Detection of fetal cells in transcervical samples and prenatal diagnosis of chromosomal abnormalities

Transcervical samples collected by lavage, aspiration, and cytobrush from women between 6 and 13 weeks of gestation were tested for the presence of fetal cells using fluorescence in situ hybridization (FISH) with probes for chromosomes X, Y, 1, and 21, and by polymerase chain reaction (PCR) amplification of DNA sequences derived from chromosomes X, Y, and 21. With a few exceptions, a good correlation was observed between the results of sexing the fetuses using FISH or PCR on transcervical cell (TCC) samples retrieved by lavage and those obtained by testing fetal (placental) tissue. In a comparative study between TCC samples collected by lavage or cytobrush, the sex of the fetus was correctly diagnosed by PCR amplification of a Y-derived DNA sequence. Variable results were observed with samples obtained by aspiration, mainly because this procedure was found to be more prone to failure to remove thick mucus without previous injection of physiological saline. Chromosome 21-derived small tandem repeats (STRs) of fetal origin were successfully detected in about 40 per cent of TCC samples recovered by lavage. Two cases of chromosomal abnormalities, one of trisomy 21 and one of triploidy, were detected in TCC samples in the course of our investigations.     

Molecular evidence of fetal-derived chromosome 21 markers (STRs) in transcervical samples

Transcervical cells (TCCs), collected by flushing or aspiration at 8–13 weeks of gestation, were analysed for the presence of fetal-derived DNA sequences. DNA extracted from maternal peripheral blood, TCC samples, and placental tissue was amplified by the polymerase chain reaction (PCR) to detect small tandem repeat (STR) markers specific to chromosome 21. STR products of fetal origin could be clearly observed in four TCC samples. TCC samples collected by flushing or aspiration were also analysed by fluorescent in situ hybridization (FISH) using X and Y probes simultaneously: 46,XY cells could be detected in all TCC samples obtained from mothers with male fetuses.     

Methods for the transcervical collection of fetal cells during the first trimester of pregnancy

Before termination of pregnancy, four techniques for retrieving fetal cells transcervically were investigated: uterine lavage, endocervical lavage, cytobrush and mucus aspiration. The yield of fetal cells in these samples was studied and found to be somewhat better after uterine lavage. A preliminary assessment of the safety of mucus aspiration was carried out before transcervical chorionic villus sampling (CVS) in continuing pregnancies. No difference in outcome to a control group having only CVS was found.     

Noninvasive prenatal diagnosis of β-thalassaemia using individual fetal erythroblasts isolated from maternal blood after enrichment

Single nucleated red blood cells (NRBCs) isolated from maternal circulation were used for prenatal diagnosis of β-thalassaemia. The study included 22 pregnant women in the first trimester, 6 carriers at risk for β-thalassaemia and 16 noncarriers. Methodology involved enrichment of NRBCs by magnetic cell sorting (MACS) and microdissection of single NRBCs with a laser micromanipulation system. Single-cell genotyping based on nested real-time PCR for genotyping β-globin gene mutations was performed followed by a multiplexed minifingerprinting to confirm the origin of the isolated cells and possible contamination. Two polymorphic markers (D13S314 and GABRB3) facilitated the identification of fetal NRBCs through comparison of allele sizes found in the respective parents. In this study, 224 single NRBCs were detached and transferred into individual PCR tubes. Allele amplification in at least one microsatellite marker was achieved in 128/224 cells. Minifingerprinting analysis showed that 22 cells were fetal, 26 maternal and 80 were noninformative due to ADO or homozygosity. In 6 NRBCs the β-globin gene was amplified and in 2, coming from the same pregnancy, only the paternal mutation was detected. The low PCR success when genotyping isolated NRBCs was possibly due to the poor quality of fetal NRBCs and the relatively large size of the β-globin gene product. Copyright © 2007 John Wiley & Sons, Ltd.     

Y chromosome detection by Real Time PCR and pyrophosphorolysis-activated polymerisation using free fetal DNA isolated from maternal plasma

Objectives To validate the use of Real Time PCR, a widely used technique that can detect very low levels of Y chromosomal sequence, and to assess the use of a highly sensitive PCR technique, pyrophosphorolysis-activated polymerisation (PAP), for fetal sex determination using free fetal DNA (ffDNA). Methods The fetal sex was determined by Real Time PCR in 58 pregnancies using ffDNA isolated from maternal plasma. In parallel with the Real Time PCR experiments, the presence of Y chromosome sequence was also determined using PAP on 54 isolated ffDNA samples. Results Both techniques detected Y chromosome sequence at very low levels with 98% specificity and 100% sensitivity (Real Time n = 44, PAP n = 54). Furthermore, the PAP technique was shown to be more robust than the Real Time PCR as none of the samples tested failed to meet the acceptance criteria. Combining the two techniques for male fetal sex detection from maternal blood plasma increases the sensitivity and specificity to 100% in this series. Conclusions This study shows that both Real Time PCR and PAP can be used for Y chromosome detection on ffDNA. Furthermore, by using PAP in combination with Real Time PCR more reliable early prenatal sexing can be performed using ffDNA. Copyright © 2007 John Wiley & Sons, Ltd.     

