Abstract
Background
Alterations in the copy number of mitochondrial DNA (mtDNA) play a role in the pathogenesis of mitochondrial diseases and other many common diseases. Recently, the copy number of leukocyte mtDNA has been considered to serve as a biomarker to monitor or chase such diseases. Therefore, reproducible mtDNA measurement is required.
Methods
Peripheral blood mononuclear cells were prepared by a density-based method. The mtDNA/cell was measured by quantitative realtime polymerase chain reaction.
Results
The degree of platelet contamination varied to a large extent among preparations. The mtDNA copy numbers per mononuclear cell were 269 ± 51 and 146 ± 14 in the samples before and after the platelet depletion, respectively.
Conclusion
A density-based mononuclear cell preparation causes heavy platelet contamination. The platelet depletion from a sample is particularly important for comparing the mtDNA contents between different dates or between different patients.
Introduction
The copy number of mitochondrial DNA (mtDNA) ranges from several tens to several thousands depending on the tissues. The copy number largely determines the cellular ATP production. 1 Therefore, the decrease in mtDNA causes the cellular dysfunction leading to severe symptoms such as an mtDNA depletion syndrome. 2 The drug-induced depletion of mtDNA is also known. Some antiretroviral nucleoside reverse-transcriptase inhibitors inhibit mitochondrial DNA polymerase γ, deplete mtDNA and cause hepatopathy, myopathy, neuropathy and lactic acidosis. 3 Furthermore, infection of HIV itself decreases mtDNA. 4
Peripheral blood mononuclear cells (PBMCs) that are prepared based on a density gradient method are contaminated by platelets. The platelet contamination can contribute to overestimation of mtDNA measurements. 3 ,5 Despite this fact, PBMCs are still used without removing platelets in many studies. In extreme cases, buffy coats of blood are used.
Here, we show that the measured mtDNA content of PBMCs in a standard preparation is overestimated almost two-fold because of contaminating platelets. More importantly, the variation of the mtDNA content is also two-fold higher than that in platelet-depleted samples.
Materials and methods
Blood donors
Blood was collected, in a tube containing ethylenediamine-N,N,N′,N′-tetraacetic acid (EDTA)-2Na, from healthy volunteer donors under informed consent in Kyushu University Hospital.
Separation of mononuclear cells
PBMCs were prepared with three reagents: Lymphoprep (Daiichi Kagaku, Japan), Lymphocyte Separation Medium (Cosmobio, Japan) and Lymphocyte Separation Solution (Nacalai Tesque, Japan), according to the manufacturer's instructions. The purity was higher than 95% in all preparations. The extent of platelet contamination was essentially the same for all the three reagents (results not shown), and so the Lymphoprep was used hereafter. The contaminating platelets were removed by washing the preparations three times as reported elsewhere. 6
Measurement of mitochondrial DNA/cell
PBMC cells of volume 5 × 105 were lysed with 20 μL lysis buffer (50 mmol/L Tris–HCl, pH 8.0, 1 mmol/L EDTA, 0.5% Tween-20 and 200 μg/mL proteinase K), incubated at 55°C for 2 h and at 95°C for 10 min to inactivate proteinase K. The lysate was treated with ApaI directly used for polymerase chain reaction (PCR). Copy numbers of mtDNA and genomic DNA were quantified by realtime PCR using Light Cycler (Roche Diagnostics, Tokyo, Japan) for calculating mtDNA/cell. 7 An mtDNA ATPase 6 region (nucleotide positions [nps] 8762-8918) was amplified using the following primers: 5′-TTGCCACAACTAACCTCCTC-3′ and 5′-TGTGGTAAGAAGTGGGCTAG-3′. For genomic DNA, an anti-thrombin gene (nps2570-2824) was amplified using the following primers: 5′-TCAGAACAGAAGATCCCGGA-3′ and 5′-CAAAGGTGCTCCTAACAAGG-3′.
Results and discussion
We measured the mtDNA contents before and after the platelet depletion (Table 1). The platelet contamination varied much among individuals (platelet/PBMC, 81 ± 25). The mtDNA copy numbers of PBMC were 269 ± 51 and 146 ± 14 before and after the depletion, respectively. The ratio of mtDNA copy numbers before and after the depletion in each individual was 1.9 ± 0.3 indicating that the platelet contamination indeed causes large overestimation. The platelet depletion decreases the variation of mtDNA copy number among individuals to about a half (% coefficient of variation 9.6 versus 19.0), suggesting that the platelet depletion is also important for the reproducible measurement (Table 1).
Mitochondrial DNA (mtDNA) copy number in peripheral blood mononuclear cells (PBMCs) before and after platelet depletion
Each value of mtDNA was a mean of duplicate within the same assay. CV, coefficient of variation
*mtDNA copy number per PBMC in each preparation
†Contaminating cell number of platelet per PBMC in each preparation
The relative densities of PBMCs and platelets are very close to each other. Therefore, as long as the separation depends on the density, platelets should be rather enriched together with PBMCs than separated. It is reported that there are up to 30 platelets/PBMC after a standard density-based lymphocyte isolation. 3 In fact, the number could be more than 100 in our hands (Table 1). Although the reason for the discrepancy is not clear, it could be partly because of the difference in the separation reagents. Another important finding is that the mtDNA/PBMC were very constant between individuals. The contribution of platelet contamination to the variation is not underlined so far although the overestimation is cautioned. It should be emphasized that platelet-free PBMC samples are essential for the exact and precise estimation of mtDNA content of PBMC: especially for chasing a time-course of mtDNA or judging a proper clinical state during antiviral drug treatments. By removing platelets, we would be able to set a narrow reference interval for mtDNA/PBMC and so to diagnose the mtDNA depletion even by measurement at one time-point like other usual clinical tests. For that purpose, we should examine more samples including healthy donors and patients to confirm.
Anti-CD4 or -CD8 conjugated beads and a cell sorter are good alternative ways for PBMC separation, but these methods are much more costlier than the density-based ones. Given that the difference between before and after (i.e. 123 = 269 − 146) was derived from the contaminating platelets (77 = 81 − 4), one platelet contained about 1.6 copies of mtDNA on an average. Unexpectedly, there is no report on the platelet mtDNA copy number to our knowledge. We therefore directly measured mtDNA/platelet using leukocyte-depleted platelet-rich plasma. The mtDNA/platelet was measured based on the platelet count, but not on genomic DNA because platelets do not contain nuclei. The number was 1.5. This value is close to 1.6 obtained above. We could logically correct the measured mtDNA/PBMC by determining a platelet/PBMC ratio and subtracting platelet-derived mtDNA copy number. However, the numbers are not necessarily corrected by that way for each individual case in Table 1, raising the possibility that the amount of mtDNA in platelets might vary from one individual to another. For verifying such a correction way, we need to compare a larger number of platelet-depleted and non-depleted samples.
In conclusion, (i) a density-based separation of PBMC accompanies heavy platelet contamination, (ii) the contamination indeed causes large overestimation of mtDNA/PBMC and (iii) the variation of mtDNA/PBMC becomes considerably smaller when platelets are depleted.