Biochemical Differences Among Four Inosinate Dehydrogenase Inhibitors

[CANCER RESEARCH 45.5512-5520, November 1985]

Biochemical Differences among Four Inosinate Dehydrogenase Inhibitors, Mycophenolic Acid, Ribavirin, Tiazofurin, and Selenazofurin, Studied in Mouse Lymphoma Cell Culture1

Huey-Jane Lee, Katarzyna Pawlak, Binh T. Nguyen, Roland K. Robins, and Wolfgang Sadee2

Department of Pharmaceutical Chemistry, School of Pharmacy, University of California, San Francisco, California 94143 [H-J. L, K. P., B. T. N., W. S.], and Cancer Research Center, Department of Chemistry, Brigham Young University, Provo, Utah 84602 [R. K. R.J

ABSTRACT 1.2.1.14) catalyzes the conversion of IMP to xanthylate, and it represents a key enzyme in the biosynthesis of guanine nucleo- The mechanism of the cellular toxicity of four Â¡nosinatedehy- tides. The activity of IMP dehydrogenase is positively linked to drogenase (IMP-DH) inhibitors with different antitumor and anti cellular transformation and tumor progression; therefore, this viral pharmacological profiles was investigated in mouse lym- enzyme represents a promising target of cancer chemotherapy phoma (S-49) cell culture. Drug effects on cell growth, nucleotide (1-3). Inhibition of IMP dehydrogenase results in the depletion pools, and Ã’NA and RNA synthesis were measured in the of cellular guanine nucleotides by blocking their de novo synthe presence and absence of guanine salvage supplies. Both guanine and guanosine were capable of bypassing the IMP-DH block, sis (4). Guanine nucleotides are required as substrates, activa while they also demonstrated some growth-inhibitory effects tors, or regulators in many pathways of cellular metabolism, including DNA, RNA, and protein synthesis (2). when added alone in high concentrations. All four drugs reduced Mycophenolic acid (5), ribavirin (6), selenazofurin (7, 8), and cellular guanosine triphosphate levels and caused secondary tiazofurin (9-11) (Chart 1) are all thought to exert toxic effects changes of the undine, cytidine, and adenosine triphosphate on mammalian cells primarily via IMP dehydrogenase inhibition, pools that were similar among the four drugs. However, several drug effects in addition to IMP-DH inhibition were observed since the toxicities of these drugs are largely preventable by an except with mycophenolic acid which may represent a pure IMP- exogenous supply of guanine nucleotides via the salvage path DH inhibitor. Both tiazofurin and selenazofurin interfered with the ways. Mycophenolic acid represents a prototype IMP dehydro uptake and/or metabolism of undine and thymidine tracers; genase inhibitor (5), with no other biochemical effects noted. It shows weak to moderate antitumor, antifungal, and immunosup- however, this effect appeared not to contribute to their cellular toxicity m vitro. Moreover, selenazofurin and tiazofurin impaired pressive activities (12). Despite the similar major mechanism of the utilization of exogenous guanine salvage supplies for DNA cellular toxicity, it was shown that ribavirin (13,14) represents a and RNA synthesis, and guanine was particularly ineffective in potent antiviral agent currently in clinical use with low antitumor reversing the toxic effects of tiazofurin on cell growth. This finding effects, while tiazofurin which is undergoing clinical trial (Phase is important in view of the available guanine salvage supplies in II) as an antitumor agent has low antiviral activities but acts as a vivo. Since tiazofurin, selenazofurin, and their known metabolites potent anticancer agent in certain animal tumor models (15). failed to inhibit hypoxanthine-guanine-phosphoribosyl transfer- Selenazofurin, a seleno analogue of tiazofurin, is highly effective ase, guanosine monophosphate kinase, and guanosine diphos- against both viral infections and animal tumors (16, 17); it is phate kinase in cell extracts or permeabilized cells, these drugs approximately 5 to 10 times more potent as an antitumor agent may interfere with salvage transport across cellular membranes. than tiazofurin (18). The toxic effects of mycophenolic acid and ribavirin were similarly At least two hypotheses can be proposed for the different reversed by salvage supplies of up to 200 UM guanine, which pharmacological effects of these four agents, (a) Differences in suggests that ribavirin primarily acts as an IMP-DH inhibitor under their pharmacokinetics could cause different exposure times that these conditions. This result could explain the rather low antitu- may be critical to the therapeutic outcome, (o) Some of these mor efficacy of both mycophenolic acid and ribavirin in vivo. agents produce biochemical effects at therapeutic concentra However, increasing the guanine salvage supply in the medium tions that are unrelated to IMP dehydrogenase inhibition. This above 200 ^M further reversed the toxic effects of mycophenolic paper addresses the second question by comparing the effects acid to maximum rescue, while it increased the toxicity of ribavirin of each agent on cell growth, cellular nucleotide pools, and DNA (300 UM). This finding suggests the presence of a toxic mecha and RNA synthesis of mouse S-49 lymphoma cells. These stud nism of ribavirin at higher concentrations that is dependent upon ies were performed in the presence and absence of guanine the presence of guanine supplies sufficient to fully overcome the salvage supplies that bypass the IMP dehydrogenase block. The IMP-DH inhibition. This study documents that each antimetabo- results reveal several biochemical differences among the four lite displays a unique spectrum of activities with multiple toxic IMP dehydrogenase inhibitors. targets.

INTRODUCTION MATERIALS AND METHODS

Inosinate dehydrogenase (IMP:NAD oxidoreductase, EC Reagents and Apparatus. Mycophenolic acid was provided by Eli Lilly and Co., Indianapolis, IN. Ribavirin (1-0-D-ribofuranosyl-1,2,4-tria- 1This research was supported by USPHS Grants CA-27866. CA 34304. and zole-3-carboxamide), ribavirin-5'-monophosphate, selenazofurin (2-/ÃŽ- CA-34384 from the National Cancer Institute. 2 To whom requests for reprints should be addressed. D-ribofuranosylselenazole-4-carboxamide), setenazofurin-5'-monophos- Received 9/10/84; revised 7/3/85; accepted 7/8/85. phate, tiazofurin (2-/Ã•-D-ribofuranosylthiazole-4-carboxamide), tiazofurin-