Flow sorting of fetal erythroblasts using intracytoplasmic anti-fetal haemoglobin: Preliminary observations on maternal samples

Monoclonal antibody to fetal haemoglobin (a₂γy₂) has been proposed as a fetal-specific reagent. We developed an intracellular staining protocol that combines fluorescein isothiocyanate or phycoerythrin conjugated anti-γ with the DNA binding dye Hoechst 33342 to identify and flow sort fetal erythroblasts from maternal blood. Our preliminary observations on anti-γ-positive cells sorted from four different pregnant women are described here, using fluorescence in situ hybridization (FISH) with chromosome-specific probes to identify fetal cells. Our data demonstrate that far fewer candidate fetal cells are sorted with this protocol than by current cell surface staining methods that employ the monoclonal antibody CD71. This results in increased fetal cell sorting purities. With this protocol, standard FISH techniques require modification due to the rigorous fixation with 4 per cent paraformaldehyde. Our initial data indicate the promise of this approach.     

Demonstration of spontaneously dividing male fetal cells in maternal blood by negative magnetic cell sorting and fish

We investigated a case of massive feto-maternal bleeding by using negative magnetic cell sorting (MACS) and fluorescent in situ hybridization (FISH). A 37-year-old pregnant woman had an uncomplicated amniocentesis for advanced maternal age at 16 weeks' gestation. The fetal karyotype was 46, XY. At 19 weeks' gestation, she had a minor car accident and slight vaginal bleeding. A subsequent Kleihauer-Betke test showed a 140 ml feto-maternal haemorrhage. Serial sonographic examinations indicated a normal fetus and placenta. We performed FISH analysis on maternal peripheral blood at 25 weeks. Anti-CD45 and MACS were used to deplete maternal leucocytes, enriching the proportion of fetal nucleated erythrocytes present. The isolated cells were analysed by using dual-colour FISH with X and Y specific probes. Approximately 65 800 nucleated cells were obtained after MACS depletion. A total of 234 cells were analysed by FISH. The results revealed that 70 of the nucleated cells (30 per cent) were male with one X and one Y signal. Among these cells, six male metaphases were observed in spontaneously dividing cells.     

Mid-trimester fetal sex determination from maternal peripheral blood by fluorescence in situ hybridization without enrichment of fetal cells

To determine the fetal sex on 30 women who were 16–20 weeks pregnant, about 100 000 maternal blood nucleated cells were analysed by means of fluorescence in situ hybridization (FISH) with a Y-chromosome-specific DNA probe. Cells with the hybridization signal were detected in 12 of the 30 women. All the 12 mothers gave birth to a male child. Of the other 18 women who had no Y-positive cells in the peripheral blood, 14 gave birth to a female child and four gave birth to a male child. These false-negative results probably occurred because the number of cells examined was inadequate. The data obtained in this study suggest that fetal sex determination using maternal peripheral blood with FISH is possible and that this diagnostic method will be clinically useful when more cells are analysed.     

New technique using galactose-specific lectin for isolation of fetal cells from maternal blood

To isolate fetal cells from maternal blood, we developed a new method based on galactose-bearing conjugation. Nucleated red blood cells (NRBCs), which highly express galactose on their surface, were selectively attached to a substrate coated with a galactose-containing polymer via soybean agglutinin (SBA), a galactose-specific lectin. Cord blood samples were used to evaluate enrichment efficacy of NRBCs by this method. Blood samples were obtained from 131 pregnant women between 6 and 27 gestational weeks. After preliminary condensation of fetal cells by Ficoll gradient centrifugation, NRBCs were enriched using galactose-positive selection by adjusting SBA concentration. We isolated one to severalhundred NRBCs (mean±SD, 7.8±8.5) in 2.3 ml of peripheral blood samples from 96% of pregnantwomen. The isolated NRBCs were analyzed by a Y-chromosome FISH probe in eight cases carrying male fetuses. Y-signals were detected in all eight cases and more than half of the NRBCs wereoffetal origin. The study demonstrates that our new method using galactose-specific lectin provides effective enrichment of fetal NRBCs allowing non-invasive prenatal diagnosis. Copyright © 2002 John Wiley & Sons, Ltd.     

