| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
Departments of1 Chemical Pathology and 2 Obstetrics and Gynaecology, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, New Territories, Hong Kong Special Administrative Region, China
aaddress correspondence to this author at: Department of Chemical Pathology, The Chinese University of Hong Kong, Prince of Wales Hospital, 30-32 Ngan Shing St., Shatin, New Territories, Hong Kong Special Administrative Region, China; fax 852-2194-6171, e-mail loym{at}cuhk.edu.hk
The discovery of fetal DNA in maternal plasma has opened up new possibilities for noninvasive prenatal diagnosis (1)(2)(3)(4)(5)(6). A recent report(7) indicated that the proportion of fetal DNA in maternal plasma can be dramatically enriched through the adoption of a blood-processing protocol involving the addition of formaldehyde to maternal blood samples. Dhallan et al. (7) suggested that these observations might be the result of several factors, including a reduction in background maternal DNA by minimization of maternal cell lysis through formaldehyde-mediated cell membrane stabilization and the use of a gentle centrifugation protocol, as well as the preservation of fetal DNA through nuclease inhibition by formaldehyde.
In view of the profound implications of the study (7), we aimed to validate and investigate the underlying mechanisms of the reported phenomenon. To assess the effects of several contributory factors, our study was conducted in three successive stages. In the first part of the study, we aimed to verify the effects of the previously published protocol (7) on total DNA concentrations in plasma from nonpregnant individuals. In the second part of the study, we evaluated the effects of formaldehyde addition on total and fetal DNA concentrations in maternal plasma in relation to the time of blood processing (0, 6, and 24 h after blood collection). In the last part of the study, we investigated whether the reported enrichment in circulating fetal DNA concentrations (7) might be a consequence of the imprecision of the analytical method chosen by the authors.
All participants were recruited with informed consent from the Prince of Wales Hospital, Hong Kong, with institutional ethics approval. In the first part of the study, blood (24 mL) was collected into 3-mL EDTA tubes from each of eight nonpregnant volunteers and divided into four equal portions immediately after venesection. One portion was processed by a previously described two-step centrifugation protocol (8), and the remaining three portions were centrifuged by the gentle centrifugation protocol described in the previous formaldehyde study (7). One of the latter three portions was not treated further, whereas formaldehyde was added to the other two portions before or after centrifugation. For every 3 mL of blood, we added 75 µL of 10% neutral-buffered solution containing formaldehyde (40 g/L). This evaluation protocol was designed to assess whether the gentle centrifugation protocol and formaldehyde addition each contributed independently to the minimization of cell lysis and, thus, the decrease in total plasma DNA concentration. Plasma supernatants were carefully removed, and DNA was isolated by use of a QIAamp DNA Blood Midi Kit (Qiagen) and subjected to real-time quantitative PCR (RQ-PCR) for ß-globin (9) on a 7900 Sequence Detector (Applied Biosystems).
Statistical analyses were carried out using Sigma Stat 3.0 (SPSS).
For plasma samples obtained immediately after venesection, ß-globin DNA concentrations did not differ significantly among portions processed by the centrifugation protocols with or without formaldehyde addition before or after centrifugation (P = 0.615, Friedman; Table 1
). As previous studies reported that circulating DNA concentrations would increase substantially after undue delays in blood processing (10)(11), the second part of the study was designed to investigate the interrelationships between formaldehyde addition, plasma DNA concentrations, and delayed blood processing in pregnant women.
|
Pregnant women carrying male fetuses as identified by fetal ultrasound scanning were recruited for part 2 of the study. Blood (12 mL) was collected into 3-mL EDTA tubes from each of 13 and 17 pregnant women in the second (gestational age, 17–20 weeks) and third (gestational age, 36–41 weeks) trimesters of pregnancy, respectively. Portions of the whole blood specimens were centrifuged either immediately or after 6 or 24 h of storage at 4 °C. One half of each portion was treated with formaldehyde. Plasma SRY and ß-globin DNA was quantified by RQ-PCR (9). We compared the differences among the SRY, ß-globin, and fractional fetal DNA concentrations in the maternal plasma aliquots processed at the three different time-points with or without formaldehyde addition statistically were compared by Friedman repeated-measures ANOVA on ranks followed by all-pairwise post hoc analysis using the Dunn method to identify the specific groups that differed from the others.
