Mitochondrial DNA spectra of single human CD34⁺ cells, T cells, B cells, and granulocytes

Mitochondria play many important roles in cellular homeostasis, including the production of adenosine triphosphate (ATP) by the process of oxidative phosphorylation (the respiratory chain); metabolism of amino acids, fatty acids, cholesterol, steroids, heme, and nucleotides; and the release of death-promoting factors including cytochrome c leading to apoptosis. The generation of cellular energy in the form of ATP is probably the most crucial function of mitochondria.^1,2  Hundreds of mitochondria are present within each cell's cytoplasm, and 2 to 10 copies of mitochondrial DNA (mtDNA) are packaged in each mitochondrion. Human mtDNA is a double-stranded circular molecule containing 16 569 base pairs (bp) encoding for 37 genes: the 12S and 16S ribosomal RNA genes, the 22 transfer RNA genes required for mitochondrial protein synthesis, and 13 polypeptide genes that generate subunits of the respiratory chain.³  However, most mitochondrial respiratory chain polypeptides are coded by nuclear DNA (nDNA). Mitochondrial DNA is distinguished from nDNA by several characteristics, including the properties of heteroplasmy, maternal inheritance, relaxed replication, and mitotic segregation.²  In addition, mtDNA has a modified genetic code⁴  and a paucity of introns, and it lacks histone protection.⁵  Oxidative phosphorylation for ATP generation is believed to be the main source of the reactive oxygen species (ROSs), such as superoxide anions, hydrogen peroxide, and hydroxyl radicals, produced in mitochondria. Chronic ROS exposure induces oxidative damage to mitochondrial and cellular macromolecules, including DNA, proteins, and lipids.⁶  The 10- to 20-fold higher mutation rate in mtDNA compared with nDNA is believed to be caused by its proximity to sites of ROS generation, lack of histones, and limited DNA repair capacity,⁶  though recent data have shown that excision repair mechanisms operate for mammalian mtDNA.^7,8 

Hundreds of mtDNA mutations (deletions, insertions, and base substitutions) have been identified in many degenerative diseases; other mutations that impact mitochondrial function have been associated with nuclear genes. Thus, mitochondrial dysfunction can stem from a mitochondrial maternal inheritance pattern, a nuclear Mendelian inheritance pattern, or a combination of them.⁹  A large deletion of mtDNA is a hallmark of Pearson syndrome, a congenital disorder, one manifestation of which is sideroblastic anemia.¹⁰  Somatically acquired mtDNA mutations also have been linked to aging, degenerative diseases, cancer, and autoimmunity.⁹  MtDNA mutations were reported in apparently acquired sideroblastic anemia and in myelodysplastic syndromes in general,¹¹  though our laboratory was unable to confirm these results.¹²  Somatic mtDNA mutations are attributed mainly to mitochondrial oxidative damage, which increases with age.⁶  In the aging process, individual postmitotic cells initially acquire a single mutant mtDNA among an overwhelming majority of wild-type species (heteroplasmy), and probably different mutant mtDNA clonally expand to homoplasmy in parallel in individual cells through relaxed replication, leading to mitochondrial dysfunction.^13,14  In mitotic tissues, such as bone marrow, intracellular random genetic drift is probably caused by relaxed replication and mitotic segregation.¹³  Computer modeling has suggested that random processes are sufficient to explain the incidence of mtDNA mutations in dividing cells.¹⁵  Moreover, single-cell analysis of mtDNA in proliferating epithelial tissue showed high proportions of age-associated homoplasmic mutant cells.^16,17 

We have described an age-dependent accumulation of mtDNA mutations leading to a high degree of mtDNA sequence heterogeneity in normal marrow and blood CD34⁺ cells, but not in umbilical cord blood.¹⁸  The accumulation of mtDNA mutations over time and the clonal expansion of cells with somatic mtDNA mutations appear to be relatively common in marrow progenitor cells and possibly even in marrow stem cells.¹⁹  In our previous studies, CD34⁺ individual cells were cloned during a brief in vitro culture period, after which their mtDNAs were sequenced.^18,19  Single-cell analysis—the examination of individual cells without culture— ensures that even limited growth in vitro does not introduce mutations; furthermore, some mature blood cells are either cumbersome to cultivate (T and B cells) or will not undergo mitosis (granulocytes). We now have established a method for mtDNA sequence analysis in single cells, obtained by flow cytometric sorting and analyzed using gene amplification with primers specific for the control region of mtDNA. In the present study, single blood CD34⁺ cells, granulocytes, T cells, and B cells were obtained from the same healthy donors who had volunteered for our previous studies, thus permitting us to examine changes in the polymorphic spectra of the mtDNA control region over time. We also used the single-cell method to assess the mtDNA sequence in differentiated cells derived from lymphohematopoietic stem cells.

