STAMP: a multiplex sequencing method for simultaneous evaluation of mitochondrial DNA heteroplasmies and content

Xiaoxian Guo; Yiqin Wang; Ruoyu Zhang; Zhenglong Gu

doi:10.1093/nargab/lqaa065

. 2020 Oct 19;2(4):lqaa065. doi: 10.1093/nargab/lqaa065

STAMP: a multiplex sequencing method for simultaneous evaluation of mitochondrial DNA heteroplasmies and content

Xiaoxian Guo ^1,², Yiqin Wang ^2,², Ruoyu Zhang ³, Zhenglong Gu ^4,^✉

PMCID: PMC7573736 PMID: 33134911

Abstract

Human mitochondrial genome (mtDNA) variations, such as mtDNA heteroplasmies (the co-existence of mutated and wild-type mtDNA), have received increasing attention in recent years for their clinical relevance to numerous diseases. But large-scale population studies of mtDNA heteroplasmies have been lagging due to the lack of a labor- and cost-effective method. Here, we present a novel human mtDNA sequencing method called STAMP (sequencing by targeted amplification of multiplex probes) for measuring mtDNA heteroplasmies and content in a streamlined workflow. We show that STAMP has high-mapping rates to mtDNA, deep coverage of unique reads and high tolerance to sequencing and polymerase chain reaction errors when applied to human samples. STAMP also has high sensitivity and low false positive rates in identifying artificial mtDNA variants at fractions as low as 0.5% in genomic DNA samples. We further extend STAMP, by including nuclear DNA-targeting probes, to enable assessment of relative mtDNA content in the same assay. The high cost-effectiveness of STAMP, along with the flexibility of using it for measuring various aspects of mtDNA variations, will accelerate the research of mtDNA heteroplasmies and content in large population cohorts, and in the context of human diseases and aging.

INTRODUCTION

The human mitochondrial genome (mtDNA) is a 16.6 kb circular genome encapsulated in the inner membrane of mitochondria. It encodes 22 tRNA and 2 rRNA genes used for mitochondrial protein synthesis as well as 13 evolutionarily conserved proteins in four of the five mitochondrial oxidation phosphorylation (OXPHOS) protein complexes (1). Although the primary and well-recognized function of mtDNA is encoding genes for energy production, a broad array of new biological processes and disorders associated with mtDNA have received increasing attention in the past decade (2,3).

Recent deep sequencing studies have shown that mtDNA mutations are much more prevalent in human tissues than previously thought (4,5). Given the multicopy nature of mtDNA in a single cell, mtDNA mutations can arise and co-exist with the wild-type allele in a state called heteroplasmy. mtDNA heteroplasmies can increase in fraction through clonal expansion in cells and tissues (6), without affecting mitochondrial function until their abundance reaches a certain threshold. At an intermediate fraction, a single disease-causing mtDNA mutation may lead to mitochondrial morphological changes and decreased transcription of mtDNA, recapitulating the mild mitochondrial dysfunction in diseases like diabetes and autism (7). At a relatively high fraction, it may induce global changes of gene expression involved in signal transduction, epigenomic regulation and pathways implicated in neurodegenerative diseases (7). Accordingly, the varying fraction and abundance of mtDNA mutations, as well as their tissue sources, may give rise to distinct downstream phenotypes, which poses a challenge for mtDNA studies.

In addition to mtDNA sequence variations, the total number of mtDNA molecules in a cell, known as mtDNA copy number (i.e. mtDNA content), can also impact mitochondrial function (8). Altered mtDNA content in peripheral tissues is frequently reported in patients with neuropsychiatric disorders, and has been shown to be affected by stressful life events (9). Recently, several large-scale prospective studies found correlations between low mtDNA content in blood and age-related chronic diseases, such as cardiovascular diseases, illustrating that mtDNA content can serve as a biomarker for age-related decline of mitochondrial function and a predictor for adverse health outcomes in humans (10,11).

Lately, large-scale population studies on human mtDNA have been facilitated by widely available datasets from genome-wide sequencing projects. Off-target reads from whole-exome sequencing (WES) studies can be used to assess mtDNA sequence variations and medium-fraction heteroplasmies (12,13). Likewise, whole-genome sequencing (WGS) with deep and uniform coverage on mtDNA can be used to measure mtDNA copy number and low-fraction heteroplasmies in tissues (14,15). However, WES and WGS are not cost-effective for investigators to study mtDNA with a large sample size. Even if the genomic datasets have already been produced, analyses of mtDNA are often restricted by their original study design, which rarely allow investigators to study important characteristics of mtDNA, such as the temporal and tissue dynamics of mtDNA heteroplasmies and content.

mtDNA-targeted sequencing is an economical alternative to genome-wide methods. The main strategy is to isolate and enrich mtDNA from the total genomic background, and thus focus the sequencing capacity on mtDNA reads (16–18). These methods normally start with amplification of mtDNA from total genomic DNA (16,17). Alternatively, mtDNA sequencing libraries can be enriched from total genomic sequencing libraries by using hybridization capture baits derived from mtDNA (18). Sequencing libraries containing short mtDNA fragments and adaptors are subsequently generated from the polymerase chain reaction (PCR) products by using commercially available kits for sequencing library preparation, such as Nextra (16,17) and QIAseq (19). However, as these library preparation protocols are optimized for processing large, linear genomic DNA, their use for the short 16.6 kb circular mtDNA would dramatically increase the overall cost of mtDNA sequencing (20). To overcome this limitation, Nunez et al. developed a method to add sequencing adaptors directly to the short DNA fragments with T4 DNA ligase, after PCR amplification of mtDNA (21). Although this method can be applied to human mtDNA at a low cost, it has drawbacks that are similar to other methods, since mtDNA enrichment and library construction involve multiple steps and reaction plates, which incurs extra labor and increases the possibility of sample contamination during DNA purification and transferring.

Moreover, most of these mtDNA-enrichment strategies depend on high numbers of PCR cycles to increase mtDNA content, which inevitably introduce amplification errors and related mutation artifacts (22). These errors can lead to false discovery of mtDNA heteroplasmies, as most of them are of low fraction at a tissue level. A previous study showed that even at a sequencing coverage as high as 20 000-fold (X) on mtDNA, the majority of the mtDNA sites were polymorphic with a variant allele fraction (VAF) of ≥0.1%, most of which could result from PCR or sequencing errors (16,23).

Other methods that, conversely, reduce the level of the nuclear genome by using exonuclease V to digest the linear nuclear DNA (24), or by isolating mitochondria with differential centrifugation or magnetic beads for immunoprecipitation (25) require a large amount of starting materials, which is not suited for large-scale population studies with limited DNA sources or frozen tissue biospecimens. Importantly, by processing only mtDNA, the mtDNA-targeted sequencing methods lose valuable information on mtDNA levels in relation to nuclear DNA in the sample, making them unable to quantify mtDNA content in the same assay. Therefore, a cost-effective, accurate and flexible mtDNA sequencing method is urgently needed, especially for studying mtDNA variations in large populations.

In the current study, we developed a novel mtDNA-targeted sequencing approach called STAMP (sequencing by targeted amplification of multiplex probes) which relies on multiplex single-stranded oligonucleotide probes to capture and subsequently sequence the entire 16.6 kb human mtDNA. Furthermore, we extended STAMP to allow mtDNA content quantification in the same assay. Here, we discuss the design, workflow, cost-effectiveness and best practices of using STAMP for studying mtDNA heteroplasmies and content.

MATERIALS AND METHODS

Study samples

Two HapMap lymphoblast cell lines (sample 1: NA12751 and sample 2: NA18523) were purchased from Coriell institute. Upon receiving them, the lymphoblast cell lines were revived and cultured, at 37°C with 5% CO₂, in RPMI 1640 medium containing 15% fetal bovine serum (VWR Life Science Seradigm, Inc.) and 1 × Antibiotic-Antimycotic (Thermo Fisher Scientific, Inc.). Total genomic DNA of these two samples was obtained using Wizard Genomic DNA Purification Kit (Promega, Inc.) as per the manufacturer's instructions. The concentration of purified DNA was quantified by using a Qubit dsDNA HS assay kit (Thermo Fisher Scientific, Inc.). The five DNA sample mixtures were created by combining total genomic DNA of these two HapMap samples at relative ratios of 1:199, 1:99, 5:95, 20:80 and 50:50 (NA12751 versus NA18523). The lymphoblast cell line samples from 200 healthy individuals were collected in REGISTRY (26), and DNA were extracted as per REGISTRY protocol (26). Consent were obtained from all individuals enrolled in REGISTRY at site (26).

Design of STAMP

We designed single-stranded oligonucleotide probes to capture human mtDNA with an extension-ligation (EL) reaction (Figure 1). The extension probe has three parts, a 3′ extension arm with sequence complementary to the mtDNA target, a unique molecular tag used for tracking the capturing event, and a common PCR primer annealing region used for PCR amplification of the captured target. The ligation probe has a 5′-phosphorylated ligation arm with sequence complementary to the mtDNA target, along with another 20-nt common PCR primer annealing region at its 3′ end (Figure 1B).

To identify the required number of EL probe pairs and their mtDNA target locations, we first performed BLASTN search of human mtDNA sequences against the latest human reference genome (assembly GRCh38). We required that the resulting mtDNA segments be distinguishable from high-similarity segments derived from the human nuclear genome (Supplementary Tables S1 and 2). Given the maximum sequencing read length of available Illumina sequencing platforms, we also required that the lengths of mtDNA targets should be around 400bp, so that they can be fully sequenced by using 2 × 250 or 2 × 300 paired-end reads, while the overlap between the paired-end reads is minimal.

