Deoxyribonucleoside triphosphate pools and differential thymidine sensitivities of cultured mouse lymphoma and myeloma cells

The effects of various concentrations of thymidine on DNA synthesis and deoxyribonucleoside triphosphate contents of a highly thymidine-sensitive cultured mouse lymphoma cell line (WEHI-7) and a relatively resistant mouse myeloma cell line (HPC-108) have been studied by 32P-labelling techniques. DNA synthesis in the myeloma cells was inhibited by thymidine at concentrations of 10(-3) M or greater, while DNA synthesis in the lymphoma cells was inhibited by concentrations 30-fold lower, consistent with the 25-fold difference between the two cell lines in sensitivity to growth inhibition by thymidine. Thymidine caused marked elevation of the dTTP and dGTP pools, slight elevation or no change in the dATP pool and a marked decrease in the dCTP pool in cells of both lines. The greater resistance of HPC-108 cells to thymidine inhibition was related to the finding that they normally contained a much higher concentration of dCTP than did the WEHI-7 cells. Pool size measurements on thymidine-treated (10(-4) M) cells of an additional seven sensitive lymphoma and six relatively resistant myeloma cell lines indicated that in all 15 lines studied, with one exception, a critical concentration of dCTP of about 32 nmol per ml of cell volume was required for the maintenance of normal rates of DNA synthesis. The dCTP content found normally in the lymphoma cells was only a little above this concentration. Amongst the myeloma lines, three contained similarly low levels of dCTP, but were more resistant to thymidine inhibition probably because of their inefficient production of dTTP from thymidine. Cells of the other four myeloma lines (including HPC-108) normally contained much higher dCTP concentrations. The mechanism of thymidine action was explained by reference to the known allosteric properties of ribonucleotide reductase.     

Deoxyribonucleotide pools in mouse-fibroblast cell lines with altered ribonucleotide reductase.

Mutant cells lines of 3T6 mouse fibroblasts, resistant to thymidine and deoxyadenosine, have an altered allosteric regulation of the enzyme ribonucleotide reductase (Meuth, M. and Green, H., Cell, 3, 367, 1974). Compared to 3T6, these lines contain larger pools of deoxynucleoside triphosphates, in particular deoxycytidine triphosphate, but show a normal rate of DNA synthesis. Addition of thymidine or deoxyadenosine to 3T6 cells results in large accumulations of the corresponding triphosphates and a dramatic decrease in the dCTP pool, concomitant with inhibition of DNA synthesis. Addition of thymidine to the mutant cell lines also leads to an increase in the dTTP pool but does not result in a depletion of dCTP or inhibition of DNA synthesis. Addition of deoxyadenosine only leads to a small increase of the dATP pool. In general the change in the allosteric regulation of bibonucleotide reductase is reflected in the deoxynucleotide pools.     

Deoxythymidine nucleotide metabolism in Bacillus subtilis W23 infected with bacteriophage SP1Oc: preliminary evidence that dTMP in SP10c DNA is synthesized by a novel, bacteriophage-specific mechanism.

Despite the fact that mature SP10c DNA contains dTMP, the acid-soluble fraction of infected cells contained no dTTP during the interval of phage replication. However, infected cells contained normal cellular levels of dATP, dGTP, and dCTP. Upon infection of deoxythymidine-starved Bacillus subtilis M160 (a deoxythymidine-requiring mutant of B. subtilis W23), mature phage DNA with a normal dTMP content was made. SP10c codes for an enzyme that seems to catalyze the tetrahydrofolate-dependent transfer of 1-carbon fragments to the 5 position of dUMP. The transfer of 1-carbon fragments is not accompanied by oxidation of tetrahydrofolage to dihydrofolate, implying that the enzyme in question is not a dTMP synthetase. It is proposed that dTMP in mature SP10c DNA is derived by the postreplicational modification of some other nucleotide and not by the direct incorporation of dTTP into DNA.     

Lack of specific correlation of the deoxycytidine triphosphate pool level with rate of DNA synthesis.

