Background: Mitochondrial DNA (mtDNA) mutations cause a wide range of serious genetic diseases with maternal inheritance. Because of the high transmission risk and the absence of therapy in these disorders, at-risk couples often ask for prenatal diagnosis (PND). However, because heteroplasmy load (coexistence of mutant and wild-type mtDNA) may vary among tissues and with time, the possibility that a single fetal sample may not reflect the whole neonate impedes prenatal diagnosis of mtDNA diseases.
Methods: We performed 13 prenatal diagnoses for the NARP (neurogenic weakness, ataxia, retinitis pigmentosa) m.8993T-->G mtDNA mutation (p.Leu156Arg) in the ATP synthase subunit 6 gene. Analyses were performed on chorionic villous (CVS) and/or amniocyte samples carried out at various stages of pregnancy, using a method enabling quantification of low DNA amounts.
Results: Maternal mutant loads ranged from 0 to 75% in blood and had no predictive value for the fetus status, except for women with no detectable mutant DNA, whose fetuses were consistently mutation-free. In 8/13 PND, mutant load was <30%. These children are healthy at 2-7 years of age. In 5/13 PND, mutant load ranged from 65 to 100%, and parents preferred to terminate the pregnancies (15-22 weeks of gestation). Single-cell analysis of 20 trophoblastic cells and 21 amniocytes isolated from two affected fetuses found an average mutant load close to the overall CVS and amniocyte mutant load, despite striking intercellular variation. The m.8993T-->G mutant loads, assessed in 7, 17, 11, and 5 different tissues from 4 terminations, respectively, were identical in all tissues from a given individual (mean (SD) 78 (1.2)%, 91 (0.7)%, 74 (2)%, and 63 (1.6)% for the 4 fetuses, respectively).
Conclusions: Our results indicate that the placental/amniotic mutant loads do reflect the NARP mutant mtDNA load in the whole fetus, even when the sample amount is small, and suggest that heteroplasmy level remains stable during pregnancy, at least after 10 weeks of gestation. Although these data establish the feasibility of PND for this mutation, assessing more precisely the correlation between mutant load and disease severity should further help in interpreting PND results.