As a result, we found that the entire 16.6 kb human mitochondrial genome could be captured by using as low as 46 pairs of EL probes. We then placed the pairs of EL probes on the heavy and light strands of mtDNA alternatingly, to minimize the physical interference of adjacent probes in the multiplex reaction (Figure 1A). The locations and lengths of the extension and ligation arms in each of the EL probe pairs were further adjusted to ensure similar melting temperatures, around 55°C and similar GC-content, around 50%, and to avoid overlap with common mtDNA polymorphisms (population frequency > 1%) at the 3′ end of the extension arm and the phosphorylated end of the ligation arm (Figure 1B). In the end, we obtained 46 pairs of EL probes with a mtDNA target size ranging from 400 to 450 bp to capture human mtDNA (Supplementary Table S1).

mtDNA sequencing with STAMP

The experimental and computational protocols of STAMP used in the current study for sequencing mtDNA are listed as follows:

Step 1. mtDNA capture with extension-ligation (EL) probes

The oligos of EL probe pairs for each of the 46 mtDNA target regions and five nDNA target regions were column-synthesized, at 25 nanomole scale with standard desalting purification (Integrated DNA Technologies, Inc.). Equal amount of the 51 EL probe pairs were first pooled for mtDNA capture. In order to achieve uniform sequencing coverage on mtDNA, we repeated our pilot experiment from step 1 to step 5 with a different probe pool. In each round, we adjusted the relative ratios of mtDNA probes in the probe pool according to the number of consensus reads obtained in the last round until the number of consensus reads from almost all mtDNA (>98%) probes surpassed 20% of the mean coverage. Hybridization reactions were performed on 50 ng genomic DNA with 4 µl EL probe mix and 1 × Ampligase buffer in a 10 μl volume. Thermal conditions included 10 min at 95°C for denaturation, followed by a decrease of 1°C per min to 55°C and 20 h at 55°C for hybridization. A total of 6 μl gap-filling mix including 0.1 mM dNTPs, 0.6M Betaine, 0.1M (NH₄)₂SO₄, 0.5 units of Tsp DNA polymerase and 0.5 units of Ampligase in 1 × Ampligase buffer was then added to the reaction mixture, which was incubated at 55°C for another 20 h for gap filling. The time for hybridization and gap-filling was determined empirically to ensure effective capture of DNA by multiplexed probes, which is in good agreement with other hybridization-based approaches (27,28).

Step 2. PCR amplification of capture products

We used a dual indexing strategy to pool sample libraries for parallel sequencing (29). Each indexing primer comprised P5 or P7 Illumina adapter sequences, an 8-nt index sequence, a 13- or 14-nt pad sequence, and a universal sequence designed to anneal with the common termini of EL probes (Figure 1B) (29). PCR amplification was performed on 1.5 μl of capture product in a 50 μl PCR reaction with 1 × Phusion HF buffer, 0.5 μM of p5i5 and p7i7 indexing primers, 0.2 mM dNTP and 1 unit of Phusion Hot-Start II DNA polymerase (Thermo Scientific, Inc.). PCR thermal conditions were 30 s at 98°C for initial denaturation, followed by 25 cycles of 10 s at 98°C, 15 s at 65°C and 15 s at 72°C. The size and integrity of PCR products were visually verified by agarose gel electrophoresis.

Step 3. Library purification

The obtained PCR products were purified and filtered by using AMPure XP magnetic beads with double size selection (Beckman Coulter, Inc.). In brief, 0.25 volume of beads was first used to bind DNA of >700 bp in the PCR products, after which the supernatant was transferred to a fresh tube. An extra 0.4 volume of beads was added to bind DNA of >500 bp in the supernatant. After the beads were washed and dried, DNA bound to these beads which contained PCR products of size in the range of 550–650 bp were eluted with 10 mM Tris–HCl, pH8.5 and were quantified with Qubit^® 2.0 Fluorometer (Life Technologies, Inc.). Equal amounts of purified PCR products from different samples were pooled before parallel sequencing.

Step 4. Massively parallel sequencing

The sample libraries were sequenced with customized sequencing primers and 2 × 250 paired-end reads on Illumina sequencing flow cells. The Read 1 primer contained the13-nt pad sequence and the 20-nt universal sequence (TGCACGTCATCTACAGTAGGTCGGTGCGTAGGT), reverse complement of the sequence at the 3′ terminus of the ligation probe. The Read 2 primer contained the 14-nt pad sequence and the 20-nt universal sequence (CTCACTGGAGTTCAAGGGACGATGAGTGGCGATG) of the extension probe. The Index primer was the reverse complement of the Read 2 primer sequence (CATCGCCACTCATCGTCCCTTGAACTCCAGTGAG), which along with the complementary adapter sequences on the flow cell was used to read the dual sample indices (29). Cluster generation, image processing and sequencing for samples of the current study were processed on MiSeq or HiSeq 2500 in the rapid run mode. Phi-X DNA library was spiked in at 5% to increase the complexity of the STAMP sequencing libraries.

Step 5. Sequencing data processing

A Python pipeline was developed to process and align paired-end reads generated from STAMP (Supplementary Note B). In brief, paired-end reads were first demultiplexed into files of individual samples based on the i5 and i7 index sequences. For each individual sample, paired-end reads were sorted into 51 clusters of capture products according to the arm region sequences identified at the locations of EL probes. The arm region sequences and the molecular barcode were trimmed from the paired-end reads, which were recorded in the read alignment files as annotations. To minimize complications from NUMTS in mtDNA read alignment, paired-end reads were first aligned to the reference human genome containing both nuclear DNA (genome assembly GRCh 38) and mtDNA (Revised Cambridge Reference Sequence, rCRS) sequences downloaded from ftp://ftp.1000genomes.ebi.ac.uk, using Burrows-Wheeler aligner (bwa mem, version 0.7.17) (30). Paired-end reads, annotated as having one of the 46 mtDNA EL probes, were marked as potential NUMTS in the alignment file if they could also be aligned to nuclear DNA with MAPQ ≥ 10. Paired-end reads were aligned in a second round to a modified version of rCRS which had the final 120 bp copied to the start to accommodate alignment of D-loop-region reads with the D10 probes. Paired-end reads that could not be aligned to the target region specified by their arm region sequences were removed. The remaining reads were locally realigned by using freebayes (version 1.1.0) (31) and their base qualities were subsequently recalibrated by using samtools (version 1.6) (32).

For paired-end reads with the same molecular barcode, the base information called at corresponding sites of the alignments was merged by using a Bayesian approach to generate a consensus read representing the captured mtDNA product. The same method was also used to merge base information within the overlapping region of the paired-end reads. The sequences of consensus reads were compared to a collection of known NUMTS sequences in the reference genome obtained from BLASTN search of the 46 mtDNA segments captured with EL probes, as well as their variant sequences harboring common polymorphisms (minor allele fraction > 1%) identified in the 1000 Genomes project (Supplementary Table S2). This list of NUMTS sequences can be further extended to include sequences of rare or low-frequency, polymorphic NUMTS identified in large populations (33). A consensus read was marked as potential NUMTS if it showed a lower pairwise edit distance to NUMTS sequences than to the sample's major mtDNA sequence, or if it was constructed from paired-end reads already annotated as NUMTS according to BWA alignment. Finally, consensus reads were converted to single-end reads, along with their base quality information and stored in a bam file for each individual sample.

mtDNA variant detection

mtDNA variants were determined by using consensus reads with MAPQ ≥ 20 and BAQ ≥ 30. Consensus reads marked as NUMTS or showing an excess of mismatches (>5 in the coding region and >8 in the D-loop region) compared to the individual's major mtDNA sequence were also excluded from analysis. Because the maximum number of nucleotide difference between sample 1 and sample 2 in our pilot study is 11 (probe D9, Supplementary Table S5), we retained consensus reads with mismatches up to 11 in the analysis of data from sample mixtures. To reduce false positive calls of heteroplasmies, variants were subject to a list of quality filters, including (i) ≥100× depth of coverage with ≥70% of the bases having BAQ ≥ 30; (ii) not in low-complexity regions (nt 302–316, nt 512–526, nt 16184–16193) or low-quality sites (nt 545, 16224, 16244, 16249, 16255 and 16263); (iii) ≥5 minor alleles detected; (iv) a log likelihood quality score of the variant ≥5; (v) comparable VAFs (Fisher's exact test P ≥ 10⁻⁴ and fold change ≤5) computed using consensus reads constructed with or without duplicate paired-end reads; (vi) the detected number of minor alleles significantly larger than the expected number of errors, which was estimated at a rate of 0.02% in STAMP (Exact Poisson test, P < 0.01/16569).

Functionalities for raw read processing, paired-end read alignment, consensus read calling and variant detection have been implemented in the STAMP tool kit. The list of commands of the STAMP tool kit, arguments and example code can be also found in Supplementary Note B.

mtDNA content evaluation with quantitative PCR

The relative mtDNA content of 126 lymphoblast samples, with enough genomic DNA after STAMP sequencing, was measured by using a quantitative PCR-based assay (34). The PCR reactions were performed as per manufacturer's instructions (The Detroit R&D, Inc.). In brief, 15 ng total genomic DNA was amplified with mtDNA or nDNA target primers and SYBR green PCR master mix in a 20 μl PCR reaction. Thermal conditions included 10 min at 95°C, followed by 40 cycles of 15 s at 95°C and 60 s at 60°C. For each sample, both mtDNA and nDNA targets were amplified twice in a total of four PCRs. Results from duplicates were averaged to compute mean Ct values for mtDNA and nDNA targets. The differences between them (ΔC_T) were then normalized to that of a positive control sample measured on the same 96-well plate, by using ΔΔC_T method, to obtain qPCR-CN. qPCR-CN from 10 samples that failed in any of the four PCRs, and/or had a difference in C_T values of over three cycles between experimental duplicates, were excluded from analysis.