The levels of the four deoxyribonucleoside triphosphate pools and the distribution of cells in the various phases of the cell cycle have been examined in Chinese hamster cells as thymidine, present as a regular constituent in the growth medium, was removed in stages. The results indicate that: 1. Duration of the DNA synthetic phase was lengthened when thymidine was removed from the growth medium. 2.Temporally correlated with lengthening of the DNA synthetic phase upon thymidine removal was a 7-fold increase in level of the dCTP pool, reduction in the dGTP pools, and little or no change in dATP pool. 3.Radioactive labeling procedures indicated that expansion of the dCTP pool could be completely accounted for by increased ribonucleotide reductase activity and that the dTTP pool switched from a largely exogenous thymidine source to endogenous dTTP synthesis as the extracellular thymidine concentration was reduced. 4.Deoxyuridine and thymidine were apparently transported by the same system in Chinese hamster cells, while deoxycytidine was transported by a different system. Although deoxycytidine transport was unaffected by thymidine, phosphorylation of intracellular deoxycytidine compounds to the triphosphate level was stimulated by thymidine. Cytidine transport was not significantly affected by thymidine.     

Diverse deoxyribonucleotide profiles in cultured human cells with differential sensitivity to thymidine

Intracellular deoxyribonucleotide pools were examined before and after thymidine treatment in highly sensitive T-lymphoid cells, relatively resistant B-lymphoid cells and moderately sensitive melanoma cells. Among the 4 cell lines studied, proportions of the 4 deoxyribonucleotide pools varied appreciably while ribonucleotide profiles were similar. The ratio of dGTP to dCTP increased with sensitivity to thymidine. Increase in dTTP levels with increasing thymidine concentration was dependent on sensitivity of cells to thymidine and was accompanied by reduction in the dCTP pool. dGTP levels increased as did dTTP levels in all cells, while dATP pool expansion correlated with thymidine sensitivity. The results indicate an additional aspect of purine deoxyribonucleotide involvement in the growth inhibitory effects of thymidine.     

Sensitivity of a mutator gene in Chinese hamster ovary cell to deoxynucleoside triphosphate pool alterations.

The Thy- mutants of Chinese hamster ovary cells have a 5- to 10-fold elevated pool of deoxycytidine 5'-triphosphate (dCTP) and are auxotrophic for thymidine as an apparent consequence of a single mutation. thy is also a mutator gene, elevating the spontaneous rate of mutation 5- to 200-fold for at least two genetic markers. Previous experiments suggested that this mutator activity was caused by the elevated pool of dCTP in Thy- cells. To test this, the dCTP and deoxythymidine 5'-triphosphate (dTTP) pools were manipulated by altering the external concentration of thymidine in the growth medium. The rate of mutation at one genetic locus, ouabain resistance, was directly related to cellular dCTP content. At the highest level of dCTP the rate in one Thy- strain was approximately 200 times that of wild-type cells. However, the relationship between dCTP content and the rate of mutation at the ouabain locus was different for two mutator strains and wild-type cells. The rate of mutation at a second locus, thioguanine resistance, was increased approximately 10-fold over wild type regardless of the dCTP-dTTP pools. These experiments suggest that the mutator activity of thy is clearly related to dCTP content, but the dCTP level alone does not appear to be the cause of the mutator.     

Biochemical characterization of the hamster thy mutator gene and its revertants

The thy- mutator phenotype of Chinese hamster ovary cells is distinguished by increased intracellular levels of dCTP, auxotrophy for thymidine, and elevated spontaneous mutational rates. To determine the biochemical lesion responsible for this complex phenotype, enzymes responsible for the synthesis of dCTP and dTTP were investigated. Levels of ribonucleotide reductase and dCMP deaminase were identical in mutant and wild type strains. In contrast, CTP synthetase activity in extracts from thy- strains was consistently altered in that 50% of enzyme activity was resistant to feedback inhibition by CTP. Additionally, thy- strains obtained by DNA transfection also had CTP-resistant CTP synthetase. Thy+ revertants lost the resistant enzyme, and total activity was reduced. CTP-resistant CTP synthetase was regained in thy- mutants reselected from thy+ revertants, but in these strains all activity was resistant. These experiments demonstrate that the thy- mutator phenotype is a consequence of a mutation of CTP synthetase and suggest that one pathway of reversion to the wild type state is by loss or inactivation of the mutant allele rendering the revertants hemizygous for the gene.     

Deoxyribonucleoside triphosphate pools in synchronized human cells infected with herpes simplex virus types 1 and 2.

