Structural determinants in human DNA polymerase gamma account for mitochondrial toxicity from nucleoside analogs

Although antiviral nucleoside analog therapy successfully delays progression of HIV infection to AIDS, these drugs cause unwelcome side-effects by inducing mitochondrial toxicity. We and others have demonstrated that the mitochondrial polymerase, DNA polymerase gamma (pol gamma), participates in mitochondrial toxicity by incorporating these chain-terminating antiviral nucleotide analogs into DNA. Here, we explore the role of three highly conserved amino acid residues in the active site of human pol gamma that modulate selection of nucleotide analogs as substrates for incorporation. Sequence alignments, crystal structures and mutagenesis studies of family A DNA polymerases led us to change Tyr951 and Tyr955 in polymerase motif B to Phe and Ala, and Glu895 in polymerase motif A was changed to Ala. The mutant polymerases were tested for their ability to incorporate natural nucleotides and the five antiviral nucleoside analogs currently approved for antiviral therapy: AZT, ddC, D4T, 3TC and carbovir. Steady-state kinetic analysis of the pol gamma derivatives with the normal and antiviral nucleotides demonstrated that Tyr951 is largely responsible for the ability of pol gamma to incorporate dideoxynucleotides and D4T-MP. Mutation of Tyr951 to Phe renders the enzyme resistant to dideoxynucleotides and D4T-TP without compromising the activity of the polymerase. Alteration of Glu895 and Tyr955 to Ala had the largest effect on overall polymerase activity with normal nucleotides, producing dramatic increases in K(m(dNTP)) and large decreases in k(cat). Mutation of Tyr955 in pol gamma causes the degenerative disease progressive external ophthalmoplegia in humans, and we show that this residue partially accounts for the ability of pol gamma to incorporate D4T-MP and carbovir. Alteration of Glu895 to Ala slightly increased discrimination against dideoxynucleotides and D4T-TP. The mechanisms by which pol gamma selects certain nucleotide analogs are discussed.     

Toxicity of nucleoside analogues used to treat AIDS and the selectivity of the mitochondrial DNA polymerase

Incorporation of nucleoside analogues by the mitochondrial DNA polymerase has been implicated as the primary cause underlying many of the toxic side effects of these drugs in HIV therapy. Recent success in reconstituting recombinant human enzyme has afforded a detailed mechanistic analysis of the reactions governing nucleotide selectivity of the polymerase and the proofreading exonuclease. The toxic side effects of nucleoside analogues are correlated with the kinetics of incorporation by the mitochondrial DNA polymerase, varying over 6 orders of magnitude in the sequence zalcitabine (ddC) > didanosine (ddI metabolized to ddA) > stavudine (d4T) > lamivudine (3TC) > tenofovir (PMPA) > zidovudine (AZT) > abacavir (metabolized to carbovir, CBV). In this review, we summarize our current efforts to examine the mechanistic basis for nucleotide selectivity by the mitochondrial DNA polymerase and its role in mitochondrial toxicity of nucleoside analogues used to treat AIDS and other viral infections. We will also discuss the promise and underlying challenges for the development of new analogues with lower toxicity.     

TDP1 repairs nuclear and mitochondrial DNA damage induced by chain-terminating anticancer and antiviral nucleoside analogs

Chain-terminating nucleoside analogs (CTNAs) that cause stalling or premature termination of DNA replication forks are widely used as anticancer and antiviral drugs. However, it is not well understood how cells repair the DNA damage induced by these drugs. Here, we reveal the importance of tyrosyl–DNA phosphodiesterase 1 (TDP1) in the repair of nuclear and mitochondrial DNA damage induced by CTNAs. On investigating the effects of four CTNAs—acyclovir (ACV), cytarabine (Ara-C), zidovudine (AZT) and zalcitabine (ddC)—we show that TDP1 is capable of removing the covalently linked corresponding CTNAs from DNA 3′-ends. We also show that Tdp1^−/− cells are hypersensitive and accumulate more DNA damage when treated with ACV and Ara-C, implicating TDP1 in repairing CTNA-induced DNA damage. As AZT and ddC are known to cause mitochondrial dysfunction, we examined whether TDP1 repairs the mitochondrial DNA damage they induced. We find that AZT and ddC treatment leads to greater depletion of mitochondrial DNA in Tdp1^−/− cells. Thus, TDP1 seems to be critical for repairing nuclear and mitochondrial DNA damage caused by CTNAs.     

