[CANCER RESEARCH 44, 2534-2539. June 1984] Cytotoxic and Biochemical Effects of Thymidine and 3-Deazauridine on Human Tumor Cells1 Arnold Lockshin, John T. Mendoza, Beppino C. Giovanella,2 and John S. Stehlin, Jr. St. Joseph Hospital Laboratory for Cancer Research, Houston, Texas 77002 ABSTRACT cells, thymidine-resistant human malignant B-cells have much lower ratios of these activities (6, 19). The cause of thymidine- Cytotoxicity and perturbations of the deoxyribonucleoside tri- induced inhibition of DNA synthesis in mammalian cells has been phosphate pools caused by thymidine were studied in thymidine- ascribed to depletion of the intracellular dCTP pool brought about sensitive and -resistant human tumor cells. Incubation with 1 mw by allosteric effects of dTTP on ribonucleotide reducíase(2, 33, thymidine reduced cell viability by more than 90% in the three 36). Other investigators have concluded that the decrease in the sensitive cell lines (two melanomas and one adrenal carcinoma) dCTP pool is too small to account for growth inhibition of L1210 and reduced the growth rate without decreasing the viability of murine leukemia cells (14) and that high levels of dTTP (and/or resistant LO melanoma cells. Thymidine (1 HIM)greatly increased other dNTPs3) interact with a regulatory protein for DNA polym- the ratio of the deoxythymidine 5'-triphosphate to deoxycytidine 5'-triphosphate pools in the sensitive cells compared to LO cells erase (37). Yet another interpretation is that imbalance of the pyrimidine dNTP pools brought about by supranormal levels of and also caused larger relative increases in the pool sizes of deoxyguanosine 5'-triphosphate and deoxyadenosine 5'-tri- exogenous thymidine causes misincorporation of bases into DNA, which leads to mutagenicity and cytotoxicity (4). phosphate in the sensitive compared to the resistant cells. 3-Deazauridine inhibits the growth of microbial and murine 3-Deazauridine, known to inhibit synthesis of deoxycytidine 5'-triphosphate and cytidine 5'-triphosphate in other cell lines, tumor cells in culture and of transplantable mouse tumors in vivo (5, 39). In patients with acute leukemia, infusions of 3-deazauri potentiated the cytotoxicity of thymidine for thymidine-sensitive dine resulted in hematological improvement but no significant BE melanoma and LO cells. In LO cells, 3-deazauridine (50 UM) responses (41). 3-Deazauridine 5'-triphosphate, an intracellular decreased the intracellular pool of deoxycytidine 5'-triphosphate metabolite of 3-deazauridine, inhibits CTP synthetase from mu to the level obtained with 1 rnw thymidine. Lower concentrations rine tumor cells and from calf liver, thereby lowering intracellular of deoxycytidine as compared to cytidine were required to pro levels of CTP and indirectly those of dCTP (5, 26, 27). A second tect BE and LO cells against the cytotoxicity of thymidine plus ary effect of 3-deazauridine in these cells is inhibition by 3- 3-deazauridine. Deoxycytidine also was more effective than was deazauridine 5'-diphosphate of the reduction of CDP to dCDP cytidine in preventing loss of cell viability after exposure to mediated by ribonucleoside reducíase (5). Cytidine and less thymidine or to 3-deazauridine individually. In these human mel effectively deoxycytidine or uridine counteract growth inhibition anoma cells, ribonucleotide reducíase may be a major site of caused by 3-deazauridine in murine adenocarcinoma Ca755 and action of thymidine, of 3-deazauridine, and of both drugs in in L1210 cells, whereas Ihymidine or deoxyuridine do not have combination. this effect (5,26). 3-Deazauridine is not incorporated into nucleic These results indicate that in human tumor cells the cytotoxic acids (39). effect of thymidine correlates with greater perturbations of the The major known effects of thymidine and 3-deazauridine pyrimidine deoxyribonucleoside 5'-triphosphate pools and that metabolites on pyrimidine nucleotide biosynthesis are presenled thymidine and 3-deazauridine, which independently reduce the intracellular levels of deoxycytidine 5'-triphosphate, act synergis- in Chart 1. Other biochemical effecls include feedback inhibilion by dTTP of Ihymidine kinase (15) and dCMP deaminase (25) and tically against human tumor cells. inhibilion by 3-deazauridine and ils 5'-monophosphate nucleo- lide of cylidine deaminase and dCMP deaminase, respectively INTRODUCTION (10). We are exploring the potenlial Iherapeulic value of combining High concentrations of thymidine selectively kill certain human fhymidine wilh agenls lhal deplete the dCTP pool. Using human tumor cells In vitro (21), and continual infusions of thymidine tumor cell lines, we are interesled in determining (a) if differences cause regression of human tumor heterotransplants in nude in Ihe inlracellular dNTP pools are consislenl wilh sensilivily and (athymic) mice (22). Clinical evidence of antitumor effect has resislance to thymidine, and (b) if drugs which deplete dCTP been observed with T-cell leukemias, lymphomas, and metastatic pools