MATERIALS AND METHODS

Clinical samples

To identify confined placental mosaicism cases, we first screened 50 women whose infants had intra-uterine growth retardation (birthweight <5th percentile for gestational age) of unknown cause. Routine screening of other risk factors for intra-uterine growth retardation (for example, fetal infection, TORCH syndrome, pre-eclampsia, or cord factors) was negative in all cases. Neonatal and parental blood was collected and tissue samples were obtained from at least two sites of the placenta at delivery. Standard karyotyping was then carried out. Cellular DNA was extracted from leucocytes of the neonates and their parents, and cell free DNA was collected from maternal plasma (1.6 ml) before delivery and one month after delivery, using a Qiagen blood mini kit (Qiagen, Hilden, Germany). Placental DNA was extracted after delivery. All these study protocols were approved by the Committee for the Ethical Issues on Human Genome and Gene Analysis in Nagasaki University, and written informed consent was obtained from all the women.

Genotyping

We genotyped DNA samples at various dinucleotide repeat polymorphism loci (ABI PRISM linkage mapping set MD10, Applied Biosystems, Foster City, CA) using GeneAmp PCR System 9700 (Applied Biosystems). Each genotyping was repeated at least three times to avoid erroneous results. PCR was cycled 40 times at 95°C for 10 s, 55°C for 30 s, and 72°C for 30 s, in a 20 μl mixture containing 50 ng genomic DNA, 10 pM of each fluorescent primer and reverse primer, 250 M dNTP, 0.5 U Ampli Taq Gold DNA polymerase (Applied Biosystems), and 10×PCR buffer (Applied Biosystems). PCR products were resolved on 5% denaturing polyacrylamide gels using an ABI 377 DNA sequencer (Applied Biosystems). Genotypes were analysed using an ABI GeneScan 3.1.2 and Genotyper 2.5 software (Applied Biosystems). Sensitivity testing of the procedure performed prior to the PCR-based genotyping revealed that male specific alleles were clearly detected in a 1/1000:1/10 000 male:female DNA pool.

RESULTS AND DISCUSSION

Nine of the 50 cases examined were diagnosed to have confined placental mosaicism, which is defined as identification of the same aneuploidy in at least two separate cultures—that is, a normal karyotype in lymphocytes of the neonates with intra-uterine growth retardation, and a chromosome abnormality confined to their placentas. We then analysed DNA from the neonates, placentas, and parents to determine the origin of confined placental mosaicism. Three cases (Cases 4, 5, and 6) were informative; their confined placental mosaicism was composed of a normal diploid cell line and a cell line with trisomy for 22, 7, or 2, respectively (table 1). Cases 4 and 5 were primigravida, and case 6 was multiparous. Parent-child transmission patterns of polymorphic markers showed that diploid cells in all the three neonates had one of each parental allele, whereas trisomic cells in their placentas inherited one maternal and two paternal alleles. Markers used for analysis in the three cases were as follows: D22S420, D22S280, D22S423, D22S539, D22S274, D21S263, D21S1914, D20S195, D20S173, D19S209, D19S418, D7S540, D7S640, D7S513, D7S657, D7S516, D2S2333, D2S126, D2S325, D2S364, and D2S337. In case 4, four parental alleles at a polymorphic locus (D22S280) were detected: 220 bp and 222 bp alleles in the father, and 216 bp and 226 bp alleles in the mother (fig 1a). In the placenta of this case, three alleles (216 bp, 220 bp, and 222 bp) were identified—that is, one maternally derived (216 bp) and two paternally derived (220 bp and 222 bp) alleles, while in leucocytes of the neonate, only a maternal (216 bp) and a paternal (222 bp) allele were observed, the result confirming confined placental mosaicism. On the other hand, the 220 bp allele that was absent in the neonate was detected in cell free DNA from prenatal maternal plasma before delivery, but disappeared in maternal plasma after delivery (fig 1a). Similar results were obtained in cases 5 and 6 at D7S640 and D2S337, respectively (table 1, fig 1a): a paternal allele, which was present in the placenta with confined placental mosaicism but absent in neonates, was detected in maternal plasma DNA before delivery. Genotyping of 31 normal women as negative controls at the same marker loci showed normal disomic patterns (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Allelotypes of father (F), mother (M), neonate with intra-uterine growth retardation (N), placenta (P), and maternal plasma before delivery (MP1) and after delivery (MP2). Placentas of cases 4, 5, and 6 had confined placental mosaicism for trisomies 22, 7, and 2, respectively.

Paternal DNA contamination was excluded in these three cases by one paternal and one maternal allele transmissions to the fetal tissues at other loci on non-trisomic chromosomes, and by the lack of the paternal allele, which was absent in the neonate and placenta, in the maternal plasma sample for non-trisomic chromosomes (fig 1b). Retention of fetal DNA derived from previous pregnancies in maternal plasma was unlikely, because such cell free fetal DNA generally disappears from the maternal circulation within two hours, as seen in our three cases (fig 1a), and its mean half life was reported to be 16.3 min.^20,21 In addition, it was absolutely ruled out in cases 4 and 5, because they were primigravida. These results indicate that, in all three cases, a trisomic cell line of the placenta resulted from meiotic non-disjunction in the father, and one of two paternally derived chromosomes 22, 7, or 2 was subsequently lost (trisomic rescue) during an early somatic division, leading to a normal diploid cell line both in a subset of placental cells and in the neonatal cells. It is most likely that the paternally derived allele that was absent in the neonates but was detected in maternal plasma was derived from fetal DNA in the trisomic cells of the placenta. The results also provide direct evidence for the placental origin of cell free fetal DNA circulating in maternal blood, although a possibility that the results might only reflect pregnancies with confined placental mosaicism cannot be absolutely ruled out. Measurements of cell free DNA concentrations in maternal plasma may enable the placental status to be monitored during pregnancy. In other words, cell free fetal DNA in maternal plasma may become a useful material to investigate the function, characterisation, and development of the placenta during pregnancy. Also, our results showed that the presence of two paternal alleles in maternal plasma does not necessarily indicate the presence of two paternal alleles in a fetus.

An infant with intra-uterine growth retardation (case 5) was diagnosed as having Silver-Russell syndrome (MIM 180860). Although maternal uniparental disomy for chromosome 7 (UPD7) has been known to cause 10% of Silver-Russell syndrome cases,²² such UPD was excluded in this case. Therefore, the infant’s intra-uterine growth retardation may be attributed to other causes, probably to placental dysfunction.

In conclusion, the result of our study provides direct evidence for the release of placental DNA into the maternal circulation and suggests possible usefulness of such DNA for non-invasive prenatal diagnosis of genetic diseases and screening for placental dysfunction.