Deoxyribonucleoside triphosphate pools in uninfected and herpes simplex virus type 1 (HSV-1)- and HSV-2-infected KB cells were analyzed to determine whether ribonucleotide reductase functions in vivo in the presence and absence of thymidine (TdR). Previously we showed that HSV-2 replication was inhibited in KB cells blocked in their capacity to synthesize DNA by TdR. HSV-1 replication was not inhibited under these conditions. Both HSV-1 and HSV-2 induced an altered ribonucleotide reductase resistant to dTTP inhibition. Thus, the block to HSV-2 replication apparently was not at the level of reductase. However, the in vitro activity of the enzyme does not necessarily correspond to intracellular conditions. In TdR-blocked HSV-2-infected cells, we found that, while dTTP levels remained high, dCTP concentrations increased. In contrast, KB cells blocked by TdR showed increased dTTP but decreased dCTP levels. We conclude that the HSV-2 enzyme is functional in vivo and that TdR inhibits viral replication by a mechanism other than depletion of dCTP. Infection of KB cells with HSV-1 or HSV-2 altered both dATP and dGTP levels in the presence or absence of TdR. Inhibition of viral replication was not explained by changes in these pools. We suggest that, during infection, HSV-1 induces a virus function(s) not related to reductase which is resistant to TdR, whereas the corresponding HSV-2 function is sensitive. Our evidence shows that the TdR-sensitive function is not in the pathways leading to deoxyribonucleoside triphosphate and may occur at the level of DNA replication.     

Microinjection of deoxynucleotides into mouse cells. No evidence that precursors for DNA synthesis are channeled

The technique of microinjection was applied for the introduction of radioactive, phosphorylated precursors of DNA synthesis into living mouse cells in culture. Autoradiographs proved that DNA was well labeled by injected dTTP, dCTP, dATP, dGTP, and (to a minor extent) dTMP, but the efficiency was much lower when CDP, ADP, or GDP was used. For practical reasons, injections into nuclei were preferred, but injections into cytoplasm showed no principal difference with the autoradiographed nuclei. The kinetics of the uptake of the injected material agreed neatly with the calculation (from pool sizes and polymerization rate) that the intracellular deoxytriphosphates are sufficient for about 10 min of DNA synthesis. All this evidence strongly argues against the concept that precursors of DNA synthesis are channeled in vivo within a multienzyme complex and suggests a free diffusion of deoxynucleotides within the cell. Injected thymidine was less able to enter the deoxy nucleotide metabolism compared with thymidine from the culture medium. A mutant cell line deficient in thymidine kinase did not accumulate intracellular thymidine. These data indicate that thymidine kinase is a membrane associated enzyme and that uptake and phosphorylation of thymidine are coupled reactions.     

Deoxyribonucleoside triphosphate pools in mutagen sensitive mutants of Neurospora crassa

Deoxyribonucleoside triphosphate (dNTP) levels were measured in wild type Neurospora and nine mutagen-sensitive mutants, at nine different genes. Eight of these mutants are sensitive to hydroxyurea and histidine and show chromosomal instability, a phenotype which could result from altered levels of dNTPs. Two patterns were seen. Five of the mutants had altered ratios of dNTPs, with relatively high levels of dATP and dGTP and low levels of dCTP, but changes in the dTTP/dCTP ratio did not correlate with changes in spontaneous mutation levels. During exponential growth all but two of the mutants had small but consistent increases in dNTP pools compared to wild type. DNA content per microgram dry hyphae was altered in several mutants but these changes showed no correlation with the dNTP pool alterations.     

Regulation of pyrimidine deoxyribonucleotide metabolism by substrate cycles in dCMP deaminase-deficient V79 hamster cells.

A mutant V79 hamster fibroblast cell line lacking the enzyme dCMP deaminase was used to study the regulation of deoxynucleoside triphosphate pools by substrate cycles between pyrimidine deoxyribosides and their 5'-phosphates. Such cycles were suggested earlier to set the rates of cellular import and export of deoxyribosides, thereby influencing pool sizes (V. Bianchi, E. Pontis, and P. Reichard, Proc. Natl. Acad. Sci. USA 83:986-990, 1986). While normal V79 cells derived more than 80% of their dTTP from CDP reduction via deamination of dCMP, the mutant cells had to rely completely on UDP reduction for de novo synthesis of dTTP, which became limiting for DNA synthesis. Because of the allosteric properties of ribonucleotide reductase, CDP reduction was not diminished, leading to a large expansion of the dCTP pool. The increase of this pool was kept in check by a shift in the balance of the deoxycytidine/dCMP cycle towards the deoxynucleoside, leading to massive excretion of deoxycytidine. In contrast, the balance of the deoxyuridine/dUMP cycle was shifted towards the nucleotide, facilitating import of extracellular deoxynucleosides.     