Statistical analysis

All statistical analyses were carried out with R (v3.5.0). The methods used and two-tailed P-values are indicated in the main text.

RESULTS AND DISCUSSION

Effective capture of mtDNA with EL probes

In a pilot experiment, we synthesized the set of 46 pairs of EL probes (Integrated DNA Technologies, Inc.). We performed enzymatic gap-filling and ligation reactions on 50 ng genomic DNA extracted from a lymphoblast cell line sample from the HapMap project, with 115 femtomoles of EL probe mixture. PCR amplification of the captured targets, using the 20-nt common PCR primers, requires the presence of PCR primer annealing regions at both ends, after successful polymerization of nucleic acids between the hybridized EL probes, as well as ligation of the polymerized nucleic acids with the 5′ end of the ligation arm. Therefore, captured products which lacked one of the common primer sequences, due to failed hybridization of either probe with its target sequence, or no ligation at the 5′ end of the ligation arm, could not be amplified.

Without adding genomic DNA templates or necessary enzymes, no clear bands were observed in the gel electrophoresis of PCR products from the gap-filling reactions (Figure 2A). In contrast, gel electrophoresis of PCR products from the effective gap-filling reaction on genomic DNA exhibited a smear, which had a size distribution centered at about 550–600 bp, reflecting the expected sizes of PCR products comprising the target DNA, the molecular barcode, the probe arms, the common primers and the sequencing adapters (Figure 2A).

Figure 2. — Effective capture of mtDNA with EL probes. (A) Agarose gel electrophoresis of the captured products: molecular-weight markers (lane M), no genomic DNA (lane 1), no gap-filling enzymes (lane 2), genomic DNA from a HapMap lymphoblast sample (lane 3), genomic DNA from 143B.TK⁻ ρ° cell line (lane 4), genomic DNA from 143B.TK⁻ cell line (lane 5) and captured product after purification with double size selection (lane 6). (B) The relative depth of coverage of consensus reads on mtDNA for each of the 46 regions captured by EL probes (from A1 to D10 listed in Supplementary Table S1). The blue dotted line and red dashed line indicate 50 and 20% of the mean sequence coverage, respectively.

Next, we purified PCR products with AMPure XP magnetic beads (Beckman Coulter, Inc.) and sequenced them with 2 × 250 bp paired-end reads on MiSeq (Illumina, Inc.). After mapping the paired-end reads to the reference human genome, we found that all of the 46 mtDNA target regions were covered with reads at an average sequencing depth of 3512×, confirming that the set of 46 pairs of EL probes was able to capture the full length of human mtDNA (Supplementary Table S3 and Figure 2B). We further replicated these results in a second lymphoblast sample from the HapMap project (Supplementary Table S3 and Figure 2B). There was also a high correlation of coverage of regions captured by the same EL probe pair between these two samples (Pearson's r = 0.8, P = 3.3 × 10⁻¹¹).

Moreover, capturing mtDNA sequences from genomic DNA could be potentially biased by the presence of nuclear DNA regions with high sequence similarity to mtDNA (i.e. nuclear mitochondrial segments, NUMTS) (35). To reveal its influence on mtDNA capture in STAMP, we performed gap-filling and ligation reactions with the set of 46 EL probes on genomic DNA from 143B.TK⁻ mtDNA-less (ρ0) cell line, and compared its PCR products with those obtained from the parental cell line 143B.TK⁻ which contains mtDNA. The gel electrophoresis of corresponding PCR products showed that the characteristic smear was only detectable in 143B.TK⁻ cells, but was absent in 143B.TK⁻ ρ0 cells, indicating that the resulting PCR products were largely derived from captured mtDNA target regions rather than off-target NUMTS (Figure 2A).

Utilizing molecular barcodes to remove PCR duplicates and improve sequencing accuracy

Most mtDNA heteroplasmies are at low factions at the tissue level, which require a high depth of coverage of reads to reveal the presence of the variant allele and assess their fraction in relation to the wild-type allele. However, an ultra-deep read coverage (i.e. >2000×) on the small 16.6Kb human mtDNA may create errors in removing read duplicates if they are solely determined based on alignment coordinates and read sequences (36). PCR amplification of mtDNA before sequencing may also introduce biases in estimating VAF of a heteroplasmy (37). To resolve these issues, in STAMP, a 12- or 15-random-nucleotide molecular barcode was incorporated via the EL probe pairs to each of the capture products before PCR amplification, creating an identity for each capturing event (Supplementary Table S1). Therefore, paired-end reads from the same mtDNA fragment captured in STAMP, including duplicates, can be determined according to the attached barcode information (Figure 1B).

Moreover, nucleotide mismatches at corresponding sites of paired-end reads with the same molecular barcode would suggest either PCR artifacts or sequencing errors (Figure 1B). We employed a Bayesian approach to merge the base information of paired-end reads with identical molecular barcode, generating a consensus read representing the captured DNA fragment (Supplementary Note B). We found that the number of nucleotide mismatches between the consensus read and the reference mtDNA sequence significantly decreased after merging base information of paired-end reads (Kolmogorov–Smirnov test, P < 2.2 × 10⁻¹⁶, Figure 3A). Consensus reads with an excess of mismatches (number of mismatches, NM > 5) in comparison to the reference mtDNA were almost undetectable if they were constructed with duplicate paired-end reads (multiple reads with identical molecular barcode), which is 30-fold less frequent than those of consensus reads constructed from single paired-end read (Chi-squared test, P < 2.2 × 10⁻¹⁶).

Figure 3. — Accurate detection of mtDNA variants in sample mixtures. (A andC) The numbers of nucleotide mismatches between consensus reads and the reference mtDNA sequence in (A) sample 1 and (C) sample 2. (B andD) The proportions of variant alleles per base in the consensus reads after filtering out low-quality bases (BAQ < 30) and consensus reads with excessive mismatches (NM > 5) in (B) sample 1 and (D) sample 2. #PE-reads: the number of paired-end reads used to construct the consensus read. The VAFs of the mtDNA variants detected in the mixtures of sample 1 and sample 2 were depicted in E for variants at the 58 polymorphic sites and in F for variants at the five heteroplasmic sites. Each dotted line in E and F refers to the VAF changes of one variant in relation to the sample proportion indicated by the values on the x-axis. Both x- and y-axes in E are shown on a log scale, and VAFs are shown in relation to mtDNA of sample 1,

After filtering out consensus reads with an excess of mismatches (NM > 5) and bases with low quality scores (BAQ < 30), we found that the proportion of variant alleles, which encompassed PCR and sequencing errors as well as low-level heteroplasmies, was about 0.013 and 0.03% per base among consensus reads with and without duplicate paired-end reads, respectively (Figure 3B). Both values are considerably lower than the reported proportions of ∼0.1% per base in commonly-used mtDNA-targeted sequencing methods (38,39). The proportion of variant alleles, at 0.013% per base, is also close to the error rate of Tsp DNA polymerase used in the gap-filling reaction of STAMP, which can be further improved using DNA polymerase with a higher fidelity (40). We obtained similar reductions in the distribution of nucleotide mismatches, and variant allele proportions, in the consensus reads of the second sample (Figure 3C and D).

Accurate detection of mtDNA heteroplasmies and their fractions in total genomic DNA

To evaluate the sensitivity and specificity of STAMP for detecting mtDNA heteroplasmies and their fractions in total genomic DNA, we applied STAMP to a series of sample mixtures created by combining total genomic DNA from the two lymphoblast samples used in the pilot experiment at varying ratios, ranging from 1:199 to 1:1 (Supplementary Table S3). mtDNA sequences of these two lymphoblast samples differ at 59 single nucleotide sites (Supplementary Table S4). One site (nt 16 189) was in a low-complexity poly-C region of mtDNA and was excluded from the analysis of heteroplasmies. The average mean depth of coverage of consensus reads on mtDNA among the five sample mixtures was 3938× (median depth: 3284×), comparable to that of the two original samples at 3988× (median depth: 3392×; Supplementary Table S3).

We found that all 58 polymorphic sites exhibited changes in their VAFs in accordance with the ratios of DNA used to generate these sample mixtures (r > 0.997, P < 0.00017; Figure 3E and Supplementary Table S4). Of note, two polymorphic sites were located within the annealing regions of the EL probes: nt 16 519 at the second position from the 3′end of the extension arm of probe D10, and nt 13 650 at the 10th position from the 3′end of the extension arm of probe D1. These two polymorphisms did not abolish the hybridization of EL probes to the target regions (for reasons discussed in the next section), but the variance in the VAFs of the 18 heteroplasmies close to the annealing regions of probes D1 and D10 was increased compared to that of the other heteroplasmies, especially in low-fraction sample mixtures (two-sided F-test P < 1.4 × 10⁻⁵, Figure 3E). This suggests that a nucleotide mismatch between the arm regions of EL probes and their annealing regions in mtDNA can alter DNA capture efficiency, affecting the estimation of VAFs of nearby heteroplasmies in linkage with the nucleotide mismatch.

However, our pilot experiment using the sample mixtures represents an extreme scenario where all the 58 polymorphisms are in complete linkage with each other, and their alleles are separable into two haplogroups of mtDNA. In real human samples, the incidence of medium- and high-fraction heteroplasmies is usually low (41), and new heteroplasmies tend to arise in different mitochondria (42). Therefore, the variant and wild-type alleles of a heteroplasmy tend to share the same flanking mtDNA sequences. Both alleles would be captured by EL probe pairs at the same rate.

