Feasibility and impact of haplogroup matching for mitochondrial replacement treatment

Abstract

Mitochondrial replacement technology (MRT) aims to reduce the risk of serious disease in children born to women who carry pathogenic mitochondrial DNA (mtDNA) variants. By transplanting nuclear genomes from eggs of an affected woman to enucleated eggs from an unaffected donor, MRT creates new combinations of nuclear and mtDNA. Based on sets of shared sequence variants, mtDNA is classified into ~30 haplogroups. Haplogroup matching between egg donors and women undergoing MRT has been proposed as a means of reducing mtDNA sequence divergence between them. Here we investigate the potential effect of mtDNA haplogroup matching on clinical delivery of MRT and on mtDNA sequence divergence between donor/recipient pairs. Our findings indicate that haplogroup matching would limit the availability of egg donors such that women belonging to rare haplogroups may have to wait > 4 years for treatment. Moreover, we find that intra-haplogroup sequence variation is frequently within the range observed between randomly matched mtDNA pairs. We conclude that haplogroup matching would restrict the availability of MRT, without necessarily reducing mtDNA sequence divergence between donor/recipient pairs.

Keywords: haplogroup matching; mitochondrial disease; mitochondrial replacement therapy.

Figure EV1. Impact of haplogroup matching on the availability of egg donors for MRT

1. Simplified mtDNA haplogroup phylogeny showing the 31 ‘major’ mtDNA haplogroups (phylotree.org).

2. Graphical representation of global mtDNA haplogroup diversity, showing the relative estimated proportions of the most frequent mtDNA haplogroups by region (where available, AFR = Africa, SAM = South America, ASA = Southern Asia, EAS = East Asia and EUR = Europe, Dataset EV1A and B).

3. Graph showing the estimated number of women per 100,000 that proceed to egg donation (red) in relation to those that are eligible to donate (blue) for each of the commonest mtDNA haplogroups identified across European populations (Based on 316 donors, Datasets EV1A and EV2A, and assuming the haplogroup of those who donate eggs is reflective of ancestry).

4. Estimated years to recruit one volunteer per European mtDNA haplogroup based on the estimated donor availability and the estimated number of women who progress to egg donation (Datasets EV1A and EV2A).

Data information: Bar graphs show mean and standard deviation.

Figure 1. MtDNA sequence divergence within European, African, and Eurasian mtDNAs

1. A

   Boxplots of the estimated pairwise mtDNA sequence divergence for the major European, African, and Eurasian haplogroups (i.e., haplogroup‐matched mtDNAs, Dataset EV3A–C). The intra‐haplogroup mtDNA sequence divergence of each population differs significantly (one‐way ANOVA in each population P < 2.2 × 10⁻¹⁶).

2. B

   Boxplots of estimated pairwise mtDNA sequence divergence when two random sequences are selected within each population (i.e., unmatched mtDNAs, Dataset EV3A–C).

3. C

   Bar charts of the change in mean variant differences when mtDNA pairs are haplogroup matched (as in A) compared when they are randomly selected from the combined major European, African, and Eurasian haplogroups (as in B).

Data information: Boxplots show median, 25^th and 75^th percentile, with whiskers indicating 95^th upper/lower interquartile range. Dots indicate outliers. Bar charts show mean, and standard deviation. Estimates (A and B) and counts (C) are based on 7,655 European, 3,688 African and 6,857 Eurasian mtDNA sequences. Population groups were defined by mtDNA haplogroup. Tajima‐Nei's genetic distance model (Tajima‐Nei's D) was used to derive sequence divergence, where Tajima‐Nei's D 0.00006 = 1 variant difference.

Figure 2. Non‐synonymous variant differences divergence within European, African and Eurasian populations' mtDNAs

1. A

   Boxplots showing the non‐synonymous variant divergence within the major European, African, and Eurasian haplogroups (i.e., haplogroup matched mtDNAs, Dataset EV3A–C). The intra‐haplogroup mtDNA sequence divergence of each population differs significantly (one‐way ANOVA in each population P < 2.2 × 10⁻¹⁶).

2. B

   Boxplots showing the mean non‐synonymous variant differences when two random sequences are selected within each population (i.e., unmatched mtDNAs, Dataset EV3A–C).

3. C

   Bar charts showing the mean non‐synonymous variant differences when mtDNA pairs are haplogroup matched (as in A) compared when they are randomly selected from the combined major European, African, and Eurasian haplogroups (as in B).

Data information: Boxplots show median, 25^th and 75^th percentile, with whiskers indicating 95^th upper/lower interquartile range. Dots indicate outliers. Bar charts show mean and standard deviation. Estimates (A and B) and counts (C) are based on 7,655 European, 3,688 African and 6,857 Eurasian mtDNA sequences. Population groups were defined by mtDNA haplogroup. Nei‐Gojobori model was used to investigate non‐synonymous variant differences.

Figure 3. Non‐synonymous variant differences divergence within European, African and Eurasian populations' mtDNAs

1. A

   Graphs showing the average total mtDNA variation observed when donor/recipient mtDNA pairs are selected from across two populations (Dataset EV3E). The top and middle graphs show average differences when a patient belonging to an African or Eurasian‐origin haplogroup respectively is matched with an egg donor from a European haplogroup compared with a population‐matched egg donor. The lower graphs show the average differences when a patient belonging to a European haplogroup is matched with an egg donor belonging to an African, Eurasian, or European haplogroup.

2. B

   Heatmap showing the average variant differences between pairs of mtDNAs selected from all common haplogroups in each population (Dataset EV4).