Development of non-invasive fetal DNA diagnosis from maternal blood

Several attempts have been made to detect and retrieve fetal nucleated cells including nucleated erythrocytes (NRBCs), leukocytes, and trophoblasts in maternal blood. We have recently developed a new method for non-invasive fetal DNA diagnosis from maternal blood. Peripheral blood granulocytes including NRBCs were isolated by a discontinuous density gradient method using Percoll (Pharmasia). NRBCs were found and retrieved at a single cell level using a micromanipulator under a microscope. To determine whether the origin of the NRBCs was maternal or fetal, the NRBCs were analysed by polymerase chain reaction (PCR) amplification to determine the presence of a Y-chromosome-specific repeat sequence in mothers carrying male fetuses. We were successful in predicting fetal sex accurately in 10 out of 11 samples taken from maternal blood. This new technique opens up fetal DNA diagnosis from maternal blood during the first trimester of pregnancy to the whole population because there is no risk to the fetus or the mother.     

Detection of fetal HLA-DQα sequences in maternal blood: A gender-independent technique of fetal cell identification

The objective of this study was to detect fetal HLA-DQα gene sequences in maternal blood. HLA-DQα genotypes of 70 pregnant women and their partners were determined for type A1. We specifically sought couples where the father, but not the mother, had genotype A1. In 12 women, maternal blood samples were flow-sorted. Candidate fetal cells were isolated and amplified by using PCR primers specific for a paternal HLA-DQα A1 allele. Fetal HLA-DQα A1 genotype was predicted from sorted cells; amniocytes or cheek swabs were used for confirmation. Six of twelve sorted samples had amplification products indicating the presence of the HLA-DQα A1 allele; 6/12 did not. Prediction of the fetal genotype was 100 per cent correct, as determined by subsequent amplification of amniocytes or cheek swabs. We conclude that paternally inherited uniquely fetal HLA-DQα gene sequences can be identified in maternal blood. This system permits the identification of fetal cells independent of fetal gender, and has the potential for non-invasive prenatal diagnosis of paternally inherited conditions.     

Significantly higher number of fetal cells in the maternal circulation of women with pre-eclampsia

Although the pathophysiology of pre-eclampsia is unknown, several studies have indicated that abnormal placentation early in pregnancy might play a key role. It has recently been suggested that this abnormal placentation may result in transfusion of fetal cells (feto-maternal transfusion) in women with pre-eclampsia. In the present study, fetal nucleated red blood cells were isolated from 20 women with pre-eclampsia and 20 controls using a very efficient magnetic activated cell sorting (MACS) protocol. The number of male cells was determined using two-color fluorescence in situ hybridization (FISH) for X and Y chromosomes. Significantly more XY cells could be detected in women with pre-eclampsia (0.61±1.2 XY cells/ml blood) compared to women with uncomplicated pregnancies (0.02±0.04 XY cells/ml blood) (Mann–Whitney U-test, p<0.001). These results suggest that fetal cell trafficking is enhanced in women with pre-eclampsia, and this finding may contribute to the understanding of the pathophysiology of the disease. Copyright © 2001 John Wiley & Sons, Ltd.     

The time of appearance and disappearance of fetal DNA from the maternal circulation

A single copy Y-chromosome DNA sequence was amplified using the polymerase chain reaction (PCR) from the peripheral blood of 30 women who had achieved a pregnancy through an in vitro fertilization (IVF) programme. The time of conception was known precisely and was confirmed by serial ultrasound scans. Conceptions were dated as the number of weeks after fertilization plus 2, to give a time equivalent to the obstetric menstrual dating of the pregnancy (LMP). Y-chromosome-specific DNA was detected in all pregnancies with a male fetus (18/30). The earliest detection was at 4 weeks and 5 days, and the latest at 7 weeks and 1 day. Y-chromosome-specific sequences were no longer detected in any of the male pregnancies 8 weeks after delivery. No Y-chromosome sequences were detected in any of the pregnancies where only female babies were delivered. This demonstrates that fetal DNA appears in the maternal circulation early in the first trimester, that it can be identified in all pregnancies tested by 7 weeks, that it continues to be present throughout pregnancy, and that it has been cleared from the maternal circulation 2 months after parturition. Early non-invasive prenatal diagnosis for aneuploidies and inherited disorders will be possible in all pregnancies if fetal cells can be isolated free from maternal contamination (or identified accurately in the presence of maternal cells) without problems of contamination from previous pregnancies.     

