| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Centre de Diagnostic Prénatal, American Hospital of Paris, 63 bd Victor Hugo, 92200 Neuilly-sur-Seine, France
2 Maternité, Hôpital Necker-Enfants Malades, 75015 Paris, France
3 Laboratoire de Génétique Moléculaire, Faculté des Sciences Pharmaceutiques et Biologiques de Paris, 75006 Paris, France
aauthor for correspondence: fax 33-1-46-41-26-56, e-mail jean-marc.costa{at}ahparis.org
Noninvasive approaches to prenatal diagnosis can avoid the risk of fetal loss associated with invasive procedures such as chorionic villus sampling, amniocentesis, and cordocentesis. Isolation of fetal cells from maternal blood requires further improvements before it can be applied in a clinical setting (1), but the reliability of cell-free fetal DNA analysis in maternal plasma or serum is now well established (2)(3)(4). As a result, it is currently used in specialized centers for the determination of fetal sex and fetal RhD status for the management of pregnant women at risk for X-linked disorders (5) or RhD alloimmunization (6). Because fetal DNA in maternal serum is circulating in an excess background of maternal DNA, clinical applications are restricted mainly to the detection of fetal sequences distinct from the mother’s DNA sequences.
Fetal RNA in maternal blood may be an alternative source of fetal nucleic acids. Al-Mufti et al. (7) detected specific RhD mRNA in mononuclear fetal cells isolated from blood of RhD-negative pregnant women, and Lo’s group demonstrated the presence of Y-chromosome-specific (ZFY) mRNA in maternal plasma of women carrying a male fetus (8). These two applications, however, again require fetal sequences that differ from the mother’s.
We have investigated the presence in maternal blood of fetal transcripts that may have the same sequence as that of the mother. Human chorionic gonadotropin (hCG) mRNA is a good candidate because it is a pregnancy-specific polypeptide hormone produced by the placenta and is specifically expressed in the fetal syncytiotrophoblast.
We studied 43 pregnant women and 20 nonpregnant women who had previously given birth to at least one neonate. After receiving informed consent, we collected blood (2.5 mL) into PAXgeneTM tubes to reduce RNA degradation (9). All pregnancy samples were obtained before any invasive procedure during either the first (n = 23) or the second (n = 20) trimester. The mean gestational ages were 11.8 weeks (range, 9–14.5 weeks) and 19.8 weeks (range, 18–23 weeks), respectively.
We collected an additional 7 mL of blood into Vacutainer SST® tubes (Becton Dickinson). Immediately after clotting, the serum was separated by centrifugation at 3000g for 10 min at 4 °C, and 2.5 mL was transferred to PAXgene tubes.
Total blood and serum were treated according to the same protocol. RNA was isolated with the PAXgene Blood RNA reagent set (Qiagen) as recommended by the manufacturer, except that RNA was eluted in 50 instead of 100 µL of elution buffer. A DNase digestion step was systematically included during the procedure, and the integrity of the RNA was monitored by 1% agarose gel electrophoresis. Each sample extract was tested in duplicate.
Each extracted sample (5 µL) was subjected to reverse transcription in 20 µL of a reverse transcriptase mixture containing 1x reverse transcription buffer (0.5 mM each deoxynucleotide triphosphate, 5 mM MgCl2, 75 mM KCl, 50 mM Tris-HCl, pH 8.3), 2.5 µM random hexamers, 10 units of RNase inhibitor, and 15 U of Multiscribe reverse transcriptase (Applied Biosystems). The mixture was incubated for 10 min at 25 °C, followed by 60 min at 42 °C, and the reverse transcriptase was inactivated by heating at 99 °C for 10 min. A reaction mixture without the reverse transcriptase was used as a negative control.
We used 5 µL of the reverse transcription reaction for specific detection of hCGß transcripts in a real-time PCR method adapted from a previously described method (10). Amplification was carried out in a LightCycler® instrument (Roche Biochemicals). PCR reactions were set up in a final volume of 20 µL, using the Fast DNA Master Hybridization Probes Kit (Roche Biochemicals), with 0.5 µM each primer, 0.25 µM each probe, 1.25 U of uracil DNA glycosylase (Biolabs), and 4.5 mM MgCl2. After an initial denaturation step of 8 min at 95 °C, amplification was performed for 50 cycles (denaturation at 95 °C for 10 s, annealing at 60 °C for 10 s, and extension at 72 °C for 15 s). The annealing step for each sample was monitored by continuous fluorescence monitoring.