Fidelity of the human mitochondrial DNA polymerase

We have quantified the fidelity of polymerization of DNA by human mitochondrial DNA polymerase using synthetic DNA oligonucleotides and recombinant holoenzyme and examining each of the possible 16-base pair combinations. Although the kinetics of incorporation for all correct nucleotides are similar, with an average Kd of 0.8 microM and an average k(pol) of 37 s(-1), the kinetics of misincorporation vary widely. The ground state binding Kd of incorrect bases ranges from a low of 25 microM for a dATP:A mispair to a high of 360 microM for a dCTP:T mispair. Similarly, the rates of incorporation of incorrect bases vary from 0.0031 s(-1) for a dCTP:C mispair to 1.16 s(-1) for a dGTP:T mispair. Due to the variability in the kinetic parameters for misincorporation, the estimates of fidelity range from 1 error in 3563 nucleotides for dGTP:T to 1 error in 2.3 x 10(6) nucleotides for dCTP:C. Interestingly, the discrimination against a dGTP:T mismatch is 16.5 times lower than that of a dTTP:G mismatch due to a tighter Kd for ground state binding and a faster rate of incorporation of the dGTP:T mismatch relative to the dTTP:G mismatch. We calculate an average fidelity of 1 error in 440,000 nucleotides.     

Acute cytotoxicity of arabinofuranosyl nucleoside analogs is not dependent on mitochondrial DNA

The nucleoside analogs 9-β-D-arabinofuranosylguanine (araG) and 1-β-d-arabinofuranosylthymine (araT) are substrates of mitochondrial nucleoside kinases and have previously been shown to be predominantly incorporated into mtDNA of cells, but the pharmacological importance of their accumulation in mtDNA is not known. Here, we examined the role of mtDNA in the response to araG, araT and other anti-cancer and anti-viral agents in a MOLT-4 wild-type (wt) T-lymphoblastoid cell line and its petite mutant MOLT-4 ρ⁰ cells (lacking mtDNA). The mRNA levels and activities of deoxyguanosine kinase (dGK), deoxycytidine kinase (dCK), thymidine kinase 1 (TK1) and thymidine kinase 2 (TK2) were determined in the two cell lines. Compared to that in the MOLT-4 wt cells the mRNA level of the constitutively expressed TK2 was higher (p < 0.01) in the ρ⁰ cells, whereas the TK1 mRNA level was lower (p < 0.05). The enzyme activity of the S-phase restricted TK1 was also lower (p < 0.05) in the MOLT-4 ρ⁰ cells, whereas the activities of dGK, dCK and TK2 were similar in MOLT-4 wt and ρ⁰ cell lines. The sensitivities to different cytotoxic nucleoside analogs were determined and compared between the two cell lines. Interestingly, we found that the acute cytotoxicity of araG, araT and other anti-viral and anti-cancer agents is independent of the presence of mtDNA in MOLT-4 T-lymphoblastoid cells.     

Exonuclease proofreading by human mitochondrial DNA polymerase

We have examined the ability of the human mitochondrial DNA polymerase to correct errors in DNA sequence using single turnover kinetic methods. The rate of excision of single-stranded DNA ranged from 0.07 to 0.17 x s(-1), depending on the identity of the 3'-base. Excision of the 3'-terminal base from correctly base paired DNA occurred at a rate of 0.05 x s(-1), indicating that the cost of proofreading is minimal, as defined by the ratio of the k(exo) for correctly base-paired DNA divided by the rate of forward polymerization (0.05/37 = 0.14%). Excision of duplex DNA containing 1-7 mismatches was biphasic, and the rate and amplitude of the fast phase increased with the number of mismatches, reaching a maximum of 9 x s(-1). We showed that transfer of DNA from the polymerase to the exonuclease active site and back again occurs through an intramolecular reaction, allowing for a complete cycle of reactions for error correction. For DNA containing a buried mismatch (T:T followed by C:G base pairs), the 3' base was removed at a rate of 3 x s(-1). The addition of nucleotide to the reaction that is identical to the 3' base increased the rate of excision 7-fold to 21 x s(-1). We propose that the free nucleotide enhances the rate of transfer of the DNA to the exonuclease active site by interrupting the correct 3' base pair through interaction with the template base. The exonuclease contribution to fidelity is minimal if the calculation is based on hydrolysis of a single mismatch: (k(exo) + k(pol,over))/(k(pol,over)) = 10, but this value increases to approximately 200 when examining error correction in the presence of nucleotides.     

