Human mitochondrial DNA (mtDNA) encodes 13 polypeptides essential for oxidative phosphorylation. Because of the unique features of "replicative segregation" and "threshold expression" of mtDNA genetics, identification of homoplasmy versus heteroplasmy status is critical. Results from various detection methods may lead to different interpretations on formation or outcome of mtDNA mutations, such as the conclusion of somatic mutation versus genetic drift in cancers. Denaturing high-performance liquid chromatography (DHPLC) and pyrosequencing (PSQ) have recently been employed to confirm the presence of heteroplasmy of mtDNA because of their high sensitivity in detecting heteroplasmic mutations compared with direct sequencing. Moreover, PSQ has superior ability in quantifying percentage of heteroplasmy. However, there could be disagreement between these two techniques and several issues specific for mtDNA should be taken into consideration. Particularly, DHPLC analysis should be more prone to be interfered by nuclear mitochondrial sequences (Numts), if it is coamplified with mtDNA, than PSQ analysis.

High-sensitivity and high-throughput mutation detection techniques are useful for screening the homoplasmy or heteroplasmy status of mitochondrial DNA (mtDNA), but might be susceptible to interference from nuclear mitochondrial DNA sequences (NUMTs) co-amplified during polymerase chain reaction (PCR). In this study, we first evaluated the platform of SURVEYOR Nuclease digestion of heteroduplexed DNA followed by the detection of cleaved DNA by using the WAVE HS System (SN/WAVE-HS) for detecting human mtDNA variants and found that its performance was slightly better than that of denaturing high-performance liquid chromatography (DHPLC). The potential interference from co-amplified NUMTs on screening mtDNA heteroplasmy when using these 2 highly sensitive techniques was further examined by using 2 published primer sets containing a total of 65 primer pairs, which were originally designed to be used with one of the 2 techniques. We confirmed that 24 primer pairs could amplify NUMTs by conducting bioinformatic analysis and PCR with the DNA from 143B-ρ0 cells. Using mtDNA extracted from the mitochondria of human 143B cells and a cybrid line with the nuclear background of 143B-ρ0 cells, we demonstrated that NUMTs could affect the patterns of chromatograms for cell DNA during SN-WAVE/HS analysis of mtDNA, leading to incorrect judgment of mtDNA homoplasmy or heteroplasmy status. However, we observed such interference only in 2 of 24 primer pairs selected, and did not observe such effects during DHPLC analysis. These results indicate that NUMTs can affect the screening of low-level mtDNA variants, but it might not be predicted by bioinformatic analysis or the amplification of DNA from 143B-ρ0 cells. Therefore, using purified mtDNA from cultured cells with proven purity to evaluate the effects of NUMTs from a primer pair on mtDNA detection by using PCR-based high-sensitivity methods prior to the use of a primer pair in real studies would be a more practical strategy.