The median ß-globin, SRY, and fractional fetal DNA concentrations for the differently processed maternal plasma aliquots are summarized in Table 1
. The maternal plasma ß-globin concentrations for the different blood processing protocols were statistically different for samples from both the second (P = 0.02, Friedman) and third (P = 0.002, Friedman) trimesters of pregnancy. Post hoc analysis revealed that among all possible comparisons, the only significant difference was between the ß-globin concentrations in the two third-trimester maternal plasma portions processed with 24-h delay with and without formaldehyde addition (P <0.05, Dunn). The corresponding ß-globin concentrations in the second-trimester plasma aliquots processed at 24 h demonstrated the greatest difference in ranks among all possible comparisons. However, this difference did not reach statistical significance in the post hoc analysis (P >0.05, Dunn). We found no statistical difference between the maternal plasma SRY concentrations (Table 1
) for any of the plasma aliquots in either the second (P = 0.600, Friedman) or third (P = 0.697, Friedman) trimesters of pregnancy. The plasma portions treated with formaldehyde demonstrated a small (1–3.3% increase in median values; Table 1
) but nonsignificant increase in the fractional fetal DNA concentrations at each of the time-points in both the second (P = 0.237, Friedman) and third trimesters (P = 0.339, Friedman) of pregnancy (Table 1
). These data contrasted markedly with the extent of fetal DNA enrichment reported previously (7) (59% of samples with fetal DNA percentages >25% with a range extending to >50%).
To investigate the cause of the discrepancy, we compared the analytical performance of RQ-PCR (9) with the serial dilution method used in the original formaldehyde study (7) for quantification of maternal plasma DNA. Blood (12 mL) was collected from two pregnant women carrying male fetuses (gestational age, 17 and 18 weeks). Formaldehyde was added, and plasma was obtained by gentle centrifugation (7). A solution containing 4.5% male DNA [20 genome-equivalents (GE) of SRY among 450 GE of ß-globin] was also prepared by mixing male and female genomic DNA. Extracted DNA from the three specimens was separated into 10 aliquots and assayed for SRY and ß-globin. The fetal DNA percentage of each aliquot (30 aliquots in total) was determined by both the RQ-PCR method [SRY (GE/mL)/ß-globin (GE/mL) x 100] (9) and the method based on determining the highest dilutions of input DNA for which the SRY and ß-globin amplicons were detectable by gel electrophoresis (7). Six serial 1:5 dilutions (1:5 to 1:15 625) were made for all aliquots.
Ten-replicate analyses of the artificial male DNA mixture yielded median [mean; interquartile range (IQR)] fetal DNA percentages of 6.3% (6.7%; 6.0–7.3%) by RQ-PCR and 8.0% (15.7%; 1.6–40%) by serial dilution (Fig. 1
). The results based on the latter method were particularly noteworthy in that they (a) deviated more from the expected 4.5% than that determined by RQ-PCR; (b) exhibited a wide discrepancy between the mean and median; and (c) demonstrated large fluctuations on repeated measurements as evident by the wide IQR (Fig. 1
). In fact, these were signs of inaccuracy, skewed data distribution, and analytical imprecision, respectively.
|
A priori, RQ-PCR is likely to be more accurate and precise than the serial dilution method because male DNA percentages can be expressed as any value between 0% and 100% by RQ-PCR whereas, based on a fivefold dilution series, male DNA percentages can be scored only as one option among a discrete set of percentages, i.e., 0.0128%, 0.064%, 0.32%, 1.6%, 8%, or 40%. Because of the categorical nature of this scoring method and the uneven numerical distribution of the categories, the data would be skewed and inaccurately represented. For example, fetal DNA concentrations between 8% and 40% may only be scored as 8% or be promoted to the next category, namely, 40%. In fact, this phenomenon was also observed from the datasets presented in the previous formaldehyde study (7). Using a twofold dilution series, the authors reported fetal DNA concentrations of >50% in several maternal plasma samples [see Table 2 in Dhallan et al. (7)]. By virtue of the twofold dilution, a fractional fetal DNA concentration >50% would essentially be scored as 100%.
