# Mito-SiPE is a sequence-independent and PCR-free mtDNA enrichment method for accurate ultra-deep mitochondrial sequencing

### Breeding and tissue harvesting

Mice were housed in shoe box cages and fed ProLab RMH 1800 diet (PMI Nutrition International) containing 50 μg vitamin B12/kg of diet and 3.3 mg folic acid/kg of diet. Breeding mice were fed Picolab Mouse Diet 20, containing 51 μg vitamin B12/kg diet and 2.9 mg folic acid/kg of diet. Heterozygous Polgwt/D257A males and Polgwt/D257A females were mated. Homozygous mutant and wild-type progeny were aged to 6 months, at which point they were sacrificed. There were 4 Polgwt/wt mice (two male and two female) and 4 PolgD257A/D257A mice (two male and two female) used in the Polg experiments. Brain, heart, lung, liver, spleen, kidney and muscle tissue was isolated and mitochondrial DNA enrichment was performed on all tissues.

### Tissue homogenisation

Harvested tissue was placed in a homogenisation tube with 10X volume per gram of fresh homogenisation buffer i.e. 5 ml buffer for 500 mg tissue. Tissues were homogenised until no discernible whole tissue was present. The homogenate was then transferred to 1.5 ml microcentrifuge tubes and spun at 1000 × g for 1 min at 4 °C. The supernatant was transferred to a new microcentrifuge tube and spun at 12,000 × g for 10 min at 4 °C to pellet mitochondria. The mitochondrial pellet was resuspended with 100 µl of resuspension buffer for storage or for immediate DNA extraction.

### Mitochondrial DNA isolation from fresh mouse tissue

The mitochondria resuspension was added to 200 µl alkaline lysis buffer, vortexed, and placed on ice for 5 min. Potassium Acetate Buffer (150 µl) was then added and the mixture was vortexed slowly and placed on ice for 5 min. The mixture was centrifuged at 12,000 × g for 5 min at 4 °C to pellet proteins and the supernatant was decanted to a new tube. RNase (1 µg) was added to the mixture and left at room temperature for 15 min. Phenol-chloroform (500 µl) was added to each tube, inverted and placed on a shaker/rotator for 20 min. Afterwards, centrifugation at 12,000 × g for 2 min at room temperature was carried out. The aqueous (top) layer was decanted to a new tube (~450 µl from this phase was retrieved) and 40 ul sodium acetate, 1 µl glycogen (20 mg/ml) and 1200 µl 100% EtOH were added. The mixture was inverted and mixed well, then left on dry ice for 60 min. The mixture underwent centrifugation at 12,000 × g and the supernatant was removed. The pellet was finally washed twice using 70% ethanol, air-dried, and resuspended in a low-TE buffer for sequencing or regular TE buffer for (q)PCR.

### Library preparation and next-generation DNA sequencing

Libraries were generated from approximately 50 ng genomic DNA using the Accel-NGS 2 S Plus DNA Library Kit (Swift Biosciences) using five cycles of PCR to minimise PCR bias. The DNA samples were sheared by sonication (Covaris Inc., Woburn, MA) to a mean of 300 bp. Libraries were tagged with unique dual index DNA barcodes to allow the pooling of libraries and minimise the impact of barcode hopping. Libraries were pooled for sequencing on the NovaSeq 6000 (Illumina) to obtain at least 7.6 million 151-base read pairs per individual library. Sequencing data was processed using RTA version 3.4.4. DNA sequencing was carried out at the NIH Intramural Sequencing Center.

### Data processing and alignment

Fastq files were aligned to the mouse reference genome, GRCm38, using bwa mem using the default parameters47. Picard tools were used to add read groups, and to mark and remove duplicates48. Samtools was used to calculate the coverage across the nuclear and mitochondrial genome for each sample. Finally, R (v3.5.0) and ggplot2 (v3.3.0) were used for statistical analysis and subsequent visualisation of graphs49,50. Library complexity and fragment sizes were calculated using Picard tools v1.4.2 on 15 randomly-selected samples (Supplementary Data 2).