Mechanism of UV-induced deoxynucleoside triphosphate pool imbalance in CHO-K1 cells

Several laboratories have reported that exposure of cells to UV radiation results in a significant imbalance in deoxynucleoside triphosphate pool concentrations. In our CHO-K1 cells, a rapid drop in dCTP is accompanied by a rapid increase in dTTP. Examination of enzyme activities associated with synthesis/degradation of these molecules suggests that UV transiently enhances a putative dCTPase, dCMP deaminase and CdR kinase activities. This results in accumulation of excess dUMP which is probably converted to dTMP, then to dTTP. The absence of dCMP deaminase in V79 cells prohibits this rapid response in those cells. Moreover, significantly different dCMP deaminase activities were observed in CHO-K1 cells obtained from other laboratories, suggesting they, too, may respond differently to irradiation.     

Structural basis for recognition and catalysis by the bifunctional dCTP deaminase and dUTPase from Methanococcus jannaschii

Potentially mutagenic uracil-containing nucleotide intermediates are generated by deamination of dCTP, either spontaneously or enzymatically as the first step in the conversion of dCTP to dTTP. dUTPases convert dUTP to dUMP, thus avoiding the misincorporation of dUTP into DNA and creating the substrate for the next enzyme in the dTTP synthetic pathway, thymidylate synthase. Although dCTP deaminase and dUTPase activities are usually found in separate but homologous enzymes, the hyperthermophile Methanococcus jannaschii has an enzyme, DCD-DUT, that harbors both dCTP deaminase and dUTP pyrophosphatase activities. DCD-DUT has highest activity on dCTP, followed by dUTP, and dTTP inhibits both the deaminase and pyrophosphatase activities. To help clarify structure-function relationships for DCD-DUT, we have determined the crystal structure of the wild-type DCD-DUT protein in its apo form to 1.42A and structures of DCD-DUT in complex with dCTP and dUTP to resolutions of 1.77A and 2.10A, respectively. To gain insights into substrate interactions, we complemented analyses of the experimentally defined weak density for nucleotides with automated docking experiments using dCTP, dUTP, and dTTP. DCD-DUT is a hexamer, unlike the homologous dUTPases, and its subunits contain several insertions and substitutions different from the dUTPase beta barrel core that likely contribute to dCTP specificity and deamination. These first structures of a dCTP deaminase reveal a probable role for an unstructured C-terminal region different from that of the dUTPases and possible mechanisms for both bifunctional enzyme activity and feedback inhibition by dTTP.     

Deoxycytidine Triphosphate Deaminase: Identification and Function in Salmonella typhimurium

The biosynthesis of 2'-deoxyuridine monophosphate (dUMP) has been studied in a cytidine- and uracil-requiring mutant of Salmonella typhimurium (DP-55). The dUMP pool and the thymidine monophosphate (dTMP) pool of DP-55, grown in the presence of (3)H-uracil and unlabeled cytidine, are found to have the same specific activities. However, only 30% of the dUMP and the dTMP is synthesized from a uridine nucleotide. Seventy per cent is derived directly from a cytosine compound. The identification and partial purification of a Mg(2+)-dependent 2'-deoxycytidine triphosphate (dCTP) deaminase from S. typhimurium suggests that the combined action of dCTP deaminase and 2'-deoxyuridine triphosphate pyrophosphatase accounts for 70% of the dUMP, and therefore the dTMP, synthesized in vivo. The introduction of a thymine requirement (i.e., a block in thymidylate synthetase) into DP-55 results in a 100-fold increase in the size of the dUMP pool. However, the relative contribution of the uridine and cytidine pathways to dUMP synthesis is unaltered. The high dUMP pool is accompanied by extensive catabolism of dUMP to uracil. Partial thymine starvation of the cells results in a significant increase in the dUMP and dCTP pools. Moreover, an increase in the contribution of the dCTP pathway to dUMP synthesis is observed. As a result of these changes the catabolism of dUMP to uracil is augmented.     