Isolating fetal nucleated red blood cells from maternal blood: The baylor experience—1995

In our previous work we have isolated fetal cells from maternal blood and used fluorescent in situ hybridization (FISH) for chromosome-specific probes to detect aneuploidy. Current efforts in the Baylor College of Medicine programme are focusing on obtaining consistency in flow-sorting methodology and on determining sensitivity and specificity. To this end, systematic evaluation of five glycophorin A (gly A) antibodies all produced agglutination, leading us to abandon the use of gly A antibodies for positive selection of fetal cells. Conversely, we have found LDS-751 to be useful for nuclear selection. CD45 negative selection can best be accomplished by the use of flasks coated with goat antibodies against mouse antibodies. Positive selection by flow sorting for either CD71⁺ cells or gamma-globin-positive cells seems to be successful. Using these two approaches, we have recently detected male (fetal) cells in pregnancies in which the fetus was 46, XY in 10 of 18 and in 12 of 14 cases, respectively.     

Microsatellite analysis provides efficient confirmation of fetal trophoblast isolation from maternal circulation

Fetal trophoblasts can be found in maternal circulation from an early stage of pregnancy and thus provide a potential source of DNA for non-invasive prenatal diagnosis. We have developed a two-step method for trophoblast isolation between the 8th and 12th week of pregnancy. Blood was sampled from 14 women undergoing termination of pregnancy or spontaneous abortion. Immunomagnetic beads precoated with HLA class I and II, and with anti-cytokeratin-18 monoclonal antibodies, were used to remove CD8+ and other maternal cells, and to select for fetal trophoblasts, respectively. Microsatellite analysis was performed on DNA extracted from the isolated, maternal, paternal and placental cells after PCR amplification. Recovery of the trophoblasts was confirmed in 13/14 cases (93%) by the identification of an identical microsatellite pattern for fetal and placental cells. Further evidence was the presence of heterozygous alleles of both maternal and paternal origin. The correct prediction of gender in all five male fetuses was an additional confirmation of trophoblast recovery. We conclude that trophoblasts can be effectively isolated from maternal blood in the first trimester, and by using polymorphic microsatellite markers to confirm sample purity, this method has potential future application in prenatal diagnosis. Copyright © 2001 John Wiley & Sons, Ltd.     

Maternal contamination of amniotic fluid demonstrated by DNA analysis

DNA from 16 sets of samples comprising DNA from uncultured amniotic fluid cells, cultured amniotic fluid cells, fetal tissue, and maternal blood was analysed by the polymerase chain reaction (PCR) with AC-repeat primers. The analysis was performed to investigate the presence of contaminating maternal cells in amniotic fluid which would affect the reliability of DNA studies for prenatal diagnosis. In three sets, maternal contamination of uncultured amniotic fluid cells was detected. In one of the three sets, maternal contamination was present in both uncultured and cultured amniotic fluid cells. The use of amniotic fluid cells as a source of DNA for prenatal diagnosis should be limited to cases where the purity of the DNA can be demonstrated prior to the diagnostic test being performed. This limitation in the use of amniotic fluid DNA also extends to other forms of diagnosis relying on the purity of amniotic fluid samples, particularly the new in situ hybridization methods currently being developed.     

Extraction of DNA from amniotic fluid cells for the early prenatal diagnosis of genetic disease

Ten-ml samples of amniotic fluid were taken from pregnancies being terminated at 8–14 weeks' gestation. DNA was extracted from the amniotic cells by sequential centrifugation and analysed using the polymerase chain reaction (PCR). Fifteen samples were analysed for evidence of maternal contamination using Mfd5 oligo-nucleotide primers for repeat polymorphisms. Ten amniotic fluid samples were tested for the Delta-F₅₀₈ deletion characteristic of cystic fibrosis to demonstrate a diagnostic application for the technique. In each case, DNA extracted from fetal tissue from the same pregnancy was included in the controls. In 14 of the 15 cases tested with the Mfd5 primers, both the amniotic fluid DNA and the fetal DNA showed no evidence of contaminating DNA. In one case, neither the amniotic fluid cells nor the fetal cells yielded results. In nine of the ten cases tested with the Delta-F₅₀₈ primers, the amniotic fluid cell DNA provided accurate information about the genetic status of the fetus; in the tenth, the fetal DNA failed to amplify. The results indicate that adequate DNA can be extracted from amniotic fluid from 8 weeks' gestation onward and these samples are suitable for prenatal diagnosis using PCR.     

