[CANCER RESEARCH 44, 4328-4332, October 1984]

Selection of 9-/?-D-Arabinofuranosyladenine-resistant Human T-Lymphoblasts with Altered Ribonucleotide Reductase Activity1

Arnold Fridland Division oíBiochemical and Clinical Pharmacology, St. Jude Children 's Research Hospital, Memphis, Tennessee 38101

ABSTRACT drug, as well as lhal of ils new adenosine deaminase-resistant analogue, 2-fluoro-ara-A, has not been established. In vivo, ara- We sought to define the cellular activity that mediates resist A can be converted to its nucleotide derivalives (4, 9, 22, 23), ance in human leukemic cells (CCRF-CEM) to the nucleoside 9- which can compele for cellular processes requiring ATP or dATP; j3-D-arabinofuranosyladenine (ara-A). Stable mutants were ob the active form of the analogue is the triphosphate derivative (4, tained by continuous selection at ara-A concentrations of 1 or 9). In particular, ara-ATP is a potenl inhibitor of DMA polymerase- 2.5 /¿Minthe presence of the adenosine deaminase inhibitor 2'- a and ribonucleolide reducíase(8, 9, 30). Recent work (17) has deoxycoformycin. Four clones selected for further investigation also shown that, in cultured mammalian cells, incorporation of were 4- to 11-fold less sensitive to the cytotoxicity of ara-A than ara-A into DNA correlates well with ils cyloloxicily. the parental CCRF-CEM line. These clones also showed cross- Mechanisms of ara-A cyloloxicily olher than DNA synthesis resistance to deoxyadenosine and thymidine, but normal sensi inhibition have also been proposed. Rose ef a/. (25) demonstraled tivity to arabinosylcytosine and adenosine, and increased sensi that ara-A causes a selective inhibition of RNA polyadenylation tivity to the etoposide VP16-213. No change was found in the in cullured mouse fibroblasts. Also, ara-A inhibits S-adenosyl- activity of kinases that phosphorylate ara-A and the various homocysleine hydrolase (12, 15, 16), which leads lo Ihe accu nucleosides that could account for the resistant phenotype in mulation of S-adenosylhomocysteine (a polenl inhibitor of aden- these mutant lines. Resistance was associated with a 2- to 8- osine-melhionine-dependenl melhylation reaclions) in some pa- fold increase in the level of all four deoxyribonucleoside triphos- tients undergoing treatmenl wilh Ihe drug (13). Finally, ara-A is phates. The triphosphate pools in the mutants were resistant to capable of killing nondividing lymphocytes (3). the inhibition produced in wild-type cells by addition of deoxy Sludies of molecular mechanisms involved in cytotoxicity of adenosine or thymidine, although significant activation in the ara-A and its related aclive analogue have been impeded by the deoxyguanosine triphosphate pool was obtained by higher con paucily of drug-resistant mammalian cell mutante wilh specific centrations of thymidine. An examination of ribonucleotide re- enzyme defecls. ara-A-resistant mouse cells have been reported ductase in extracts of two of the mutants revealed a specific to contain a resislanl form of DNA polymerase (18). In baby alteration in the normal sensitivity of the enzyme for deoxyaden hamster kidney cells, resistance lo ara-A has been linked lo 3 osine triphosphate and adenosine triphosphate but not 9-/3-D- differenl mechanisms: (a) the loss of activily of a nucleoside arabinofuranosyladenine 5'-triphosphate. When the level of ri kinase necessary for fhe firsl step of ara-A phosphorylation (6); bonucleotide reducíaseactivity was measured, it was found that (o) a mutalion affecfing Ihe allosleric regulation of ribonucleotide the ara-A-resistant cells contained approximately twice the wild- reducíase activily; and (c) increased S-adenosylhomocysleine type level of cytidine diphosphate reducíase activity at physio hydrolase activity (5, 6). The latter cells have also shown the logical adenosine triphosphate level. This combination of in inleresling capacily for an increased rale of mulalion to 6- creased enzyme activily and alteration in sensitivily lo Ihe nu Ihioguanine resistance. cleoside triphosphates could accounl for both the changes in In an effort to idenfify Ihe inlracellular site(s) of ara-A sensitivity deoxyribonucleotide pool sizes and the resistant phenotype of in human cells, we isolated and characterized variants with Ihe presumed mutants. increased resistance to Ihe anliproliferative effect of the drug from Ihe established leukemic T-lymphoblast cell line CCRF- INTRODUCTION CEM. These variant cell lines showed cross-resistance to dThd and dAdo, normal sensitivily to ara-C and adenosine, and in ara-A2 (vidarabine) has antitumor and antiviral aclivilies (4,26, creased sensitivily to the etoposide VP16-213. Overproduction 31 ). In combination with dCF (Pentostatin), a lighl-binding inhib of deoxyribonucleotides due lo altered ribonucleotide reducíase itor of adenosine deaminase ara-A is undergoing clinical trials for aclivity may be responsible for the complex phenotype of these Ireatment of lymphoproliferative cancers (13). Despite extensive cells. biochemical studies, however, the cytotoxic mechanism of Ihe

