Does Centrifugation Cause the ex Vivo Release of DNA from Blood Cells?

Using a sex-mismatched bone marrow transplantation model, we have previously demonstrated that hematopoietic cells represent the predominant origin of cell-free plasma DNA (1). In that study, plasma samples were obtained by high-speed centrifugation followed by high-speed microcentrifugation. The comparison of the total DNA concentrations in plasma obtained by centrifugation and microcentrifugation for different numbers of times revealed no significant difference. We therefore concluded that centrifugation is not associated with damage to blood cells and artificial increases in cell-free plasma DNA concentrations. However, there remains a possibility that the first spin might have already destroyed blood cells, leading to the ex vivo release of DNA. This theoretical possibility might produce an aberrantly high proportion of cell-free plasma DNA originating from hematopoietic cells. In fact, to prevent cell lysis, the use of low-speed centrifugation, at 500g or even lower, has been a usual practice in harvesting blood cells (2)(3). We therefore investigated the effect of the use of different centrifugation speeds on the quantification of plasma DNA. Six bone marrow transplantation patients at the Department of Pediatrics of the Prince of Wales Hospital and 10 healthy individuals were recruited for the study with informed consent. The bone marrow transplantation recipients had no clinical evidence of graft rejection or graft-vs-host disease. Peripheral blood samples were collected into EDTA tubes and immediately divided into five aliquots, which were subjected to centrifugation for 10 min at 400g, 800g, 1500g, 3000g, and 16 000g (Eppendorf Centrifuge 5415D), respectively. This was then followed by filtration through a 0.22 μm filter to obtain cell-free plasma (4). DNA was extracted from 400 μL of plasma with the QIAamp Blood Kit (Qiagen) by use of the “blood and body fluid protocol” (5). We subjected 5 μL (of an elution volume of 50 μL) of extracted plasma DNA to real-time quantitative PCR for the β-globin gene as described previously (6) on a 7700 Sequence Detector (PE Applied Biosystems). The DNA concentration was expressed as genome-equivalents/mL as described previously (6). Statistical tests were carried out using the SigmaStat 2.0 software (SPSS). The median total plasma DNA concentrations harvested from blood of bone marrow transplantation patients processed at 400g, 800g, 1500g, 3000g, and 16 000g were 628, 679, 607, 574, and 780 genome-equivalents/mL, respectively (Fig. 1). No significant difference was observed among these values (Friedman repeated-measures analysis of variance on ranks, P = 0.30). The corresponding values for healthy individuals were 637, 571, 497, 610, and 551 genome-equivalents/mL, respectively (Fig. 1). Again, no significant difference was observed among these values (P = 0.12). Box plots of total DNA concentrations in the plasma aliquots obtained from bone marrow transplantation patients (top) and healthy individuals (bottom) prepared at different centrifugation speeds. Plasma DNA concentrations (genome-equivalents/mL) as determined by real-time quantitative PCR for the β-globin gene are plotted on the y axis. The centrifugation speeds (400g, 800g, 1500g, 3000g, and 16 000g) are shown on the x axis. The upper and lower limits of each box and the line across each box indicate the 75th and 25th percentiles and the median, respectively. The upper and lower horizontal bars denote the 90th and 10th percentiles, respectively. • indicate outliers. Various centrifugation speeds have been used to obtain plasma samples for circulating DNA analysis (6)(7). Although previous work has demonstrated that the release of DNA from lymphocytes is independent of mechanical effects conferred by centrifugation (8), our study provides a more in-depth investigation into the effects of centrifugation on quantification of plasma DNA on the basis of different centrifugation speeds. We can conclude that blood-processing protocols using different centrifugation speeds with subsequent filtration (4) would not lead to aberrantly high concentrations of plasma DNA. We investigated the possibility that hematopoietic cells might be more susceptible to damage after bone marrow transplantation in view of our previous conclusion, based on studies of such patients, that the predominant source of plasma DNA was hematopoietic. No significant difference was observed among the total DNA concentrations in plasma samples obtained by use of different centrifugation protocols. Thus, this present work supports the view that the release of plasma DNA in our previous study was a genuine in vivo phenomenon and that hematopoietic cells represent the predominant origin of plasma cell-free DNA (1). Our work also provides information about the compatibility of studies of plasma DNA that used different centrifugation speeds. Y.M.D. Lo is supported by the Innovation and Technology Fund (AF/90/99 and ITS/195/01).