Advances on circulating fetal DNA in maternal plasma

The discovery of cell-free fetal DNA in maternal plasma in 1997 has opened up new possibilities for noninvasive diagnosis.1 By RT-PCR, circulating fetal DNA can be detected in the plasma of pregnant women, even in the first trimester of pregnancy,2,3 and thus can be used for noninvasive prenatal diagnosis of sex-linked disorders,4–6 the RhD status of fetuses,7 and single gene disorders such as beta-thalassaemia,8,9 congenital adrenal hyperplasia,10 and achondroplasia.11 In addition, quantitative aberrations of circulating fetal DNA may indicate various pregnancy-associated disorders, including preeclampsia,12–14 preterm labor15,16 and fetal trisomy 21.17 However, there are two open questions in the application of circulating maternal DNA. One question is how to amplify fetal DNA from circulating maternal DNA. Since the quantity of background (circulating maternal) DNA may influence the detection rate of low copy number sequences,18 some experts selected proper speed of centrifugation for plasma separation,19 or used different primer and probe sets20 and added formaldehyde to the blood collection tubes.21 Although these methods can increase the fetal fraction DNA concentration in a proportion of samples to some extend, the detection rates are not improved. Another question is what can be used as the internal positive control, demonstrating that a particular maternal plasma sample actually contains a detectable concentration of circulating fetal DNA. Y chromosomal sequences from male fetuses have been used as an internal positive control.22 However, obviously, it is not applicable for female fetuses. A panel of insertion-deletion polymorphisms allows the detection of fetal-derived, paternally-inherited polymorphisms that are distinguishable from the maternal DNA background in maternal plasma,23 but may increase the complexity of the resulting diagnostic assay. In this article, we present some advances on the circulating fetal DNA in maternal plasma that may contribute to the widespread utilization of plasma DNA technology. FURTHER CHARACTERIZATION OF FETAL DNA IN MATERNAL PLASMA Detection of relatively low levels of fetal DNA sequences (as compared to maternal DNA levels) depends on the sensitivity of the assay and the amount of target fetal sequences. Several studies have confirmed that gestational age positively correlates with the amount of fetal DNA in plasma. The concentration of fetal DNA is low in the first trimester, and increases with gestational age.24 In male fetuses, the fetal DNAs are detectable from the 5th gestational week, increases during gestation (R2=0.52 by linear regression), and reach a high level in the third trimester.2 Therefore, higher detection rates can be achieved during late gestation; and proper time for diagnosis can be selected according to different diseases. Enrichment of circulating fetal DNA from the overwhelming amount of circulating maternal DNA is another topic. Various sized PCR amplicons have been used to discern the respective size distributions of circulatory fetal and maternal DNA species. The results showed that the DNA molecules in the plasma of pregnant women were longer than those in non-pregnant women; in addition, in the maternal plasma, fetal DNA molecules are generally shorter than maternal DNA molecules.25 These results have been confirmed by some different methods independently.26,27 The different size distribution of DNA in the plasma of pregnant women may open up a possible way to enrich fetal DNA by size fractionation of DNA extracted from the plasma of pregnant women. Such enrichment is useful for the prenatal diagnosis of autosomal recessive disorders and chromosomal aneuploidies. OPTIMAL MEASUREMENTS OF MALE FETAL DNA Noninvasive prenatal diagnosis in the first trimester is one of the long-sought goals in human genetics. Despite a large number of experiments have been done for the quantification of fetal DNA, data about the amount of DNA in early pregnancy are still limited.28 In 2005, Zimmermann and colleagues3 developed an assay specific to the multicopy sequence DYS 14 on the Y chromosome and compared it with the single-copy SRY sequence. Through probit regression analysis, they found that, by the DYS14 assay, measurements of male DNA had a 10-fold lower detection limit (0.4 genome equivalents) than those of SRY, and the DNA concentrations determined by the DYS14 protocal were 2.5-fold higher than those determined by the SRY protocol (range: 1.4-fold to 4.3-fold). In addition, for plasma samples from women in the first trimester of pregnancy, imprecision (CV) was 2% to 22% by amplifying DYS14, while it was 26% to 140% by using the SRY. The sensitivity and specificity of the DYS14 protocol for sex determination were both 100% (25/25 and 33/33), while those of the SRY were 75% (8/12) and 91.7% (11/12), respectively. Then, in 2006, Stanghellini and coworkers29 designed a new multi-locus probe mapping on the deleted in azoospermia (DAZ) repetitive region of the Y chromosome to assess the fetal DNA concentration in maternal plasma. They revealed that fetal DNA could be detected by the 5th gestational week, and the results were