Next, we examined the influence of applying rigid quality control filters on detecting low-fraction mtDNA variants in the three sample mixtures, created with genomic DNA ratios of 1:199, 1:99 and 5:95. We found that 5 out of the 174 (3 × 58) variants were unable to survive the quality filtering procedures described in ‘Materials and Methods’ section. Of these, three showed a low percentage of high-quality reads and two were located at sites that did not have sufficient reads containing the variant alleles (Supplementary Table S5). Moreover, all the 174 variants had VAF > 50% of the ratios of the genomic DNA in the sample mixtures (Supplementary Table S4). Therefore, STAMP’s sensitivity of identifying low-fraction heteroplasmies (VAF = 0.5–5%) is over 97%, with a cutoff for VAF at 0.25% and an average coverage of consensus reads at about 4000×.

By using these quality control filters, we found 17 other mtDNA heteroplasmies at VAF ≥ 0.25% (Supplementary Table S5). All of them were located at the five heteroplasmic sites already detected in one of the two original mtDNA samples, at a VAF from 0.4 to 2.9% (Supplementary Table S4). These five heteroplasmic sites also displayed VAF changes proportional to the ratios of the DNA in the sample mixtures (r > 0.9, P < 0.0046; Figure 3F). Therefore, the false positive rate of STAMP in detecting heteroplasmies of VAF ≥ 0.25% is under 10⁻⁴ (1/16 569) per site of mtDNA.

Application of STAMP in a population study of mtDNA heteroplasmies

To demonstrate the effectiveness and robustness of using STAMP for assessing mtDNA heteroplasmies in larger numbers of samples, we used STAMP to sequence mtDNA in 200 lymphoblast samples collected in REGISTRY (26). These 200 lymphoblast samples constituted the healthy control group of our research project on Huntington's disease, and were sequenced on HiSeq2500 along with other samples relating to this project (sequencing reads of all samples and technical replicates available at European Genome-Phenome Archive under study accession number EGAS00001004092).

We were able to generate sequencing libraries for 192 (92%) out of 208 samples, including the experimental replicates from 8 lymphoblast samples. Among the 192 samples with STAMP libraries, 190 (99%) libraries from 182 lymphoblast samples were sequenced to >1000× depth of median coverage of consensus reads on mtDNA. The average median and mean depths of coverage of consensus reads on mtDNA were 4580× and 5450×, respectively (Figure 4A). The percentages of mtDNA sites that were covered with over 20 and 50% of the mean depth of coverage of consensus reads, as indicators for read coverage uniformity, were 98 and 78%, respectively (Figure 4B and C). About 99.4% mtDNA sites were covered with >500 consensus reads and 96.3% were covered with >1000 consensus reads. Similar to the observations from the two lymphoblast samples of the HapMap project, the overall proportions of variant alleles were 0.011 and 0.031% per base of consensus reads constructed with and without duplicate paired-end reads, respectively (Figure 4D). Taken together, these results confirm that STAMP can effectively sequence mtDNA in larger-scale studies.

Figure 4. — Reliable sequencing of mtDNA in a population study. Results of STAMP sequencing performed on 182 lymphoblast samples of REGISTRY are shown. (A) Median depth of coverage of consensus reads on mtDNA used for calling mtDNA variants. (B and C) Proportions of mtDNA sites with depths of consensus read coverage greater than (B) 0.2 and (C) 0.5 times the mean value, respectively. (D) Proportions of variant alleles per base in the consensus reads used for calling mtDNA variants. (E) Correlations of VAFs of 45 mtDNA heteroplasmies identified in the STAMP replicates performed on eight samples. (F) Correlations of the relative mtDNA content measured by using STAMP and qPCR. (G) The box plots of mtDNA heteroplasmies detected in lymphoblasts of individuals aged above and under the sample median of 48 years old. (H) The box plots of the relative mtDNA content (STAMP-CN) detected in lymphoblasts of individuals aged above and under the sample median of 48 years old. VAFs in E were depicted using the major allele in replicate 1 as the reference allele.

Of note, two low-frequency mtDNA polymorphisms (nt 2626 and nt 15 758) were identified at the 3′ end of EL probes A6 and D7 in two samples. These two mtDNA polymorphisms are single base transitions from A to G or T to C which give rise to purine-pyrimidine (A-C, C-A, G-T and T-G) mismatches between the mtDNA templates and the arm regions of the EL probes. Since purine-pyrimidine mismatches in the 3′ end regions of primers have less detrimental impacts on PCR amplification than purine-purine or pyrimidine–pyrimidine mismatches (43), these two probes still produced 1248 and 268 consensus reads in the corresponding samples. These reads accounted for 0.3 and 0.4% of total consensus reads obtained, or equivalently 17 and 25% of the average proportions of consensus reads captured by EL probes A6 and D7 from mtDNA, respectively (Supplementary Figure S1).

Since >90% of known polymorphisms, as well as heteroplasmies, in mtDNA cause single base transitions and are outside the 3′ end regions of EL probes (44), the current design of EL probes can be applied to capturing the entire length of most mtDNA haplogroups worldwide. However, any population other than Europeans, that were investigated in the current study, may lead to different coverage of consensus reads on mtDNA (Supplementary Figure S1) if there are ethnicity-specific common polymorphisms or substitutions located in the arm regions of EL probes. Therefore, a pilot experiment of STAMP, aiming to identify appropriate molar ratios of EL probes that can ensure read uniformity on mtDNA for a population of interest, may be needed before large-scale applications. We further list all possible common polymorphisms (population frequency > 1%) located in the arm regions of EL probes in Supplementary Table S1. Modification of probe sequences to include degenerate bases at some or all polymorphic sites listed in Supplementary Table S1 will help improve capture efficiency and robustness when STAMP is applied to mtDNA of varying populations worldwide in the future.

Among the eight lymphoblast samples with replicated STAMP measures, we found that the major mtDNA sequences were identical between sequencing replicates. All the 45 heteroplasmies identified with VAF ≥ 1% in one sample had VAF ≥ 0.5% in the replicate, 44 (98%) of which also passed all quality control filters for heteroplasmy calling (Supplementary Table S6). Overall, the correlation between VAFs of the heteroplasmies detected in sequencing replicates was estimated to be r = 0.998 (P = 3 × 10⁻⁵³, Figure 4E). The average coefficient of variation in repeated measures of VAFs, at median site coverage of 4330×, was 5% for medium-to-high-fraction heteroplasmies (VAF ≥ 5%; median VAF = 15%) and was 17% for low-fraction heteroplasmies (0.5% ≤ VAF < 5%; median VAF = 1.4%). These values are close to the corresponding estimates of 4 and 13% computed using the sampling distribution of sample proportions at the VAF medians. These results indicate that STAMP can reliably detect heteroplasmies and quantify their fractions in genomic DNA.

Extension of STAMP to measure mtDNA content in the same assay

We further explored the possibility of modifying STAMP to enable mtDNA content quantification in the same assay. To this end, we added five pairs of EL probes to capture single-copy regions in nuclear DNA (nDNA), along with the 46 pairs of mtDNA EL probes in STAMP. These five target nDNA regions are located on different autosomal chromosomes (Figure 1A). Assuming that the amount of DNA captured by each pair of EL probes is proportional to the number of copies of the target region in the human genome, reads from the nDNA regions then can be used as a normalization factor to adjust differences in total genomic DNA input and sequencing coverage across samples.

We first evaluated the performance of the nDNA EL probe pairs in capturing their target regions relative to the mtDNA EL probe pairs. We have noted in the previous analyses that the presence of polymorphisms in the arm regions of the EL probes could influence the capture efficiency of the target region. We thus focused on a subset of the 46 EL probe pairs to compute an average number of consensus reads for mtDNA. This subset comprised 18 EL probe pairs (A5-A8, B2, B6, B7, B9, C1-C5, C7, C9, C12, D1, D5) that lack common polymorphisms in their arm regions in European populations, and showed relatively low variations in consensus read coverage across samples of the current study (Supplementary Table S1 and Figure S1).

We found that all the five nDNA probe pairs exhibited positive correlations in their consensus read numbers with that of mtDNA EL probe pairs (R² = 0.4–0.79, Supplementary Figure S2A–E). To improve reliability in estimating nDNA content in the sample, we used the three EL probe pairs targeting chromosomes 8, 14 and 19 with an R² ≥ 0.74 to compute an average number of consensus reads for nDNA. In addition, we found that the performance of EL probe pairs was not equal when capturing nDNA compared to mtDNA target regions (Supplementary Figure S2A–E), possibly due to a compact design of EL probes on mtDNA. To adjust for this difference, we computed the relative mtDNA content for STAMP (hereafter referred to as STAMP-CN) as the average consensus read number from mtDNA relative to that from nDNA, by using the equation: log₂(No. of mtDNA consensus reads)–C × log₂(No. of nDNA consensus reads). C in the equation stands for the normalization factor for nDNA consensus reads, estimated using the coefficient β from the regression of log₂(No. of mtDNA consensus reads) against log₂(No. of nDNA consensus reads), which was equal to 0.53 (Supplementary Figure S2F).

By comparing STAMP-CN with the relative mtDNA content (hereafter referred to as qPCR-CN) determined using a commercially available quantitative PCR-based assay performed on the same sample (45), we found a significant positive correlation (r = 0.54, P = 10⁻⁵, Figure 4F) between the values of STAMP-CN and qPCR-CN. This correlation is in good agreement with the result reported in a previous comparative study of mtDNA content measured by sequencing-based methods and qPCR-based methods (45). Because STAMP estimates the amount of mtDNA by directly counting reads from multiple regions in nDNA and mtDNA, STAMP-CN may be more accurate in reflecting real mtDNA content than that quantified based on fluorescence intensity from qPCR assays targeting single nDNA and/or mtDNA regions.