1Supported by NIH Research Grant CA 33017, National Cancer Institute Cancer MATERIALS AND METHODS Center Core Grant CA 21765, and American Lebanese Syrian Associated Charities. 2The abbreviations used are: ara-A, 9-/3-D-arabinofuranosyladenine; dCF, 2- Materials. Free nucleosides, nucleotides, and nucleoside analogues deoxycoformycin; dNTP, deoxyribonucleoside triphosphate; HPLC, high-pressure (other than ara-A) were purchased from Sigma Chemical Co. (St. Louis, liquid chromatography; ara-C, 1-/3-o-arabinofuranosylcytosine; ara-ATP, 9-/3-o-ara- MO). ara-A and dCF were from the Drug Development Branch of the binofuranosyladenine 5'-triphosphate; \CX, 50% inhibitory concentration; dAdo, National Cancer Institute (Bethesda, MD). VP16-213 was provided by deoxyadenosine; dThd, deoxythymidine; ara-CTP, 1-/3-rj-arabinofuranosylcytosine 5'-triphosphate. Sandoz Pharmaceutical, Basel, Switzerland. Received October 19,1983; accepted June 21,1984. Mutant Selection. The technique for cloning cells in methylcellulose

4328 CANCER RESEARCH VOL. 44

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1984 American Association for Cancer Research. ara-A Resistance in Mutant T-Lymphoblasts medium has been described in detail (10). A tetraploid variant of the RESULTS human leukemic lymphoblast line CCRF-CEM, hereafter termed CEM, was grown in Eagle's medium modified for suspension culture and supplemented with L-glutamine and 10% heat-inactivated newborn calf Selection of ara-A-resistant Lines. The IC5oof ara-A (in the serum. Mass cultures maintained at a density between 4 x 105 cells/ml presence of 4 ^M dCF) was 0.6 MM. ara-A-resistant cells were and 1 x 106 cells/ml and subcultured every 2 days were exposed to isolated by culturing CEM cells in medium containing either 1 or either 1 or 2.5 ^M ara-A with 4 MM dCF for 6 to 8 weeks. Cells that 2.5 tiM ara-A in the presence of 4 /IM dCF. Despite repeated survived were cloned in methylcellulose medium containing the same attempts, single-concentration selections with 5 or 10 /¿Mara-A drug concentrations. After 7 to 10 days, clones that appeared on the failed to provide any resistant colonies. Table 1 presents a plates were isolated and propagated in liquid culture. summary of the ICso values of various drugs for wild-type cells Growth Experiments. For rapid determination of the growth-inhibitory and 4 representatives resistant clones selected at the 2 different effects of various drugs on the resistant clones isolated in this study, 2 drug concentrations. Two clones selected with 1 UM ara-A (des x 105 cells/ml in complete growth medium were plated into 50-ml Costar ignated ara-1b and ara-1c) required 6 to 7 times more ara-A for flasks in duplicate with the appropriate drug concentrations. The flasks were incubated at 37° under 5% CO2 for 48 hr; during this time, the 50% growth inhibition than did the wild-type CEM cells. Two control doubled twice. The cell number in each flask was determined clones selected with 2.5 UM ara-A (designated ara-2.5b and ara- with a Model B Coulter Counter. 2.5c) had ara-A ICsovalues 4 and 11 times, respectively, higher Measurements of Intracellular Levels of Nucleoside Triphos- than that for wild-type cells. These resistant clones maintained phates. The CEM cells and their variants were grown in 200-ml flasks a stable phenotype during growth in nonselective medium for in complete medium without drug for about 1 week. Flasks containing 5 over 6 months, indicating that they had undergone a permanent x 105 per ml exponentially growing cells were then incubated at 37°for alteration. All 4 mutants showed cross-resistance to dAdo and 4 hr without drug (controls), with 0.5 or 2.0 rriM dAdo (in the presence dThd but normal sensitivity to ara-C and adenosine (Table 1), of 4 MM dCF), 2.0 mw dThd, or 1 or 10 ^M [3H]ara-A with 4 MM dCF. suggesting that these cells had not undergone a generalized Cells were harvested by centrifugation at room temperature, washed once with chilled isotonic phosphate-buffered saline, extracted with 0.5 transport defect for nucleosides. We then tested wild-type cells and 2 of the mutant clones, M HCICvO.1 M KPO«,and neutralized with 1 M