Mutational specificity of DNA precursor pool imbalances in yeast arising from deoxycytidylate deaminase deficiency or treatment with thymidylate

Disruption of the dCMP deaminase (DCD1) gene, or provision of excess dTMP to a nucleotide-permeable strain, produced dramatic increases in the dCTP or dTTP pools, respectively, in growing cells of the yeast Saccharomyces cerevisiae. The mutation rate of the SUP4-o gene was enhanced 2-fold by the dCTP imbalance and 104-fold by the dTTP imbalance. 407 SUP4-o mutations that arose under these conditions, and 334 spontaneous mutations recovered in an isogenic strain having balanced DNA precursor levels, were characterized by DNA sequencing and the resulting mutational spectra were compared. Significantly more (greater than 98%) of the changes resulting from nucleotide pool imbalance were single base-pair events, the majority of which could have been due to misinsertion of the nucleotides present in excess. Unexpectedly, expanding the dCTP pool did not increase the fraction of A.T----G.C transitions relative to the spontaneous value nor did enlarging the dTTP pool enhance the proportion of G.C----A.T transitions. Instead, the elevated levels of dCTP or dTTP were associated primarily with increases in the fractions of G.C----C.G or A.T----T.A. transversions, respectively. Furthermore, T----C, and possibly A----C, events occurred preferentially in the dcd1 strain at sites where dCTP was to be inserted next. C----T and A----T events were induced most often by dTMP treatment at sites where the next correct nucleotide was dTTP or dGTP (dGTP levels were also elevated by dTMP treatment). Finally, misinsertion of dCTP or dTTP did not exhibit a strand bias. Collectively, our data suggest that increased levels of dCTP and dTTP induced mutations in yeast via nucleotide misinsertion and inhibition of proofreading but indicate that other factors must also be involved. We consider several possibilities, including potential roles for the regulation and specificity of proofreading and for mismatch correction.     

Regulation of dCTP deaminase from Escherichia coli by nonallosteric dTTP binding to an inactive form of the enzyme

The trimeric dCTP deaminase produces dUTP that is hydrolysed to dUMP by the structurally closely related dUTPase. This pathway provides 70-80% of the total dUMP as a precursor for dTTP. Accordingly, dCTP deaminase is regulated by dTTP, which increases the substrate concentration for half-maximal activity and the cooperativity of dCTP saturation. Likewise, increasing concentrations of dCTP increase the cooperativity of dTTP inhibition. Previous structural studies showed that the complexes of inactive mutant protein, E138A, with dUTP or dCTP bound, and wild-type enzyme with dUTP bound were all highly similar and characterized by having an ordered C-terminal. When comparing with a new structure in which dTTP is bound to the active site of E138A, the region between Val120 and His125 was found to be in a new conformation. This and the previous conformation were mutually exclusive within the trimer. Also, the dCTP complex of the inactive H121A was found to have residues 120-125 in this new conformation, indicating that it renders the enzyme inactive. The C-terminal fold was found to be disordered for both new complexes. We suggest that the cooperative kinetics are imposed by a dTTP-dependent lag of product formation observed in presteady-state kinetics. This lag may be derived from a slow equilibration between an inactive and an active conformation of dCTP deaminase represented by the dTTP complex and the dUTP/dCTP complex, respectively. The dCTP deaminase then resembles a simple concerted system subjected to effector binding, but without the use of an allosteric site.     

DNA fragmentation and cytotoxicity from increased cellular deoxyuridylate

Previous results from this laboratory have shown that thymidylate deprivation results in dramatic elevation of intracellular dUTP and incorporation of dUMP into DNA. The goal of the present studies was to determine whether the latter changes may play a part in the associated cytotoxicity ("thymineless death"), which is ordinarily assumed to be a direct result of reduced intracellular dTTP. The approach used here was to increase intracellular dUTP without allowing dTTP to diminish and observe the effects on cell viability. dUMP pools were expanded by exposure of cells to deoxyuridine [in cell growth medium containing hypoxanthine, methotrexate, and thymidine (HAT medium)], resulting in accumulation of dUTP to levels that approached those of dTTP, which were at, or higher than, the levels in untreated cells. In conjunction with this the cells became nonviable, and newly synthesized DNA was fragmented, both of which occur with thymidylate depletion and, we assume, result from the active process of excision repair at the many uracil-containing sites in DNA. The results indicate that, although the relative importance of low dTTP remains unknown, elevated dUTP can account for the cytotoxicity caused by thymidine starvation. Most of the "dTTP" measured by the DNA polymerase assay in cells treated with methotrexate (MTX) (plus purine supplement) was, in fact, dUTP, which may explain some previous observations of only modest depression of dTTP in cells treated with MTX or similarly acting drugs.     