### Variant calling and mutation analysis

Variant calling was performed using bcftools v1.9 with ‘bcftools mpileup -f -Q 30 –skip-indels reference_fasta bam_file | bcftools call -mv’ to identify single nucleotide variants only. Filtering was performed by removing any SNVs that had a QUAL score lower than 20. The code used for alignment and variant calling is available on github (https://github.com/walshd59/mtDNAhetScripts.git). Of 66,738 variants identified across all samples in our study at an alternative allele frequency ≥0.2%, only 137 of these had an ln(Strand Odds Ratio) value ≥3. Analysis of the mutation spectrum and further characterisation/annotation of heteroplasmic variants was performed using SnpEff (v 5.1)51.

### Quantification of mtDNA copy number

Mitochondrial DNA copy number was assessed via qPCR targeting both mitochondrial and nuclear loci, as previously described52,53. Briefly, 2.5 µl LightCycler® 480 SYBR Green I Master (Roche, Molecular Systems, Inc, Germany), 2 µl of DNA (20 ng/µl) and 0.5 µl primer mix were added in triplicate to a 384-well plate and the reactions were carried out by the QuanStudio 6 Flex (Applied Biosystems, Foster City, CA, USA). The conditions were as follows: 95 °C for 5 min, 45 cycles of 95 °C for 10 s, 60 °C for 10 s and 72 °C for 20 s. A melting curve was performed using 95 °C for 5 s, 66 °C for 1 min and gradual increase to 97 °C. The mitochondrial DNA copy number was assessed using the following equation:

$$2\times {2}^{\varDelta {{{{{\rm{Ct}}}}}}}({{{{{\rm{where}}}}}}\,\varDelta {{{{{\rm{Ct}}}}}}={{{{{\rm{Ct}}}}}}({{{{{\rm{mtDNA}}}}}}\,{{{{{\rm{gene}}}}}})-{{{{{\rm{Ct}}}}}}({{{{{\rm{nDNA}}}}}}\,{{{{{\rm{gene}}}}}}))$$

(1)

The following primers were used for human mtDNA copy number: mtDNA tRNA (forward: CACCCAAGAACAGGGTTTGT, reverse: TGGCCATGGGTATGTTGTTA) nuclear DNA β2-microglobulin (forward: TGCTGTCTCCATGTTTGATGTATCT), reverse: TCTCTGCTCCCCACCTCTAAGT). Mouse mtDNA copy number was assessed using the following primers: mtDNA ND1 (forward: CTAGCAGAAACAAACCGGGC, reverse: CCGGCTGCGTATTCTACGTT) nuclear DNA HK2 (forward: GCCAGCCTCTCCTGATTTTAGTGT, reverse: GGGAACACAAAAGACCTCTTCTGG). These primers are available in the supplementary data file (Supplementary data 3).

### Long-range PCR enrichment of mtDNA

This technique was used to amplify human and mouse mitochondrial DNA in two fragments from a whole DNA extract. DNA was quantified via Nanodrop (Methods 2.2.7) unless otherwise stated. Each PCR reaction consisted of Q5 High-fidelity Polymerase (0.02 U/μl), 5X Q5 reaction buffer (1X), 10 mM dNTPs (300 μM), 5 μM Forward and Reverse primers (0.25 μM). Template DNA (100 ng) was added to each reaction except for the no-template control (NTC) but an equivalent volume of molecular biology-grade water was added instead. The temperature cycles were as follows: 1 × 30 s denature 98 °C, 25 × 10 s denature 98 °C, 30 s annealing 66 °C, 4 min 30 s elongation 72 °C, 1 × 10 min elongation 72 °C on a thermocycler. Both fragments were quantified using Qubit and mixed in equimolar ratios. The following primers were used for each fragment: lrPCR fragment 1 (forward: GGATCCTACTCTCTACAAAC, reverse: TAGTTTGCCGCGTTGGGTGG) and lrPCR fragment 2 (forward: CTACCCCCTTCAATCAATCT, reverse: CCGGTTTGTTTCTGCTAGGG). These primers are also available in the supplementary data file (Supplementary Data 3).