CANCER RESEARCH VOL. 45 NOVEMBER 1985 5512

INOSINATE DEHYDROGENASE INHIBITION

0.5 M KCI solution with a linear gradient of 0.01 M NHuridine was added to the cell MycophenolicAcid Tiazofurin suspension 30 min prior to harvest. The acid-insoluble cell pellet was precipitated by adding perchloric acid, and radioactivity was measured by liquid scintillation counting as previously described (21). Absolute DMA and RNA Synthesis Rates. Absolute rates of DNA and RNA synthesis were estimated by the incorporation into acid- insoluble material of [8-14C]guanine diluted with 50 to 400 ^M unlabeled guanine, after correction for the specific [14C]guanine activity in the medium. In the presence of high concentrations of the IMP dehydrogen ase inhibitors, exogenous guanine should represent the predominant HOCH2 guanine source for RNA and DNA. [8-14C]Guanine (1 nCi/ml), unlabeled guanine (50 to 400 MM),and the drugs were added to the cell medium 4 h prior to harvest. The acid-insoluble cell pellet was washed with water and then suspended overnight in 0.3 N KOH at room temperature to OH hydrolyze RNA to free ribonucleotides as described previously (21). Selenazofurin Acidification with 0.4 N HCIO4 precipitated DNA, while free ribonucleo Ribavirin tides derived from RNA remained in the supernatant. The radioactivity Chart 1. Chemical structures of mycophenolic acid, ribavirin, tiazofurin, and of the supernatant and pellet was measured by liquid scintillation count selenazofurin. ing. In order to assure that the specific activities of endogenous guanine 5'-monophosphate, and thiazole-4-carboxamide adenine dinudeotide nucleotide pools reach the same specific 14C activity as that of the (tiazofurin-NAD) were synthesized in the laboratory of Dr. Roland K. [14C]guanine in the medium, a requisite for the use of this approach to Robins. Guanine, guanosine, ATP, GMP, GDP, and L-a-lysophosphati- determine absolute nucleic acid synthesis rates, the specific 14Cactivity dylcholine (i_-Â«-lysolecithin)type I were purchased from Sigma, St. Louis, of the cellular GTP pool was determined by HPLC and liquid scintillation MO. All the other chemical reagents were of analytical grade. [8-14C]- counting. Cells were incubated for 4 h with 10 UM mycophenolic acid, Guanine HCI (54.4 mCi/mmol), [mef/iy/-3H]thymidine (80 Ci/mmol), and 300 /jM ribavirin, 300 /IM tiazofurin, or 40 /Â¿Mselenazofurin plus 100, [5-3H]uridine (23 Ci/mmol) were provided by ICN, Irvine, CA. [8,5'-3H]- 200, or 400 MM[14C]guanine. In each case, the specific [14C]GTP activity Guanosine 5'-monophosphate (38 Ci/mmol) was obtained from NEN, was identical to that of [14C]guanine in the medium (within experimental Boston, MA. [8-3H]Guanosine 5'-diphosphate (10.6 Ci/mmol) was pro error, Â± ~20%), which suggests that the exogenous guanine indeed vided by Amersham, Arlington Heights, IL. Cell density for growth ex represented the predominant guanine source in the cells. periments was determined on a Model ZF Coulter Counter (Coulter Preparation of Permeable Cells. S-49 cells were inoculated at a Electronics, Hialeah, FL). Radioactive analyses were carried out on a density of 2 to 5 x 105/ml and were harvested during their exponential Beckman LS-7800 liquid scintillation counter (Beckman Instruments, Palo growth (1 to 1.7 x KP/ml). Cells were permeabilized according to the Alto, CA). HPLC analyses were performed on a Model ALC/GPC 204 procedure of Miller ef al. (25) with the following modifications. The cells liquid Chromatograph equipped with a Model 6000A solvent delivery were collected by centrifugation at 320 x g for 5 min at 4Â°C,washed system (Waters Associates, Milford, MA). twice with 30 ml of ice-cold 150 HIM sucrose-80 mw KCI-35 mw 4-(2- Cell Culture. Mouse T-lymphoma (S-49) cells were grown in suspen hydroxyethyl)-1-piDerazineethanesulfonic acid-7.5 mM potassium phos- sion at 37Â°Cin Dulbecco's modified Eagle's medium supplemented with phate-4 mM MgCI2 buffer (pH 7.4), and subsequently suspended in the 10% heat-inactivted horse serum in a humidified, 10% CO2 atmosphere. same buffer at the cell density of 4.7 x 107/ml. t-a-Lysolecithin solution Their growth properties have been previously described (19, 20). (1 mg/ml) was added to give a final concentration of 150 Mg/ml. Within Growth experiments were performed in Costar tissue culture plates 2 min at 4Â°C,95 to 100% of the cells were permeable, as estimated by (24 wells, 2 ml/well) in at least triplicate sets. Cells in the logarithmic the trypan blue exclusion test. growth phase were incubated in complete medium at an initial cell density Assay of GMP and GDP Kinase A 25-Ml aliquot of the permeable cell of ~2 x 105/ml with various concentrations of the chemical agents to be suspension containing 106 cells was added to 75 n\ of the reaction tested either individually or in combination for 24 h. Cell growth after mixture prepared in 50 mw 4-(2-hydroxyethyl)-1-piperazineethanesulfonic drug exposure was determined by counting the cell number in a Coulter acid-7.5 mw potassium phosphate-10 mw MgCI2 buffer (pH 7.4). The Counter and was expressed as the percentage of control growth (21); it resulting suspension contained 3 mu ATP, 0.02 to 1.00 mM GMP, and was considered zero when the final cell density was less than the initial 0.1 MG of [3H]GMP for the GMP kinase assay. GDP kinase was assayed density. in the same mixture except that GDP replaced GMP as the enzyme HPLC9 Analysis of Nucleotide Pools. Cellular pools of DTP. CTP, substrate (0.04 to 2.6 mM, 0.1 Â¿