RNA obtained from a first-trimester chorionic villus sample was introduced during each run as a positive control for hCGß mRNA. As a control for the quality of the extracted RNA, all samples were also subjected to a reverse transcription-PCR assay targeted to actin mRNA, which is expressed by both fetal and maternal tissues. The primer (MWG-Biotech) and probe (Proligo) sequences are listed in Table 1
. The primers for hCGß mRNA are specific and do not amplify luteinizing hormone-ß (hLHß) mRNA (11)). Free hCGß protein in serum was measured on the Kryptor analyzer (Brahms).
|
Actin gene reverse transcription-PCR products were detected in all tested samples, establishing the presence of amplifiable RNA. Specific hCGß transcripts were demonstrated in 18 (78%) of the 23 first-trimester blood samples but in only 9 (45%) in the second-trimester group. Typical results are shown in Fig. 1
. hCGß mRNA was not detected in any corresponding sera or in any blood samples from nonpregnant women.
|
For samples positive for hCGß mRNA, results were expressed as the crossing point, defined as the maximum of the second derivative of the fluorescence curves. The crossing point was lower in first-trimester than second-trimester samples (38.9 vs 43.7 cycles), suggesting an 
30-fold higher concentration of hCGß mRNA in the early samples. No correlation was observed between the hCGß transcripts and the free-hCGß protein concentration in the pregnancy samples (P = 0.25).
In this study, we used an assay specific for hCGß transcripts to test whole blood or serum from nonpregnant women and from pregnant women during the first and second trimesters of pregnancy. Results showed that such transcripts were present only in whole blood samples from pregnant women, especially during the first trimester of pregnancy.
The possibility that the detected transcripts were not from the hCGß gene but from the highly homologous hLHß gene was excluded by use of specifically designed primers (10)(11)(12). hCGß transcripts have been reported in the white blood cells of nonpregnant and non-cancer patients (13), but our results (blood samples from all nonpregnant women were negative) as well as other reports (14) do not support these findings.
That the presence of hCGß mRNA in the peripheral blood reflects circulating tumor cells has previously been shown in metastatic breast cancer (14), germ-cell tumors (15), and gestational trophoblastic disease (16). Because our population was presumably not affected by cancer, the detected transcripts were most probably hCGß transcripts of fetal origin.
These results obtained on whole blood samples raise questions concerning the source of this fetal RNA. No hCGß mRNA was detected in the corresponding sera. This strongly suggests that the major part of the detected fetal RNA is circulating in a cell- or particle-associated form in maternal blood. Poon et al. (8) have shown that fetal RNA circulates mostly in a cell-free form, and Ng et al. (17) concluded that a substantial proportion of plasma mRNA species are particle-associated.
The fact that we found no fetal RNA circulating in the blood of nonpregnant women with a history of pregnancy seems to indicate that this cell- or particle-associated RNA does not persist after delivery. However, this conclusion needs to be confirmed in women with fetal microchimerisms because it has recently been shown for fetal DNA (18).
Interestingly, we observed the highest detection rate for hCGß mRNA during the first trimester of pregnancy, whereas Lo’s group found the lowest detection rate for ZFY mRNA in early pregnancy (8). Therefore, the amount of fetal RNA in maternal blood does not increase with gestational age simply as fetal DNA (19); it is also gene-specific regarding the developmental stage of fetal tissue origin. Furthermore, there was no correlation between free-hCGß protein concentrations and the detection rate for hCGß transcripts, which provides an opportunity to study the fetus at the molecular level.
This study demonstrates that tissue-specific fetal RNA sequences can be detected in maternal blood in pregnant women regardless of the genetic situation of the fetus relative to the mother.
Such noninvasive access to the gene expression pattern of fetal tissues offers new possibilities. The discovery of additional fetal RNA markers (e.g., trophoblast-produced human placental lactogen or placental growth hormones) may open new opportunities for detecting and monitoring aberrant or differentially expressed genes. Fetal RNA analysis in maternal blood may therefore provide information regarding gene expression in many physiologic and pathologic conditions, especially those involving trophoblastic tissue, even if the question of the role of such transcripts remains to be elucidated.
Acknowledgments
We are indebted to Dr. Lavergne for reviewing this manuscript.
References
-fetoprotein-messenger RNA and ß human chorionic gonadotropin-messenger RNA. Cancer Res 2000;60:3170-3174.The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  S. Kumar and A. O'Brien Recent developments in fetal medicine BMJ, April 24, 2004; 328(7446): 1002 - 1006. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