KOH. Ribonucleotides were destroyed by periodation as described by Garrett and Santi (11 ), ara-1b and ara-2.5c, for sensitivity to podophyllotoxin, a micro- and then dNTPs or ara-ATP was separated by HPLC using a laboratory tubule inhibitor, and its epipodophyllotoxin derivative VP16-213. data control HPLC system with a Whatman Partisil 10-SAX column. The As shown in Table 2, the ara-A mutants showed greater sensi dNTPs were eluted isocratically with 0.4 M ammonium phosphate and tivity to VP16-213 than did wild-type cells but only normal 1.5% acetonitrile (pH 3.0) at a flow rate of 2.0 ml/min and were monitored sensitivity to podophyllotoxin. This characteristic of the ara-A at 254 and 280 nm. Compounds were identified by comparing the mutants also proved stable when the cells were cultivated in retention times of unknown peaks with those of standards, and concen drug-free medium. We also tested the sensitivity of 2 other CEM trations of triphosphates were determined by the peak area method. cell mutants, CEM/ara-C and CEM/PZ, isolated previously for Ribonucleotide Reducíase Assay. Exponentially growing cells were their reduced sensitivity to ara-C and pyrazofurin, respectively. collected by centrifugation, washed in phosphate-buffered saline, and resuspended in 50 mw Tris-HCI (pH 7.4J-100 mw KCI. The suspension These mutants, which are defective in their deoxycytidine kinase was lysed by ultrasonification with a single 10-sec burst on a Branson and adenosine kinase activities, respectively, showed no unusual Model LB-75 soniter. The lysate was centrifuged at 10,000 x g for 30 sensitivity to either VP16-213 or podophyllotoxin (Table 2). min at 4°,and the supernatant was precipitated with streptomycin sulfate Characterization of ara-A Resistance. To examine whether (6 mg/ml). Ribonucleotide reducíase was precipitated from the super the resistance of our CEM cell mutants to ara-A could be natant by adding ammonium sulfate to 40% saturation, dissolved in 50 attributed to defective activation of ara-A to its nucleotide deriv ITIMTris-HCI (pH 7.4) containing 1 rriM dithiothreitol, and dialyzed over atives, we measured the activities of deoxycytidine kinase and night against 2 changes of the same buffer. Reductase activity was adenosine kinase, the enzymes in CEM cells involved in ara-A determined in a mixture consisting of 50 ITIM4-(2-hydroxyethyl)-1-piper- phosphorylation (29). The results shown in Table 3 indicate that azineethanesulfonic acid (pH 7.4), 3 mw ATP, 10 rtiM MgCI2, 6 mw dithiothreitol, 0.4 ITIM [14C]CDP (5000 cpm/nmol), and enzyme extract the activity of adenosine kinase was decreased about 50% in the mutant line ara-2.5c, but no changes in either activity relative containing about 1 mg of protein in a total volume of 0.15 ml. The reaction was allowed to proceed at 37°for 60 min and was stopped by to wild-type cells were observed in the other 3 resistant lines. boiling for 2 min. The deoxycytidine formed was measured as described Experiments were also carried out to determine whether the by Steeper and Steuart (27). formation of ara-A nucleotides in the mutants differed from wild

Table 1 Properties of CEM and ara-A-resistant cells The procedures for growing CEM and mutant cells in the presence of various concentrations of nucleosides are described in "Materials and Methods." Growth rate experiments began with 2 x 10s cells. After 48 hr, cells were counted, and 1C»values were determined for each cell line; values given are the average of at least 2 experiments done in duplicate. Relative resistance is expressed as the ratio of the ICsovalue for the resistant cells to that for wild-type cells. AdenosineCell

lineCEM (MM)3536resistance1.0 (MM)0.6 resistance1.0 resistance1.0 resistance1.0 resistance1.0

ara-1b 1.0 4.2 7.0 11 9.2 0.02 0.5 75 7.5 ara-1c 28 0.8 3.5 5.8 6 5.0 0.03 0.75 150 15.0 ara-2.5b 4135Relative 1.2 2.87Relative 4.6 1118Relative 9.2 0.03 0.75 110 11.0 ara-2.5cICso 1.0ara-A1C» 11.6dAdoICso(MM)1.2 15.0ara-CICso(MM)0.040.06Relative 1.5dThdICso(MM)10100Relative 10.0

OCTOBER 1984 4329

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1984 American Association for Cancer Research. A. Friulana

