E C Røyrvik1, J P Burgstaller2,3, I G Johnston4. 1. Division of Biomedical Sciences, Warwick Medical School, Gibbet Hill Road, University of Warwick, Coventry CV4 7AL, UK e.royrvik@warwick.ac.uk i.johnston.1@bham.ac.uk. 2. Biotechnology in Animal Production, Department for Agrobiotechnology, IFA Tulln, 3430 Tulln, Austria. 3. Institute of Animal Breeding and Genetics, University of Veterinary Medicine Vienna , Veterinärplatz 1, 1210 Vienna, Austria. 4. School of Biosciences, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK e.royrvik@warwick.ac.uk i.johnston.1@bham.ac.uk.
Abstract
STUDY QUESTION: Does mitochondrial DNA (mtDNA) diversity in modern human populations potentially pose a challenge, via mtDNA segregation, to mitochondrial replacement therapies? SUMMARY ANSWER: The magnitude of mtDNA diversity in modern human populations is as high as in mammalian model systems where strong mtDNA segregation is observed; consideration of haplotype pairs and/or haplotype matching can help avoid these potentially deleterious effects. WHAT IS KNOWN ALREADY: In mammalian models, substantial proliferative differences are observed between different mtDNA haplotypes in cellular admixtures, with larger proliferative differences arising from more diverse haplotype pairings. If maternal mtDNA is 'carried over' in human gene therapies, these proliferative differences could lead to its amplification in the resulting offspring, potentially leading to manifestation of the disease that the therapy was designed to avoid-but existing studies have not investigated whether mtDNA diversity in modern human populations is sufficient to permit significant amplification. STUDY DESIGN, SIZE, DURATION: This theoretical study used over 7500 human mtDNA sequences from The National Center for Biotechnology Information (NCBI), a range of international and British mtDNA surveys, and 2011 census data. PARTICIPANTS/MATERIALS, SETTING, METHODS: A stochastic simulation approach was used to model random haplotype pairings from within different regions. In total, 1000 simulated pairings were analysed using the basic local alignment search tool (BLAST) for each region. Previous data from mouse models were used to estimate proliferative differences. MAIN RESULTS AND THE ROLE OF CHANCE: Even within the same haplogroup, differences of around 20-80 single-nucleotide polymorphisms (SNPs) are common between mtDNAs admixed in random pairings. These values are sufficient to lead to substantial segregation in mouse models over an organismal lifetime, even given low starting heteroplasmy, inducing increases from 5% to 35% over 1 year. Substantial population mixing in modern UK cities increases the expected genetic differences. Hence, the likely genetic differences between humans randomly sampled from a population may well allow substantial amplification of a disease-carrying mtDNA haplotype over the timescale of a human lifetime. We report ranges and mean differences for all statistics to quantify uncertainty in our results. LIMITATIONS/REASONS FOR CAUTION: The mapping from mouse and other mammalian models to the human system is challenging, as timescales and mechanisms may differ. Reporting biases in NCBI mtDNA data, if present, may affect the statistics we compute. We discuss the robustness of our findings in the light of these concerns. WIDER IMPLICATIONS OF THE FINDINGS: Matching the mtDNA haplotypes of the mother and third-party donor in mitochondrial replacement therapies is supported as a means of ameliorating the potentially deleterious results of human mtDNA diversity. We present a chart of expected SNP differences between mtDNA haplogroups, allowing the selection of optimal partners for therapies. LARGE SCALE DATA: N/A STUDY FUNDING/COMPETING INTERESTS: The authors report no external funding sources or conflicts of interest.
STUDY QUESTION: Does mitochondrial DNA (mtDNA) diversity in modern human populations potentially pose a challenge, via mtDNA segregation, to mitochondrial replacement therapies? SUMMARY ANSWER: The magnitude of mtDNA diversity in modern human populations is as high as in mammalian model systems where strong mtDNA segregation is observed; consideration of haplotype pairs and/or haplotype matching can help avoid these potentially deleterious effects. WHAT IS KNOWN ALREADY: In mammalian models, substantial proliferative differences are observed between different mtDNA haplotypes in cellular admixtures, with larger proliferative differences arising from more diverse haplotype pairings. If maternal mtDNA is 'carried over' in human gene therapies, these proliferative differences could lead to its amplification in the resulting offspring, potentially leading to manifestation of the disease that the therapy was designed to avoid-but existing studies have not investigated whether mtDNA diversity in modern human populations is sufficient to permit significant amplification. STUDY DESIGN, SIZE, DURATION: This theoretical study used over 7500 human mtDNA sequences from The National Center for Biotechnology Information (NCBI), a range of international and British mtDNA surveys, and 2011 census data. PARTICIPANTS/MATERIALS, SETTING, METHODS: A stochastic simulation approach was used to model random haplotype pairings from within different regions. In total, 1000 simulated pairings were analysed using the basic local alignment search tool (BLAST) for each region. Previous data from mouse models were used to estimate proliferative differences. MAIN RESULTS AND THE ROLE OF CHANCE: Even within the same haplogroup, differences of around 20-80 single-nucleotide polymorphisms (SNPs) are common between mtDNAs admixed in random pairings. These values are sufficient to lead to substantial segregation in mouse models over an organismal lifetime, even given low starting heteroplasmy, inducing increases from 5% to 35% over 1 year. Substantial population mixing in modern UK cities increases the expected genetic differences. Hence, the likely genetic differences between humans randomly sampled from a population may well allow substantial amplification of a disease-carrying mtDNA haplotype over the timescale of a human lifetime. We report ranges and mean differences for all statistics to quantify uncertainty in our results. LIMITATIONS/REASONS FOR CAUTION: The mapping from mouse and other mammalian models to the human system is challenging, as timescales and mechanisms may differ. Reporting biases in NCBI mtDNA data, if present, may affect the statistics we compute. We discuss the robustness of our findings in the light of these concerns. WIDER IMPLICATIONS OF THE FINDINGS: Matching the mtDNA haplotypes of the mother and third-party donor in mitochondrial replacement therapies is supported as a means of ameliorating the potentially deleterious results of human mtDNA diversity. We present a chart of expected SNP differences between mtDNA haplogroups, allowing the selection of optimal partners for therapies. LARGE SCALE DATA: N/A STUDY FUNDING/COMPETING INTERESTS: The authors report no external funding sources or conflicts of interest.
Authors: Joerg P Burgstaller; Thomas Kolbe; Vitezslav Havlicek; Stephanie Hembach; Joanna Poulton; Jaroslav Piálek; Ralf Steinborn; Thomas Rülicke; Gottfried Brem; Nick S Jones; Iain G Johnston Journal: Nat Commun Date: 2018-06-27 Impact factor: 14.919
Authors: Hanne Hoitzing; Payam A Gammage; Lindsey Van Haute; Michal Minczuk; Iain G Johnston; Nick S Jones Journal: PLoS Comput Biol Date: 2019-06-26 Impact factor: 4.475