Material and Methods
We measured the frequency of the ten pathogenic mtDNA mutations often found in patients with mitochondrial disease (m.1555A→G, m.3243A→G, m.3460G→A, m.7445A→G, m.8344A→G, m.8993T→G, m.11778G→A, m.13513G→A, m.14459G→A, m.14484T→C) in ∼3000 sequential umbilical-cord-blood samples from north Cumbria in England.12 Consent for inclusion into the study exceeded 80%. Ethical approval to investigate mtDNA mutations within this cohort was granted by the West Cumbria Local Research Ethics Committee (Project 381). Previous epidemiological and nuclear genetic studies established that this cohort is a random sample of the population with numerous nuclear alleles in Hardy-Weinberg equilibrium.13

Haplogroup Determination
The ten major European mtDNA haplogroups were determined in 344 random samples from the cohort as described in the literature,14 with the use of modified primer sequences. This confirmed that the study group was representative of the UK population (Table 1). Mitochondrial haplogroups were also determined in samples that harbored mtDNA point mutations and in their mothers, with the same approach used.

High-Throughput Genotyping
The population was genotyped by primer extension of multiplex PCR products with the detection of the allele-specific extension products by matrix-associated laser desorption/ionization time of flight (MALDI-TOF; Sequenom MassARRAY, San Diego, CA). Assays were designed with Sequenom Assay Design software v2.0.7.0, resulting in five multiplex assays. The allelotyping assay was followed according to manufacturer's instructions, with modifications. At the primary PCR step, DNA was amplified under the following conditions: initial denaturation of 95°C for 15 minutes, then 30 cycles of denaturation at 95°C for 20 seconds, annealing at 60°C for 30 seconds, and extension at 72°C for 1 minute. Finally, there was a further extension at 72°C for 3 minutes before the samples were cooled and stored at 4°C. A homogeneous MassEXTEND (hME) reaction mix containing appropriate hME EXTEND mix (1× buffer with 0.225 mM d/ddNTPs), 2.7 μM MassEXTEND primer, and 0.576 U ThermoSequenase (GE Healthcare), made up to a final volume of 2 μL with anH2O, was added to each SAP-cleaned PCR product. The microplate was then thermocycled as follows: initial denaturation of 94°C for 2 minutes, then 55 cycles of denaturation at 94°C for 5 seconds, annealing at 52°C for 5 seconds, and extension at 72°C for 5 seconds before cooling to 4°C. The sample microplate and a 384 SpectroCHIP were loaded onto the deck of the Samsung Nanodispenser. 15 nL of solution from the sample microplate was transferred onto the chip, which was read by a Bruker Autoflex Mass Spectrometer system. Data was collected with the use of SpectroACQUIRE v3.3.1.3 software and visualised with the use of MassARRAY Typer v3.4 TyperAnalyzer software. The sensitivity of the MALDI-TOF MS assay for each mutation was assessed with mixed cloned mtDNA fragments in duplicate for both uniplex and multiplex reactions (Table S1, available online).
The assay detected ≥ 10% mutated mtDNA in each case, and for some mutations, the detection threshold was considerably lower (Table S1). For nine mutations, positive calls were confirmed by direct sequencing and last-cycle fluorescent PCR-RFLP from an independent aliquot of DNA, which established the percentage mutated mtDNA (see below). The m.1555A→G mutation was confirmed by cloning and sequencing from a separate aliquot of DNA from the same subject.

Measurement of mtDNA Heteroplasmy
Positive calls were confirmed in an independent aliquot from the original DNA sample with the use of last-cycle fluorescent PCR-RFLP as described previously.15 This also established the percentage mutated mtDNA. For the m.1555A→G mutation, heteroplasmy was quantified by cloning and sequencing of 23 independent mtDNA fragments from an independent aliquot from the original DNA sample. Primers and restriction enzymes and digests are shown in Table S2.

MtDNA Sequencing
The mitochondrial noncoding control region (D-loop) was sequenced by PCR of four overlapping segments, with the use of forward and reverse M13-tagged primer pairs. Primer sequences are shown in Table S3. PCR products were treated with ExoSAP (ExoSAP-IT, GE Healthcare, USA) and sequenced on a fluorescent genetic analyzer (Beckman-Coulter CEQ 8000) with the standard dideoxy chain-termination method (Beckman-Coulter Quickstart). The sequence data were compared to the revised Cambridge reference sequence (rCRS)16 with the use of CEQ Sequence Analysis v2.3.13 analysis software (Beckman-Coulter). This allowed identification of D-loop changes and comparison of positive samples.

Statistical Analysis
Frequencies were compared by calculation of empirical P-values with a Monte Carlo-based simulation approach based on the method of Roff and Bentzen.17 Exact 95% confidence intervals were calculated by the Clopper-Pearson method.18