Age-dependent increase of heteroplasmy incidence in lymphoblast samples

With both mtDNA heteroplasmies and mtDNA content measured in the same sample, we then evaluated how age impacts these mtDNA characteristics in lymphoblast samples as compared to what we previously observed in mtDNA of blood samples, using the WGS dataset of the UK10K project (14).

We identified 1007 heteroplasmies of VAF ≥ 1% across the entire length of mtDNA in 182 lymphoblast samples of REGISTRY (Supplementary Figure S3A). The average number of heteroplasmies per sample was 5.5 (range: 0–15). A total of 180 (99%) lymphoblasts possessed at least one heteroplasmy in mtDNA (Supplementary Figure S3B). Similar to our previous study on lymphoblast mtDNA (5), the number of mtDNA heteroplasmies identified in lymphoblasts was consistently greater than those of whole blood, at about one heteroplasmy of VAF ≥ 1–2% (14,16), implying that mtDNA of lymphoblasts may enrich for pre-existing variants in somatic cells that are undetectable at a tissue level.

At the variant level, 760 (88%) of the 862 identified heteroplasmic sites were unique to one of the 182 lymphoblast samples Supplementary Figure S3C). Over half (54%) of the heteroplasmic sites did not overlap with known mtDNA polymorphisms (a population frequency < 0.01%), and another 20% were found to overlap only with rare polymorphisms in less than 0.1% of the general population (Supplementary Figure S3D). The base changes of heteroplasmies showed a high transition to transversion ratio at 15. This suggests that the dominant mutational force underlying heteroplasmies is nucleotide misincorporation by polymerase gamma or deamination of bases in mtDNA, consistent with mtDNA mutation patterns identified in blood (46).

We first performed Student's t-test to compare mtDNA heteroplasmies and content between individuals aged above and under the sample median of 48 years old. We found increased mtDNA heteroplasmy incidence and decreased mtDNA content in lymphoblast samples in the older group (mean age: 55 years old) compared to those in the younger group (mean age: 41 years old). But only mtDNA heteroplasmy incidence showed a significant difference between these two age groups (Figure 4G; Cohen's d = 0.42, P = 0.0055). The lack of a significant association of mtDNA content with age (Figure 4H; Cohen's d = −0.12, P = 0.44 for STAMP-CN; Cohen's d = −0.07, P = 0.71 for qPCR-CN) might be due to insufficient statistical power, since only a mild annual decrease of 0.4 mtDNA copies (0.24% of the population average in 1511 individuals, P = 0.0097) in blood was noted in our previous study (14). Furthermore, cell culture condition might affect mtDNA content (47), which overwrites the real association between age and mtDNA copy number that was observed in blood.

By using a Poisson regression model in the analyses of heteroplasmy incidence, we found an annual rate of increase of 1.2% (95% CI: 0.5−2.0%; P = 0.00063) for heteroplasmies of VAF ≥ 1% in lymphoblast samples (Table 1, model 1). In line with our previous findings in blood (14). the increase of heteroplasmy incidence in lymphoblast samples with age was not affected by changes in mtDNA content (Table 1, model 2). Moreover, we found a similar age-dependent increase of heteroplasmy incidence after focusing on unique variants detected in our dataset, meaning that random genetic events in mtDNA, such as replication errors or drift, are largely responsible for the accumulation of mtDNA heteroplasmies during aging (Table 1, model 3). Significant age effects were also obtained using heteroplasmies of higher VAFs (VAF ≥ 2% or VAF ≥ 5%, Table 1).

Table 1.

Age-dependent increase of mtDNA heteroplasmy incidence in lymphoblast samples

	Model 1		Model 2		Model 3
mtDNA heteroplasmy	Beta [95% CI]	P	Beta [95% CI]	P	Beta [95% CI]	P
VAF ≥ 1%	0.012 [0.005–0.020]	0.00063	0.012 [0.005–0.019]	0.00073	0.014 [0.005–0.022]	0.0012
VAF ≥ 2%	0.017 [0.007–0.027]	0.00087	0.017 [0.007–0.026]	0.00091	0.020 [0.009–0.031]	0.00043
VAF ≥ 5%	0.027 [0.013–0.042]	0.00025	0.027 [0.013–0.042]	0.00026	0.037 [0.020–0.054]	2.7 × 10⁻⁵

Open in a new tab

The associations between heteroplasmy incidence and age were computed by using Poisson log-linear model with adjustment for sex and sequencing coverage in model 1, and for sex, sequencing coverage, and the relative mtDNA content (STAMP-CN) in model 2 and model 3. In model 3, only heteroplasmies that occurred once in the 182 lymphoblast samples were used to compute the incidence.

The observed rate of increase of heteroplasmy incidence in lymphoblast samples is in line with the estimate in our previous study of blood, showing a 59% increase in the heteroplasmy incidence between individuals in the youngest and oldest age groups, spanning an average of 44 years (14). These results indicate that the age-dependent accumulation of mtDNA heteroplasmies may be conserved in lymphoblast samples, after immortalization of B lymphocytes by Epstein–Barr virus and short cultivation of the cell lines. Therefore, lymphoblast samples may serve as a useful genetic resource for studying age-related mtDNA mutation spectra in the hematopoietic system, and their contributions to mitochondrial dysfunction in diseases associated with aging.

CONCLUSION

In the current study, we present a novel human mtDNA targeted sequencing method, STAMP, which enables simultaneous assessment of mtDNA sequence variations and mtDNA content. Our method streamlines the experimental workflow with multiplex capture of human mtDNA and nDNA, and generates high-quality sequencer-ready libraries. This novel methodology eliminates the error-prone steps of transferring reagents and DNA samples, reduces the risk of DNA contamination and enables mtDNA sequencing in thousands of samples.

Importantly, with high cost-effectiveness (Supplementary Figure S4 and Table S7) and a flexible design, STAMP can be used to study mtDNA variations at different scales and to determine mtDNA heteroplasmies at different fraction levels (Supplementary Table S8; discussed in Supplementary Note A). Given the 0.01–0.03% error rates of STAMP (Figure 3B and D), STAMP can be used to detect heteroplasmies of fractions as low as <0.5%, with deeper sequencing coverage (Supplementary Figure S5). Thus, STAMP can be used in studies of somatic mtDNA mutations in tissue specimens, which is currently unachievable by using other mtDNA-targeted sequencing methods, or whose experimental cost is prohibitive, when a large number of samples need to be sequenced.

Furthermore, we provide in the current study the related experimental details and computational solutions to assist the application of STAMP in future human mtDNA studies. Accordingly, the insights gained from these studies will transform our understanding of the role of mtDNA in aging and age-related diseases of humans.

DATA AVAILABILITY

The STAMP tool kit used for processing sequencing reads from STAMP is available for download at https://github.com/mtstamp/stamp.

Paired-end reads and mtDNA consensus reads of REGISTRY samples have been submitted to the European Genome-Phenome Archive under the EGA accession number EGAS00001004092.

Supplementary Material

lqaa065_Supplemental_File

lqaa065_supplemental_file.docx^{(1.3MB, docx)}

ACKNOWLEDGEMENTS

We thank NIH, the CHDI Foundation and the European Huntington's Disease Network for their support in this research project. We thank Andrew Cheung for assisting in mtDNA sequencing experiments. We also thank Drs Chunyu Liu, Qi Sun, Paul Soloway, Haiyuan Yu and Yiping Wang for critical reading and comments on the manuscript.

Author Contributions: X.G. and Z.G. designed STAMP with inputs from Y.W. and R.Z. X.G. performed mtDNA sequencing experiments and Y.W. provided computational tools. Y.W., X.G. and Z.G. analyzed the data and wrote the paper with inputs from R.Z. All authors read and approved the final manuscript.

Notes

Present address: Ruoyu Zhang, Regeneron Pharmaceuticals, Inc, Tarrytown, NY 10591, USA.

Contributor Information

Xiaoxian Guo, Division of Nutritional Sciences, Cornell University, Ithaca, NY 14853, USA.

Yiqin Wang, Division of Nutritional Sciences, Cornell University, Ithaca, NY 14853, USA.

Ruoyu Zhang, Division of Nutritional Sciences, Cornell University, Ithaca, NY 14853, USA.

Zhenglong Gu, Division of Nutritional Sciences, Cornell University, Ithaca, NY 14853, USA.

SUPPLEMENTARY DATA

Supplementary Data are available at NARGAB Online.

FUNDING

NIH[R01AI085286]; CHDI Foundation.

Conflict of interest statement. None declared.