In addition, the analytical result and the performance of the serial dilution method can be a function of the arbitrary choice of the dilution factor. Although previous work discussed above used 2- and 5-fold dilutions, the effect of large dilution factors can be illustrated by considering 10-fold dilutions. The artificial DNA mixture was additionally analyzed by 10-replicate measurements by a series of four 10-fold dilutions, where the male DNA percentage can be expressed only as 0.02%, 0.2%, 2%, or 20%. As expected, the male DNA percentages deviated further from the expected (median, 0.2%; mean, 24.3%; IQR, 0.2–20%), with the skewness becoming more severe (Fig. 1
). These data could be interpreted in completely opposite ways depending on whether the median (much lower than the true value of 4.5%) or the mean (much higher than the true value of 4.5%) was chosen for presentation.
The issue associated with analytical imprecision was equally applicable to the estimation of fetal DNA percentages. Repeated measurements of the two second-trimester maternal plasma samples by RQ-PCR yielded a median of 3.9%, mean of 3.7%, and IQR of 2.1–5.3% and a median of 5.2%, mean of 5.0%, and IQR of 2.8–7.2%, respectively. However, results for the fivefold serial dilution method were as follows: median, 4.8%; mean, 11.1%; IQR, 1.6–8%; and median, 1.6%; mean, 3.4%; IQR, 1.6–8%, respectively (Fig. 1
).
We therefore conclude that, in our hands, gentle centrifugation did not confer any observable advantage and formaldehyde addition did not yield the previously reported dramatic increases in fractional concentrations of fetal DNA (7). The latter could have resulted from the imprecise estimation of the fetal DNA percentages in maternal plasma when the previously reported serial dilution method was used.
Acknowledgments
This work was supported by an Earmarked Research Grant (CUHK4395/03M) from the Research Grants Council of the Hong Kong Special Administrative Region, China. Y.M.D.L. and R.W.K.C. hold patents and patent applications on aspects of fetal nucleic acid analysis from maternal plasma.
Footnotes
1 these authors contributed equally 
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  E C W Hung, R W K Chiu, and Y M D Lo Detection of circulating fetal nucleic acids: a review of methods and applications J. Clin. Pathol., April 1, 2009; 62(4): 308 - 313. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. F. Wright and H. Burton The use of cell-free fetal nucleic acids in maternal blood for non-invasive prenatal diagnosis Hum. Reprod. Update, January 1, 2009; 15(1): 139 - 151. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. M. F. Lun, N. B. Y. Tsui, K. C. A. Chan, T. Y. Leung, T. K. Lau, P. Charoenkwan, K. C. K. Chow, W. Y. W. Lo, C. Wanapirak, T. Sanguansermsri, et al. Noninvasive prenatal diagnosis of monogenic diseases by digital size selection and relative mutation dosage on DNA in maternal plasma PNAS, December 16, 2008; 105(50): 19920 - 19925. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. M. F. Lun, R. W. K. Chiu, K. C. Allen Chan, T. Yeung Leung, T. Kin Lau, and Y. M. Dennis Lo Microfluidics Digital PCR Reveals a Higher than Expected Fraction of Fetal DNA in Maternal Plasma Clin. Chem., October 1, 2008; 54(10): 1664 - 1672. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. M. Dennis Lo and R. W. K. Chiu Noninvasive Prenatal Diagnosis of Fetal Chromosomal Aneuploidies by Maternal Plasma Nucleic Acid Analysis Clin. Chem., March 1, 2008; 54(3): 461 - 466. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. M. D. Lo, F. M. F. Lun, K. C. A. Chan, N. B. Y. Tsui, K. C. Chong, T. K. Lau, T. Y. Leung, B. C. Y. Zee, C. R. Cantor, and R. W. K. Chiu From the Cover: Digital PCR for the molecular detection of fetal chromosomal aneuploidy PNAS, August 7, 2007; 104(32): 13116 - 13121. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. T.Y. Chung, R. W.K. Chiu, K.C. A. Chan, T. K. Lau, T. N. Leung, L. W. Chan, and Y.M. D. Lo Detrimental Effect of Formaldehyde on Plasma RNA Detection Clin. Chem., June 1, 2005; 51(6): 1074 - 1076. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and by the semiquantitative method with 5-fold (
) or 10-fold (
) dilutions.