Nuclear-mitochondrial DNA segments resemble paternally inherited mitochondrial DNA in humans

Abstract

Several strands of evidence question the dogma that human mitochondrial DNA (mtDNA) is inherited exclusively down the maternal line, most recently in three families where several individuals harbored a 'heteroplasmic haplotype' consistent with biparental transmission. Here we report a similar genetic signature in 7 of 11,035 trios, with allelic fractions of 5-25%, implying biparental inheritance of mtDNA in 0.06% of offspring. However, analysing the nuclear whole genome sequence, we observe likely large rare or unique nuclear-mitochondrial DNA segments (mega-NUMTs) transmitted from the father in all 7 families. Independently detecting mega-NUMTs in 0.13% of fathers, we see autosomal transmission of the haplotype. Finally, we show the haplotype allele fraction can be explained by complex concatenated mtDNA-derived sequences rearranged within the nuclear genome. We conclude that rare cryptic mega-NUMTs can resemble paternally mtDNA heteroplasmy, but find no evidence of paternal transmission of mtDNA in humans.

Fig. 1. Bioinformatic pipeline to detect the mixed haplotypes from 11,867 trios.

a Sample quality control. b Search for the putative trios carrying the mixed haplotype pattern. I. Distribution of the informative trios where at least one variant (variant allele fraction (VAF) > 5%) is detected in the father, but not in the mother. II. Distribution of trios where at least one variant was shared in the fathers and their offspring. c Number of homoplasmic variants in 32 fathers. Yellow bars represent fathers’ homoplasmic variants also observed in the mothers and their offspring; pink bars represent the homoplasmic variants also observed in their offspring, but not detected in the mothers. Blue bars represent homoplasmic variants only observed in the fathers which are not detected in either the mothers or their offspring. d Pedigrees of the seven families showing mixed haplotypes. The symbols with lines represent the individuals carrying the mixed haplotypes.

Fig. 2. Mixed haplotype patterns and NUMTs observed in two families.

a Examples of the mixed haplotype patterns showing Family 2 and Family 6. Father and both offspring show the mixed haplotypes with the similar HFs in each family. The mixed haplotype patterns observed in the other five families are shown in Supplementary Fig. 2. b Screenshots from Integrative Genomics Viewer (IGV) showing aligned reads corresponding to the rare NUMTs. The alignment of the discordant and split reads corresponding to the NUMTs on the nuclear DNA (left) and mtDNA (right) in two families. Teal bars indicate the aligned reads which mapped to the nuclear DNA where their mates mapped to the mtDNA (left). The chromosomes on which their mates are found are shown in different colours (right). The genome position, repeats and segmental duplications tracks from UCSC genome browser are shown at the bottom. IGV Screenshots from all aligned reads corresponding to the rare NUMTs on the nuclear DNA are shown in Supplementary Fig. 3. IGV screenshots of the other five families are shown in Supplementary Fig. 4. c Circos plots show the observed variants and NUMTs in two families. Circles from the outside to the inside indicate the following: (1) position of a variant on the mtDNA; (2) regions corresponding to the different mtDNA genes; (3) variants identified in the mother where the radial axis corresponds to the VAF; (4) variants identified in the father; (5) variants identified in the offspring, proband (left) and sibling (right) are shown, respectively; (6) NUMTs observed in the family, proband (left) and sibling (right) are shown, respectively.

Fig. 3. Strategy for NUMT detection. NU – nuclear genome, MT – mitochondrial genome.

nu-mt split reads: one end of the split read maps to nuclear DNA and the other end maps to mtDNA-derived sequence. mt-mt split reads: two ends of the same split read map to two locations on the mtDNA-derived sequences. nu-mt breakpoint: the breakpoint between joined nuclear and mtDNA-derived sequences. mt-mt breakpoint: the breakpoint joint two separate mtDNA-derived sequences. For a detailed explanation see Methods.

Fig. 4. Defining complex mega-NUMTs.

a Split reads align to both mtDNA-derived and nuclear DNA sequences (top). Discordant reads are paired reads where one end aligns to mtDNA-derived sequences and the other end aligns to nuclear DNA sequences (middle). Possible constructed concatemer is shown at the bottom with observed supporting split reads. The positions of breakpoints (bp) are shown on both nuclear DNA (top) and mtDNA (bottom). b IGV screenshots showing the reads not properly aligned to mtDNA-derived sequences. The reads are coloured by pair orientation. Many read pairs with anomalous pair orientations in the fathers and probands support the mtDNA-derived sequence rearrangement. In Family 5 and Family 7, the fathers and offspring carried the same NUMT, they also carried the same nuclear-mtDNA junctions and the junctions within mtDNA-derived sequences. The defined junctions by split reads from the other five families are included in Supplementary Fig. 8.

Fig. 5. Characteristics of mega-NUMTs.

a Model showing the formation of mega-NUMTs and our strategy for their detection in whole genome sequence data. OH_, origins of heavy-strand replication; OL_, origins of light-strand replication. b Combined box and swarm plots show the estimated number of copies of the mtDNA-derived fragment within the NUMT in seven families. The middle “box” represents the median, lower and upper quartile of the data. The upper and lower whiskers represent the data outside the middle 50%. The dots represent the informative variants included in the mixed haplotype (Methods, Supplementary Table 4). c Modelling of the estimated variant allelic fraction for a NUMT at different true mtDNA sequencing depths. Modelling was based on whole genome nuclear sequencing (WGS) depths seen in our dataset (35×, 40×, 45× and 50×), and the corresponding variant fraction based on the number of copies of mtDNA-derived fragment within the NUMT. 95% confident intervals are shown for one copy and 20 copies. d Detected variant allelic fraction for each of the seven families related to the true mtDNA per-base sequence depth. Upper and lower symbols show the trend of variant allelic fraction estimated from WGS depth 50× with 20 copies of mtDNA-derived fragments and WGS depth 35× with 1 copy of mtDNA-derived fragment.

Fig. 6. Validation of NUMTs using long-read (Oxford Nanopore PromethION) sequencing.

Example shown is a concatenated mtDNA NUMT detected by both long-read sequencing and short-read sequencing. mtDNA and nuclear genome reference sequences are shown on the top in red and navy. Aligned short reads shown in orange and long reads in light blue. The sequenced genome included concatenated mtDNA sequences (fragment 1: mt 14803-14977 (+) and fragment 2: 12864-12714 (−)) inserted into chr5: 32,338,583. The two mtDNA fragments are coloured in red and plink. The sequences of the long reads aligned to two mtDNA fragments are also highlighted in red and pink. The remaining sequences aligned to nuclear genome in the same region. The IGV screenshot shows both long-read (top) and short-read (bottom) alignments. The insertion point on the nuclear genome is highlighted by the red box. The yellow bars on the long-read alignment are the large insertions from mtDNA sequences. Teal bars on the short-reads alignment are the discordant reads which one read aligned to nuclear genome and their mate reads aligned to mtDNA genome.