Table 2 addition of exogenous dThd decreased the intracellular levels of Effect of podophyllotoxin and its epipodophyllotoxin analogue VP16-213on the dCTP in the wild-type cells, but not in the resistant lines (Table growth rate of OEMand ara-A-resistantcells 5). Intracellular levels of dGTP and dATP increased similarly in 1C»valueswere determined as described in Table 1. Values are the average of 2 determinations per clone done in duplicate. the wild-type and mutant lines, except for the dATP level in mutant line ara-2.5c, which was not affected. PodophyllotoxinCell Since the data on dNTP pool sizes suggested that the lesion lineWild resistance1.0 (HM)600210 resistance1.00 was probably at the level of ribonucleotide reducíase, we as sayed this enzyme activity in wild-type cells and the mutants type ara-1b 22 1.2 0.35 ara-1b and ara-2.5c (Chart 1, A and B). In the wild-type cells, 9 ara-2.5c 17 0.9 140 0.23 UMdATP inhibited the reduction of CDP to dCDP by about 50%, CEM/ara-C 23 1.3 580 0.96 CEM/PZ(On(DM)1819Relative 1.1VP16-2131C»660Relative 1.10 whereas 100 nM dATP was required to inhibit the enzyme activity of the 2 mutants by 50% (Chart 1/1). Moreover, sensitivity of the reducíase to stimulation by ATP was different in the resistant Tabled Nucleosidekinase activities in OEMand ara-A-resistantcells cells. In the wild type, CDP reducíase activity was maximal in Supematants of homogenized cells were assayed for kinase activities as de the presence of 3 mM ATP, whereas stimulalion of Ihe enzyme scribed previously (32). The units of specific activity are nmol of nucleoside activity in both ara-1b and ara-2.5c was maximal at approxi monophosphateformed per mg of protein per 30 min. Values are the averageof 2 mately 0.8 mM ATP (Chart 10). The apparent affinity constanl determinations for each cell line. (K™)for CDP reduction was Ihe same (35 to 48 JIM) in the wild protein)Ce« Specific activity (units/mg type and ara-1b and ara-2.5c (data not shown). lineCEM Udinekinase6.7 kinase190 Since Ihe enzyme in Ihe mulanl lines has altered sensilivily lo

ara-1b 5.5 182 Tables ara-1c 165 8.2 dNTPpools in ara-A-resistantcells after incubation with dAdo and dThd ara-2.5b 6.1 185 ara-2.5cDeoxycy 5.9Adenosme 69 Wild-type and mutant cells were incubated in 50 ml of complete medium at a concentration of 5 x 106cells/ml for 4 hr with no further additions or with 0.5 mM dAdo plus 4 MMdCF, 2 mM dAdo plus 4 MMdCF, or 2 mMdThd. After incubation, cells were harvested, and dNTPs were quantitated by HPLC as described in Table 4 "Materials and Methods." The results represent a typical experiment repeated Accumulationof ara-ATPin wild-type and mutant cell lines The values in ara-ATP were established under the experimental conditions twice. described under "Materials and Methods." Values are the average of 2 determina dNTP pools (nmol/10* cells) tions for each cell line. CelllineWild ara-ATP levels (nmol/10' cells) at ara-A concentra ofCell tions typeara-1bara-2.5bara-2.5cAdditionNonedAdo (O.SfdAdo lineWild MM1.0 MM18.6 (2)dThd (2)NonedAdo type ara-1b 2.8 31.0 ara-1c 1.8 29.2 (0.5)dAdo ara-2.5b 2.2 28.8 ara-2.5c1 (2)dThd 1.710 17.2 (2)NonedAdo type. To test this possibility, wild-type and mutant lines were (0.5)dAdo incubated with 1 and 10 U.Mara-A in the presence of 4 UMdCF, (2)dThd the soluble pools were extracted, and the levels of ara-A nucleo- (2)NonedAdo tides were analyzed by HPLC. As can be seen in Table 4, the levels of ara-ATP accumulation after 4-hr incubation were in (0.5)dAdo (2)dThd creased in the 4 resistant cell lines, relative to the wild type, (2)dCTP9534242325193736283526282737dTTP4715106088816114210581631091131492136144141919dATP243116107763108358474605797332565280309206613961194dGTP209911660521439143322833110995102300 indicating that ara-A resistance in the mutants was not due to a 3 Numbers in parentheses,concentration (mM)of addition. defective phosphorylating of ara-A. We also compared the intracellular levels of dNTP pools in wild-type cells and the mutants ara-1b, ara-2.5b, and ara-2.5c (Table 5). Under normal culture conditions, the levels of all 4 dNTPs were appreciably higher in the ara-A mutants than in wild- type cells. The addition of 0.5 or 2 mM dAdo to wild-type cells resulted in increases of dATP levels from 24 to about 3000 nmol/ 109 cells; pool sizes of the other 3 dNTPs were reduced from 50 to 36% of control levels. Addition of 2 rnw exogenous dAdo to the mutant lines ara-1b and ara-2.5b likewise resulted in large increases of intracellular dATP and concomitant decreases in O IO 20 30 40 50 100 O dGTP, but the levels of dCTP and dTTP did not decrease as dATP,uM ATP, mM they did in the wild-type cells. In the mutant line ara-2.5c, there Chart 1. CDP reducíaseactivityat different dATP and ATP concentrations in was a less dramatic increase in dATP with the addition of 2 mw wild-type (•)andara-A-resistant mutants (ara-1b, G; ara-2.5c, O). A. inhibition by dATP; B, activation by ATP. Enzyme activity was determined as described under exogenous dAdo, and none of the other dNTP levels was af "Materials and Methods." The results are the average of 2 separate experiments fected. The accumulation of intracellular dTTP resulting from the done in duplicate.

