Allele Frequency Selection and No Age-Related Increase in Human Oocyte Mitochondrial Mutations

Mitochondrial dysfunction can cause a multitude of systemic diseases, and thus, the origin of such dysfunction is of great interest to many clinicians and researchers. Part of mitochondrial dysfunction can be explained by mutations in mitochondrial DNA (mtDNA), which can be measured both in somatic and germline cells. As the average maternal age continues to increase, determining if mutations in mtDNA increase with age is an important aspect of reproductive health care. This study was designed to assess the frequency of de novo mutations in mtDNA in oocytes based on age, compared with the frequency of mutations in somatic cells. This study used high-quality, full-length mtDNA sequences for 22 individuals between the ages of 20 and 42, obtaining samples from blood, saliva, and oocytes. Duplex sequencing was used to identify variants in mtDNA; median mtDNA enrichment was 85.4%, 0.7%, and 4.0% and a median duplex consensus sequence (DCS) depth of 1440X, 78X, and 158X for oocytes, blood, and saliva, respectively. There were 3525 high-confidence variants identified across all samples. Across age groups and geographic regions, there were significantly fewer mutations in oocytes compared with blood and saliva. When predicting the probability of a mutation in mtDNA as a function of age, there were significant increases in mutation frequency in blood and saliva, but not in oocytes (P<0.001, P<0.001, P=0.688, respectively). Comparing different age groups, mutation frequency was higher in blood and saliva in older women compared with younger women, but there was no significant difference in oocyte mutation frequency. Within oocyte mtDNA, mutations were more frequent in the D-loop than in the "coding region" for each age group, consistent with previous literature. This remained true in several different analyses of regions of mtDNA, though no significant differences were found related to age. A total of 109 de novo mutations were located at confirmed disease-associated sites, with 32 variants in 29 positions. Lower proportions of disease-related mutations were present in oocytes compared with blood or saliva, suggesting preferential selection for oocytes against mutations. Hotspots for variants were identified in each type of cell, showing 58 sites overlapping between cell types; 40 of these were in the D-loop, 4 in rRNA, 4 in tRNA, and 10 in protein-coding regions. This supports the previous finding of high mutation frequency in the D-loop for oocytes. These results indicate that there is no increase in mutation frequency in oocytes as women age, though there was an increase in mutation frequency in both blood and saliva. This contrasts with previous literature in animal models that showed either continuous increases in mutation frequency with age or increases up to a certain age after which there was a plateau. The age group of this study was limited to 20-42, meaning that there could be an increase in mutation frequency that occurs either before or after this time period in humans. This is the first study to use duplex sequencing, and thus the most accurate to date in terms of pinpointing variants and mutation sites. Future research should focus on validating these results with similar methods as well as implementing duplex sequencing of pedigrees to form a more complete picture.