NEW YORK — Genetic variants in mitochondrial DNA influence numerous complex human traits and diseases, a new analysis has found.
The mitochondrial genome is 16,569 base pairs large, and while genetic variants within it have been known to affect mitochondrial function, their effect on common, complex diseases has been less explored. One reason is that the tools needed to analyze the mitochondrial genome have not been well developed until recently, according to Joanna Howson, a researcher at the University of Cambridge.
She and her colleagues conducted a study on hundreds of thousands of samples from the UK Biobank using genotyping arrays that included 265 mtDNA variants. Using modified algorithms and calling procedures, Howson and her colleagues conducted mtDNA phenome-wide association studies for nearly 900 different traits. As they reported in Nature Genetics on Monday, they uncovered hundreds of novel mtDNA-phenotype associations, including ones affecting liver function and height.
"Our work has outlined a new approach to analyzing the mitochondrial genome that we hope will be helpful for others looking to investigate mitochondrial DNA in common complex conditions," Howson wrote in an email.
As part of the UK Biobank, 488,377 individuals underwent genotyping for 265 mtSNVs. The researchers tweaked existing quality control workflows to handle the mtSNVs dataset and included a four-step procedure of pre-recalling QC, manual recalling, post-recalling QC, and imputation of mtSNVs not included on the genotyping array. In all, following QC and imputation, they examined 378,696 mtSNV-trait associations, encompassing both binary and quantitative traits, with up to 473 mtSNVs.
Through this, they uncovered 260 new mtSNV-trait associations, including with multiple sclerosis and type 2 diabetes, as well as with liver biomarkers and height.
For instance, the researchers uncovered new associations between a variant in MT-ATP6 and type 2 diabetes, and between one in MT-ND5 and multiple sclerosis.
At the same time, the researchers linked 23 mtSNVs with aspartate aminotransferase, or AST, and nine mtSNVs with alanine aminotransferase, or ALT, levels, both of which are among the classic biomarkers of liver function. The lead AST-associated variant was linked to an increase in AST levels, while a missense variant in MT-CYB was associated with a reduction in ALT levels.
Howson noted that they were expecting mitochondrial variants to affect metabolic diseases, as mitochondria are a key part of cellular energy metabolism. As the liver also has a role in metabolism, she added that uncovering ties between mitochondrial DNA with liver function biomarkers is not unexpected.
The researchers also uncovered five mtSNVs and a rare noncoding variant linked to reduced height. The common variants were linked to a small, about 0.8-millimeter decline in height, while the rare variant was associated with about a 4.3-millimeter reduction in height.
"I was surprised that we observed an effect with height initially," Howson said. She added, though, that as "height is a highly polygenic trait and when we consider that some patients with rare inherited mitochondrial diseases tend to have shorter stature, perhaps it is less surprising."
Across traits, the researchers found that rare mtSNVs tended to have larger effect sizes than common variants. For instance, a common variant in MT-ND3 was linked to an odds ratio of 1.15 for multiple sclerosis, while rare variants in MT-ND5 and MT-TD had odd ratios between 1.65 and 2.06.
Eventually, Howson and her colleagues hope to better understand the interactions between the mitochondrial and nuclear genomes and how they influence disease. That, she said, would "ultimately help in future medicines development."