to certain nutritional, therapeutic, or environmental influences.
- Hagenbeek, F. A., Pool, R., van Dongen, J., Draisma, H. H. M., et al. Heritability estimates for 361 blood metabolites across 40 genome-wide association studies. Nature Communications 2020. https://dx.doi.org/10.1038%2Fs41467-019-13770-6. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6946682/
- Gieger, C., Geistlinger, L., Altmaier, E., Hrabe´ de Angelis, M., Kronenberg, F., et al. Genetics meets metabolomics: a genome-wide association study of metabolite profiles in human serum. PLoS Genetics, 4(11): e1000282 DOI: 10.1371/journal.pgen.1000282
- Gallois, A., Mefford, J., Ko, A., Vaysse, A., et al. A comprehensive study of metabolite genetics reveals strong pleiotropy and heterogeneity across time and context. Nature Communications 2019. 10: 4788. DOI: 10.1038/s41467-019-12703-7. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6803661/
- Kastenmuller, G., Raffler, J., Gieger, C., and Suhre, K. Genetics of human metabolism: an update. Human Molecular Genetics 2015; 24(R1). https://dx.doi.org/10.1093%2Fhmg%2Fddv263. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4572003/