Peripheral blood (PB) from 5 healthy adult donors was collected after informed consent and according to protocols approved by the Institutional Review Board of the National Heart, Lung, and Blood Institute. Sequence analyses of the mtDNA control region from CD34⁺ clones and single granulocytes have been published and are included here for comparative purposes only.^18,19 

Mononuclear cells and granulocytes from PB were separated by density gradient centrifugation. After washing with phosphate-buffered saline (PBS), the number of cells was adjusted to 1 × 10⁷ cells/mL with PBS containing 0.5% bovine serum albumin (BSA; Invitrogen, Carlsbad, CA). Fifteen microliters of anti-CD34 phycoerythrin (PE)–conjugated monoclonal antibody, anti-CD3 allophycocyanin (APC)–conjugated antibody, or anti-CD20 fluorescein isothiocyanate (FITC) antibody (BD Biosciences, San Jose, CA) were added to 100 μL mononuclear cell suspension, and 15 μL anti-CD33 PE-conjugated antibody (BD Biosciences) was added to a 100-μL granulocyte suspension. After incubation for 30 minutes at 4°C, cells were washed and resuspended in 0.5 mL PBS with 0.5% BSA.

Cell sorting was performed on a MoFlo Cytometer (Dako-Cytomation, Fort Collins, CO), using 100 mW of the 488-nm line of an argon laser (I-90; Coherent, Palo Alto, CA) for excitation. Forward scatter was the triggering parameter. Single-cell deposition was accomplished using the CyClone automated cloner (Dako-Cytomation) in the 0.5 single-drop mode with gating based on forward scatter and fluorescence. Single CD34⁺ cells, T cells, B cells, and granulocytes were placed into each well of an optical 96-well reaction plate (MicroAmp; Applied Biosystems, Foster City, CA) containing 50 μL of 1 × lysis buffer (10 mM Tris-HCl [pH 8.0], 50 mM KCl, 100 μg/mL Proteinase K, and 1% Triton X-100).

Single cells in the lysis buffer were incubated at 56°C for 30 minutes to liberate total DNA, and then proteinase K was inactivated by incubation at 96°C for 5 minutes in the 96-well reaction plate. Lysates were briefly centrifuged and stored at -20°C.

Cell lysates of single cells were subjected to amplification of mtDNA using the LA polymerase chain reaction (PCR) kit (TaKaRa LA Taq; TaKaRa Bio, Madison, WI). Two-step nested PCR amplification was performed with outer and inner pairs of primers to generate sufficient template from the single CD34⁺ cells, lymphocytes, and granulocytes for sequencing of the mtDNA control region. Primer pairs for PCR amplification of the mtDNA control region were previously described.¹⁹  The primary PCR mixture contained 400 μM each dNTP,2 U LA Taq (TaKaRa LA Taq), 0.8 μM outer primers, and 5 μL cell lysates. Primary PCR amplification was performed in an optical 96-well reaction plate (MicroAmp) using the GeneAmp PCR system 9700 (Applied Biosystems) as follows: one cycle of 96°C for 1 minute, then 35 cycles of 96°C for 30 seconds, 52°C for 50 seconds, and 72°C for 1 minute, with a 5-second increase per cycle, and ending with one cycle of 72°C for 5 minutes. Secondary PCR was performed in 50 μL reaction mixture containing 400 μM each dNTP, 2 U LATaq,0.8 μM inner nested primers, and 3 μL primary PCR product under the same amplification conditions described. Secondary PCR samples were purified using the QIA quick PCR purification kit (Qiagen, Valencia, CA).

