| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Thomas Jefferson University, Department of Medicine, Division of Rheumatology, Philadelphia, PA 19107;
aaddress correspondence to this author at: Thomas Jefferson University, Department of Medicine, Division of Rheumatology, Room 509, Bluemle Lifesciences Bldg., 233 South 10th St., Philadelphia, PA 19107-5541, e-mail Carol.Artlett{at}mail.tju.edu
Our laboratory and others have demonstrated the presence of microchimeric cells in the peripheral blood of nonpregnant patients with systemic sclerosis (SSc) (1)(2)(3)(4). In addition, we have demonstrated the presence of maternal microchimeric cells in the peripheral blood of patients with juvenile idiopathic inflammatory myopathies (5), an observation confirmed in a subsequent study (6). However, some recent reports describe the failure to demonstrate the presence of male microchimeric cells in the peripheral blood of women with SSc or myopathies (7)(8)(9)(10). These latter studies used PCR amplification (9) or nested PCR (7)(8)(10) of Y-chromosome DNA sequences in whole peripheral blood.
Various PCR methods have been used for detection of microchimeric cell DNA in peripheral blood. These methods have been applied primarily to studies of organ transplantation or prenatal diagnosis (11)(12)(13)(14). However, the reliable detection and quantification of minute amounts of microchimeric DNA within a large pool of recipient DNA has been problematic. The need to limit the amount of autologous DNA in PCR analyses is likely to be an important determinant of the lower limit of detection for minor nonautologous cell populations. Numerous methods have been used for the detection of fetal cells for prenatal diagnosis in the peripheral blood of pregnant women; however, magnetic cell sorting is the most commonly used (15). Magnetic cell sorting for fetal erythroblasts has been used extensively in the analysis of peripheral blood of pregnant women (16)(17)(18), and this approach (18)(19) enriched the fetal cells 
200-fold (17).
We examined whether magnetic cell sorting to isolate specific cell types used successfully in the detection of microchimeric fetal cells in pregnant women (18)(19) would improve the limit of detection for microchimeric cells in nonpregnant women. For this purpose, peripheral blood was magnetically sorted for CD4+ and CD8+ T cells, and DNA was then extracted from the isolated cell populations and subjected to real-time PCR. The results obtained from the magnetically sorted cells were compared with real-time PCR analyses of DNA extracted from whole peripheral blood.
The study was approved by the Thomas Jefferson University Institutional Review Board. We obtained 20 mL of whole peripheral blood by venipuncture from 60 nonpregnant women. Magnetic cell sorting was performed on 10 mL of blood, and the remainder was used to for DNA extraction, as described below, and constituted the whole-peripheral-blood DNA fraction. All individuals had given birth to at least one son and had never received a blood transfusion. A separate control group included five nulligravid women who had never received a blood transfusion.
Fresh whole peripheral blood was collected by venipuncture and subjected to magnetic cell sorting for CD4+ and CD8+ T cells as described previously (3), according to the manufacturer’s protocol. The supernatant containing the cells not adhering to the vessel walls was aspirated and saved while the tube was still in the magnet. This sample provided a "negatively selected" population. With the tube still in the magnet, we pipetted 5 mL of phosphate-buffered saline into the tube, aspirated it off, and discarded it. The cells remaining in the tube were resuspended in 500 µL of methanol–acetic acid (3:1 by volume) and constituted the positively selected cells. Both positively and negatively selected cells were fixed in 1 mL of methanol–acetic acid (3:1 by volume) and stored for DNA extraction. The positively selected populations were >98% pure by immunofluorescent staining.
DNA was extracted from the magnetically sorted cell populations with use of the QIAamp Blood Kit (Qiagen) according to the manufacturer’s protocol. Total DNA was quantified on a spectrophotometer and stored at -70 °C until analyzed. We extracted DNA from whole-peripheral-blood cells, using a slight modification of the method of Miller et al. (20) as described in detail previously (1). The amount of recovered DNA was resuspended to yield a final concentration of 0.5 g/L. The purified DNA was stored at -70 °C until analyzed.
Y-chromosome-specific DNA sequences were detected by real-time PCR (ABI Sequence Detector 7700), with oligonucleotide primers as described previously (21)(22). All PCR amplifications were performed in duplicate, and each analysis contained positive and negative controls for Y-chromosome DNA. A calibration curve was obtained by amplifying samples containing serial dilutions of 0, 5, 25, 125, 625, and 3125 cell-equivalents of male DNA. The resulting quantification in the positively and negatively selected cell populations was normalized for 500 000 cells.
Results obtained from amplification of DNA from magnetically sorted cells and from whole peripheral blood were evaluated using the 
2 test with Yates correction. Because the data were nonparametric, the Mann–Whitney rank-sum test (GraphPad InStat statistical program; GraphPad Software) was used to compare the quantity of microchimeric DNA in magnetically sorted cells vs whole peripheral blood. The results are expressed as medians with interquartile ranges. The limit of detection was determined by the total number of microchimeric cells detected in the magnetically sorted populations divided by the total number of microchimeric cells detected in the whole-peripheral-blood DNA samples.
