Computational models elucidate dynamics in the intergenerational inheritance of mtDNA mutations

Abstract

Mitochondria, widely studied for their crucial role in cell bioenergetics, carry independent genomes that are inherited in offspring maternally. Understanding of the dynamics of the mitochondrial genome (mtDNA) is critical to further study of a large host of diseases (such as MELAS and LHON) that are caused by mtDNA mutations. One of the key steps in the inheritance of mtDNA is the germline bottleneck in the development and proliferation of primordial germ cells, during which process the mtDNA copy number of these cells fluctuates greatly. This leads to key alterations in and increased variance of mutation frequencies and allows complex patterns of selection at the organellar and cellular level. Furthermore, there can be varying degrees or patterns of selection that are present throughout the bottleneck process. Gaining an understanding of how these pressures interact with other factors such as genetic drift and Muller’s ratchet can better elucidate the inheritance of mtDNA mutations. To date, however, there is little consensus on how this process occurs and though numerous models have been developed to explain trends in experimental data, a thorough analysis of the different components and parameters of these models and how they relate to biological systems is lacking. Drawing from extensive intergenerational datasets of pathogenic mutations in mice and humans, we develop models based on trends that appear in the transmission of mtDNA between mothers and their offspring. The combination of diverse datasets comprising high fidelity single molecule‐sequence and NGS data allows sufficient resolution to support detailed, multivariate analysis. We derive novel metrics for describing these data and create biologically informed simulations that can flexibly account for constant, variable, and multiphasic modes of selection. We apply these models to datasets generated from simulating bottlenecked cell proliferation, using previously proposed and newly derived bottleneck frameworks and population dynamics. The resulting mutational profiles allow deeper understanding of the mechanisms underlying the biological data and provide a foundation for further development of computational approaches to studying germline mtDNA. Our simulations thus represent a substantial step forward in the understanding of the complex inheritance of mitochondrial mutations.