CANCER RESEARCH VOL. 45 NOVEMBER 1985 5513

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1985 American Association for Cancer Research. INOSINATE DEHYDROGENASE INHIBITION acetic acid solution supplemented with 1% sodium pyrophosphate- After 10 min, acid-insoluble material was collected by centrifugation at 13,000 x g for 20 s, washed twice with 1 ml of 5% trichloroacetic acid, and solubilized with 0.25 ml of 0.3 N KOH. The radioactivity of these hydroly- sates was measured to check for potential tracer incorporation into RNA. The supematants were neutralized with an equal volume of 0.5 M tri-n- octylamine in Freon (26, 27) and analyzed for acid-soluble products of the enzymatic reaction by HPLC with UV detection (Partisi! 10-SAX, 250- mm x 4.6-mm column; Whatman). A 25-//I sample was applied on the column, and GMP, GDP, and GTP were eluted with 0.1, 0.4, and 0.8 M ammonium phosphate buffer, pH 3.6 (each step, 6 min; flow rate, 2 ml/ min). One-mi fractions were collected and counted for 3H radioactivity. GMP kinase activity was calculated from the amount of [3H]GMP radio activity recovered in the GDP and GTP HPLC fractions and incorporated into acid-precipitable material. GDP kinase activity was calculated from "0 0.4 O.t 0 10 20 the amount of pHjGDP radioactivity associated with the GTP pool and MYCOPHENOLIC ACID.jiM RIBAVIRIN, MM cellular macromolecules. The latter radioactivity was tower than 0.5% of the radioactivity of the acid-soluble pool in each case. Assay of HGPRTase. S-49 lymphoma cells were harvested in the log CELL GROWTH 0â€”0 ONA SYNTHESIS phase of cell growth by centrifugation. The cells were washed by centnfugation 2 times in phosphate-buffered saline (pH 7.4). Washed D-d HNA SYNTHESIS cells were suspended into 50 HIM Tris buffer, pH 7.4, and homogenized in a sonteator (Model W-375; Heat Systems-Ultrasonics, Inc., Rain View, NY) for 10 s. Centrifugation of this crude extract at 20,000 x g for 20 min (4Â°C)yielded a dear supernatant. HGPRTase activities were deter mined according to a procedure described previously (28). [8-14C]Gua- nine (0.1 uCi) was incubated with 50 p\ of the supernatant containing 2 x 10* cell extract, 1.0 mM dithtothreitol, 1.0 mM freshly dissolved 5- phosphoribosyl-1 -pyrophosphate, 50 mM Tris buffer (pH 7.4), 5 mM MgCI2, and bovine serum albumin (2.4 mg/ml) in a total volume of 150 M!.After 30 min at 37Â°C,the samples were filtered under vacuum suction through PEI cellulose disks that were rinsed with 4 x 10 ml of 10 mM sodium acetate and once with 10 ml of distilled water. The PEI cellulose 0 2 0 10 20 disks were dried, and the radioactivity was measured by liquid scintillation SELENAZOFURIN, liM TIAZOFURIN, liM counting. The same four drugs and their metabolites as listed for the Chart 2. The toxicity of mycophenolic acid, ribavirin, selenazofurin, and tiazo GMP and GDP kinase assay were added in the beginning of the incu furin on S-49 cell growth (â€¢)andDNA (O) and RNA (D) synthesis. Cell growth was bation period at concentrations of 50 and 250 Â¿/M. determined over a 24-h incubation period. DNA Â¡13H|thymidme incorporation per 30 min) and RNA (pH [undine incorporation per 30 min) synthesis were determined after a 4-h incubation period. Each point was assayed in duplicate or triplicate in 2 to 4 independent experiments. RESULTS

Effects on Cell Growth, DMA, and RNA Synthesis. The the tracers into DNA and RNA, 30 MM guanosine was able to effects of mycophenolic acid, ribavirin, selenazofurin. and tiazo- completely reverse the effects of mycophenolic acid and ribavirin furin on cell growth and tracer incorporation into DMA and RNA over a large concentration range up to 10 times their toxic in S-49 cells are shown in Chart 2. The 50% growth-inhibitory concentration. The toxic concentration is defined here as that concentrations were: 0.3 MMmycophenolic acid; 10 MMribavirin; concentration causing zero cell growth over 24 h. However, 1.5 MM selenazofurin; and 7.5 MM tiazofurin. For mycophenolic when increasing the concentrations of selenazofurin and tiazo acid and ribavirin, the concentrations required to inhibit | 'H|- furin, the incorporation of tracers into DNA and RNA decreased thymidine uptake into DNA (measured at 4 h) were lower than despite the presence of 30 MM guanosine. Selenazofurin and those required to inhibit cell growth over 24 h. Similar results tiazofurin showed similar effects on the incorporation of tracers were obtained for ONA uptake at 24 h (data not shown here; into DNA and RNA at the same concentrations, although selen see Ref. 23). In contrast, for selenazofurin and tiazofurin, the azofurin was severalfold more toxic to the S-49 cells. Chart 4 inhibition of DNA tracer uptake paralleled cell growth inhibition. was plotted on an equitoxic concentration scale for better com All four agents were less effective in inhibiting [3H]uridine uptake parison of the effects on tracer uptake at therapeutically relevant into RNA than [3H]thymidine uptake into DNA. concentrations of each drug. Guanosine at 30 MM maximally rescued the toxic effects on The effects of the four drugs on cell growth in the presence of cell growth of all four agents (at the level causing zero growth guanosine rescue showed a different pattern (Chart 4). The over 24 h) and of three of the drugs (excluding tiazofurin) on rescue obtained with 30 MMguanosine (~70% of control growth) tracer incorporation into DNA and RNA (Chart 3). In order to test was not reversed when increasing mycophenolic acid and selen whether either of these four agents displays a toxic mechanism azofurin far above their toxic concentrations (zero cell growth; 1 that cannot be rescued by guanosine, cell growth and tracer and 4 MM,respectively). In contrast, increasing the concentration uptake into RNA and DNA were measured in the presence of 30 of tiazofurin to 10 times the toxic concentration gradually re MMguanosine and increasing doses of each drug. The results of duced the guanosine rescue to 30% of control cell growth, and this experiment are shown in Chart 4. For the incorporation of ribavirin completely abolished cell growth at high concentration

CANCER RESEARCH VOL. 45 NOVEMBER 1985 5514

GUANOSINE I Â»M)