Rapid prenatal diagnosis of 14 cases of triploidy using fish with multiple probes

Fluorescence in situ hybridization (FISH) of chromosome-specific probes to interphase nuclei can rapidly identify aneuploidies in uncultured amniotic fluid cells. Using DNA probe sets specific for chromosomes 13, 18, 21, X, and Y, we have identified 14 fetuses where the hybridization pattern was consistent with a triploid chromosome constitution. In each case, the identification of fetal abnormalities by ultrasound examination initiated a request for rapid determination of ploidy status via prenatal FISH analysis of uncultured amniocytes. FISH produced a three-signal pattern for the three autosomes in combination with signals indicating an XXX or XXY sex chromosome complement. This hybridization pattern was interpreted to be consistent with triploidy. Results were reported to the physician within 2 days of amniocentesis and subsequently confirmed by cytogenetics. These cases demonstrate the utility of FISH for rapid prenatal identification of triploidy, particularly when fetal abnormalities are seen with ultrasonographic examination.     

First-trimester fetal sex determination in maternal serum using real-time PCR

An Erratum has been published for this article in Prenatal Diagnosis 22(13) 2002, 1241. Fetal sex prediction can be achieved using PCR targeted at the SRY gene by analysing cell-free fetal DNA in maternal serum. Unfortunately, the results reported to date show a lack of sensitivity, especially during the first trimester of pregnancy. Therefore, determination of fetal sex by maternal serum analysis could not replace karyotype analysis following chorionic villus sampling. A new highly sensitive real-time PCR was developped to detect an SRY gene sequence in maternal serum. Analysis was performed on 121 pregnant women during the first trimester of pregnancy (mean gestational age: 11.8 weeks). Among them, 51 had at least one previous male-bearing pregnancy. Results were compared with fetal sex. SRY PCR analysis of maternal serum was in complete concordance with fetal sex. Among the 121 pregnant women, 61 were bearing a male fetus and 60 a female fetus. No false-negative results were observed. Furthermore, no false-positive results occurred, even though 27 women carrying a female fetus during the current pregnancy had at least one previous male-bearing pregnancy. This study demonstrates that a reliable, non-invasive sex determination can be achieved by PCR analysis of maternal serum during the first trimester of pregnancy. This non-invasive approach for fetal sex prediction should have great implications in the management of pregnant women who are carriers of an X-linked genetic disorder. Prenatal diagnosis might thus be performed for male fetuses only, avoiding invasive procedures and the risk of the loss of female fetuses. Copyright © 2001 John Wiley & Sons, Ltd.     

Maternal cell contamination in cultured chorionic villi: Comparison of chromosome Q-polymorphisms derived from villi,fetal skin,and maternal lymphocytes

Maternal cell contamination of chorionic villi (CV) samples used for first trimester prenatal diagnosis can cause obvious and/or unrecognized diagnostic dilemmas. The purpose of this investigation is to assess the frequency of maternal cell contamination (MCC) in chorionic villus samples and to evaluate selected parameters which might predict where contamination is more likely to have occurred. Maternal lymphocytes, chorionic villi from ultrasonically directed transcervical catheter aspiration, and fetal tissue were obtained at 8–11 weeks gestation from 45 patients undergoing elective termination. Quinacrine (Q) banded metaphases were compared from duplicate direct preparations of chorionic villi; cultured chorionic villi, fetal fibroblast tissue cultures, and maternal lymphocyte cultures. Q-polymorphisms in metaphase chromosomes were 100 per cent concordant between fetal tissue and direct CV preparation. However, evidence for maternal cell contamination occurred in 13.1 per cent of cultured chorionic villi preparations where polymorphisms were found to be identical between maternal and cultured CV and both distinct from fetal tissue preparations. Where MCC was identified, it was noted that CV cell cultivation interval was prolonged (24.2±6.8 days) compared with non-contaminated cultures (14.1±4.4 days) (p <0.05). We conclude that maternal cell contamination is a significant problem with chorionic villus sampling. Where direct preparations are not employed or when cultures are ‘slow growing’, MCC may be a significant and unrecognized complication re: fetal diagnosis. Direct preparations, multiple cultures, quinacrine banding, and maternal Q-polymorphism comparisons can minimize diagnostic dilemmas secondary to maternal cell contamination. Q-polymorphism comparisons between maternal and fetal chromosomes should be included in all instances where cultured chorionic villi are utilized for fetal diagnosis and where direct preparations are not available.