Incorporation of nucleoside analogs into nuclear or mitochondrial DNA is determined by the intracellular phosphorylation site

Nucleoside analogs used in cancer chemotherapy and in treatment of virus infections are phosphorylated in cells by nucleoside and nucleotide kinases to their pharmacologically active form. The phosphorylated nucleoside analogs are incorporated into DNA and cause cell death or inhibit viral replication. Cellular DNA is replicated both in the nucleus and in the mitochondria, and nucleoside analogs may interfere with DNA replication in both these subcellular locations. In the present study we created a cell model system where nucleoside analogs were phosphorylated, and thereby pharmacologically activated, in either the nucleus, cytosol, or mitochondria of cancer cells. The system was based on the reconstitution of deoxycytidine kinase (dCK)-deficient Chinese hamster ovary cells with genetically engineered dCK targeted to the different subcellular compartments. The nucleoside analogs phosphorylated by dCK in the mitochondria were predominantly incorporated into mitochondrial DNA, whereas the nucleoside analogs phosphorylated in the nucleus or cytosol were incorporated into nuclear DNA. We further show that the nucleoside analogs phosphorylated in the mitochondria induced cell death by an apoptotic program. These data showed that the subcellular site of nucleoside analog phosphorylation is an important determinant for incorporation of nucleoside analogs into nuclear or mitochondrial DNA.     

Fidelity of nucleotide incorporation by human mitochondrial DNA polymerase

We have examined the fidelity of polymerization catalyzed by the human mitochondrial DNA polymerase using wild-type and exonuclease-deficient (E200A mutation) forms of recombinant, reconstituted holoenzyme. Each of the four nucleotides bind and incorporate with similar kinetics; the average dissociation constant for ground state binding is 0.8 microm, and the average rate of polymerization is 37 x s(-1), defining a specificity constant kcat/Km = 4.6 x 10(7) x m(-1) x s(-1). Mismatched nucleotides show weaker ground-state nucleotide binding affinities ranging from 57 to 364 microm and slower rates of polymerization ranging from 0.013 to 1.16 x s(-1). The kinetic parameters yield fidelity estimates of 1 error out of 260,000 nucleotides for a T:T mismatch, 3563 for G:T, and 570,000 for C:T. The accessory subunit increases fidelity 14-fold by facilitating both ground-state binding and the incorporation rate of the correct A:T base pair compared with a T:T mismatch. Correctly base-paired DNA dissociates from the polymerase at a rate of 0.02 x s(-1) promoting processive polymerization. Thus, the mitochondrial DNA polymerase catalyzed incorporation with an average processivity of 1850, defined by the ratio of polymerization rate to the dissociation rate (37/0.02) and with an average fidelity of one error in 280,000 base pairs.     

Differential incorporation and removal of antiviral deoxynucleotides by human DNA polymerase gamma

Mitochondrial toxicity can result from antiviral nucleotide analog therapy used to control human immunodeficiency virus type 1 infection. We evaluated the ability of such analogs to inhibit DNA synthesis by the human mitochondrial DNA polymerase (pol gamma) by comparing the insertion and exonucleolytic removal of six antiviral nucleotide analogs. Apparent steady-state K(m) and k(cat) values for insertion of 2',3'-dideoxy-TTP (ddTTP), 3'-azido-TTP (AZT-TP), 2',3'-dideoxy-CTP (ddCTP), 2',3'-didehydro-TTP (D4T-TP), (-)-2',3'-dideoxy-3'-thiacytidine (3TC-TP), and carbocyclic 2',3'-didehydro-ddGTP (CBV-TP) indicated incorporation of all six analogs, albeit with varying efficiencies. Dideoxynucleotides and D4T-TP were utilized by pol gamma in vitro as efficiently as natural deoxynucleotides, whereas AZT-TP, 3TC-TP, and CBV-TP were only moderate inhibitors of DNA chain elongation. Inefficient excision of dideoxynucleotides, D4T, AZT, and CBV from DNA predicts persistence in vivo following successful incorporation. In contrast, removal of 3'-terminal 3TC residues was 50% as efficient as natural 3' termini. Finally, we observed inhibition of exonuclease activity by concentrations of AZT-monophosphate known to occur in cells. Thus, although their greatest inhibitory effects are through incorporation and chain termination, persistence of these analogs in DNA and inhibition of exonucleolytic proofreading may also contribute to mitochondrial toxicity.     

Quantitative determination of antiviral nucleoside analog in DNA

A technique for the analysis of the amount of an antiviral nucleoside analog incorporated into DNA, utilizing enzymatic digestion of DNA, followed by high-performance liquid chromatography is described. The cells or tissue samples were treated with perchloric acid to inactivate the nucleases, then digested with pronase in the presence of EDTA. DNA was purified by CsCl centrifugation followed by Sephadex chromatography and treatment with deoxyribonuclease 1 and venom phosphodiesterase. The deoxyribonucleoside monophosphates and the monophosphate of the nucleoside analog liberated from DNA were separated and quantitated by HPLC analysis and measurement of radioactivity. This assay is more sensitive, specific, and precise than the determination of DNA density shift. It is also applicable for nucleoside analogs which do not change the density of DNA either because of their structure or their very small degree of incorporation.     