It M GUANOSINE (Â»M) GUANOSINE l UM) Chart 3. GuarÃosmerescue of the toxicity of mycophenolic acid, ribavirin, tiazofurin, and selenazofurin on S-49 cell growth (A) and tracer incorporation into DMA (B) and RNA (C). Cell growth and DNA and RNA incorporation were measured as described in the legend of Chart 2. The concentrations of mycophenolic acid Chart 4. The effects of mycophenolic acid (MA), ribavirin (RV), selenazofurin (MA), ribavirin (RV), selenazofurin (SZ), and tiazofurin (TZ) were 0.8, 30, 4, and 30 (SZ), and tiazofurin (TZ) on S-49 cell growth (A) and tracer incorporation into DNA MM,respectively. These concentrations were sufficient to completely suppress net (B) and RNA (C) were determined in the presence of a guanosine rescue dose (30 cell growth over 24 h. Each point was assayed in duplicate or triplicate in 2 to 4 MM).Cell growth and DNA and RNA incorporation were measured as described in independent experiments. The dashed line indicates the effects of guanosine only. the legend of Chart 2. The abscissa was scaled to reflect equitoxic concentrations The control (100%) value represents incubations in the absence of drug and of each agent. Cells incubated in the absence of drug and guanosine (QUO) served guanosine. as the control, although 30 Â«IMguanosineslightly increased tracer uptake by 10 to 20% and decreased cell growth by 20 to 30% (O, ordinate). Because of changing de novo and salvage guanine fluxes with increasing drug concentrations, neither despite the presence of 30 MMguanosine. The observed cellular cells grown in the absence or presence of 30 UM guanosine can serve as the correct control values for each condition. However, plotting the data against toxicity of ribavirin at the higher concentrations stands in contrast incubation with 30 Â«tMguanosine as the control did not alter the general pattern of to the lack of any effect on tracer uptake into DNA and RNA in the graph and, hence, did not affect the conclusions. the presence of 30 /JMguanosine. Drug Effects on Cellular Ribonucleotide Triphosphate presented in Charts 5 and 6. Here, the depletion of GTP was Pools. The four IMP-DH inhibitors were incubated with S-49 reversed only for mycophenolic acid and ribavirin, but not for cells at their toxic concentrations (zero cell growth) and at 10- tiazofurin, while selenazofurin was intermediate. The finding that fold their toxic concentrations; the pools of UTP, CTP, ATP, and tiazofurin was more potent than selenazofurin (at equitoxic con GTP were determined after 4 h by HPLC (Table 1). There were centrations) in preventing the restoration of the cellular GTP pool no dramatic differences noticed among the effects of mycophen by exogenous guanine was reproduced in a separate experiment olic acid, ribavirin, tiazofurin, and selenazofurin at the lower with 300 iiM tiazofurin or 40 UM selenazofurin plus 100, 200, or concentration (zero cell growth). GTP pools were sharply re 400 pu guanine added to the medium. Finally, ribavirin (300 /IM) duced, while UTP and CTP pools were increased. Changes of plus guanine (200 /Â¿M)appeared to reduce the UTP, CTP, and the ATP pools were small in all cases. These results are com ATP pools (single determination). parable to previous findings, e.g., with tiazofurin (29). The addi Drug Effects on Cellular [3H]Uridine Uptake. The incorpora tion of 30 MMguanosine brought the GTP pools to 2 to 3 times tion of [3H]uridine into acid-soluble material is shown in Table 2. those of control cells, in both the presence and absence of the Both mycophenolic acid and tiazofurin reduced the 3H activities four drugs at the lower concentration studied. Also the effects in RNA (see Charts 2 and 3) and in the soluble nucleotide pools of the low drug concentrations on UTP, CTP, and ATP pools (UMP + deoxyuridine, UDP, and UTP). However, a marked were minimal in the presence of guanosine. difference was observed between the effects of mycophenolic The high drug concentrations caused similar changes of the acid (1 /Â¿M)and tiazofurin (30 /Â¿M)in the presence of 30 ^M ribonucleotide pools, with the possible exception of tiazofurin guanosine that afforded maximum rescue of cell growth. While which caused generally lower levels (single determination). Qua- mycophenolic acid had little effect on tracer uptake into acid- nine (200 UM) was used in combination with high drug concen soluble pools in the presence of guanosine, tiazofurin inhibited trations (Table I) in order to compare the results with those uptake by more than 50%. Similar results were obtained with

CANCER RESEARCH VOL. 45 NOVEMBER 1985 5515

Table 1 Drugs effects on the pool sizes of ribonudeoside Mphosphates Cells were analyzed after 4-h drug exposure. % of control ATP CTP GTP UTP Control' 100 100 100 100

Mycophenolic acid, 1 MM*" 101 + 15 134 Â±17 24 Â± 5 139 Â±17 Ribavirin, 30 Â¡M 131 161 21 184 Tiazofurin, 30 MM" 110 Â±12 136 Â±15 40 Â± 8 140 Â±16 Selenazofurm. 4 Â»w, 91 116 29 124

Mycophenolic acid, 1 MM, + 72 Â±11 76 Â±14 196 Â±37 70 Â±11 guanosine, 30 MM Ribavirin, 30 MM, + guano- 131 131 343 116 sine, 30 MM Tiazofurin, 30 urn, + guano- 98 Â±10 100 Â±13 254 Â±12 92 + 14 sine, 30 MM Selenazofunn, 4 MM, + gua- 111 110 294 97 nosine, 30 MM 500 Guanosine, 30 MMC 87 Â± 4 92 Â± 5 221 Â±22 77 Â± 4 Chart 6. The effect of guanine added to the medium on cell growth over 24 h in the presence of high concentrations of the four IMP-DH inhibitors [10 MM Mycophenolic acid, 10 MM" 90 Â± 2 112 Â± 2 23+4 111 Â± 1 mycophenolic acid (MA). 300 MMribavirin (RV), 40 MM selenazofunn (SZ), and 300 Ribavirin, 300 MM 111 172 17 164 MMtiazofurin (TZ)]. (Compare with Chart 5.) The dashed line represents the effects Tiazofurin, 300 MM 67 122 10 106 of guanine (GUA) only. The control (100%) value represents incubations in the Selenazofurm. 40,/M 105 155 18 164 absence of drug and guanine. Bars, SD.

Mycophenolic add, 10 MM,+ 95 100 234 94 guanme. 200 MM [3H]thymidine tracer uptake into DNA and thymidine nucleotides Ribavirin, 300 MM, + gua- 53 63 217 52 nine, 200 MM (data not shown). Tiazofurin, 300 MM,+ gua- 113 151 32 170 Absolute DNA and RNA Synthesis Rates. When the de novo nine, 200 MM Setenazofurin, 4 MM, + gua- 126 131 63 146 synthesis of GTP is blocked with high levels of the IMP dehydro nine, 200 MM genase inhibitors, the intracellular GTP pool approaches the specific activity of guanine in the medium (measured by HPLC; Guanine, 200 MM 102 110 254 101 see "Materials and Methods"), and isotope incorporation into a Mean of 5 determinations. The pool sizes of control were: ATP, 373 nmol/10" cells; CTP, 30 nmol/1 Of cells; GTP, 25 nmol/IO8 cells; and UTP, 98 nmol/1011cells. RNA/DNA then should provide a measurement of the absolute b Mean Â±SD of 4 determinations. ' Mean Â±SD of 3 determinations. rates of RNA/DNA synthesis when long tracer incubation periods " Mean Â±range of 2 determinations. are used (4 h). A similar principle has been previously applied to the incorporation of thymidine into DNA (30). Furthermore, we have noticed that [14C]guanine at 300 to 400 /JM in the medium apparently also inhibits de novo guanine nucleotide synthesis, since [14C]guanine incorporation into DNA and RNA was similar in the presence and absence of 10 Â«Mmycophenolic acid, and the specific [14C]GTP activity was similar to that of [14C]guanine in the medium, even in the absence of an IMP-DH inhibitor. The estimated absolute DNA and RNA synthesis rates measured in the presence of the four inhibitors and increasing concentrations of guanine in the medium are shown in Chart 5. For mycophenolic acid and ribavirin, 100 UM guanine was sufficient to increase DNA and RNA synthesis to maximal levels (~10 to 15 and 30 to 40 nmol/108 cells/30 min, respectively), while selenazofurin and tiazofurin did not allow maximal rates at this guanine concentra tion. Maximal DNA and RNA synthesis rates (i.e., those observed in the presence of guanine alone) were achieved for all four agents when increasing the guanine level in the medium to 400 flM. Effect of Guanine Rescue on Cell Growth in the Presence of High Concentrations of IMP-DH Inhibitors. Cell growth over 100 200 400 Cone. Of Guanine ' uÂ« 24 h was measured in the presence of high concentrations of the IMP-DH inhibitors [10 times the toxic concentration (= zero Chart 5. Absolute DNA and RNA synthesis rates were estimated with [14C]- guanine added to the medium in the presence of high IMP dehydrogenase inhibitor cell growth)] and increasing levels of guanine as described in the levels. Cells were incubated for 4 h in the presence of |"*Ciguanme (1 ,