4330 CANCER RESEARCH VOL. 44

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1984 American Association for Cancer Research. ara-A Resistance in Mutant T-Lymphoblasts

15r synlhesis in Ihe mulanls as il did in Ihe wild lype. II seems likely lhat this lessening of pool size changes is responsible, at least in part, for the resistant phenotype of Ihese mulanls. There is considerable evidence from Iransplantable tumors and from cultured cells indicaling thai ara-ATP accumulation is primarily responsible for ihe growth inhibilion and cyloloxicily of ara-A (4, 9, 10, 17, 23). However, still unclear is whelher ribo- nucleolide reducíaseis a largel for ihe killing action of this agent. Generally, DNA polymerase in mammalian cells is much more sensitive to inhibition by ara-ATP than ribonucleotide reducíase (4, 9, 30). Chang and Cheng (8) demonslraled ara-ATP to act as a competilive inhibitor, with respect to ATP, of ribonucleotide reductase isolated from the human Molt 4 T-lymphoblast line and determined Ihe K for ara-ATP to be aboul 15 limes higher 200 0 200 lhan that for DNA polymerase-« («1^M). However, the intracel- AraATP, lular ara-ATP concentrattion of human lymphoblasloid cells Chart2. Ribonudeotide reducíase activity of wild type (circles), ara-1b (squares), and ara-2.5c (triangles) in the presence of varying ara-ATP concentra treated wilh ara-A can exceed 200 ^M (22) and, under these tions. The enzyme activity was determined at 3 rriM (closed symbols) and 0.5 mu conditions, both DNA polymerase and ribonucleotide reducíase (open symbols) ATP, respectively. Each point is the slope determined from a linear time course of enzyme activity. should be inhibited. The isolation of varianls from human leukemia cells with an both dATP and ATP, we compared the sensitivity of CDP reduc alteration in ribonucleotide reducíase activity has allowed us lo íasein ara-1 b, ara-2.5c, and wild type to ara-ATP. As shown by ask questions about the role of Ihis enzyme in Ihe cytotoxicity the results depicted in Chart 2, the ICso of ara-ATP («166 MM) of ara-A. For example, Ihe ribonucleolide reducíase in our 2 mutants, ara-1 b and ara-2.5c, is refractory to complete inhibition appeared to be similar in mutant extracts and wild type. However, as shown by the data, the extracts from the mutant lines had by dATP, a general allosteric inhibitor of the reduclion of all 4 roughly twice the enzyme activity per cell as did extracts for wild ribonucleolides lo Iheir corresponding deoxyribonucleolides (28), type in the presence of 3 mw ATP and about 5-fold more with and simultaneously is 3-fold more sensilive lo aclivalion by ATP 0.5 DIM ATP added to the incubation. Previous studies in CEM than the parental cells. However, in vitro examination of Ihe enzyme from bolh mutanl and parental cells showed no detecl- cells have shown that the physiological level of ATP in these cells is approximately 2.4 mu.3 Thus, the ara-A resistance in able difference in sensilivily lo ara-ATP inhibilion (Chart 2). ara- these mutants appeared to be associated with the presence of A phosphorylalion to the respective Iriphosphale, Ihe presumed increased level of ribonucleotide reducíaseactivity at the phys active metabolite, was also determined in mutant lines