Deoxyribonucleoside triphosphate metabolism and the mammalian cell cycle: Effects of thymidine on wild-type and dCMP deaminase-deficient mouse S49 T-lymphoma cells

The size of the dCTP pool has been implicated as a possible regulator of DNA synthesis. In this investigation we correlate large intracellular variations in deoxyribonucleoside triphosphate levels to the growth rates and cell-cycle kinetics of mouse S49 T-lymphoma cells. Wild-type and a mutant line AzidoC-100-5, lacking dCMP-deaminase activity resulting in a 10-fold expanded dCTP pool were studied and compared using flow cytometry, centrifugal elutriation and nucleoside triphosphate determinations. An increase in the dCTP pool was closely correlated to the passage of cells from G1 to S phase in both cell types. Addition of thymidine to wild-type and mutant cells resulted in an accumulation of cells in early S phase, concomitant with a decreased dCTP level. Mutant cells excreted large amounts of deoxycytidine into the medium which partially protected the cells from thymidine inhibition. The doubling times for the mutant and wild-type cells were very similar but the mutant had a somewhat prolonged S phase and shortened G1 phase compared with the wild-type cells. Large changes in the DNA precursor levels were produced by addition of thymidine to mutant cultures. This gave no change in the growth rate but a somewhat shortened S phase and prolonged G1. The biochemical background for these effects is discussed.     

A bifunctional dCTP deaminase-dUTP nucleotidohydrolase from the hyperthermophilic archaeon Methanocaldococcus jannaschii

By the sequential action of dCTP deaminase and dUTPase, dCTP is converted to dUMP, the precursor of thymidine nucleotides. In addition, dUTPase has an essential role as a safeguard against uracil incorporation in DNA. The putative dCTP deaminase (MJ0430) and dUTPase (MJ1102) from the hyperthermophilic archaeon Methanocaldococcus jannaschii were overproduced in Escherichia coli. Unexpectedly, we found the MJ0430 protein capable of both reactions, i.e. hydrolytic deamination of the cytosine ring and hydrolytic cleavage of the phosphoanhydride bond between the alpha- and beta-phosphates. When the reaction was followed by thin layer chromatography using [3H]dCTP as substrate, dUMP and not dUTP was identified as a reaction product. In the presence of unlabeled dUTP, which acted as an inhibitor, no label was transferred from [3H]dCTP to the pool of dUTP. This finding strongly suggests that the two consecutive steps of the reaction are tightly coupled within the enzyme. The hitherto unknown bifunctionality of the MJ0430 protein appears beneficial for the cells because the toxic intermediate dUTP is never released. The MJ0430 protein also catalyzed the hydrolysis of dUTP to dUMP but with a low affinity for the substrate (Km >100 micro m). According to limited proteolysis, the C-terminal residues constitute a flexible region. The other protein investigated, MJ1102, is a specific dUTPase with a Km for dUTP (0.4 micro m) comparable in magnitude with that found for previously characterized dUTPases. Its physiological function is probably to degrade dUTP derived from other reactions in nucleotide metabolism.     

Mitochondrial deoxynucleotide pools in quiescent fibroblasts: a possible model for mitochondrial neurogastrointestinal encephalomyopathy (MNGIE)

Mitochondrial (mt) DNA depletion syndromes can arise from genetic deficiencies for enzymes of dNTP metabolism, operating either inside or outside mitochondria. MNGIE is caused by the deficiency of cytosolic thymidine phosphorylase that degrades thymidine and deoxyuridine. The extracellular fluid of the patients contains 10-20 microM deoxynucleosides leading to changes in dTTP that may disturb mtDNA replication. In earlier work, we suggested that mt dTTP originates from two distinct pathways: (i) the reduction of ribonucleotides in the cytosol (in cycling cells) and (ii) intra-mt salvage of thymidine (in quiescent cells). In MNGIE and most other mtDNA depletion syndromes, quiescent cells are affected. Here, we demonstrate in quiescent fibroblasts (i) the existence of small mt dNTP pools, each usually 3-4% of the corresponding cytosolic pool; (ii) the rapid metabolic equilibrium between mt and cytosolic pools; and (iii) the intra-mt synthesis and rapid turnover of dTTP in the absence of DNA replication. Between 0.1 and 10 microM extracellular thymidine, intracellular thymidine rapidly approaches the extracellular concentration. We mimic the conditions of MNGIE by maintaining quiescent fibroblasts in 10-40 microM thymidine and/or deoxyuridine. Despite a large increase in intracellular thymidine concentration, cytosolic and mt dTTP increase at most 4-fold, maintaining their concentration for 41 days. Other dNTPs are marginally affected. Deoxyuridine does not increase the normal dNTP pools but gives rise to a small dUTP and a large dUMP pool, both turning over rapidly. We discuss these results in relation to MNGIE.