[CANC ER RESEARCH 50. 108-112. January 1, 1990] On the Paradoxically Concentration-dependent Metabolism of 6-Mercaptopurine in WEHI-3b Murine Leukemia Cells1 Jan Liliemark,2 Birgitta Pettersson, Britt Engberg, Pierre I.afolie. MichèleMasquelier, and Curt Peterson Department of Medicine, Karolinska Institute at Huddinge Hospital, S-141 86 Huddinge ¡J.L., B. E.J, and Department of Clinical Pharmacology, Karolinska Hospital, S-104 01 Stockholm ¡B.P.. P. L.. A/. A/.. C. PJ, Sweden ABSTRACT The aim of the present investigation was to delineate the cellular metabolism of 6-MP in vitro as a tool to optimize the The intracellular metabolism of 6-mercaptopurine (6-MP) was studied in a murine leukemia cell line, \VEHI-3b. Cells were incubated 3 to 24 h design of future clinical studies. with 10 nM to 50 MM6-MP. Nucleotides were extracted with perchloric acid, and the 6-thiopurine nucleotides were isolated on mercurial cellu MATERIALS AND METHODS lose. The endogenous ribonucleotides in the perchloric acid extracts as well as 6-thiopurinc nucleotides were separated and quantified with anión Chemicals. 6-[14C]Mercaptopurine, 55 Ci/mol. was purchased from exchange high-performance liquid chromatography. The concentration of CEA, Siren, France, and elemental sulfur-35, 30 Ci/mol, from Amer- 6-thioinosinate (6-TIMP) and 6-thioxantinate (6-TXMP) increased with sham. 6-TIMP, 6-thioinosine 5'-diphosphate, 6-thioinosine S'-triphos an increasing 6-MP dose. The concentration of the 6-thioguanosine phate, 6-TXMP, and 6-TGN were synthesized by incubation of eryth- nucleotides (6-TGN) increased with 6-MP concentrations between 10 ii\i rocytes with 6-MP, 6-thioxanthinc, and 6-thioguanine, respectively, as and I MM.However, further increase in 6-MP concentration led to a previously described (6). 6-[i5S]Mercaptopurine was synthesized as decrease in the formation of 6-TGN. At 50 MM6-MP, the concentration described by Morávek(7). From the crude preparation, [-"SJ6-MP was of 6-thioguanosine S'-triphosphate was one fifth of that seen at 1 JJM. separated from labeled sulfur by HPLC to 98% radiochemical purity. The incorporation of 6-|'5S|mercaptopurine into DNA was also slightly The fractions containing [J5S]6-MP were collected, lyophilized, recon higher at 1 MMcompared with SOMM.The cytocidal effect on clonogenic stituted in methanol, and stored at -70°C.Tiazofurin was generously cells was one log greater at l /IM 6-MP compared with 50 u\t 6-MP. provided by Dr. Ven Narayanan, National Cancer Institute, Bethesda, The decrease of 6-TGN was accompanied not only by an increased 6- MD. All other chemicals were purchased from Sigma. I IMP concentration but also by an inhibition of the purine tie novo Cell Lines. WEHI-3Ã’, a murine leukemia cell line with myelomono- synthesis and consequently by a decrease of the cellular ATP concentra cytic features, was kindly provided by Dr. Malcolm A. S. Moore tion. The ATP concentration in the cells treated with 1 ^M 6-MP could (Memorial Sloan-Kettering Cancer Center, New York, NY). Cells were be reduced to the level seen in cells treated with 50 MM6-MP by regularly monitored for Mycoplasma infection. Cells were cultured in simultaneous incubation with 0.3 MMantimycin A. This decrease of ATP RPMI-1640 and supplemented with 10% fetal calf serum at 37°Cand concentration was accompanied by a reduction of 6-TGN and to a lesser extent of 6-TXMP. These experiments suggest that the "self-limiting" 5% CO2. Penicillin G (50 IU/ml) and streptomycin (50 Mg/ml) were added to prevent bacterial and fungal infections. Cells were recultivated phenomenon in the metabolism of 6-MP might be caused by a depletion twice a week and found to grow exponentially between 5 x 10*and 1.5 of ATP by inhibition of purine de novo synthesis presumably by 6-TIMP. x IO6 cells/ml. All experiments were performed with IO8 cells at a concentration of IO6cells/ml except where otherwise stated. All results are derived from duplicate experiments unless otherwise stated. INTRODUCTION Colony Formation in Soft Agar. Cells (IO1 to IO5)were plated in 1 ml 6-MP3 has been used in the treatment of acute leukemias of McCoy's Medium A5 in 0.3% agar and 15% fetal calf serum. The feeder layer (1 ml) consisted of McCoy's Medium A5 in 0.5% agar. since the 1950s. Today it is a cornerstone in the maintenance treatment of acute lymphoblastic leukemia. Despite 30 years of The number of colonies (>40 cells) was counted in triplicate plates on clinical use, the knowledge of its metabolism and mechanism Day 8. Plates with 20 to 500 colonies were accepted. The plating efficiency in the controls was approximately 40%. of action is sparse. The importance of the intracellular conver Determination of the 6-MP Concentration in Culture Medium. One- sion of 6-MP to phosphorylated metabolites has been empha mi aliquots of culture medium were collected after centrifugaron (550 sized (1). It has also been shown that the incorporation of 6- x g, 5 min) before and after the incubations. 