CANCER RESEARCH VOL. 45 NOVEMBER 1985 5516

Table 2 pHJUridine (30-min pulse) uptake into cellular nucleotide pools, 4 h after the start ol the incubations (1.6 x 1(f cells/50 iJ). HPLC eluent fractions were collected and analyzed for 3H activity. Uridine and UMP were not separated under the assay conditions.

30 MM,+ guano- phenolic acid, 1 MM+ sine, furin,30 sine, acid,1 guanosine,30 30MM3,998 MM1,396 30MM1,654 MM1,627 MM3,574 Uridine (and UMP) UDP 1,279 1,159 529 420 6995,764Mycophenolic912 UTPControl4,945 14,614Guano-9,955Tiazo-5,295dpm/sampteTiazofurin,4,098Myco- 9,044

by measuring the rate of disappearance of guanine in the medium slight decrease of GMP kinase activity (approximately 15%) was that was found to contain guanase activity. After 24 h of incu observed by the 5'-monophosphate metabolites of the three bation, more than 50% of guanine (200 /Â¿M)wasmetabolized to nucleoside drugs and the NAD-analogue of tiazofurin, however, xanthine. Similarly guanosine was partially converted to guanine not in a dose-dependent manner. by purine nucleoside phosphorylase activity in the medium. Similarly, none of the drugs and their metabolites (at 50 MM The results on guanine rescue of cell growth (Chart 6) show and 250 MM) had a measurable effect on HGPRTase in cyto- that 300 MM tiazofurin inhibited guanine rescue below 200 MM, plasmic cell extracts. Assays were carried out in duplicate for while mycophenolic acid was ineffective and selenazofurin was two independent experiments, and [3H]GMP product formation intermediate. The ribavirin curve was biphasic with maximum cell ranged between 90 and 110% of the control value when the growth reached at 150 MMguanine and a complete loss of rescue drugs were added. above 200 MMguanine. Maximum guanine rescue from ribavirin toxicity on cell growth was limited to 40 to 50%. Guanosine was DISCUSSION also tested for its ability to rescue the cell growth toxicity caused by high concentrations of the four IMP-DH inhibitors. The effects We have selected four potent IMP dehydrogenase inhibitors of guanosine were identical to those observed with guanine (mycophenolic acid, ribavirin, tiazofurin, and selenazofurin) that (albeit for guanosine at 5- to 10-fold-lower concentrations than display different pharmacological profiles as antiviral and antitu- guanine), and therefore, the results are not shown here. mor agents. Moreover, the action of each agent is dependent Effects of Drugs on GMP and GDP Kinases and on upon different mechanisms of metabolic activation, while biolog HGPRTase. The last two enzymes, GMP and GDP kinase, in ical effects other than IMP dehydrogenase inhibition have only the biosynthesis of GTP, the substrate for RNA synthesis, display been established for ribavirin (31). Mycophenolic acid acts di a high capacity relative to the guanine usage in nucleic acid rectly on IMP dehydrogenase without any further effects re synthesis. While the RNA synthesis rate in S-49 cells amounted ported, while its metabolites are inactive (32). In contrast, ribavirin to approximately 30 to 40 nmol/108 cells/30 min (see above) (10 requires phosphorylation to its 5'-monophosphate by adenosine pmol/106 cells/min), the maximum rate of GTP synthesis by GDP kinase for activity (33). IMP dehydrogenase is regulated in vivo kinase was determined here as 27 nmol/min/106 cells with Km = by guanosine monophosphate, and ribavirin monophosphate is 919 MMin a permeabilized cell preparation. The efficiency of GTP 50 to 100 times more potent than GMP as an inhibitor of IMP generation in the cells is limited by the lower activity of GMP dehydrogenase (34). Moreover, ribavirin assumes a conformation kinase, Vâ„¢Â«=1 nmol/min/106 cells and Km = 36 MM,if one also strikingly similar to guanosine as determined by single crystal X- considers the much lower GMP than GDP pool in the intact cell. ray crystallography (35). It is hence phosphorylated to ribavirin These results were obtained by fitting straight lines through triphosphate which inhibits 5'-cap formation of mRNA (31). This Lineweaver-Burk plots of the GMP kinase (r = 0.915) and the secondary effect of ribavirin leads to the accumulation of mRNAs GDP kinase (r = 0.908) assay data. The GDP produced by GMP either inert or impaired in protein synthesis (36), since the 5'- kinase was immediately transformed to GTP by GDP kinase with terminal 7-methylguanosine in mRNA is required for its efficient close to 100% yield during the first 5 min of incubation, and only translation (37). However, it remains unknown to which extent 20% of GDP was accumulated during longer incubations (data this mechanism contributes to ribavirin toxicity against mamma not shown). lian cells. The nucleosides and 5'-monophosphate nucleotides No significant inhibition of the activity of GMP and GDP kinase of both tiazofurin and selenazofurin are relatively poor inhibitors by the drugs, mycophenolic acid, ribavirin, tiazofurin, and selen of IMP dehydrogenase; however, both agents were found to be azofurin, and their respective metabolites (see "Materials and incorporated into NAD+ in lieu of nicotinamide to yield the dinu- Methods") was observed. Assay of the inhibitory effects of the cleotide metabolites tiazofurin-NAD and selenazofurin-NAD. drugs on the enzyme activities was performed over a 5-min time These "NAD" analogues are potent inhibitors of IMP dehydro period at a substrate concentration close to the Km value, i.e., genase (8-11,18). No further major effects have been reported 0.04 HIM GMP or 1 rriM GDP. The drugs were added to the for tiazofurin and selenazofurin that could account for their incubation mixture over a large concentration range including different pharmacological spectrum, except for a reduction of highly toxic levels that cause complete inhibition of S-49 cell NAD+ pools that was caused by tiazofurin and could contribute growth during 24-h incubations. Under these conditions, only a to its cell toxicity (38).