and parental cells. The cellular concentration of ara-ATP formed iological ATP concentration. varied from 1 to 3 nmol/109 cells and 18 to 30 nmol/109 cells in cultures incubated with 1 and 10 MMexogenous ara-A, respec DISCUSSION tively. Based on a mean CEM cell volume of about 1100 eu pu Resistance to ara-A by our CCRF-CEM mutants appears to (22), this would correspond lo a cellular ara-ATP concentration result from increased dNTP pool sizes and altered ribonucleotide of approximately 2 and 30 UM, respectively. Assuming uniform intracellular dislribulion of nucleotides, these levels of ara-ATP reducíaseactivily. The finding lhal Ihese mutants showed normal sensitivily lo ara-C and resistance to excess concentrations of are 5- lo 80-fold lower than lhal required for inhibilion of ribo nucleotide reductase in either mutanl or wild-lype cells (Chart 2). dThd, however, suggesls lhal ihe molecular mechanism of this resistance is different from that reported previously in baby Taken logelher, iherefore, Ihese data suggest the lethal effect hamster kidney cells, which demonstraled cross-resistance to of ara-A in these T-lymphoblasls does not involve a direct action ara-C (6). of ara-ATP on ribonucleotide reductase. The increased dNTP pool sizes appeared to correlate reason Anolher possibility is lhal the alteration in ribonucleotide re ably well wilh increased resistance lo ara-A. For example, ara- ductase is mediated by an indirect action of ara-A on nucleotide 1b cells, with about 4 limes as much dATP as wild-type cells, pools. Thus, ara-A, once phosphorylaled to ara-ATP, can inhibil showed 7-fold increased resistance to ara-A, and mutanl ara- DNA polymerase reaction and cause accumulation of dNTPs 2.5c, wilh about 8 times as much dATP as wild-type cells, because of lack of incorporation into DNA. Since dNTPs may be exhibited almost 12-fold more resistance to ara-A. While dNTP toxic lo cells, a mutalion in ribonucleolide reductase which pools were not greatly elevated (2- lo 8-fold), Ihe mutants were lessens the effecls of high concenlrations of dATP on the sizes of the other dNTP pools would increase competition wilh ara-A generally refractory to the normal inhibitory effects that excess nucleosides have on wild-type cells. When pool sizes are ex nucleolides for binding wilh the targel enzyme (e.g., DNA polym pressed in nmol/109 cells, equivalenl exogenous concentrations erase) at its killing site and thus increase Ihe resistance of Ihe cells to ara-A. This, in turn, would render the cells cross-resistant of dAdo and dThd produced larger dATP and dTTP pools in the mutants than in wild-type cells (Table 4). Despite Ihese large to both dAdo and dThd cytotoxicily, since Ihe elaborate regula increases in dATP and dTTP levels, however, Ihe dCTP pool tory system of CDP reducíasewould be largely bypassed. size did noi decrease as il did in Ihe wild lype. Consequently, The resulls of olher sludies have also suggested that nucleo tide pool imbalances may be involved in the cytotoxicity of ara- Ihe supply of triphosphates did noi become limiting for DNA A in mammalian cells. North (21) has reported recently that 3A. Fridland. unpublished observations. following addition of ara-A, Ihe levels of dNTPs increased al leasl