Incorporation and replication of 8-oxo-deoxyguanosine by the human mitochondrial DNA polymerase

To assess the role of oxidative stress on the replication of mitochondrial DNA, we examined the kinetics of incorporation of 8-oxo-7,8-dihydroguanosine (8-oxodG) triphosphate catalyzed by the human mitochondrial DNA polymerase. Using transient state kinetic methods, we quantified the kinetics of incorporation, excision, and extension beyond a base pair containing 8-oxodG. The 8-oxodGTP was incorporated opposite dC in the template with a specificity constant of 0.005 microM(-1) s(-1), a value approximately 10,000-fold lower than that for dGTP. Once incorporated, 96% of the time 8-oxodGMP was extended by continued polymerization rather than being excised by the proofreading exonuclease. The specificity constant for incorporation of 8-oxodGTP opposite a template dA was 0.2 microM(-1) s(-1), a value 13-fold higher than incorporation opposite a template dC. The 8-oxodG:dA mispair was extended rather than excised at least 70% of the time. Examination of the kinetics of polymerization with 8-oxodG in the template strand also revealed relatively low fidelity in that dCTP would be incorporated only 90% of the time. In nearly 10% of events, dATP would be incorporated, and once incorporated dA (opposite 8-oxodG) was extended rather than excised. The greatest fidelity was against a dTTP:8-oxodG mismatch affording a discrimination value of only 1800. These data reveal that 8-oxodGTP is a potent mutagen. Once it is incorporated into DNA, 8-oxodGMP codes for error prone DNA synthesis. These reactions are likely to play important roles in oxidative stress in mitochondria related to aging and as compounded by nucleoside analogs used to treat human immunodeficiency virus infections.     

Functional human mitochondrial DNA polymerase gamma forms a heterotrimer

Mitochondrial DNA polymerase gamma (pol gamma) is responsible for replication and repair of mtDNA and is mutated in individuals with genetic disorders such as chronic external ophthalmoplegia and Alpers syndrome. pol gamma is also an adventitious target for toxic side effects of several antiviral compounds, and mutation of its proofreading exonuclease leads to accelerated aging in mouse models. We have used a variety of physical and functional approaches to study the interaction of the human pol gamma catalytic subunit with both the wild-type accessory factor, pol gammaB, and a deletion derivative that is unable to dimerize and consequently is impaired in its ability to stimulate processive DNA synthesis. Our studies clearly showed that the functional human holoenzyme contains two subunits of the processivity factor and one catalytic subunit, thereby forming a heterotrimer. The structure of pol gamma seems to be variable, ranging from a single catalytic subunit in yeast to a heterodimer in Drosophila and a heterotrimer in mammals.     

Purification and identification of subunit structure of the human mitochondrial DNA polymerase.

The mitochondrial DNA polymerase of HeLa cells was purified 18,000-fold to near homogeneity. The purified polymerase cofractionated with two polypeptides that had molecular mass of 140 and 54 kDa. The 140-kDa subunit was specifically radiolabeled in a photoaffinity cross-linking assay and is most likely the catalytic subunit of the mitochondrial DNA polymerase. The purified enzyme exhibited properties that have been attributed to DNA polymerase gamma and shows a preference for replicating primed poly(pyrimidine) DNA templates in the presence of 0.5 mM MgCl2. As in the case of mitochondrial DNA polymerases from other animal cells, human DNA polymerase gamma cofractionated with a 3'----5' exonuclease activity. However, it has not been possible to determine if the two enzymatic activities reside in the same polypeptide. The exonuclease activity preferentially removes mismatched nucleotides from the 3' end of a duplex DNA and is not active toward DNA with matched 3' ends. These properties are consistent with the notion that the exonuclease activity plays a proofreading function in the replication of the organelle genome.     

Origins of human mitochondrial point mutations as DNA polymerase gamma-mediated errors

Mitochondrial mutational spectra in human cells, tissues and derived tumors for bp 10,030-10,130 are essentially identical, suggesting a predominant mutagenic role for endogenous processes. We hypothesized that errors mediated by mitochondrial DNA polymerase gamma were the primary sources of mutations. Point mutations created in this sequence by human DNA pol gamma in vitro were thus compared to the eighteen mutational hotspots, all single base substitutions, previously found in human tissues. The set of concordant hotspots accounted for 83% of these in vivo mutational events. About half of these mutations are insensitive to prolonged heating of DNA during PCR and half increase proportionally with heating time at 98 degrees C. Primary misincorporation errors and miscopying errors past thermal denaturing products such as deaminated cytosines (uracils) thus appear to be of approximately equal importance. For the sequence studied, these data support the conclusion that, endogenous error mediated by DNA pol gamma constitutes the primary source of mitochondrial point mutations in human tissues.