CANCER RESEARCH VOL. 45 NOVEMBER 1985 5517

Drug Effects on Cell Growth, Tracer Incorporation into while mycophenolic acid was ineffective, and selenazofurin Nucleic Acids and Nucleotide Pools. A careful comparison of showed only a small inhibitory effect. This result suggests that the effects of the four drugs under study on cell growth, DNA, ribavirin and tiazofurin produce effects unrelated to IMP-DH and RNA synthesis revealed systematic differences among these inhibition at high concentrations. The mentioned interference of mRNA capping could be responsible for ribavirin's toxicity (31). agents. Although tracer incorporation into DNA and RNA does not reveal their absolute synthesis rates, any differences in the The recently reported reduction of the cellular NAD+ pool could pattern caused by the four similar agents would suggest different account for the additional toxic mechanism of tiazofurin (38). mechanisms of action. We had previously shown that mycc- Another possible mechanism involves interference of guanine phenolic acid most strongly affects [3H]thymidine incorporation salvage by tiazofurin. at 4 and 24 h (22). Ribavirin caused a pattern of DNA-RNA-cell Interference of Guanine Salvage by Tiazofurin and Selen growth inhibition that is identical to that of mycophenolic acid azofurin. Streeter and Miller (39) have previously suggested that (Chart 2), which could suggest that both agents primarily act as tiazofurin, but not selenazofurin, inhibits GDP kinase as well as pure IMP dehydrogenase inhibitors at the concentrations stud IMP dehydrogenase. The inhibition of GDP kinase by tiazofurin ied. In contrast, tiazofurin and setenazofurin did not preferentially could explain the relative inability of guanine salvage supply to inhibit [3H]thymidine incorporation over cell growth. This result overcome the drug's toxicity, and therefore, this mechanism clearly indicates different biological mechanisms among the four should be considered in detail. Their method to assay enzyme drugs. The effects of the four agents on DTP, CTP, ATP, and activities in the intact cell is based on the procedure of Crabtree GTP pools were similar (in the absence of guanine rescue) (Table and Henderson (40). In principle, a metabolic precursor (e.g., 1), suggesting the absence of any overt effects on nucleotide |MC|glycine) is used to label each nucleotide pool. Relative metabolism under these conditions, other than those caused by enzyme activities are then calculated from the 14Cactivity of the IMP-DH inhibition. precursor pool (e.g., GDP) and the 14C activities of all product Drug Effects on Tracer Metabolism. The different results on pools (e.g., GTP). However, this approach suffers from at least tracer incorporation into nucleic acids could arise from selective two systematic errors that may influence the interpretation of effects of either agent on tracer uptake and metabolism in the results, (a) Pool size changes of the precursor pool are not being cells. In order to address this question, we have studied the considered, so that the specific 14Cactivity of the precursor pool effects of the four drugs on cell growth and tracer metabolism in is unknown. (Â£>)Notall product pools are measured, particularly the presence of guanine salvage supplies. Guanosine in the RNA which contains a substantial fraction of the "C activity and culture medium can effectively bypass the IMP dehydrogenase the synthesis of which is inhibited by the drugs under study. In block, thereby, largely rescuing the cells from toxiaty. Maximum fact, on the basis of the same approach, Snyder et al. (41) have rescue of cell growth inhibition for each drug (at a level that suggested that mycophenolic acid also inhibits GDP kinase in prevents cell growth over 24 h) was obtained at 30 UMguanosine addition to IMP-DH. The rather different chemical nature of (Chart 3). However, 30 MM guanosine failed to restore |3H|- tiazofurin and mycophenolic acid makes it unlikely that both thymidine and [3H]uridine incorporation into nucleic acid (Chart agents affect IMP-DH and GDP kinase to similar extents. 3) in the presence of a toxic concentration of tiazofurin (zero cell Therefore, we have measured any possible effects of the drugs growth), which suggests that tiazofurin inhibits tracer uptake on GMP kinase and GDP kinase in permeabili/ed cells under into the cell and/or their subsequent phosphorylation. This con well-defined conditions. Neither of the nucleosides, ribavirin, clusion was supported by the analysis of [3H]uridine uptake into tiazofurin, and selenazofurin, nor their 5'-monophosphate me uracil nucleotide pools in the presence of either mycophenolic tabolites, nor the NAD analogue of tiazofurin, nor mycophenolic acid or tiazofurin (Â±30UM guanosine). In the absence of guano acid showed any appreciable inhibition of the two enzymes. This sine, both tiazofurin and mycophenolic acid reduced tracer la result does not entirely rule out the possibility that GDP kinase beling of the acid-soluble pools, possibly because of the in is indeed inhibited in the intact cells by tiazofurin as suggested creased DTP pools. However, in the presence of guanosine that (39), although all known major metabolites were included with normalizes DTP pools (Table 1), only tiazofurin, but not myco the permeabilized cell assay. The different assay conditions must phenolic acid, reduced [3H]uridine uptake into acid-soluble pools. be considered. Since tiazofurin caused a similar reduction of 3H activity in each Inhibition of HGPRTase by tiazofurin could also account for of the uridine + UMP, UDP, and UTP pools (in the presence of the relative inability of guanine rescue to overcome its toxicity. guanosine; Table 2), it is possible that the interference of tracer However, none of these agents, including tiazofurin, its monophosphate, and NAD" analogue, affected HGPRTase activity at uptake occurred at the level of the membrane transport. It can be concluded, however, that in the presence of tiazofurin and or below 250 MM. An alternative explanation is the inhibition of selenazofurin, incorporation rates of uridine and thymidine into HGPRTase by elevated IMP levels, as suggested for tiazofurin nucleic acids cannot be used to assess ONA and RNA synthesis. by Lui ef al. (29). However, it is difficult to understand why this Experients with increasing concentrations of tiazofurin and se indirect effect did not occur with the guanine rescue for all IMP- lenazofurin in the presence of 30 /Â¿Mguanosine demonstrated DH inhibitors. that both drugs inhibited tracer uptake with equal potency (Chart In order to further address the question of a possible interfer 4). ence by the drugs with the utilization of salvageable guanine Guanine-Guanosine Rescue from Drug Toxicity. Further supplies in the medium, a question of importance for the activity differences in the biological effects of the four agents are revealed of these drugs in intact animals or patients, we have studied by their effect on cell growth in the presence of guanine rescue guanine utilization in the presence of highly toxic drug levels (10 supplies. High concentrations of both ribavirin and tiazofurin were times the level that causes zero cell growth over 24 h) (Chart 5). capable of reversing the rescue effects of 30 MM guanosine, Although in the presence of all four drugs 400 MM guanine