REFERENCES

1. Wallace D.C., Chalkia D. Mitochondrial DNA genetics and the heteroplasmy conundrum in evolution and disease. Cold Spring Harb. Perspect. Biol. 2013; 5:a021220. [DOI] [PMC free article] [PubMed] [Google Scholar]
2. Alston C.L., Rocha M.C., Lax N.Z., Turnbull D.M., Taylor R.W. The genetics and pathology of mitochondrial disease. J. Pathol. 2017; 241:236–250. [DOI] [PMC free article] [PubMed] [Google Scholar]
3. Hudson G., Gomez-Duran A., Wilson I.J., Chinnery P.F. Recent mitochondrial DNA mutations increase the risk of developing common late-onset human diseases. PLoS Genet. 2014; 10:e1004369. [DOI] [PMC free article] [PubMed] [Google Scholar]
4. Payne B.A.I., Wilson I.J., Yu-Wai-Man P., Coxhead J., Deehan D., Horvath R., Taylor R.W., Samuels D.C., Santibanez-Koref M., Chinnery P.F. Universal heteroplasmy of human mitochondrial DNA. Hum. Mol. Genet. 2013; 22:384–390. [DOI] [PMC free article] [PubMed] [Google Scholar]
5. Ye K., Lu J., Ma F., Keinan A., Gu Z. Extensive pathogenicity of mitochondrial heteroplasmy in healthy human individuals. Proc. Natl. Acad. Sci. U.S.A. 2014; 111:10654–10659. [DOI] [PMC free article] [PubMed] [Google Scholar]
6. Greaves L.C., Nooteboom M., Elson J.L., Tuppen H.aL., Taylor G.a, Commane D.M., Arasaradnam R.P., Khrapko K., Taylor R.W., Kirkwood T.B.L. et al. Clonal expansion of early to mid-Life mitochondrial DNA point mutations drives mitochondrial dysfunction during human ageing. PLoS Genet. 2014; 10:e1004620. [DOI] [PMC free article] [PubMed] [Google Scholar]
7. Picard M., Zhang J., Hancock S., Derbeneva O., Golhar R., Golik P., O’Hearn S., Levy S., Potluri P., Lvova M. et al. Progressive increase in mtDNA 3243A>G heteroplasmy causes abrupt transcriptional reprogramming. Proc. Natl. Acad. Sci. U.S.A. 2014; 111:E4033–E4042. [DOI] [PMC free article] [PubMed] [Google Scholar]
8. Clay Montier L.L., Deng J.J., Bai Y. Number matters: control of mammalian mitochondrial DNA copy number. J. Genet. Genomics. 2009; 36:125–131. [DOI] [PMC free article] [PubMed] [Google Scholar]
9. Cai N., Chang S., Li Y., Li Q., Hu J., Liang J., Song L., Kretzschmar W., Gan X., Nicod J. et al. Molecular signatures of major depression. Curr. Biol. 2015; 25:1146–1156. [DOI] [PMC free article] [PubMed] [Google Scholar]
10. Ashar F.N., Moes A., Moore A.Z., Grove M.L., Chaves P.H.M., Coresh J., Newman A.B., Matteini A.M., Bandeen-Roche K., Boerwinkle E. et al. Association of mitochondrial DNA levels with frailty and all-cause mortality. J. Mol. Med. 2014; 93:177–186. [DOI] [PMC free article] [PubMed] [Google Scholar]
11. Ashar F.N., Zhang Y., Longchamps R.J., Lane J., Moes A., Grove M.L., Mychaleckyj J.C., Taylor K.D., Coresh J., Rotter J.I. et al. Association of mitochondrial DNA copy number with cardiovascular disease. JAMA Cardiol. 2017; 21205:1247–1255. [DOI] [PMC free article] [PubMed] [Google Scholar]
12. Picardi E., Pesole G. Mitochondrial genomes gleaned from human whole-exome sequencing. Nat. Methods. 2012; 9:523–524. [DOI] [PubMed] [Google Scholar]
13. Wang Y., Picard M., Gu Z. Genetic evidence for elevated pathogenicity of mitochondrial DNA heteroplasmy in Autism Spectrum Disorder. PLoS Genet. 2016; 12:e1006391. [DOI] [PMC free article] [PubMed] [Google Scholar]
14. Zhang R., Wang Y., Ye K., Picard M., Gu Z. Independent impacts of aging on mitochondrial DNA quantity and quality in humans. BMC Genomics. 2017; 18:890. [DOI] [PMC free article] [PubMed] [Google Scholar]
15. Grandhi S., Bosworth C., Maddox W., Sensiba C., Akhavanfard S., Ni Y., LaFramboise T. Heteroplasmic shifts in tumor mitochondrial genomes reveal tissue-specific signals of relaxed and positive selection. Hum. Mol. Genet. 2017; 26:2912–2922. [DOI] [PMC free article] [PubMed] [Google Scholar]
16. Rebolledo-Jaramillo B., Su M.S.-W., Stoler N., McElhoe J.A., Dickins B., Blankenberg D., Korneliussen T.S., Chiaromonte F., Nielsen R., Holland M.M. et al. Maternal age effect and severe germ-line bottleneck in the inheritance of human mitochondrial DNA. Proc. Natl. Acad. Sci. U.S.A. 2014; 111:15474–15479. [DOI] [PMC free article] [PubMed] [Google Scholar]
17. Marquis J., Lefebvre G., Kourmpetis Y.A.I., Kassam M., Ronga F., De Marchi U., Wiederkehr A., Descombes P. MitoRS, a method for high throughput, sensitive, and accurate detection of mitochondrial DNA heteroplasmy. BMC Genomics. 2017; 18:326. [DOI] [PMC free article] [PubMed] [Google Scholar]
18. Maricic T., Whitten M., Pääbo S. Multiplexed DNA sequence capture of mitochondrial genomes using PCR products. PLoS One. 2010; 5:e14004. [DOI] [PMC free article] [PubMed] [Google Scholar]
19. Ring J.D., Sturk-Andreaggi K., Alyse Peck M., Marshall C. Bioinformatic removal of NUMT-associated variants in mitotiling next-generation sequencing data from whole blood samples. Electrophoresis. 2018; 39:2785–2797. [DOI] [PubMed] [Google Scholar]
20. Rohland N., Reich D. Cost-effective, high-throughput DNA sequencing libraries for multiplexed target capture. Genome Res. 2012; 22:939–946. [DOI] [PMC free article] [PubMed] [Google Scholar]
21. Nunez J.C.B., Oleksiak M.F. A cost-effective approach to sequence hundreds of complete mitochondrial genomes. PLoS One. 2016; 11:e0160958. [DOI] [PMC free article] [PubMed] [Google Scholar]
22. Kebschull J.M., Zador A.M. Sources of PCR-induced distortions in high-throughput sequencing data sets. Nucleic Acids Res. 2015; 43:e143. [DOI] [PMC free article] [PubMed] [Google Scholar]
23. Stoler N., Arbeithuber B., Guiblet W., Makova K.D., Nekrutenko A. Streamlined analysis of duplex sequencing data with Du Novo. Genome Biol. 2016; 17:180. [DOI] [PMC free article] [PubMed] [Google Scholar]
24. Jayaprakash A.D., Benson E.K., Gone S., Liang R., Shim J., Lambertini L., Toloue M.M., Wigler M., Aaronson S.A., Sachidanandam R. Stable heteroplasmy at the single-cell level is facilitated by intercellular exchange of mtDNA. Nucleic Acids Res. 2015; 43:2177–2187. [DOI] [PMC free article] [PubMed] [Google Scholar]
25. Gould M.P., Bosworth C.M., McMahon S., Grandhi S., Grimberg B.T., LaFramboise T. PCR-free enrichment of mitochondrial DNA from human blood and cell lines for high quality next-generation DNA sequencing. PLoS One. 2015; 10:e0139253. [DOI] [PMC free article] [PubMed] [Google Scholar]
26. Orth M., Handley O.J., Schwenke C., Dunnett S.B., Craufurd D., Ho A.K., Wild E., Tabrizi S.J., Landwehrmeyer G.B. Observing huntington's disease: the European Huntington's Disease Network's REGISTRY. PLoS Curr. 2011; 2:RRN1184. [DOI] [PMC free article] [PubMed] [Google Scholar]
27. Diep D., Plongthongkum N., Gore A., Fung H.-L., Shoemaker R., Zhang K. Library-free methylation sequencing with bisulfite padlock probes. Nat. Methods. 2012; 9:270–272. [DOI] [PMC free article] [PubMed] [Google Scholar]
28. Yoon J.-K., Ahn J., Kim H.S., Han S.M., Jang H., Lee M.G., Lee J.H., Bang D. microDuMIP: target-enrichment technique for microarray-based duplex molecular inversion probes. Nucleic Acids Res. 2015; 43:e28. [DOI] [PMC free article] [PubMed] [Google Scholar]
29. Kozich J.J., Westcott S.L., Baxter N.T., Highlander S.K., Schloss P.D. Development of a dual-index sequencing strategy and curation pipeline for analyzing amplicon sequence data on the MiSeq Illumina sequencing platform. Appl. Environ. Microbiol. 2013; 79:5112–5120. [DOI] [PMC free article] [PubMed] [Google Scholar]
30. Li H., Durbin R. Fast and accurate long-read alignment with Burrows–Wheeler transform. Bioinformatics. 2010; 26:589–595. [DOI] [PMC free article] [PubMed] [Google Scholar]
31. Garrison E., Marth G. Haplotype-based variant detection from short-read sequencing. 2012; arXiv doi:20 July 2012, preprint: not peer reviewedhttps://arxiv.org/abs/1207.3907.
32. Li H., Handsaker B., Wysoker A., Fennell T., Ruan J., Homer N., Marth G., Abecasis G., Durbin R.1000 Genome Project data processing subgroup The Sequence Alignment/Map format and SAMtools. Bioinformatics. 2009; 25:2078–2079. [DOI] [PMC free article] [PubMed] [Google Scholar]
33. Wei W., Pagnamenta A.T., Gleadall N., Sanchis-Juan A., Stephens J., Broxholme J., Tuna S., Odhams C.A., Fratter C., Turro E. et al. Nuclear-mitochondrial DNA segments resemble paternally inherited mitochondrial DNA in humans. Nat. Commun. 2020; 11:1740. [DOI] [PMC free article] [PubMed] [Google Scholar]
34. Santos J.M., Tewari S., Goldberg A.F.X., Kowluru R.A. Mitochondrial biogenesis and the development of diabetic retinopathy. Free Radic. Biol. Med. 2011; 51:1849–1860. [DOI] [PMC free article] [PubMed] [Google Scholar]
35. Albayrak L., Khanipov K., Pimenova M., Golovko G., Rojas M., Pavlidis I., Chumakov S., Aguilar G., Chávez A., Widger W.R. et al. The ability of human nuclear DNA to cause false positive low-abundance heteroplasmy calls varies across the mitochondrial genome. BMC Genomics. 2016; 17:1017. [DOI] [PMC free article] [PubMed] [Google Scholar]
36. Zhou W., Chen T., Zhao H., Eterovic A.K., Meric-Bernstam F., Mills G.B., Chen K. Bias from removing read duplication in ultra-deep sequencing experiments. Bioinformatics. 2014; 30:1073–1080. [DOI] [PMC free article] [PubMed] [Google Scholar]
37. Aird D., Ross M.G., Chen W.-S., Danielsson M., Fennell T., Russ C., Jaffe D.B., Nusbaum C., Gnirke A. Analyzing and minimizing PCR amplification bias in Illumina sequencing libraries. Genome Biol. 2011; 12:R18. [DOI] [PMC free article] [PubMed] [Google Scholar]
38. Rebolledo-Jaramillo B., Su M.S.-W., Stoler N., McElhoe J.a., Dickins B., Blankenberg D., Korneliussen T.S., Chiaromonte F., Nielsen R., Holland M.M. et al. Maternal age effect and severe germ-line bottleneck in the inheritance of human mitochondrial DNA. Proc. Natl. Acad. Sci. U.S.A. 2014; 111:15474–15479. [DOI] [PMC free article] [PubMed] [Google Scholar]
39. Li M., Rothwell R., Vermaat M., Wachsmuth M., Schröder R., Laros J.F.J., van Oven M., de Bakker P.I.W., Bovenberg J.A., van Duijn C.M. et al. Transmission of human mtDNA heteroplasmy in the Genome of the Netherlands families: support for a variable-size bottleneck. Genome Res. 2016; 26:417–426. [DOI] [PMC free article] [PubMed] [Google Scholar]
40. Lee D.F., Lu J., Chang S., Loparo J.J., Xie X.S. Mapping DNA polymerase errors by single-molecule sequencing. Nucleic Acids Res. 2016; 44:e118. [DOI] [PMC free article] [PubMed] [Google Scholar]
41. Li M., Schröder R., Ni S., Madea B., Stoneking M. Extensive tissue-related and allele-related mtDNA heteroplasmy suggests positive selection for somatic mutations. Proc. Natl. Acad. Sci. U.S.A. 2015; 112:2491–2496. [DOI] [PMC free article] [PubMed] [Google Scholar]
42. Morris J., Na Y.-J., Zhu H., Lee J.-H., Giang H., Ulyanova A.V, Baltuch G.H., Brem S., Chen H.I., Kung D.K. et al. Pervasive within-mitochondrion single-nucleotide variant heteroplasmy as revealed by single-mitochondrion sequencing. Cell Rep. 2017; 21:2706–2713. [DOI] [PMC free article] [PubMed] [Google Scholar]
43. Stadhouders R., Pas S.D., Anber J., Voermans J., Mes T.H.M., Schutten M. The effect of primer-template mismatches on the detection and quantification of nucleic acids using the 5′ nuclease assay. J. Mol. Diagn. 2010; 12:109–117. [DOI] [PMC free article] [PubMed] [Google Scholar]
44. Ding J., Sidore C., Butler T.J., Wing M.K., Qian Y., Meirelles O., Busonero F., Tsoi L.C., Maschio A., Angius A. et al. Assessing mitochondrial DNA variation and copy number in lymphocytes of ∼2, 000 Sardinians using tailored sequencing analysis tools. PLoS Genet. 2015; 11:e1005306. [DOI] [PMC free article] [PubMed] [Google Scholar]
45. Tsang J.C.H., Yu Y., Burke S., Buettner F., Wang C., Kolodziejczyk A.A., Teichmann S.A., Lu L., Liu P. Single-cell transcriptomic reconstruction reveals cell cycle and multi-lineage differentiation defects in Bcl11a-deficient hematopoietic stem cells. Genome Biol. 2015; 16:178. [DOI] [PMC free article] [PubMed] [Google Scholar]
46. Kennedy S.R., Salk J.J., Schmitt M.W., Loeb L.A. Ultra-sensitive sequencing reveals an age-related increase in somatic mitochondrial mutations that are inconsistent with oxidative damage. PLoS Genet. 2013; 9:e1003794. [DOI] [PMC free article] [PubMed] [Google Scholar]
47. Zhang R., Nakahira K., Choi A.M.K., Gu Z. Heteroplasmy concordance between mitochondrial DNA and RNA. Sci. Rep. 2019; 9:12942. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

