Recently, two reports [R. E. Davis et al. (1997) Proc. Natl. Acad. Sci. USA 94, 4564-4569 and E. Fahy et al. (1997) Nucleic Acids Res. 25, 3102-3109] described a series of heteroplasmic mitochondrial DNA (mtDNA) mutations in the genes encoding two cytochrome c oxidase subunits (CO1 and CO2) which segregated in higher abundance with Alzheimer's disease subjects than controls. Using mtDNA-depleted NT2 cells, we provide further evidence that these two reports are erroneously based on a PCR artifact arising from the amplification of nuclear DNA encoded mtDNA pseudogenes (mtDNA psi s). Our findings are similar, but not identical, to other recent studies of these putative mtDNA psi sequences. This sequence variability may indicate that multiple mtDNA psi s, all of comparatively recent evolutionary origin are involved. While such pseudogenes are interesting in that they provide a molecular evolutionary "snapshot" of human ancestral mtDNA, it is unlikely that they play any role in the etiology of Alzheimer's disease.