CANCER RESEARCH VOL. 45 NOVEMBER 1985 5518

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1985 American Association for Cancer Research. INOSINATE DEHYDROGENASE INHIBITION brought DMA and RNA synthesis to maximum levels (-10 to 15 8. Jayaram, H. N., Ahluwalia, G. S., Dion, R. L., Gebeyehu, G., Marquez, V. Eâ€ž Kelly, J. A., Robins, R. K., Cooney, D. A., and Johns, D. G. Conversion of 2- and 30 to 40 nmol/108 cells/30 min, respectively) that are ex .>'D-nbofuranosylselenazole-4-car box amide to an analogue of NAD with potent pected under normal conditions, guanine was less potent at IMP dehydrogenase-inhibitory properties. Biochem. Pharmacol., 32: 2633- lower concentrations to reverse the DNA and RNA effects of 2636,1983. 9. Cooney, D. A., Jayaram, H. N., Gebeyehu, G., Betts, C. R., Keltey, J. A., selenazofurin and particularly tiazofurin when compared to riba- Marquez, V. E., and Johns, D. G. The conversion of tiazofurin to an analogue virin and mycophenolic acid (Chart 5). of NAD with potent IMP dehydrogenase-inhibitory properties. Biochem. Phar macol., 37:2133-2136,1982. The ability of guanine to restore GTP pools and rescue cell 10. Jayaram, H. N., Dion, R. L., Glazer, R. I., Johns, D. G., Robins, R. K., growth was greatly impaired in the case of tiazofurin (300 ^M) Srivastava, P. C., and Cooney, D. A. Initial studies on the mechanism of action relative to mycophenolic acid, while selenazofurin was interme of a new oncolytic thiazole nucleoside. 2-/3-rMibofuranosylthiazote-4-carboxamide (NSC 286193). Biochem. Pharmacol.. 37: 2371-2380,1982. diate. Since these two drugs and their metabolites had no effect 11. Saunders, P. P., Kuttan, R., Lai, M. M., and Robins, R. K. Action of 2-0-0- on the enzymes participating in guanine salvage (HGPRTase, nbofuranosylthiazole-4-carboxamide (tiazofurin) in Chinese hamster ovary and variant cell lines. Mol. Pharmacol., 23: 534-539,1983. GMP, and GDP kinase), it is likely that the transport of guanine 12. Williams, R. H., Lively, D. H., DeLong, D. C., Cline, J. C., Sweeney, M. J., and guanosine across the cell membrane is inhibited, as ob Poore, G. A., and Larsen, S. H. Mycophenolic acid: antiviral and antitumor served with undine and thymidine transport. The potent inhibition properties. J. Antibiot. (Tokyo), 27:463-464,1968. 13. Sidwell, R. W., Huffman, J. H., Khare, G. P., Witkowski, J. T., Allen, L. B., and of guanine salvage by tiazofurin may contribute to the rather Robins, R. K. Broad-spectrum antiviral activity of virazole: 1-.Â¡-n-nbofuranosyl- high level of toxicities observed in clinical trials. 1,2,4-triazole-3-carboxamide. Science (Wash. DC), 777:705-706,1972. Effects of Guanine Salvage on Ribavirin Toxicity. Guanine 14. Witkowski. J. T., Robins, R. K., Sidwell, R. W., and Simon, L. N. Design, (as well as guanosine) displayed biphasic effects on ribavirin's synthesis, and broad spectrum antiviral activity of 1-.Â¡-o-ribofuranosyl-1.2,4- triazole-3-carboxamide and related nucleosides. J. Med. Chem., 14: 1150- cell toxicity (Chart 6). Lower concentrations of guanosine (opti 1154,1972. mal, 15 Â¡ÃM)andguanine (optimal, 150 UM) rescued the cell 15. Robins, R. K., Srivastava. P. C., Narayanan, V. L., Plowman, L., and Paull, K. D. Tiazofurin, a novel potential antitumor agent for lung tumors and mÃ©tas growth to ~40% of control growth. In these lower guanine tases. J. Med. Chem., 25:107-108,1982. dosage ranges, ribavirin behaves exactly like the "pure" IMP-DH 16. Kirsi, J. J., North, J. A., McKeman, P. A., Murray, B. N., Canonico, P. G., Huggins, J. W., Srivastava, P. C., and Robins, R. K. Broad-spectrum antiviral inhibitor mycophenolic acid, and no additional toxic mechanism activity of 2-.<-o-ribofuranosylselenazole-4-carboxamide, a new antiviral agent. of ribavirin is apparent. However, higher levels of guanine salvage Antimicrob. Agents Chemother., 24: 353-361,1983. maximally reverse mycophenolic acid toxicity, but they also allow 17. Srivastava, P. C., and Robins, R. K. Design, synthesis, and amitumor activity of 2-,>-D-nbofuranosylselenazole-4-carboxamide and related derivatives. J. the secondary mechanism of ribavirin toxicity (Chart 4) to be fully Med. Chem., 26: 445-448,1983. expressed. The mechanism of this guanine-dependent toxicity 18. Streeter, D. G., and Robins, R. K. Comparative in vitro studies of tiazofurin of ribavirin remains unclear. Since mRNA capping involves a and a selenazole analog. Biochem. Biophys. Res. Commun., 775: 544-550, 1983. guanine residue, it is conceivable that a sufficient amount of 19. Ulman, B., Cohen, A., and Martin. D. W., Jr. Characterization of a cell culture guanine must be available in order to fully express any toxic model for the study of adenosine deaminase- and purine nucleoside phospho- effects of ribavirin that stem from its proposed interference with rylase-deficient immunologie disease. Cell, 9: 205-211,1976. 20. miman, B., Gudas, L. J., Cohen, A., and Martin, D. W., Jr. Deoxyadenosine the capping mechanism (31). metabolism and cytotoxicity in cultured mouse T-lymphoma cells: a model for In conclusion, a biochemical comparison among the four drugs, immunodeficiency disease. Cell, 14:365-375,1978. 21. Cohen, M. B., Maybaum, J., and SadÃ©e,W. Guanine nucleotide depletion and ribavirin, mycophenolic acid, tiazofurin, and selenazofurin, sup toxicity in mouse T-lymphoma (S-49) cells. J. Btol. Chem., 256: 8713-8717, ports the notion that inhibition of IMP dehydrogenase indeed 1981. represents their major mode of toxicity against mammalian cells. 22. Cohen, M. B., and SadÃ©e,W. Contributions of the depletion of guanine and adenine nucteotides to the toxicity of purine starvation in the mouse T- However, each agent (with the exception of mycophenolic acid) lymphomacell line. Cancer Res., 43:1587-1591,1983. displays a different pattern of additional effects that may contrib 23. Nguyen, B. T., El Sayed, Y. M., and SadÃ©e,W. Interaction among the distinct ute to their different pharmacological spectrum. Because of the effects of adenine and guanine depletion in mouse lymphoma cells. Cancer Res., 44: 2272-2277,1984. current clinical interest in these agents, systematic biochemical 24. Pogolotli. A. L., and Santi, D. V. High-pressure liquid chromatography-ultravi differences need to be studied in detail with the expectation that olet analysis of intracellular nucleotides. Anal. Biochem., 726:335-345,1982. each new IMP dehydrogenase inhibitor may have a different 25. Miller, M. R., Castello!. J. J., Jr., and Pardee, A. B. A permeable animal cell preparation for studying macromolecular synthesis. DNA synthesis and the mode of action. rote of deoxyribonucleotides in S phase initiation. Biochemistry, 77: 1073- 1080,1978. 26. Khym, J. X. An analytical system for rapid separation of tissue nucleotides at REFERENCES low pressures on conventional aniÃ³nexchangers. Clin. Chem., 27:1245-1252, 1975. 1. Robins, R. K. Nucleoside and nucleotide inhibitors of IMP dehydrogenase as 27. Maybaum, J., Klein, F. K., and SadÃ©e,W. Determination of pyrimidine ribotide potential antitumor inhibitors. Nucleosides & Nudeotides, 1: 35-44,1982. and deoxynbotide pools in cultured cells and mouse liver by HPLC. J. Chro- 2. Weber, G. Enzymes of purine metabolism in cancer. Clin. Biochern.. 76: 57- matogr, 188:149-158,1980. 63. 1983. 28. Olsen, A. S., and Milman, G. Hypoxanthine phosphoribosyl transferase from 3. Weber, G. Biochemical strategy of cancer cells and the design of chemother Chinese hamster brain and human erythrocytes. Methods Enzymol.. 57: 543- apy. Cancer Res., 43: 3466-3492, 1983. 549.1978. 4. Lowe, J. K., Brox, L, and Henderson, J. F. Consequences of inhibition of 29. Lui, M. S., Faderan, M. A., Ã¼epmeks.J. J., Natsumeda. Y., oiah. E., Jayaram. guanine nucleotide synthesis of mycophenolic acid and virazole. Cancer Res., H. N., and Weber, G. Modulation of IMP dehydrogenase activity and guanylate 37: 736-743,1977. metabolism by tiazofurin (2-,Â¡-D-ribofuranosylthiazo!e-4-carboxamide). J. Biol. 5. Franklin, T. J., and Cook, J. M. The inhibition of nucleic acid synthesis by Chem., 259: 5078-5082,1984. mycophenolic acid. Biochem. J., 773: 515-524,1969. 30. Nicander, B., and Retehard, P. Dynamics of pyrimidine deoxyribonucleoside 6. Smith, C. M., Fontenelle, L. J., Muzik, H., Paterson, A. R. P., Unger, H., Brox, triphosphate pools in relationship to DNA synthesis in 3T6 mouse fibroblasts L. W., and Henderson, J. F. Inhibitors of inosinate dehydrogenase activity in Proc. NaÂ«.Acad. Sci. USA, 80:1347-1351,1983. Ehrlich asertes tumor cells in vitro. Btochem. Pharmacol., 23: 2727-2735, 31. Goswami, B. B., Borek, E., Sharma, 0. K., Fujitaki, J., and Smith, R. A. The 1974. broad spectrum antiviral agent ribavirin inhibits capping of mRNA. Biochem. 7. Lucas, D. L., Robins, R. K., Knight, R. D., and Wright, D. G. Induced maturation Biophys. Res. Commun., 89: 830-836,1979. of the human promyelocylic leukemia cell line, HL-60, by 2-0-o-ribofuranosyl- 32. Sweeney, M. J., Hoffman, D. H., and Esterman, M. A. Metabolism and selenazole-4-carboxamide. Biochem. Biophys. Res. Commun., 115:971-980, biochemistry of mycophenolic acid. Cancer Res., 32: 1803-1809, 1972. 1983. 33. Willis, R. C.. Carson, D. A., and Seegmiller. J. E. Adenosine kinase initiates