reliable by the 8th gestational week with an accuracy of 100% in fetal sex determination. Shortly later, a Y chromosome highly polymorphic short tandem repeat (Y-STR) multiplex PCR amplification system was developed,30,31 and used for genotyping 9 Y chromosome specific STR loci (DYS426, DYS393, DYS460, DYS391, DYS389I, DYS456, H4, DYS388, and DYS458), whose allele sizes were less than 180 bp in length. An average of 7.3 Y-STR specific alleles was detected in each of the 30 plasma DNA samples from women pregnant with male fetus, and none of the 9 Y-STR specific alleles was detected in 34 samples from those pregnant with female fetus. The results indicate that the 9 Y-STR loci with short fragment sized alleles are highly polymorphic, which was suitable for the male fetal markers. The discovery of the Y chromosome specific STR loci largely widens the range of detection in early prenatal diagnosis. In addition, by microarray coupled with emulsions PCR, Ge and colleagues32 detected 6 Y chromosome specific sequences in genes of SRY, DYS and DYZ as the markers of fetal DNA, with which the earliest plasma sample was collected on the 31st day after pregnancy. The sensitivity and specificity for sex determination were 95.2% (40/42) and 100% (18/18). In short, all these developments on the optimal measurements of male fetal DNA lay the foundation for the noninvasive prenatal diagnosis, especially in early pregnancy. PLASMA EPIGENETIC MARKERS FOR NONINVASIVE PRENATAL DIAGNOSIS In 2002, a completely different approach to the development of a universal fetal DNA marker was proposed.33 The investigators postulated that it might be possible to find epigenetic differences between the fetal and maternal DNA molecules that are present in maternal plasma.33 Epigenetic events include stable and heritable (or potentially heritable) changes in gene expression that do not entail a change in DNA sequence.34 The role of epigenetics in etiology has been increasingly recognized, since more and more evidence showing that genes subject to genomic imprinting. Of the epigenetic phenomena, DNA methylation has been well investigated. By studying an imprinted locus, Poon and colleagues33 found that unmethylated maternally inherited alleles in the fetus are different from the unmethylated alleles of the mother; and fetal-derived maternally inherited alleles can be detected in maternal plasma. Based on the results, we may discriminate the fetal DNA molecules from the maternal ones in maternal plasma. Maspin gene, a tumor suppressor gene expressed in the placenta, is unmethylated in placental cells, but is methylated in maternal blood cells.35,36 DNA methylation plays an important role in regulating the tissue-specific expression of maspin.37 Data have shown that hematopoietic cells are the predominant source of plasma DNA, while placenta is an important source of fetal nucleic acid in maternal plasma.38–41 Since the pattern of DNA methylation of maspin in the placenta is different from that in the hematopoietic cells, we may use unmethylated maspin (U-maspin) as a fetal marker in maternal plasma. By using the methylation-specific PCR,42 U-maspin sequences have been detected in maternal plasma samples obtained from women pregnant with fetuses of both sexes in all trimesters.35 Chim and coworkers35 found that the median concentration of U-maspin DNA was 17.7 copies/ml, 51.6 copies/ml, and 194.5 copies/ml for the first, second, and third trimesters, respectively, with statistically significant differences among them (Kruskal-Wallis test, P=0.00091). Moreover, the concentration of U-maspin sequences decreases quickly to an almost undetectable level within 24 hours after delivery (P=0.002, Wilcoxon signed-rank test).35 In their experiments, median concentration of U-maspin in the maternal plasma of preeclamptic group (median, 737.7 copies/ml) was 5.7-fold higher than that of control (median, 130.3 copies/ml, P=0.01158, LOGXACT logistic regression by matched case-control analysis). In addition, through single nucleotide polymorphism (SNP) genotyping, they further revealed that the unmethylated maspin sequences detected in maternal plasma contain the fetal SNP genotype, demonstrating that the epigenetic approach can be utilized to develop universal fetal DNA markers in all trimesters, irrespective of fetal gender and genetic polymorphic differences between the mother and fetus. With the launching of the Human Epigenome Project,43 we believe that further epigenetic markers for prenatal diagnosis will be developed over the next few years. And more technologies, which allow the large-scale performance of methylation analysis, will be improved to replace methylation-specific PCR-based technologies,44,45 since methylation-specific PCR assays are relatively complex and time-consuming. Recently, a number of mass spectrometry-based methods for methylation analysis have been developed.46–48 It is possible that some of them could be modified to detect fetal epigenetic markers in maternal plasma. In conclusion, based on these advances on circulating fetal DNA in maternal plasma, it is very likely that maternal plasma DNA analysis will play an important role in prenatal diagnosis and monitoring.49,50