BigDye Terminator version 3.1 ready reaction kit and ABI Prism 3100 Genetic Analyzer (Applied Biosystems) were used for direct sequencing of the amplified mtDNA PCR products. Sequencing primers have been described.¹⁹  mtDNA sequences experimentally obtained were compared with the Revised Cambridge Reference Sequence (RCRS) (http://www.mitomap.org/mitomap/mitoseq.html)²⁰  using the Blast2 program (http://www.ncbi.nlm.nih.gov/blast/bl2seq/bl2.html) and the database search tool, MitoAnalyzer (http://www.cstl.nist.gov/biotech/strbase/mitoanalyzer.html, 2001),²¹  to determine polymorphisms and mutations.

mtDNA mutations in all donors were confirmed by reanalysis of the first PCR product. For cells showing a sequence difference from the donor's corresponding aggregate mtDNA sequence, 2 or more independent PCR reamplifications were performed, followed by sequencing. Amplification and sequencing from the original single-cell lysates were not feasible for granulocytes because of poor recovery of DNA and the appearance of apparently new base substitutions (see “Discussion”).

To confirm heteroplasmy and mixed nucleotide signals in the mtDNA control region resulting from the existence of a deletion, PCR products were separated by agarose gel electrophoresis and extracted from the gels using the QIAquick gel extraction kit (Qiagen). The purified PCR products were inserted into the pCR2.1-TOPO vector and transformed into competent Escherichia coli (TOP10 cells) using the TOPO TA cloning kit (Invitrogen). Recombinant plasmids were isolated from 10 to 20 white bacterial colonies and sequenced.

One-way analysis of variance (ANOVA) was performed to examine whether the mtDNA heterogeneity (total frequency, unique differences, base substitutions, and length alterations of the polymeric C tracts) observed in single CD34⁺ cells, lymphocytes, and granulocytes from the 5 healthy donors constituted significant statistical differences (P < .05).

We analyzed new blood samples from 5 of the 6 healthy donors examined previously; donors 1 to 5 here correspond to donors 1 and 3 to 6 in the published study.^18,19  The aggregate cell mitochondrial genotype of each donor was compared with the mtDNA heterogeneity found in single CD34⁺ cells, T cells, B cells, and granulocytes. There was marked variation in the number of nucleotide changes in the control region of aggregate cells' mtDNA among the 5 healthy volunteers, with a range from 6 (donor 1) to 24 (donor 3), as previously reported (Table 1).^18,19  Fifty-nine mtDNA sequence variants were noted in the control region among the 5 donors; 58 of these variants were listed in the polymorphism database (http://www.mitomap.org), and the nucleotide variant (517A>G) was again found in donor 5 (donor 6 in the previous study^18,19 ). Donor 2 had length variations of the homopolymeric C tracts at nucleotide positions 303 to 315 in the aggregate genotype (Table 1).

Individual CD34⁺ cells, T cells, B cells, and granulocytes from the 5 healthy donors were isolated by flow cytometric sorting, and the mtDNA control region was amplified (see “Materials and methods”). We performed nested mtDNA PCR amplification of 96 single cells of CD34⁺ cells, T cells, B cells, and granulocytes for each donor. PCR amplification efficiency of the individual CD34⁺ cells, T cells, B cells, and granulocytes was 95.2% (457 of 480 cells), 95.4% (458 of 480 cells), 92.7% (445 of 480 cells), and 79.4% (381 of 480 cells), respectively (Table 2). Although there were variations in PCR efficiency among the donors, the PCR efficiency of granulocytes was lower than that for other cell types (Table 2). By using cell lysis buffer rather than heating for DNA extraction, the PCR amplification efficiency of granulocytes in this study (79.4%) was greatly improved compared with the 7.5% found previously.¹⁹ 

A total of 457 CD34⁺ cells, 458 T cells, 445 B cells, and 381 granulocytes from each of the 5 donors were subjected to sequencing of the 1121-bp mtDNA control region, known to include mutational hot spots, in order to assess the heterogeneity of individual cell mtDNA sequences.^22,23  A very small number of sequences (1 from CD34⁺ cells, 2 from T cells, 3 from B cells, and 5 from granulocytes) were discarded because of likely contamination (inferred from the absence of the donor's identifying polymorphisms present in the respective aggregate sequence). Our results, based on the analysis of 456 CD34⁺ cells, 456 T cells, 442 B cells, and 376 granulocytes from the 5 donors, showed that 37.9% ± 3.6% (mean ± SD) of CD34⁺ cells, 36.4% ± 14.1% of T cells, 36.0% ± 10.7% of B cells, and 47.7% ± 7.4% of granulocytes had mtDNA heterogeneities distinct from the donor's corresponding aggregate mtDNA sequences (Tables 3, 4). Although the amount of mtDNA heterogeneity varied among the cell types, only the greater amount of heterogeneity detected in the granulocytes was statistically significant when compared with CD34⁺ cells (P < .05).