CANCER RESEARCH VOL. 45 NOVEMBER 1985 5519

the major route of ribavirin activation in a cultured human cell line. Proc. Nati. mRNA. J. Bid. Chem., 253:1710-1715,1978. Acad. Sci. USA, 75: 3042-3044,1978. 38. Liepraeks,J. J., Faderan,M. A., Lui, M. S., and Weber, G. Tiazofunn-induced 34. Streeter, 0. G., Witkowski, J. T., Khare, G. P., Sidwell, R. W.. Bauer. R. J , selective depression of NAD content in hepatoma 3924A. Biochem. Biophys. Robins, R. K., and Simon, L. N. Mechanismof action of virazote,a new broad- Res. Commun., Ã•22:345-349,1984. spectrum antiviral agent. Proc. Nati. Acad. Sci. USA, 70:1174-1178,1973. 39. Streeter, D. G., and Miller, J. P. The in vitro inhibition of punne nucleotide 35. Prusmer. P., and Sundaralingam. M. Broad-spectrum antiviral properties of biosynthesisby 2-/3-D-ribofurarx>sylthiazole-4-carboxamide.Biochem.Biophys. Res. Commun., 703:1409-1415,1981. virazote. Nat. New Bio).,244:116-117,1973. 40. Crabtree,G. W., and Henderson,J. F. Rate-limitingsteps in the interconversion 36. Sarvar, N., and Stellar, V. Virazote prevents production of Sindbis virus and of punne ribonucleotides on Ehrlich ascites tumor cells in vitro. Cancer Res., virus-induced cytopathic effect in Aedes albopictus cells. Virology, 91: 267- 37:985-991,1971. 282,1978. 41. Snyder,F.F., Henderson,J. F.,and Cook, D. A.Inhibition of purinemetabolism- 37. Muthuknshnan. S., Moss, B., Cooper, J. A., and Maxwell, E. S. Influence of computer-assisted analysis of drug effects. Biochem. Pharmacol.,27: 2351- 5'-terminal cap structure on the initiation of the translation of vaccinia virus 2357,1972.

CANCER RESEARCH VOL. 45 NOVEMBER 1985 5520

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1985 American Association for Cancer Research. Biochemical Differences among Four Inosinate Dehydrogenase Inhibitors, Mycophenolic Acid, Ribavirin, Tiazofurin, and Selenazofurin, Studied in Mouse Lymphoma Cell Culture

Huey-Jane Lee, Katarzyna Pawlak, Binh T. Nguyen, et al.

Cancer Res 1985;45:5512-5520.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/45/11_Part_1/5512

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/45/11_Part_1/5512. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.