OCTOBER 1984 4331

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1984 American Association for Cancer Research. A. Fridland 2- to 3-fold in HeLa cells. In addition, a great majority of mutants 7. Chang, C-C., Boezi, J. A., Warren, S. T., Sabourin,C. L. K., Uu, P. K., Glatzer, in mammalian cells isolated with DMA polymerase-a inhibitors, L., and Trosko, J. E. Isolation and characterization of a UV-sensitivehyper- mutable aphidicoim-resistantChinese hamster cell line. Somatic Cell Genet., such as aphidicolin and ara-C, have also been shown to have 7:235-253,1981. 8. Chang, C-H., and Cheng, Y-C. Effects of deoxyadenosineIriphosphale and 9- altered dNTP pool sizes, and the results were interpreted as 0-D-arabinofuranosyladenine-5'-lriphosphaleon human ribonucleotide reduc effects on nucleotide pool imbalances (2, 20, 24). íasefromMolt-4F cells and the concept of "self-polenlialion." Cancer Res., It is unclear why, despite increased levels of dCTP, our CEM 40: 3555-3558,1980. mutants exhibited normal sensitivity to ara-C cytotoxicity (Table 9. Cozzarelli, N. R. The mechanism of action of inhibitors of DNA synthesis. Annu. Rev. Biochem., 46: 641-668,1977. 1). Recently, Chang and coworkers (7,19) isolated an aphidicolin- 10. Dow, L. W., Bell, D. E., Poulakos,L., and Fridland,A. Differencesin metabolism resistant Chinese hamster fibroblast that contained a dCTP pool and cytotoxicily belween 9-j3-D-arabinofuranosyladenineand9-0-o-arabinofur- anosyl-2-fluoroadenine in human leukemic lymphoblasls. Cancer Res., 40: 4 times larger than wild type and, like our CEM mutants, that 1405-1410,1980. exhibited normal sensitivity to ara-C. It was demonstrated that, 11. Garrett, C., and Santi, D. V. A rapid and sensitive high pressure liquid in addition to a lesion in ribonucleotide reducíase, this mutant chromatography assay for deoxyribonucleoside triphosphate in cell extracls. Anal. Biochem., 99: 268-273,1979. contained an alteration in DMA polymerase-a resulting in 10-fold 12. Hershfield, M. S. Apparenl suicide inaclivalion of human lymphoblasl S- increased affinity for ara-CTP and dCTP but not their purine adenosyl-homocysteine hydrolase by 2'-deoxyadenosine and adenine arabi- counterparts (19). Resistance to ara-A similarly could lead to an noside. J. Bid. Chem., 254: 22-25,1979. 13. Hershfield, M. S., Kredich, N. M., Keller, C. A., Mitchell, B. S., Kurtzberg, J., alteration in the regulation of the enzymes involved in DNA Kinney, T. R., and Palletta, J. M. S-Adneosylhomocysleine calabolism and basis of acquired resislance during treatment of T-cell acute lymphoblastic synthesis. More complete and detailed studies will be necessary leukemia wilh 2'-deoxycoformycin alone and in combination wilh 9-/3-D-arabi- to investigate this possibility. nofuranosyladenine.Cancer Res., 43: 3451-3458,1983. The CEM mutants also snowed another interesting character 14. Kalwinsky, D. K., Look, A. T., Ducore, J., and Fridland,A. Effecls of eloposide istic in having an increased sensitivity to the etoposide VP16- (VP16-213) on cell-cycle traverse, DNA synthesis, and DNA strand size in human lymphoblasts. Cancer Res., 43: 1592-1597,1983. 213. This sensitivity is probably not caused directly by increased 15. Kredich, N. M., and Hershfield,M. S. Perturbationsin S-adenosylhomocysteine dNTP pools in the cells, since addition of various concentrations and S-adenosylmelhioninemelabolism: effects of transmelhylation. Adv. En and combinations of deoxynucleosides to wild-type cells caused zyme Regul., 18:181-191,1980. 16. Kredich, N. M., and Martin, D. W., Jr. Role of S-adenosylhomocysteinein a concomitant increase in dNTPs without altering their sensitivity adenosine-mediatedtoxicily in cultured mouse T lymphoma cells. Cell, 72: to VP16-213 (data not shown). It is possible, however, that this 931-938,1977. 17. Kufe, D. W., Major, P. P., Munroe, D., Egan, M., and Herrick, D. Relationship hypersensitivity is related to the mechanism of ribonucleotide belween incorporation of 9-/3-D-arabinofuranosyladenineinL1210 DNA and reducíase activity. Recent studies have shown that VP16-213 cytotoxicily. Cancer Res., 43: 2000-2004,1983. has a radiomimetic effect on chromosomes and induces both 18. LePage, G. A. Resislance lo 9-/3-o-arabinofuranosyladenineinmurine tumor cells. Cancer Res., 38: 2314-2316,1978. single- and double-strand breaks in cellular DNA (14, 32). Fur 19. Uu, P. K., Chang, C-C., Trosko, J. E., Dube, D. K., Martin, G. M., and Loeb, thermore, free radicals may be involved in the activation of the L. A. Mammalianmutalor mulanl wilh an aphidicoim-resistantDNApolymerase drug in mammalian cells (32). Akerblom ef al. (1) have shown a. Proc. Nail. Acad. Sci. USA,80:797-801,1983. 