lqaa065_Supplemental_File

lqaa065_supplemental_file.docx^{(1.3MB, docx)}

Data Availability Statement

The STAMP tool kit used for processing sequencing reads from STAMP is available for download at https://github.com/mtstamp/stamp.

Paired-end reads and mtDNA consensus reads of REGISTRY samples have been submitted to the European Genome-Phenome Archive under the EGA accession number EGAS00001004092.

[B1] 1. Wallace D.C., Chalkia D. Mitochondrial DNA genetics and the heteroplasmy conundrum in evolution and disease. Cold Spring Harb. Perspect. Biol. 2013; 5:a021220. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B2] 2. Alston C.L., Rocha M.C., Lax N.Z., Turnbull D.M., Taylor R.W. The genetics and pathology of mitochondrial disease. J. Pathol. 2017; 241:236–250. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B3] 3. Hudson G., Gomez-Duran A., Wilson I.J., Chinnery P.F. Recent mitochondrial DNA mutations increase the risk of developing common late-onset human diseases. PLoS Genet. 2014; 10:e1004369. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B4] 4. Payne B.A.I., Wilson I.J., Yu-Wai-Man P., Coxhead J., Deehan D., Horvath R., Taylor R.W., Samuels D.C., Santibanez-Koref M., Chinnery P.F. Universal heteroplasmy of human mitochondrial DNA. Hum. Mol. Genet. 2013; 22:384–390. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B5] 5. Ye K., Lu J., Ma F., Keinan A., Gu Z. Extensive pathogenicity of mitochondrial heteroplasmy in healthy human individuals. Proc. Natl. Acad. Sci. U.S.A. 2014; 111:10654–10659. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B6] 6. Greaves L.C., Nooteboom M., Elson J.L., Tuppen H.aL., Taylor G.a, Commane D.M., Arasaradnam R.P., Khrapko K., Taylor R.W., Kirkwood T.B.L. et al. Clonal expansion of early to mid-Life mitochondrial DNA point mutations drives mitochondrial dysfunction during human ageing. PLoS Genet. 2014; 10:e1004620. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B7] 7. Picard M., Zhang J., Hancock S., Derbeneva O., Golhar R., Golik P., O’Hearn S., Levy S., Potluri P., Lvova M. et al. Progressive increase in mtDNA 3243A>G heteroplasmy causes abrupt transcriptional reprogramming. Proc. Natl. Acad. Sci. U.S.A. 2014; 111:E4033–E4042. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B8] 8. Clay Montier L.L., Deng J.J., Bai Y. Number matters: control of mammalian mitochondrial DNA copy number. J. Genet. Genomics. 2009; 36:125–131. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B9] 9. Cai N., Chang S., Li Y., Li Q., Hu J., Liang J., Song L., Kretzschmar W., Gan X., Nicod J. et al. Molecular signatures of major depression. Curr. Biol. 2015; 25:1146–1156. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B10] 10. Ashar F.N., Moes A., Moore A.Z., Grove M.L., Chaves P.H.M., Coresh J., Newman A.B., Matteini A.M., Bandeen-Roche K., Boerwinkle E. et al. Association of mitochondrial DNA levels with frailty and all-cause mortality. J. Mol. Med. 2014; 93:177–186. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B11] 11. Ashar F.N., Zhang Y., Longchamps R.J., Lane J., Moes A., Grove M.L., Mychaleckyj J.C., Taylor K.D., Coresh J., Rotter J.I. et al. Association of mitochondrial DNA copy number with cardiovascular disease. JAMA Cardiol. 2017; 21205:1247–1255. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B12] 12. Picardi E., Pesole G. Mitochondrial genomes gleaned from human whole-exome sequencing. Nat. Methods. 2012; 9:523–524. [DOI] [PubMed] [Google Scholar]

[B13] 13. Wang Y., Picard M., Gu Z. Genetic evidence for elevated pathogenicity of mitochondrial DNA heteroplasmy in Autism Spectrum Disorder. PLoS Genet. 2016; 12:e1006391. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B14] 14. Zhang R., Wang Y., Ye K., Picard M., Gu Z. Independent impacts of aging on mitochondrial DNA quantity and quality in humans. BMC Genomics. 2017; 18:890. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B15] 15. Grandhi S., Bosworth C., Maddox W., Sensiba C., Akhavanfard S., Ni Y., LaFramboise T. Heteroplasmic shifts in tumor mitochondrial genomes reveal tissue-specific signals of relaxed and positive selection. Hum. Mol. Genet. 2017; 26:2912–2922. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B16] 16. Rebolledo-Jaramillo B., Su M.S.-W., Stoler N., McElhoe J.A., Dickins B., Blankenberg D., Korneliussen T.S., Chiaromonte F., Nielsen R., Holland M.M. et al. Maternal age effect and severe germ-line bottleneck in the inheritance of human mitochondrial DNA. Proc. Natl. Acad. Sci. U.S.A. 2014; 111:15474–15479. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B17] 17. Marquis J., Lefebvre G., Kourmpetis Y.A.I., Kassam M., Ronga F., De Marchi U., Wiederkehr A., Descombes P. MitoRS, a method for high throughput, sensitive, and accurate detection of mitochondrial DNA heteroplasmy. BMC Genomics. 2017; 18:326. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B18] 18. Maricic T., Whitten M., Pääbo S. Multiplexed DNA sequence capture of mitochondrial genomes using PCR products. PLoS One. 2010; 5:e14004. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B19] 19. Ring J.D., Sturk-Andreaggi K., Alyse Peck M., Marshall C. Bioinformatic removal of NUMT-associated variants in mitotiling next-generation sequencing data from whole blood samples. Electrophoresis. 2018; 39:2785–2797. [DOI] [PubMed] [Google Scholar]