We compared a method in which peripheral blood cells were magnetically sorted before PCR and a method in which PCR was performed on DNA extracted from whole peripheral blood for their ability to detect microchimeric cells (Table 1
). The numbers of microchimeric cells detected in the positive and negative magnetic cell-sorting fractions were combined to yield the total male cells detected. We found that magnetic sorting of cells before PCR amplification of Y-chromosome sequences increased the frequency of detection of microchimeric cells compared with PCR of whole-peripheral-blood DNA. In the 60 samples analyzed by both methods, microchimeric cells were detected in the magnetically sorted cells in 49 of 60 (82%) samples compared with 14 of 60 (23%) whole-peripheral-blood samples. The difference was statistically significant (P <0.0001). A separate control group consisting of five nulliparous women was consistently negative for microchimeric cells in the magnetically sorted cells and by PCR of whole-peripheral-blood DNA.
|
Quantitative assessment showed that the median number of microchimeric cells detected was substantially lower in the whole peripheral blood than in the magnetically sorted cells. We found that magnetically sorted cell samples had a median of 13.0 (interquartile range, 1–77.8) microchimeric cells per 500 000 autologous cells, whereas whole peripheral blood had 0 (interquartile range, 0–0) per 500 000 autologous cells. The difference obtained between the two methods was highly statistically significant (P <0.0001). The limit of detection for microchimeric cells was 250-fold lower in whole peripheral blood DNA than in magnetically sorted cell DNA. This limiting factor was determined by dividing the total number of microchimeric cells detected in the magnetically sorted cells by the total number of microchimeric cells detected in whole peripheral blood.
The ability to accurately detect microchimeric cells is critical to determine the role that microchimeric cells play in the pathogenesis of SSc and other autoimmune diseases. The results from this study demonstrate that PCR of whole-peripheral-blood DNA is not adequate for the detection of microchimeric cells.
Several investigators (7)(8)(9)(10) have failed to detect microchimeric cells or DNA in patients with SSc by amplification of the Y chromosome. Furthermore, our initial studies demonstrated that only 46% of patients with SSc and 4% of controls were positive for microchimerism when whole peripheral blood was analyzed by PCR (1). In the present study, the detection of microchimeric cells in whole peripheral blood was statistically similar to these previously published results. We found that the detection of microchimeric cells in the samples was reduced by 60% when whole peripheral blood was used compared with magnetically sorted cells. We also observed a dramatic reduction in the quantification of microchimeric cells when we used DNA from whole peripheral blood compared with magnetically sorted cells for PCR and found that enrichment of cell populations by magnetic cell sorting increased the detection rate of the microchimeric cells by 250-fold. This lower limit of detection was similar to that observed in pregnant women (17).
Our results suggest that the negative findings in some previously published studies (7)(8)(9)(10) may have resulted from technical difficulties in detecting minute amounts of donor DNA within a large background of autologous DNA. The peripheral blood cells were sorted for CD4 and CD8, although these are relatively nonspecific enhancement strategies. We believe that substantial enhancement of detection is stochastically dependent on the PCR. We also believe that PCR can be enhanced by increasing the ratio of primer to DNA. The limitations with whole-peripheral-blood DNA may be attributable to the absolute amount of DNA that can be incorporated into the PCR and the ability of the primer to find and anneal to the specific sequence. Magnetic cell sorting before PCR enriches the microchimeric DNA within the PCR and provides the additional benefit that positively and negatively selected cell populations can be analyzed separately. We sorted the peripheral blood cells for CD4+ and CD8+ T cells for a previous study (3). Sorting for other cells, e.g., CD3, before PCR may be just as effective; this postulation, however, needs to be tested.
Some of the differences observed between magnetic cell separation and whole peripheral blood may be attributable to the different methods of DNA extraction. The positively selected cells were insufficient in number to be extracted by the salting-out method used for whole peripheral blood, which requires large volumes, and even if scaled down, the yield of DNA would have been insufficient. Furthermore, extracting DNA from the positively selected cells with use of the QIAamp Blood Kit would have overloaded the column with DNA, in which case the DNA would be contaminated with proteins. Because we used two different DNA extraction methods, we analyzed DNA from a total of 500 000 cells. In addition, it is unlikely that microchimeric DNA and/or cells would have been selectively lost in either of the extraction methods.
The frequency of microchimeric cells found in many of the peripheral blood samples examined was comparable to that found in early pregnancy, where magnetic cell sorting is used extensively to isolate microchimeric fetal cells for prenatal diagnosis. The high numbers of microchimeric cells may be attributable to immune activation because some of these women have autoimmune diseases; however, it is striking that microchimeric cells are not detected as easily in the whole peripheral blood from the same women. These results suggest that PCR amplification of whole-peripheral-blood DNA for the detection of microchimeric cells is not suitable possibly because of the high background of autologous cells. Isolation of cells by magnetic cell separation before PCR increases the detection of male microchimeric cells by a factor of >250.
Acknowledgments
This work was supported by a grant from the Arthritis Foundation (to C.M.A.), NIH Grant R29 AR 45399 (to C.M.A.), and NIH Grant R01 AR 19616 (to S.A.J.). Dr. Lori A. Cox was supported by NIH Training Grant T3 AR 07583. J. Bruce Smith is supported in part by a grant from the Mona Schneidman Trust.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