20. Meulh, M. Deoxyribonucteotidepools in mouse fibroblasl cell lineswilh altered that hydroxyurea resistance in mouse fibroblasts is associated ribonucleotidereducíase.Eur.J. Biochem., 77: 39-43, 1976. with increased generation of free radical species required for 21. North, T. W. Effecls of 9-/3-o-arabinofuranosyladenineand 1-0-o-arabinofu- ranosylcylosine on levels of deoxyribonucleic acid precursors in uninfecled ribonucleotide reducíase activity. It is possible, therefore, that and herpes simplex virus-infectedcells. Biochem.Pharmacol.,32:3862-3864, the hypersensitivity of our CEM mutants is due to an increased 1983. activation of VP16-213 by overproduction of free radicals in the 22. Plunkett, W., Chubb, S., Alexander, L., and Monlgomery, J. A. Comparisonof Ihe loxicily and melabolism of 9-/3-D-arabinofuranosyl-2-fluoroadenineand9- ara-A-resistant cells. Regardless of the exact mechanism in /3-D-arabinofuranosyladenineinhuman lymphoblasloid cells. Cancer Res., 40: volved, combination chemotherapy of ara-A, or similarly acting 2349-2355,1980. analogue, with VP16-213 might be an effective way to reduce 23. Plunkett, W., and Cohen, S. S. Melabolism of 9-0-D-arabinofuranosyladenine by mouse fibroblasls. Cancer Res., 35: 415-422,1975. the incidence of resistant cells which would escape therapy with 24. Robert de Sainl Vincent B., and Buttin, G. Sludies on 1-/3-o-arabinofurano- either drug. sylcylosine-resislant mutants of Chinese hamster fibroblasts: joint resistance to arabinofuranosylcylosineand to excess Ihymidine. Somatic Cell Genet., 7: 67-82,1979. REFERENCES 25. Rose, K. M., Leonard, T. B., and Carter, T. H. Effects of adeninenucleosides on RNA synthesis in adenovirus-infectedcells. Mol. Pharmacol.,22: 517-523, 1. Akerblom, L., Ehrenberg, A., Graslund, A., Lankinen, H., Reichard, P., and 1982. Thelander,L. Overproduction of the free radical of ribonucleotidereducíase¡n 26. Schabel, F. M. The antiviral aclivily of 9-0-D-arabinofuranosyladenine(ara-A). hydroxyurea-resistant mouse fibroblast 3T6 cells. Proc. Nati. Acad. Sci. USA, Chemolherapy, 73: 321-338, 1968. 78:2159-2163,1981. 27. Steeper, J. J., and Steuart, C. D. A rapid assay for CDP reducíaseactivityin 2. Ayusawa, D., Iwata, K., and Seno, T. Alteration of ribonucleotidereducíasein mammaliancell extracts. Anal. Biochem., 34: 123-130,1970. aphidicoim-resistantmutants of mouse FM3A cells with associated resistance 28. Thelander, L., and Reichard, P. Reduction of ribonucleotides. Annu. Rev. to arabinosyladenineand arabinosyfcytosine.Somatic Cell Genet., 7: 27-42, Biochem., 48: 133-158,1979. 1981. 29. Verhoef, V. L., Sarup, J., and Fridland, A. Activation of 9-ß-o-arabinofurano- 3. Carson, D. A., Wasson, D. B., Lakow, E., and Kamatani,N. Possiblemelabolic syladenine in human leukemic cells: identification with mutants deficient in basis for the different immunodeficientslates associatedwith genetic deficien nucleosidekinases. Cancer Res., 47: 4478-4483,1981. cies of adenosine deamlnase and purine nucleoside. Proc. Nati. Acad. Sci. 30. White, E. L., Shaddix, S. C., Brockman, R. W., and Bennett, L. L. Comparison USA, 79: 3848-3852, 1982. of the activity of 9-0-D-arabinofuranosyl-2-fluoroadenineand9-0-D-arabinofu- 4. Cass, C. E. 9-/3-D-Arabinofuranosyladenine(ara-A).In: F. E. Hahn (ed.), Anti ranosyladenineon target enzymes from mouse tumor cells. Cancer Res., 42: biotics, Mechanismof Action of Antieukaryotic and Antiviral Compounds, Vol. 2260-2265,1982. 5, pp. 85-109. Berlin: Springer-Verlag, 1979. 31. Whilley, R. J., Soong, S. J., Dohn, R., Galasso,G. J., Ch'ien, L. T., and Alford, 5. Chan, V. L., Guttman, S., and Juranka, P. Mutalor genes of baby hamster C. A. Adenine arabinosideIherapy of biopsy-proved herpes simplex encepha kidney cells. Mol. Cell. Biol., 1: 568-571 , 1981. litis. N. Engl. J. Med., 297: 289-294, 1977. 6. Chan, V. L., and Juranka, P. Isolation and preliminarycharacteristics of 9-0-0- 32. Wozniak, A. J., and Ross, W. E. DNA damage as a basis for 4'-demethylepi- arabinofuranosyladenine-resistantmutants of baby hamster kidney cells. So podophylloloxin-9-(4,6-O-elhylidene-/3-D-glucopyranoside)(eloposide)cytotox matic Cell Genet., 7: 147-160, 1981. icity. Cancer Res., 43:120-124, 1983.

4332 CANCER RESEARCH VOL. 44

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1984 American Association for Cancer Research. Selection of 9-β-d-Arabinofuranosyladenine-resistant Human T-Lymphoblasts with Altered Ribonucleotide Reductase Activity

Arnold Fridland

Cancer Res 1984;44:4328-4332.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/44/10/4328

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/44/10/4328. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1984 American Association for Cancer Research.