6-MP was determined MP into DNA as 6-thioguanine is one but not the only prereq with HPLC as previously described (8). uisite for its lethal action (2). In 1983 it was reported that there Determination of Endogenous Nucleotides and Intracellular 6-MP is a paradoxical relationship between the 6-MP concentration Metabolites. The method used was modified from that published by and the lethal effect on clonogenic cells in vitro (3). The cyto Lavi and Holcenberg (9). After incubation, cells were cooled in ice cidal effect was in fact lower with a higher concentration of 6- water and washed once with PBS. One aliquot was removed for cell MP. There seems to be a lack of systematical studies of the counting and the rest was centrifuged, 550 x g, 5 min. PCA, at a final metabolism of 6-MP as a function of dose. Furthermore, the concentration of 0.4 Mwas added to the cell pellet, and the supernatant formation of intracellular metabolites of 6-MP in leukemic cells was collected after centrifugation. The pellet was washed with PCA other than 6-TIMP, 6-TXMP, or 6-TGMP has been mentioned (0.8 M) and the supernatants were pooled and neutralized with KOH/ NH4H2PO4, 1.2/0.5 M. Two mg of mercurial cellulose (10) and 200 only briefly (4, 5). pmol of 4-thiouridine 5'-monophosphate were added to the PCA extract, vortexed for 30 s, and centrifuged at 1000 x g, 5 min. The Received 5/24/89: revised 9/25/89; accepted 10/3/89. supernatant was stored at -20°C until the analysis of endogenous The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in nucleotides. The pellet was washed 3 times with PBS. After addition accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1Supported by grants from the Swedish Cancer Society, the Cancer Society in of 300 u\ of mercaptoethanol (20 mM) and centrifugation, the super natant was collected, and 100 M'were injected on to the HPLC column Stockholm, the Swedish Medical Research Council (Project No. 08640), the Swedish Medical Society, and the Swedish Child Cancer Foundation. with a CMA/240 autoinjector (Carnegie Medicine, Inc., Stockholm, - To whom requests for reprints should be addressed. Sweden). The nucleotides were separated on a Whatman Partisil-10 1The abbreviations used are: 6-MP, 6-mercaptopurine; 6-TIMP. 6-lhioinosi- SAX aniónexchange column. At a flow rate of 3 ml/min, a gradient natc; 6-TXMP. 6-thio.\antinate; 6-TGN. 6-thioguanosine nucleotides; HPLC. high-performance liquid chromatography; PBS, phosphate-buffered saline; 6- was run between 100% KCI/K2HPO4, 0.01/0.005 M, pH 6, and 100% TGMP, 6-thioguanylate: 6-TGDP, 6-thioguanosine 5'-diphosphate; 6-TGTP, 6- KCI/K2HPO4, 0.5/0.25 M, pH 5.5, over 20 min following Curve 2 on thioguanosine 5' Iriphosphatc. a Milton-Roy CM4000 solvent delivery system, followed by 10-min 108 Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 1990 American Association for Cancer Research. THE PARADOXICAL METABOLISM OF 6-MP isocratic elution with 100% of the latter buffer. The metabolites were Baseline separation was achieved among all the 6-thioribonu- quantified by UV absorbance at 340 nm with a Milton-Roy SM 3100 cleotide metabolites. 6-Thioinosine 5'-di- and triphosphate absorbance detector. The peak areas were integrated with a Macintosh could also be separated from the 6-TGN, 6-TIMP, and 6- SE computer with the ChroMac 354 chromatography software con TXMP. The 5'-di- and triphosphate 6-thioinosine were, how nected with a DCU interface (Drew Scientific, Ltd.. London, United ever, not found in any sample. The intersample variation (SD) Kingdom). The response factors were determined by comparing the was less than 20%. The recovery of the internal standard 4- integrated area and the dpm of the metabolites in triplicate samples when cells were incubated with [UC]6-MP. thiouridine 5'-monophosphate was >90%. The limits of detec The PCA extracts were used for the HPLC assay of endogenous tion for the metabolites were 3 to 15 pmol at 340 nm. Due to nucleotides as previously described (11). The intracellular concentration the purity of the samples after the mercurial cellulose workup of endogenous nucleotides and 6-MP metabolites was determined by step, the detection limit in terms of concentration was depend dividing the amount of nucleotidc with the cell number x mean volume ent on the amount of cells used.