### Mitochondrial isolation via Qiagen QProteome™ kit

HepG2 cells (2 × 106) were collected, counted and pelleted when 80% confluency was reached. The supernatant was removed and the pellet was resuspended in the lysis buffer provided in the kit. Homogenisation and mitochondrial isolation were then carried out as per the manufacturers’ protocol. Briefly, the cell pellet was resuspended in 1.5 ml of ice-cold Disruption Buffer by pipetting up and down using a 1 ml pipet tip. The cell disruption was completed using a blunt-ended needle and a syringe. The mixture was centrifuged at 1000 × g for 10 min at 4 °C to pellet proteins and the supernatant was decanted to a new tube. Afterwards, centrifugation at 6000 × g for 2 min at room temperature was carried out. This pellet was then resuspended in 200 μl PBS and 20 μl proteinase K. DNA extraction was then performed on the mitochondrial isolate via Qiagen DNeasy™ Blood and Tissue kit using the manufacturer’s protocol.

### Plasmid-Safe™ digest for mtDNA enrichment

Whole DNA extractions were treated with Plasmid-Safe ATP-dependent exonuclease (Lucigen) as per the manufacturer’s protocol. Briefly, a plasmid-safe solution was created using 42 μl sterile water, 2 μl 25 mM ATP, 5 μl 10X reaction buffer and 1 μl Plasmid-Safe DNase. This DNase targets linear molecules and, as such, does not degrade intact mitochondrial DNA as it is circular. The solution was added to the DNA extracted using the QIAprep miniprep and incubated at 37 °C for 1 h. The DNase was then deactivated with a 70 °C incubation for 30 min.

### HepG2 culture conditions

HepG2 cells (Merck; 85011430-1VL) were cultured in Dulbecco’s modified Eagle medium (DMEM) with 10% supplementation of foetal bovine serum in a 5% CO2 incubator. Cells were cultured in 10 ml of media in T75 flasks. The cells were passaged by washing with 5 ml PBS (1X) followed by incubation in 2 ml of 0.25% Trypsin-EDTA at 37 °C for 5 min. Trypsinisation was inhibited by adding 4 ml DMEM. Cells were then collected via centrifugation at 500×g for 5 min before being counted.

### Qiagen QIAprep miniprep

HepG2 cells (2 × 106) were collected, counted and pelleted when 80% confluency was reached. The supernatant was removed and the pellet was resuspended in lysis buffer provided in the kit. DNA isolation was performed as per the manufacturer’s protocol, using a silica membrane to capture the mtDNA, which was subsequently collected in 100 μl the provided elution buffer.

### Statistics and reproducibility

All statistical analyses included in this paper were carried out in R (version 4.1.1) and the software package rstatix (version 0.7.0). Sample sizes are described within each experimental figure. Wilcoxon signed-rank tests were performed to compare mtDNA copy number from unenriched and enriched samples from different tissues (n = 14 for each group). Wilcoxon signed-rank tests were also performed to compare the effect of lrPCR amplification and Mito-SiPE on mitochondrial heteroplasmy in Polgwt/wt and PolgD257A/D257A (n = 24 for each genotype). A Student’s t-test was used for comparing the effect of lrPCR and Mito-SiPE on heteroplasmy in C57BL6 wild-type mice (n = 6 for the lrPCR group, n = 12 for the Mito-SiPE group).

### Solutions

The following solutions were used: Homogenisation Buffer (0.25 M Sucrose, 10 mM EDTA, 30 mM Tris-HCl, pH = 7.5), Resuspension Buffer (10 mM Tris, 0.15 M NaCl, 10 mM EDTA, pH = 8.0), Alkaline Lysis Buffer (0.18 N NaOH, 1% SDS, prepared fresh), Potassium Acetate Buffer (3 M potassium, 5 M acetate), and Low-TE buffer (10 mM Tris-HCl, 0.1 mM EDTA, pH = 8.0).

### Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.