enhance Ihe cyloloxicily of Ihymidine. We report here our melanomas when plasma levels of about 1 mw thymidine have studies wilh 4 human cell lines, 3 derived from melanomas and been maintained (3, 7, 19, 20). Thymidine must be converted to one from an adrenal carcinoma. A preliminary report of some of dTTP in order to cause cytotoxicity or cytostasis. The sensitivity Ihese findings has been published (24). of human malignant T-cells to thymidine has been attributed to the high levels of thymidine-phosphorylating activity relative to dephosphorylating activity in these cells. Compared to these T- MATERIALS AND METHODS 1 Supported by The Stehlin Foundation for Cancer Research and by the Sid Cells. The human melanoma cell lines CA, BE, and LO were obtained from this laboratory (13), and the SW-13 human adrenal tumor cells were Richardson Foundation, Dallas, TX. 2 To whom requests for reprints should be addressed. Received November 14, 1983; accepted March 14, 1984. 3The abbreviation used is: dNTP, deoxyribonucleoside 5'-triphosphate. 2534 CANCER RESEARCH VOL. 44 Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 1984 American Association for Cancer Research. Thymidine and 3-Deazauridine 3 Deazaundine Thymidine polycytidylate copolymer was used for the dCTP assays. The polymeri zation reactions were conducted in 96-well culture plates, and aliquots (90 ííl)ofeach reaction mixture were spotted on discs of Whatman GFA RNA ||X 1/\dCTPA1ÕtDeoxycytidineglass fiber filters (pretreated with cold 5% trichloroacetic acid containing \*¿f 1% inorganic sodium pyrophosphate) placed on a manifold. After 10 min, ^v DAUOP each dish was washed 12 times with the cold trichloroacetic acid- /CTP \ICTP SynthetaseX T inorganic sodium pyrophosphate solution followed by 4 washes with MiT. 'JTP «}tCytidineI| 100% methanol and 3 with diethyl ether. For each cell line, dNTP DAUTP \\ determinations were performed on extracts from 3 separate experiments tRibonucleolíde with at least 2 flasks/treatment. Reducíasettt/dTTP/DNA RESULTS Effect of Thymidine on Growth Rate and Cell Viability. At plating densities of 500,000 cells/75 sq cm flask, the logarithmic Chart 1. Major effects of thymidine and 3-deazauridine metabolites on pyrimi- phase growth rates of all 4 cell lines were approximately the dme nudeotide biosynthesis, x, enzyme inhibition. See text for details. DAUDP, 3- same (doubling times ranging from 29 to 33 hr). Under these deazauridine 5'-diphosphate; DAUTP, 3-deazauridine 5'-tríphosphate. conditions, 48-hr incubation of CA, BE, or SW-13 cells with 1 HIMthymidine prevented cell division and decreased cell viability from Dr. A. Leibovitz (23). Cells were maintained as monolayer cultures in Eagle's minima! essential medium containing penicillin (50 units/ml), by more than 90%. LO cells divided more slowly in the presence of thymidine (see below), but their viability was enhanced after streptomycin (50 ng/ml), and supplemental 2 mw glutamine, all from 24-hr incubation and was not impaired for up to 72 hr (Table 1). Grand Island Biological Co. (Grand Island, NY), and 10% heat-inactivated fetal calf serum from Reheis Chemical Co. (Phoenix, AZ). Cultures were Effect of Thymidine on Intracellular dNTP Levels. Chart 2 incubated at 37°in a humidified 5% CO2 atmosphere. Periodic tests for shows the changes in the dNTP pools caused by incubating Mycoplasma contamination were negative. these cell lines with 1 mwi thymidine. After 24 hr, the dTTP pools Chemicals. [mef/7y/-3H]dTTP (60 Ci/mmol), [5-3H]dCTP (30 Ci/mmol), of the sensitive lines increased 60-fold or more, whereas the [8-3H]dGTP, (18 Ci/mmol), and [8-3H]dATP (17 Ci/mmol) were purchased relative increase in the LO cells was less than half these values. •%. from ICN Clinical Radioisotopes Div. (Irvine, CA). Lyophilized noncovalent double-stranded copolymers of deoxypolyadenylate-deoxypolythymidy- Table 1 late, deoxypolyinosinate-deoxypolycytidylate. and deoxypolyguanylate- Effect of 1 mu thymidine on viability of human tumor cells deoxypolycytidylate were obtained from P-L Biochemicals, Inc. (Milwau Cells (500.000) were plated in 75-sq cm culture flasks. After 24 hr, fresh medium kee, Wl), and Micrococcus /úfeosDNA polymerase (EC 2.7.7.7) (specific with or without 1 mu thymidine was added. The cells were harvested at the times indicated and were replated for colony formation. activity, 469 units/mg protein) was purchased from Miles Laboratories control8)CA0.38 viability (fraction of (Elkhart, IN). Thymidine, 3-deazauridine, and other nonlabeled chemicals Thymidine treatment were from Sigma Chemical Co. (St. Louis, MO). (hr)24 Growth and Cytotoxicity Assays. Cells were seeded in 75-sq cm ±0.06" ±0.03 ±0.07 ±0.06 culture flasks and allowed to plate for 24 hr, at which time the medium 48 0.08 ±0.04 0.05 ±0.01SW-130.740.08 ±0.02LO1.361.05 ±0.10 was removed and replaced with fresh medium with or without drugs. At 72Cell 0.006 ±0.003BE0.24 0.95 ±0.04 indicated times, cells from one-half of the flasks were harvested with 8 Incubation without thymidine.