[B20] 20. Rohland N., Reich D. Cost-effective, high-throughput DNA sequencing libraries for multiplexed target capture. Genome Res. 2012; 22:939–946. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B21] 21. Nunez J.C.B., Oleksiak M.F. A cost-effective approach to sequence hundreds of complete mitochondrial genomes. PLoS One. 2016; 11:e0160958. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B22] 22. Kebschull J.M., Zador A.M. Sources of PCR-induced distortions in high-throughput sequencing data sets. Nucleic Acids Res. 2015; 43:e143. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B23] 23. Stoler N., Arbeithuber B., Guiblet W., Makova K.D., Nekrutenko A. Streamlined analysis of duplex sequencing data with Du Novo. Genome Biol. 2016; 17:180. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B24] 24. Jayaprakash A.D., Benson E.K., Gone S., Liang R., Shim J., Lambertini L., Toloue M.M., Wigler M., Aaronson S.A., Sachidanandam R. Stable heteroplasmy at the single-cell level is facilitated by intercellular exchange of mtDNA. Nucleic Acids Res. 2015; 43:2177–2187. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B25] 25. Gould M.P., Bosworth C.M., McMahon S., Grandhi S., Grimberg B.T., LaFramboise T. PCR-free enrichment of mitochondrial DNA from human blood and cell lines for high quality next-generation DNA sequencing. PLoS One. 2015; 10:e0139253. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B26] 26. Orth M., Handley O.J., Schwenke C., Dunnett S.B., Craufurd D., Ho A.K., Wild E., Tabrizi S.J., Landwehrmeyer G.B. Observing huntington's disease: the European Huntington's Disease Network's REGISTRY. PLoS Curr. 2011; 2:RRN1184. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B27] 27. Diep D., Plongthongkum N., Gore A., Fung H.-L., Shoemaker R., Zhang K. Library-free methylation sequencing with bisulfite padlock probes. Nat. Methods. 2012; 9:270–272. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B28] 28. Yoon J.-K., Ahn J., Kim H.S., Han S.M., Jang H., Lee M.G., Lee J.H., Bang D. microDuMIP: target-enrichment technique for microarray-based duplex molecular inversion probes. Nucleic Acids Res. 2015; 43:e28. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B29] 29. Kozich J.J., Westcott S.L., Baxter N.T., Highlander S.K., Schloss P.D. Development of a dual-index sequencing strategy and curation pipeline for analyzing amplicon sequence data on the MiSeq Illumina sequencing platform. Appl. Environ. Microbiol. 2013; 79:5112–5120. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B30] 30. Li H., Durbin R. Fast and accurate long-read alignment with Burrows–Wheeler transform. Bioinformatics. 2010; 26:589–595. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B31] 31. Garrison E., Marth G. Haplotype-based variant detection from short-read sequencing. 2012; arXiv doi:20 July 2012, preprint: not peer reviewedhttps://arxiv.org/abs/1207.3907.

[B32] 32. Li H., Handsaker B., Wysoker A., Fennell T., Ruan J., Homer N., Marth G., Abecasis G., Durbin R.1000 Genome Project data processing subgroup The Sequence Alignment/Map format and SAMtools. Bioinformatics. 2009; 25:2078–2079. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B33] 33. Wei W., Pagnamenta A.T., Gleadall N., Sanchis-Juan A., Stephens J., Broxholme J., Tuna S., Odhams C.A., Fratter C., Turro E. et al. Nuclear-mitochondrial DNA segments resemble paternally inherited mitochondrial DNA in humans. Nat. Commun. 2020; 11:1740. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B34] 34. Santos J.M., Tewari S., Goldberg A.F.X., Kowluru R.A. Mitochondrial biogenesis and the development of diabetic retinopathy. Free Radic. Biol. Med. 2011; 51:1849–1860. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B35] 35. Albayrak L., Khanipov K., Pimenova M., Golovko G., Rojas M., Pavlidis I., Chumakov S., Aguilar G., Chávez A., Widger W.R. et al. The ability of human nuclear DNA to cause false positive low-abundance heteroplasmy calls varies across the mitochondrial genome. BMC Genomics. 2016; 17:1017. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B36] 36. Zhou W., Chen T., Zhao H., Eterovic A.K., Meric-Bernstam F., Mills G.B., Chen K. Bias from removing read duplication in ultra-deep sequencing experiments. Bioinformatics. 2014; 30:1073–1080. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B37] 37. Aird D., Ross M.G., Chen W.-S., Danielsson M., Fennell T., Russ C., Jaffe D.B., Nusbaum C., Gnirke A. Analyzing and minimizing PCR amplification bias in Illumina sequencing libraries. Genome Biol. 2011; 12:R18. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B38] 38. Rebolledo-Jaramillo B., Su M.S.-W., Stoler N., McElhoe J.a., Dickins B., Blankenberg D., Korneliussen T.S., Chiaromonte F., Nielsen R., Holland M.M. et al. Maternal age effect and severe germ-line bottleneck in the inheritance of human mitochondrial DNA. Proc. Natl. Acad. Sci. U.S.A. 2014; 111:15474–15479. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B39] 39. Li M., Rothwell R., Vermaat M., Wachsmuth M., Schröder R., Laros J.F.J., van Oven M., de Bakker P.I.W., Bovenberg J.A., van Duijn C.M. et al. Transmission of human mtDNA heteroplasmy in the Genome of the Netherlands families: support for a variable-size bottleneck. Genome Res. 2016; 26:417–426. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B40] 40. Lee D.F., Lu J., Chang S., Loparo J.J., Xie X.S. Mapping DNA polymerase errors by single-molecule sequencing. Nucleic Acids Res. 2016; 44:e118. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B41] 41. Li M., Schröder R., Ni S., Madea B., Stoneking M. Extensive tissue-related and allele-related mtDNA heteroplasmy suggests positive selection for somatic mutations. Proc. Natl. Acad. Sci. U.S.A. 2015; 112:2491–2496. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B42] 42. Morris J., Na Y.-J., Zhu H., Lee J.-H., Giang H., Ulyanova A.V, Baltuch G.H., Brem S., Chen H.I., Kung D.K. et al. Pervasive within-mitochondrion single-nucleotide variant heteroplasmy as revealed by single-mitochondrion sequencing. Cell Rep. 2017; 21:2706–2713. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B43] 43. Stadhouders R., Pas S.D., Anber J., Voermans J., Mes T.H.M., Schutten M. The effect of primer-template mismatches on the detection and quantification of nucleic acids using the 5′ nuclease assay. J. Mol. Diagn. 2010; 12:109–117. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B44] 44. Ding J., Sidore C., Butler T.J., Wing M.K., Qian Y., Meirelles O., Busonero F., Tsoi L.C., Maschio A., Angius A. et al. Assessing mitochondrial DNA variation and copy number in lymphocytes of ∼2, 000 Sardinians using tailored sequencing analysis tools. PLoS Genet. 2015; 11:e1005306. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B45] 45. Tsang J.C.H., Yu Y., Burke S., Buettner F., Wang C., Kolodziejczyk A.A., Teichmann S.A., Lu L., Liu P. Single-cell transcriptomic reconstruction reveals cell cycle and multi-lineage differentiation defects in Bcl11a-deficient hematopoietic stem cells. Genome Biol. 2015; 16:178. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B46] 46. Kennedy S.R., Salk J.J., Schmitt M.W., Loeb L.A. Ultra-sensitive sequencing reveals an age-related increase in somatic mitochondrial mutations that are inconsistent with oxidative damage. PLoS Genet. 2013; 9:e1003794. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B47] 47. Zhang R., Nakahira K., Choi A.M.K., Gu Z. Heteroplasmy concordance between mitochondrial DNA and RNA. Sci. Rep. 2019; 9:12942. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

STAMP: a multiplex sequencing method for simultaneous evaluation of mitochondrial DNA heteroplasmies and content

Xiaoxian Guo

Yiqin Wang

Ruoyu Zhang

Zhenglong Gu

Abstract

INTRODUCTION

MATERIALS AND METHODS

Study samples

Design of STAMP

Figure 1.

mtDNA sequencing with STAMP

Step 1. mtDNA capture with extension-ligation (EL) probes

Step 2. PCR amplification of capture products

Step 3. Library purification

Step 4. Massively parallel sequencing

Step 5. Sequencing data processing

mtDNA variant detection

mtDNA content evaluation with quantitative PCR

Statistical analysis

RESULTS AND DISCUSSION

Effective capture of mtDNA with EL probes

Figure 2.

Utilizing molecular barcodes to remove PCR duplicates and improve sequencing accuracy

Figure 3.

Accurate detection of mtDNA heteroplasmies and their fractions in total genomic DNA

Application of STAMP in a population study of mtDNA heteroplasmies

Figure 4.

Extension of STAMP to measure mtDNA content in the same assay

Age-dependent increase of heteroplasmy incidence in lymphoblast samples

Table 1.

CONCLUSION

DATA AVAILABILITY

Supplementary Material

ACKNOWLEDGEMENTS

Notes

Contributor Information

SUPPLEMENTARY DATA

FUNDING

REFERENCES

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases