Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Differential Patterns of Intracellular Metabolism of 2′,3′-Didehydro-2′,3′-dideoxythymidine and 3′-Azido-2′,3′-dideoxythymidine, Two Potent Anti-human Immunodeficiency Virus Compounds

Differential Patterns of Intracellular Metabolism of 2′,3′-Didehydro-2′,3′-dideoxythymidine and... Vol. 264, No. 11, Issue of ’ April pp. 61274133,1989 15, THE JOURNAL OF BIOLOGICAL CHEMISTRY Printed in U. S.A. 0 1989 by The American Society for Biochemistry and Molecular Biology, Inc. Differential Patterns of Intracellular Metabolism of 2’,3’-Didehydro- 2’,3‘-dideoxythymidine and 3’-Azido-2’,3’-dideoxythymidine, Two Potent Anti-human Immunodeficiency Virus Compounds* (Received for publication, October 21, 1988) Jan Balzarini, Piet Herdewijn, and Erik De Clercq From the Rega Institute for Medical Research, Katholieke Universiteit Leuven, B-3000 Leuuen, Belgium 3’-Azido-2‘,3’-dideoxythymidine (AZT) and 2’,3’- didehydro-2’,3’-dideoxycytidine (ddeCyd, ddddcyd, D4C) (2- didehydro-2‘,3’-dideoxythymidine (D4T) are potent 4), 3‘-azido-2‘,3‘-dideoxythymidine (AzddThd, AZT) (5-8), and selective inhibitors of human immunodeficiency 2’,3’-didehydro-2’,3’-dideoxythymidine (ddeThd, ddddThd, virus replication in MT-4 and ATH8 cells. They are D4T) (3,9, lo), 3’-azido-2’,3’-dideoxyuridine (AzddUrd) (11, also inhibitory to the replication of murine retrovi- 12), 3’-fluoro-2’,3‘-dideoxythymidine (FddThd) (6, 12-14), ruses, i.e. Moloney murine sarcoma virus-induced and 3’-fluoro-2’,3’-dideoxyuridine (FddUrd) (12)) have been transformation of C3H cells. In MT-4 cells AZT is recognized as potent and selective inhibitors of the replication readily phosphorylated to its 5’-monophosphate, while of human immunodeficiency virus (HIV), the etiologic agent the 5‘-di- and 5“triphosphates are generated to a 200- of the acquired immunodeficiency syndrome (AIDS). The 600-fold lower extent than the 5’-monophosphate. 2‘,3‘-dideoxyribonucleoside analogues are assumed to be tar- D4T is phosphorylated in MT-4 cells to its 5’-mono- geted at the virus-encoded RNA-directed DNA polymerase phosphate at a 300-600-fold lower extent than AZT. (reverse transcriptase). To exert their antiretroviral activity, The phosphorylation of AZT in the thymidine kinase- the ddNs should be phosphorylated intracellularly to their 5’- deficient cell line (Raji/TK-) is severely depressed, triphosphate (ddNTP) metabolites. The phosphorylation while D4T phosphorylation is only slightly diminished steps are catalyzed by cellular enzymes, as has been shown in Raji/TK- as compared to Raji/O cells. D4T has a 10- for AZT in H9, ATH8, and Molt/4F cells (7,15,16), for ddCyd fold lower affinity for phosphorylation by crude MT-4 in ATH8 and Molt/4F cells (16-19), and for FddThd in cell extracts than AZT (IC,,,, 142 and 14 PM, respec- Ehrlich ascites carcinoma, MT-4, CEM, and H9 cells (13, 14, tively), and the Vmax for phosphorylation of D4T is only 5% that of AZT. D4T is phosphorylated by MT-4 cell 20). The nucleoside kinases (ie. dThd kinase and dCyd ki- extracts about 180-fold less efficiently than AZT nase) convert the ddNs (i.e. AZT and ddCyd, respectively) to (VmaJKm, 0.06 for D4T, as compared to 11 for AZT), their 5’-monophosphates. However, the affinity of the ddNs and this is consistent with the differences found in the for their respective nucleoside kinases may differ considerably amounts of phosphorylated products of D4T and AZT from one compound to another (i.e. K, value of AZT for H9 formed in intact MT-4 cells. The 5’-triphosphates of cell dThd kinase, 3 FM (15); K, value of FddThd for Ehrlich AZT and D4T are equipotent in their inhibitory effects ascites carcinoma cell dThd kinase, 11.2 PM (20); K,,, value of on the reverse transcriptases from human immunode- ddCyd for Molt/4F cell and KB cell dCyd kinase, 200 and 220 ficiency virus and Moloney murine leukemia virus. PM, respectively (7, 19)) and determines the eventual intra- cellular levels of the ddNTPs. In an attempt to gain a better insight in the metabolism and mechanism of antiretrovirus action of the ddNs, we Several pyrimidine 2’,3’-dideoxynucleoside analogues investigated the intracellular phosphorylation of two struc- (ddNs)’ (i.e. 2’,3’-dideoxycytidine (ddCyd, D2C) (1, 2), 2‘,3‘- turally related and potent anti-HIV compounds (AZT and D4T) in MT-4 cells. This cell line is exquisitely sensitive to * These investigations were supported in part by the AIDS Basic the inhibitory effects of 2‘,3‘-dideoxythymidine analogues Research Programme of the European Community and by grants (ie. AZT and D4T) on HIV replication and would therefore from the Belgian Fonds voor Geneeskundig Wetenschappelijk On- seem useful to investigate in parallel the intracellular metab- derzoek (Projects 3.0040.83 and 3.0097.87) and the Belgian Gecon- certeerde Onderzoeksacties (Project 85/90-79). The costs of publica- olism of AZT and D4T. D4T has not previously been studied tion of this article were defrayed in part by the payment of page for its metabolic fate within the cell. charges. This article must therefore be hereby marked “aduertise- rnent” in accordance with 18 U.S.C. Section 1734 solely to indicate MATERIALS AND METHODS this fact. ’ The abbreviations used are: ddNs, pyrimidine 2’,3’-dideoxynu- Cells-The origin and cultivation of MT-4, ATH8, and C3H cells cleoside analogues; ddCyd, D2C, 2’,3’-dideoxycytidine; ddeCyd, have been described previously (21-23). MT-4 cells were a gift from ddddCyd, D4C, 2‘,3‘-didehydro-2‘,3’-dideoxycytidine; AzddThd, Dr. N. Yamamoto (Yamaguchi University, Yamaguchi, Japan). Both AZT, 3’-azido-2’,3’-dideoxythymidine; ddeThd, ddddThd, D4T, MT-4 and ATH8 cells represent immortalized helper/inducer T cell 2’,3’-didehydro-2’,3’-dideoxythymidine; AzddUrd, 3’-azido-2’,3’-di- clones, obtained by cloning a normal T4 cell line in the presence of deoxyuridine; FddThd, 3’-fluoro-2’,3’-dideoxythymidine; FddUrd, tumor cells derived from a patient with adult T cell leukemia or 3’-fluoro-2’,3’-dideoxyuridine; HIV, human immunodeficiency virus; lethally irradiated HTLV-I-producing MJ tumor cells, respectively. MSV, Moloney murine sarcoma virus; MLV, Moloney murine leu- C3H cells represent a continuously growing murine embryo fibroblast kemia virus; HPLC, high performance liquid chromatography; AZT- cell line. Cultivation and characterization of the Fhji/O cell line and TP, AZT 5”triphosphate; D4T-TP, D4T 5”triphosphate; ddThd, its dThd kinase-deficient derivative (Raji/TK-) has been described 2’,3’-dideoxythymidine; AZT-MP, AZT 5’-monophosphate; AZT- previously (24). DP, AZT 5’diphosphate; D4T-MP, D4T 5’-monophosphate; D4T- Viruses-Human immunodeficiency virus (HIV) was obtained DP, D4T 5“diphosphate. from the culture supernatant of H9 cells persistently infected with 6127 6128 Phosphorylation of D4T and AZT in MT-4 Cells HTLV-IIIB (derived from a pool of American patients with AIDS) pared from exponentially growing MT-4 cells, which were first washed and kindly provided by Dr. R. C. Gallo (National Cancer Institute, twice with phosphate-buffered saline at 4 “C and then homogenized Bethesda, MD) (25). Moloney murine sarcoma virus (MSV) was by sonication. The suspensions were clarified by centrifugation at prepared from tumors obtained in 3-day-old NMRI mice that were 100,000 X g for 60 min. In the enzyme experiments, [methyL3H] inoculated intramuscularly with the virus (26). Moloney murine leu- dThd, [methyL3H]AZT, or [meth~l-~H]D4T served as the radiola- kemia virus (MLV) was from Electronucleonics (Bethesda, MD). beled substrates. Substrate concentrations were as follows: for Compounds-3‘-Azido-2‘,3’-dideoxythymidine (AZT) and 2’,3’- [methyL3H]dThd, 5, 10, 25, and 50 p~; for [meth~l-~H]AZT, 2, 5, 10, didehydro-2’,3’-dideoxythymidine (D4T) were synthesized according 20, and 40 p~; and for [methyL3H]D4T, 20,40,100,200, and 250 p~. to previously published methods (6). Their formulae are shown in The K,,, and V,, values for the radiolabeled substrates were derived Fig. 1. The 5”triphosphate derivative of AzddThd was synthesized from Lineweaver-Burk plots, using a linear regression analysis pro- as described earlier (27,28). The 5”triphosphate of D4T was prepared gram. In the competition assays, [meth~l-~H]AZT and [methyL3H] following essentially the same synthetic procedure. The other re- D4T were incubated in the presence of 10 p~ unlabeled dThd or agents used were of the highest quality available. dTTP. The other experimental conditions used for the dThd kinase Radio~hemicals-[methyl-~H]AZT (specific radioactivity, 10 Ci/ assay were as previously described (31). mmol) and [rnethyL3H]D4T (specific radioactivity, 20 Ci/mmol) were Inhibition of the HIV- or MLV-associated reverse transcriptase by obtained from Moravek Biochemicals Inc., Brea, CA. [methyL3H] the 5”triphosphates of AZT (AZT-TP) and D4T (DIT-TP) was dThd (specific radioactivity, 40 Ci/mmol) was from the Radiochem- determined as previously described (27). Inhibition of reverse tran- ical Centre Amersham (Amersham, United Kingdom). scriptase by the test compounds was estimated during the period that Antiretrouiral Assays-The procedures for measuring anti-HIV the reverse transcriptase activity increased linearly. In the MLV activity in ATH8 and MT-4 cells and anti-MSV activity in C3H cells reverse transcriptase experiments, the endogenous viral RNA served have been described previously (2, 3, 22, 29, 30). as the template. In the HIV reverse transcriptase assays, exogenous Cytostatic Assays-Cytostatic effects of the compounds were as- poly(A) .oligo(dT)lz-la served as the template. DNA polymerase a was sessed by measuring inhibition of cell proliferation. The experimental obtained from calf thymus (Pharmacia, Uppsala, Sweden). The re- procedures have been described previously (3, 12). Briefly, exponen- action mixture (40 pl) contained 4 pl of activated DNA (0.5 mg/ml), tially growing MT-4, CEM, and H9 cells were seeded in 200-p1 4 p1 of bovine serum albumin (5 mg/ml), 2 pl of a mixture of dGTP, microplate wells and incubated in the presence of varying concentra- dCTP, and dATP, 2 mM), 4 pl of [meth~l-~H]dTTP (4 pCi; 30 p~), tions of the test compound (50,000 cells/well). After an incubation 6 p1 of distilled water, 4 pl of a mixture containing 200 mM Tris-C1 period of 3,4, and 5 days the cell number was determined in a Coulter (pH 7.5), 30 mM MgClZ, and 2 mM dithiothreitol, 8 pl of the appro- counter (12). ATH8 cells were seeded at lo5 cells/culture tube (2 ml/ priate inhibitor concentrations (final concentrations: 500, 100, 20, 4 tube) in the presence of the test compound, and after 7 days of pM) and 12 pl of enzyme (0.024 unit). Determination of the Lipophilicity of Test Compounds-To estimate incubation the number of living cells was determined in a blood cell the lipid solubility of 2‘-deoxythymidine (dThd), 2’,3’-dideoxythy- counting chamber using the trypan blue dye exclusion technique (3). Metabolism of [methyL3H1AZT and [meth~l-~H]D4T in Human midine (ddThd), D4T, and AZT, the partition of the compounds MT-4 and Raji Cells-The metabolism of the radiolabeled compounds between 1-octanol and 10 mM potassium phosphate buffer, pH 7.4, was monitored according to previously established procedures (16, was measured as previously described (30). Briefly, a 50 pM concen- 17). Briefly, the MT-4 cells were seeded at 2-4 X 10’ cells/ml and tration of the test compound in 10 mM potassium phosphate buffer incubated with varying concentrations of the radiolabeled compounds was thoroughly mixed with an equal volume of 1-octanol for 30 min. (see footnote to Table 111). At different time intervals (see footnote Then, the mixture was further equilibrated at room temperature for to Table IV), cells were centrifuged, washed, and precipitated with 60 min, UV absorption was measured for the aqueous and alcoholic cold trichloroacetic acid (10%). After centrifugation, the supernatants liquid phases, and the percentage of the test compound present in each liquid phase was calculated based on their absorption maxima. were neutralized with tri-n-octylamine in Freon. HPLC analysis of Alternatively, the RF values of dThd, ddThd, D4T, and AZT were the neutralized cell extracts was carried out using a Partisil-SAX determined by thin layer chromatography on Silica Gel M5735 with radial compression column. A linear gradient of 0.007 M ammonium a mixture of ch1oroform:methanol (92.5:7.5). dihydrogen phosphate (pH 3.80) to 0.25 M ammonium dihydrogen + 0.50 M KC1 (pH 4.5) was used. The different fractions Tumor Formation in NMRI Mice Inoculated Intramuscularly with phosphate Moloney Murine Sarcoma Virus (MSV)-NMRI mice were inoculated of the eluate were assayed for radioactivity in a toluene-based scin- tillant. at 2 days after birth with MSV (day 0). AZT and D4T were admin- Phosphorylation of [methyL3H]D4T and [methyL3H]AZT was also istered daily by the intraperitoneal route at a dose of 125 or 25 mg/ examined in Raji/O cells and its dThd kinase-deficient Raji/TK- kg/day, starting 2 h before virus inoculation (day 0) and continued until day 5. Tumor appearance and mortality associated herewith counterpart, and H9 cells. The procedure was essentially the same as that described for the MT-4 cells. were recorded daily. There were 10 mice per group. Effect of AZT and D4T on the Intracellular Radiolabeled [methyl- 3H]dTMP, [meth~l-~HIdTDP, and [meth~l-~HIdTTP Pools in MT-4 RESULTS Cells-MT-4 cells were seeded at 400,000 cells/ml in 5-ml tissue Antiretroviral and Cytotoxic Effects of AZT and D4T“AZT culture bottles for 24 h in the presence of different concentrations of and D4T were evaluated for their antiretroviral effects in AZT or D4T (0, 50, 500 p~). Then, 0.04 p~ [methyL3H]dThd (10 human MT-4 and ATH8 cells infected with HIV and murine pCi/5-ml tissue culture bottle) was added to the cell cultures, and 2 h, the trichloroacetic acid-soluble intracellular material was after C3H cells infected with MSV (Table I). Both AZT and D4T analyzed by thin layer chromatography in a mixture of isopropyl proved to be potent and highly selective inhibitors of HIV alcohol:NH3:HzO (63:l) and by HPLC using the same separation replication in MT-4 cells. AZT showed an ED50 (50% effective method as described above. dose) of 0.005 PM, whereas D4T was 10-fold less active. Since Enzyme Assays-Cell extracts containing dThd kinase were pre- the CDS0 (50% cytotoxic dose) of AZT and D4T was 8.1 and 19 PM, respectively, AZT could be considered as 3-fold more selective as an anti-HIV agent than D4T in MT-4 cells. In 0 A ..JJ.. ATH8 cells, however, AZT and D4T were about 480- and 80- fold less effective against HIV than in MT-4 cells, while their cytotoxic effects were diminished only by a factor of 5. In murine C3H cells AZT was 100-fold more effective in inhib- iting MSV cell transformation than was D4T (EDso, 0.023 and 2.1 PM, respectively), while neither compound was toxic for C3H cells at 200 PM (Table I). When the cytostatic effects of AZT and D4T were evaluated against a number of human T cell lines (i.e. MT-4, H9, CEM, FIG. 1. Structural formulae of 3’-azido-2’,3‘-dideoxythy- ATHS), remarkable differences were noted (Table 11). AZT midine (AxddThd, AZT) and 2’,3’-didehydro-2’,3’-dideoxy- was considerably more inhibitory to the growth of MT-4 and thymidine (ddeThd, Dan. Phosphorylation of D4T and AZT in MT-4 Cells 6129 TABLE I Inhibitory effects of AZT and D4T on HIV-induced cytopathogenicity in human MT-4 and ATH8 cells and MSV-induced transformation of murine C3H cells MT-4 cells C3H cells Compound EDm’ EDmc MTCd PM PM PM 0.023 f 0.01 >200 AZT 0.005 f 0.001 8.1 f 1.8 >200 19 f 3.6 2.1 f 0.7 D4T 0.05 f 0.001 a 50% effective dose, or dose required to inhibit HIV-induced destruction of MT-4 or ATH8 cells by 50%. 50% cytotoxic dose, or dose required to inhibit MT-4 or ATH8 cell proliferation by 50%. 50% effective dose, or dose required to inhibit MSV-induced transformation of C3H cells by 50%. Minimal toxic concentration, or lowest dose required to cause a microscopical alteration of normal C3H cell morphology. e Data taken from Ref. 3. TABLE I11 TABLE I1 Phosphorylation of 1 [methyl-3H]AZT and 1 p~ Imeth~l-~HI Cytostatic effects of AZT and D4T on MT-4, ATH8, H9, D4T in MT-4 cells as a function of incubation time and CEM cells ~~ ~ Phosphorylated products” 50% cytostatic dose Time of Compound Cell line incubation 5’-Monophosphate 5”Diphosphate 5’-Triphosphate nmlllOg celb PM PM AZT MT-4 53 f9 42 f 7 36f 10 AZT H9 3.5 0.68 CEM 582 f 73 >lo00 >lo00 ATH8 40 24 0.80 48 0.72 D4T MT-4 112 f 13 84 & 12 41 f 18 72 95 0.43 0.15 H9 264 f 70 243 f 19 228 f 13 CEM 138f 56 143 f 15 158f 29 D4T ATH8 110 0.06 0.23 7 0.29 a Data taken from Ref. 3. Due to the long time that the ATH8 cells required for proliferation, the inhibitory effects of the compounds on 48 0.24 ATH8 cell growth were measured only after 7 days of incubation. Data represent average values for two separate experiments. ATH8 cells than for CEM or H9 cells. The antiproliferative effect of AZT on MT-4 cells slightly increased upon longer incubation times (Table 11). Less striking differences were noted in the cytostatic effects of D4T on the different T cell lines. The CD, values of D4T for H9, CEM, and ATH8 cells ranged from 112 to 264 PM. The antiproliferative effect of D4T on MT-4 cells significantly increased upon longer incu- bation times (Table 11). Phosphorylation of [methyl-3HJAZT and [methyL3H]D4T in MT-4 Cells following Different Incubation Times-The metabolism of radiolabeled AZT and D4T by MT-4 cells was followed upon incubation of the cells with 1 PM radiolabeled compounds for 3.5, 7, 24, 48, and 72 h (Table 111). A typical HPLC chromatogram for the intracellular phosphorylation of [methyC3H]AZT and [methyL3H]D4T is depicted in Fig. 2. AZT was extensively metabolized to its 5”monophosphate derivative (AZT-MP) in MT-4 cells. Upon incubation with 1 PM AZT, the intracellular AZT-MP levels measured after 3.5- F r a c t i on nun brr 48 h ranged from 100 to 164 nmol/lOg cells and slightly decreased to 95 nmol/lOg cells after 72 h. In contrast, rela- FIG. 2. Ion exchange (Partisil-SAX) HPLC elution profile of a 10% trichloroacetic acid extract of MT-4 cells incubated tively small levels of the 5’-diphosphate (AZT-DP) (1.3-1.9 for 24 h with 1 p~ [methyZ-’H]AZT (M) or 1 phi [methyl- nmol/lOg cells) and 5’-triphosphate (AZT-TP) (0.68-0.80 *H]D4T (.---). The column was equilibrated and developed nmol/lOg cells) were detected within the first 48 h of incuba- with 0.007 M ammonium phosphate (pH 3.80) for 6 min, followed by tion; and after 72 h, the AZT 5‘-di- and 5”triphosphate levels a linear gradient to 0.25 M ammonium phosphate + 0.50 M KC1 (pH decreased to 0.43 and 0.15 nmol/lOg cells, respectively. The 4.5) over the next 20 min, and finally by 20-min isocratic elution with accumulation of AZT 5‘-monophosphate in MT-4 cells con- the latter buffer. trasts sharply with the low levels found for the D4T 5’- monophosphate (D4T-MP) (Table 111). The D4T 5”triphos- following 24-h incubation with MT-4 cells is presented. No phate (D4T-TP) levels were similar to those detected for metabolites other than the 5’-mono-, 5’-di-, and 5”triphos- D4T-MP. The intracellular DIT-TP levels were only 2-3-fold phates of AZT and D4T were detected. Identification of the lower than the AZT-TP levels after 24-48 h incubation, and 5’-mono-, 5‘-di-, and 5”triphosphate metabolites of AZT and the 72-h levels of DIT-TP were similar to the AZT-TP levels D4T was ascertained by spiking the cell extracts with the (Table 111). chemically synthesized 5’-mono- and 5”triphosphates of AZT In Fig. 2, the phosphorylation pattern of AZT and D4T and D4T. The radiolabeled peaks that were tentatively iden- 0.03 0.14 72 0.15 0.07 0.35 24 0.21 0.10 0.23 0.09 0.24 3.5 0.47 147 1.34 100 1.96 164 1.53 0.71 155 1.30 >loo0 >loo0 >loo0 Day3 Day4 Day5 Dayl“ 4.1‘ 110‘ 2.4‘ 40‘ CDmb EDw” CDmb cells ATHB 6130 Phosphorylation of D4T and AZT in MT-4 Cells TABLE IV Phosphorylation of Im~thyl-~HlAZT and [methyL3H1D4T in MT-4 cells upon 24-h incubation as a function of the input concentrations of the nucleosides Initial Compound concentration Nucleoside 5'-Monophosphate 5"Diphosphate 5"Triphosphate P'U nm01/10~ cells" AZT 50 5.6 1.6 10 35 377 4.0 1.1 1 .o 0.53 1.7 26 1.6 0.54 0.37 1.6 21 1.4 0.45 D4T 50 93 2.4 10 19 0.77 1.0 1.0 1.1 2.0 0.13 0.16 0.22 0.25 Data represent average values for two separate experiments. TABLE V TABLE VI Phosphorylation of 1 p~ [meth~l-~H]AZT and 1 p~ [rneth~l-~H] Effect of dThd and dCyd on the phosphorylation of 1 pM [methyL3H] D4T in human lymphoblast Raji/O and Raji/TK- cells upon 24-h AZT and 1 pM [rnethyL3H]D4T in MT-4 cells incubation 5'-Phosphorylated metabolites" Concentration of 5'- phosphorylated metabolites" Compound Upon addition of Compound 250~~ dThd+ 25aM 1OOOpM Raji/O RaiiiTK- dThd dCyd 1000 p~ dCyd nm01/10~ cells nmol/109 cells AZT 2.52 0.704 AZT 4.25 2.88 2.15 1.62 AZT-MP 31.7 0.053 AZT-MP 100 42.0 AZT-DP 1.01 C0.005 AZT-DP 1.96 1.12 AZT-TP AZT-TP D4T 1.91 1.77 D4T D4T-MP 0.038 0.025 D4T-MP 0.007 DIT-DP 0.011 0.006 D4T-DP 0.098 C0.005 0.005 0.006 D4T-TP DIT-TP C0.005 Data represent average values for two separate experiments. Values are the mean of two separate experiments. added to the cell extracts, the D4T 5'-tri- and 5"diphosphate peaks significantly diminished in size, whereas the D4T 5'- monophosphate peak increased (data not shown). Phosphorylation of [meth~l-~H/AZT and [methyL3H/D4T in MT-4 Cells as a Function of Different Input Concentra- tions-MT-4 cells were incubated with [ methyL3H]AZT and [methyL3H]D4T at concentrations ranging from 0.25 to 50 pM. Formation of the 5'-mono-, 5'-di-, and 5"triphosphate metabolites of AZT and D4T increased with higher initial concentrations of the nucleosides. While the intracellular levels of AZT-MP increased pro- portionally with increasing concentrations of the input AZT, or about 60-fold when the initial AZT concentration increased from 0.37 to 50 p~, AZT 5'-diphosphate and 5"triphosphate levels increased only by a factor of 4-5-fold under the same .4 I experimental conditions. In contrast, when the initial concen- QE Ql 0.2 f 1 ldThd 1-1 or l/AZT~--l(phf'Ij tration of D4T was raised from 0.25 to 50 p~, phosphorylated products (5'-mOnO-, 5'-di-, and 5"triphosphate) were formed " " "" + "" + "" ~ "" ~ "" dsam 0.02 a03 QW (105 at a 30-80-fold greater extent (Table IV). The D4T-TP levels I/ 0CTl~-*l(pM) were inferior to the AZT-TP levels if the initial concentration of the nucleosides was lower than 1 p~, equal to the AZT-TP FIG. 3. Double-reciprocal (Lineweaver-Burk) plots for levels if the initial nucleoside concentrations were 1-10 p~ dThd (o"o), AZT (W), and D4T (A---A) phospho- rylation by MT-4 cell extracts. and superior to the AZT-TP levels if the initial nucleoside concentrations were 50 pM. Essentially the same results were obtained in H9 cells (data not shown). tified as the 5'-mono- and 5"triphosphate of AZT and D4T Metabolism of [meth~l-~H/AZT and [methyL3H1D4T in coincided with the authentic 5'-mono- and 5'-triphosphates Raji/O and dThd Kinase-deficient Raji/TK- Cells-Phospho- of AZT and D4T on the chromatogram. Upon treatment of rylation of [ methyk3H]AZT and [ methyL3H]D4T was exam- the cell extracts with alkaline phosphatase, the radiolabeled ined in the human B-lymphoblast Raji/O cell line and its peaks identified as the phosphorylated derivatives of AZT dThd kinase-deficient counterpart Raji/TK- (Table V). and D4T disappeared while the nucleoside forms accumulated AZT was phosphorylated in Raji/O cells about 2-3-fold less (data not shown). Similarly, after phosphodiesterase had been 0.231 0.019 0.014 0.017 0.015 0.201 0.041 0.025 1.35 0.621 0.762 0.658 0.795 0.427 0.313 0.022 0.721 0.007 0.717 0.036 18.6 1.19 0.44 0.03 0.06 0.06 1.7 2.3 70 3.1 1.3 4.0 219 1212 Phosphorylation of D4T and AZT in MT-4 Cells 6131 TABLE VI1 Zntracellular [methyL3H]dTTP levels and ratios of intracellular [methyl-3HldTMP/([methyl-3HldTDP + [methyl- 3H]dTTP) levels in MT-4 cells preincubated with AZT and D4T prior to pulse labeling with 0.04 pM [methyl-3Hl dThd for 2 h Data represent average values for two to three separate experiments. [meth~l-~H)dTMP Ratio Intracellular levels of [methyL3H]dTTP Compound [methyL3H]dTDP + [methyL3H]dTTP concentration D4T AZT D4T AZT PM pmolllOQ cells 160 0.71 0.71 0 160 164 8.54 0.96 50 24.5 273 17.0 1.77 500 3.6 TABLE VI11 and D4T, it appeared that AZT was phosphorylated to a Inhibitory effects of AZT-TP and DIT-TP on the actiuity of MLV- similar extent as dThd, while D4T was phosphorylated about and HZV-associated reverse transcriptase and DNA polymerase a 200-fold less efficiently. 50% inhibitory concentration Effect of dThd and dCyd on the Phosphorylation of [methyl- 3H]AZT and [methyL3H]D4T in MT-4 Cells-As shown in DNA Compound Table VI, addition of 250 p~ dThd (in the presence of 1000 30-min polymerase a 30-min p~ dCyd to avoid cytotoxicity of dThd) resulted in a dramatic assay assay 30-min assay assay decrease of the levels of phosphorylated metabolites of AZT and D4T. Under our experimental conditions, the AZT 5‘- AZT-TP” 1.15 f 0.4 0.51 f 0.3 0.43 >500 mono- and 5”triphosphate levels decreased by 80- and 35- 175 f 117 D4T-TP 1.63 f 0.4 1.18 f 0.2 0.84 fold, respectively; the D4T 5’-monophosphate levels de- Data taken from Ref. 27. creased by 30-fold; and the D4T-TP levels even fell below the detection limit (0.005 nmol/lOg cells) (Table VI). Addition of TABLE IX a lower concentration of exogenous dThd (25 p~) or addition Partition coefficients (P) between 1-octanol and 10 mMpotassium of 1000 p~ dCyd also resulted in a significant but less marked Rf values in chloroform:methanol(92.5:7.5) on phosphate buffer and diminution of the phosphorylated AZT and D4T pools than silica gel TLC for dThd, ddThd, D4T. and AZT that observed with 250 p~ dThd. Compound RF* P” Effect of Preincubation of MT-4 Cells with AZT and D4T dThd 0.124 f 0.059 0.067 f 0.005 on the Intracellular Phosphorylation of 0.04 phi [methyL3HI ddThd 0.339 f 0.088 0.233 f 0.005 dThd-MT-4 cells were incubated for 24 h with different D4T 0.325 f 0.091 0.154 f 0.008 concentrations of AZT or D4T prior to a pulse labeling of the AZT 0.431 & 0.082 0.964 f 0.038 cells with 0.04 p~ [meth~l-~H]dThd (10 pCi/4 X lo6 cells/ “Data represent average values (fS.D.) for at least two to three culture) for 2 h. Then, the intracellular ratios of [methyL3H] separate experiments. dTMP/( [ methyL3H]dTDP + [ methyL3H]dTTP) were deter- mined. Without inhibitor, the ratio of radiolabeled dTMP/ efficiently than in MT-4 cells, and D4T was phosphorylated (dTDP + dTTP) was 0.71 (Table VII). This corresponds to 10-fold less efficiently in Raji/O than MT-4 cells (compare 160, 32, and 137 pmol of [meth~l-~HIdTTP, [methyL3H] the data in Tables IV and V). In the dThd kinase-deficient dTDP, and [methyL3H]dTMP per lo9 MT-4 cells, respec- cell line Raji/TK-, AZT was 500-1000-fold less extensively tively. However, increasing initial concentrations of AZT phosphorylated to its 5’-mOnO-, 5‘-di-, and 5”triphosphate resulted in a dramatic increase of the [meth~l-~H]dTMP/ than in Raji/O cells. AZT-DP levels were even under the ( [methyb3H]dTDP + [ methyL3H]dTTP) ratio (17 in the pres- detection limit (0.005 nmol/lOg cells). In contrast, D4T phos- ence of 500 p~ AZT). In contrast, 500 p~ D4T afforded a phorylation proceeded at about the same rate in Raji/TK- as slight increase of the ratios of radiolabeled dTMP/(dTDP + in Raji/O cells (Table V). These data suggest a crucial role for dTTP) (Table VII). cellular dThd kinase in the phosphorylation of AZT but not Inhibitory Effects of AZT-TP and DIT-TP on HIV-associ- D4T in Raji cells. ated Reverse Transcriptase and DNA Polymerase a-AZT- Phosphorylation of Lmeth~l-~HJdThd, [methyL3HJAZT, and TP and D4T-TP were evaluated for their inhibitory effects [methyL3H]D4T by MT-4 Cell Extracts-Different concentra- on the reverse transcriptases of MLV and HIV and DNA tions of [ methyL3H]dThd, [ methyL3H]AZT, and [ meth~l-~H] polymerase a. The initial concentration of [ methyL3H]dTTP D4T were incubated with MT-4 cell extracts under assay in the reaction mixture was 1 p~. AZT-TP and D4T-TP were conditions stable for optimal dThd kinase activity. Michaelis- strongly inhibitory to the reverse transcriptases. Their 50% Menten constants (K,) and maximal velocity ( Vmax) values inhibitory concentrations were in the range of 0.5-1.5 pM for were calculated from Lineweaver-Burk diagrams (Fig. 3). the MLV reverse transcriptase and in the range of 0.02-0.1 dThd and AZT were phosphorylated to a comparable extent p~ range for HIV reverse transcriptase (Table VIII). AZT- by the MT-4 cell extracts, while D4T was phosphorylated much less efficiently. The V,,, values for phosphorylation of MP had no effect on the enzyme, even at a concentration of 1000 p~ (27). AZT-TP and D4T-TP were considerably less dThd, AZT, and D4T were 190, 153, and 8.2 nmol/mg of protein/h, respectively. The K, values were 13.5, 13.8, and inhibitory to DNA polymerase a than MLV and HIV reverse 142 p~, respectively. Addition of 10 p~ dThd or 10 ~LM dTTP transcriptase. Their 50% inhibitory concentrations were >500 considerably decreased the phosphorylation of radiolabeled and 175 pM, respectively. AZT or D4T by MT-4 cell extracts (data not shown). When Lipophilicity of AZT and D4T-To estimate the lipid solu- the V,,./K, ratio (14, 11, and 0.06, respectively) was taken bility of AZT, D4T, dThd, and ddThd, their RF values were as a parameter for the phosphorylation rate of dThd, AZT, determined by thin layer chromatography in a mixture of PM PM PM PM MLV HIV 6132 Phosphorylation of D4T and AZT in MT-4 Cells ch1oroform:methanol (Table IX). The RF values of AZT, D4T, D4T and AZT phosphorylation by MT-4 cell extracts. This dThd, and ddThd were 0.43,0.33,0.12, and 0.34, respectively. suggests that dThd kinase is responsible for the phosphoryl- The partition of the test compounds between 1-octanol and ation of AZT and D4T. However, the phosphorylation pat- potassium phosphate buffer was also measured (Table IX). terns obtained for D4T and AZT in Raji/O and Raji/TK- cells AZT was far more lipophilic than D4T (partition coefficient implicate dThd kinase in the activation of AZT but not D4T. (P):0.964 and 0.154, respectively). The lipophilicity of D4T Furthermore, D4T is equally cytostatic for Raji/O and Raji/ was higher than that of dThd but lower than that of ddThd TK- cells and inhibit Simian AIDS-related retrovirus (SRV)- (Table IX). induced syncytium formation almost to a similar extent in Raji/O and Raji/TK- cells (data not shown). These observa- DISCUSSION tions, again, argue against the role of dThd kinase in the The 2’,3’-dideoxythymidine analogues AZT and D4T are phosphorylation of D4T. Thus, it is unclear which enzyme(s) potent inhibitors of HIV and MSV replication in vitro and are responsible for the phosphorylation of D4T. Whatever show a marked “therapeutic index” in MT-4 cells (ratio of enzyme is responsible, it is highly susceptible to inhibition by the compound concentration required to inhibit cell growth dThd as well as dTTP. by 50% to the concentration required to inhibit virus repli- The phosphorylation data obtained in MT-4 extracts are cation by 50%). Although structurally related, AZT and D4T in agreement with our observation that (i) phosphorylation show a totally different pattern of intracellular phosphoryla- of D4T and AZT in intact MT-4 cells is severely suppressed tion. In contrast to AZT that accumulates mainly as its 5’- by dThd, and (ii) both the cytostatic and antiretroviral activ- monophosphate resulting in relatively small levels of the 5’- ity of D4T and AZT in MT-4 cells are reversed by the addition di- and 5’-triphosphate derivatives, D4T does not accumulate of dThd. The latter findings also suggest that a phosphoryl- as it 5’-monophosphate. Due to the low extent by which D4T ated product of D4T and AZT, presumably their 5”triphos- is initially phosphorylated to its 5’-monophosphate, the D4T- phate derivatives, accounts for the biological activity of the TP levels that are eventually achieved are lower than the compounds. One of the factors that may contribute to the AZT-TP levels, at least when starting with an input nucleo- decreased intracellular formation of phosphorylated D4T side concentration of 1 p~ or less. However, when the initial products in the presence of dThd (and dCyd) may be the dose of AZT and D4T was increased, anabolism of D4T to competition of dThd with D4T for the catalytic site of D4T D4T-TP was facilitated to a relatively greater extent over kinase. Additional inhibition may occur after the conversion AZT-TP formation from AZT, so that at input nucleoside of dThd to dTTP, because dTTP could act as a potent concentrations greater than 10 PM the levels obtained for (allosteric) inhibitor of the putative D4T kinase. Also, inter- D4T-TP exceeded those of AZT-TP. ference with the cellular transport (uptake) of D4T may be Furman et al. (15) demonstrated that AZT-MP accumula- taken in consideration to explain the decreased anabolism of tion causes a severe inhibition of dTMP kinase resulting in D4T in the presence of dThd. Should this be the case, how- an efficient blockage of its further phosphorylation to AZT- ever, it is surprising that 25 p~ dThd would afford a similar DP. From our findings, one may assume that when the initial inhibitory effect on the D4T transport system as 1000 PM AZT concentrations are higher than 1 p~, dTMP kinase of dCyd. MT-4 cells must be blocked to a significant extent. Conse- AZT is active as an inhibitor of HIV replication in MT-4 quently, increase of the initial AZT concentration from 1 to cells and of C3H cell transformation by MSV at lower con- 50 p~ would not lead to a proportional increase in the centrations than is D4T. Since AZT-TP and D4T-TP are formation of AZT-DP and AZT-TP. D4T is apparently unable equally inhibitory to the HIV- and MLV-associated reverse to block its own phosphorylation, and, hence, increase in the transcriptases, this greater potency of AZT against HIV rep- initial D4T concentration results in a fairly proportional lication can most probably be attributed to its more efficient increase in the levels of all phosphorylated products of D4T phosphorylation within the cells to the 5”triphosphate. In- (Table IV). Additional evidence that D4T-MP unlike AZT- deed, higher 5’4riphosphate levels are achieved within MT-4 MP does not significantly inhibit dTMP kinase is obtained cells for AZT than for D4T if the nucleosides are added to from our pulse-labeling experiments in which MT-4 cells were the cells at a concentration of 0.25-0.37 p~. Due to the preincubated with AZT or D4T and then labeled with [methyl- relatively low specific radioactivity of both compounds, phos- 3H]dThd. A block at the level of dTMP kinase should result phorylation of the compounds at lower (antivirally active) in an increase of intracellular radiolabeled dTMP and a concentrations could not be accurately determined. concomitant decrease of the radiolabeled dTTP pools com- The observation that HIV reverse transcriptase has a very pared to control. In contrast to AZT, D4T did not lead to a high affinity for AZT-TP and D4T-TP, this affinity being significant accumulation of [ methyL3H]dTMP and depletion several orders of magnitude higher than the affinity for cel- of [methyG3H]dTTP pools. The fact that D4T, unlike AZT, lular DNA polymerase a, may explain the selectivity and does not interfere with dTMP kinase and thus fails to affect potency of both AZT and D4T as an antiretroviral chemo- the salvage pathway of dThd, may be clinically advantageous therapeutic agent in vitro. In this respect, D4T-TP proved 3- in that a shut-off of this pathway, because of the ensuing fold more inhibitory to DNA polymerase a than AZT-TP. reduction in dTTP pool levels, may lead to substantial cyto- In view of the propensity of HIV to infect and damage the toxicity. central nervous system, potential treatment schedules for It has been established that AZT and dThd are phospho- AIDS patients should address the ability of candidate anti- rylated equally well by the cytosol dThd kinase (15, 16). The AIDS drugs to cross the blood brain barrier. Preliminary K, and Vmax values of dThd kinase are very similar for both evidence suggests that AZT favorably influences the course compounds. We now find that D4T is phosphorylated by MT- of neurological disorders associated with AIDS (32). Com- 4 cell extracts at a much higher K,,, and with a significantly pounds with high lipophilicity may be expected to cross the lower Vmax than AZT and dThd. The phosphorylation of D4T blood brain barrier more easily than polar compounds. AZT in MT-4 cells is sensitive to inhibition by dThd, and so is the proved clearly more lipophilic than dThd (P, 0.964 and 0.067, phosphorylation of AZT. dTTP, a well known allosteric in- respectively). D4T showed a higher partition coefficient than hibitor of dThd phosphorylation by dThd kinase, also inhibits dThd, but its lipophilicity was inferior to that of AZT. Also, Phosphorylation of D4T and AZT in MT-4 Cells 6133 10. Lin, T.-S., Schinazi, R. F., and Prusoff, W. H. (1987) Biochem. D4T had a higher partition coefficient value than ddCyd (P, Phurmncol. 36,2713-2718 0.050). Since it has been established that dThd as well as 11. Schinazi, R. F., Chu, C. K., Ahn, M.-K., and Sommadossi, J.-P. ddCyd cross the blood brain barrier (33, 34), one may specu- (1987) J. Cell. Biochem. Suppl. 11D, 74 late from our lipophilicity data that D4T also does so. 12. Balzarini, J., Baba, M., Pauwels, R., Herdewijn, P., and De Clercq, Finally, we found that D4T is considerably less active than E. (1988) Biochem. Phurmacol. 37, 2847-2856 13. Matthes, E., Lehmann, Ch., Scholz, D., Rosenthal, H. A., and AZT as an antiretroviral agent in mice. Treatment of the Langen, P. (1988) Biochim. Biophys. Res. Commun. 153,825- MSV-infected newborn mice for 5 subsequent days with AZT at 125 mg/kg/day increased the mean tumor initiation time 14. Balzarini, J., Matthes, E., Meeus, P., Johns, D. G., and De Clercq, by 2-fold, and resulted in a 71% survival of the mice at day (1988) Proceedings of the VI International Symposium on E. 25. In contrast, D4T treatment at 125 and 25 mg/kg/day only Human Purine & Pyrimidine Metabolism, Hakone, Japan, July resulted in a modest delay of tumor formation, without a 17-21, 1988, in press 15. Furman, P. A,, Fyfe, J. A., St. Clair, M. H., Weinhold, K., Rideout, dramatic effect on the increase of the survival rate of the mice J. L., Freeman, G. A., Nusinoff Lehrman, S., Bolognesi, D. P., (data not shown). This decreased efficiency may, at least in Broder, S., Mitsuya, H., and Barry, D. W. (1986) Proc. Natl. part, be related to the low efficiency by which D4T is con- Acad. Sci. U. S. A. 83,8333-8337 verted by the cellular phosphorylating enzymes to its 5’- 16. Balzarini, J., Pauwels, R., Baba, M., Herdewijn, P., De Clercq, monophosphate and eventually 5’-triphosphate in the murine E., Broder, S., and Johns, D. G. (1988) Biochem. Phurmacol. model. The in vivo data are consistent with the in vitro data 37,897-903 obtained in MSV-infected murine C3H cells. D4T was 100- 17. Cooney, D. A., Dalal, M., Mitsuya, H., McMahon, J. B., Nadkarni, M., Balzarini, J., Broder, S., and Johns, D. G. (1986) Biochem. fold less efficient in the latter system than AZT. Pharmncol. 35,2065-2068 In conclusion, this is the first report in which the cellular 18. Balzarini, J., Cooney, D. A,, Dalal, M., Kang, G-J., Cupp, J. E., metabolism and kinetic properties of D4T were investigated S., and Johns, D. G. (1987) Mol. Phur- De Clercq, E., Broder, and compared to those of AZT. A close correlation was found mmol. 32,798-806 between the antiviral and cytostatic activity of the compounds 19. Starnes, M. C., and Cheng, Y. (1987) J. Biol. Chem. 262, 988- and their metabolism to their 5”triphosphate form. D4T has 991 20. Langen, P., Kowollik, G., Etzold, G., Venner, H., and Reinert, H. unique metabolic features in that it does not accumulate as (1972) Acta Biol. Med. Ger. 29,483-494 its 5’-monophosphate and generates similar levels of 5‘- 21. Mitsuya, H., Guo, H.-E., Cowman, L., Megson, M., Reitz, M. S., mono-, 5’-di-, and 5”triphosphate. The fact that D4T does Jr., and Broder, S. (1984) Science (Wmh. D.C.) 225, 1484- not seem to block dTMP kinase may give D4T a potential edge over AZT. 22. Pauwels, R., De Clercq, E., Desmyter, J., Balzarini, J., Goubau, P., Herdewijn, P., Vanderhaeghe, H., and Vandeputte, M. Acknowledgments-We thank Ann Absillis, Lizette van Berckelaer, (1987) J. Virol. Methods 16, 171-185 23. Harada, S., Koyanagi, Y., and Yamamoto, N. (1985) Science 229, Miette Stuyck, Ria Van Berwaer, and Luk Kerremans for excellent technical assistance. The dedicated editorial help of Christiane Cal- 563-566 lebaut is highly appreciated. 24. Balzarini, J., De Clercq, E., Torrence, P. F., Mertes, M. P., Park, J. S., Schmidt, C. L., Shugar, D., Barr, P. J., Jones, A. S., Verhelst, G., and Walker, R. T. (1982) Biochem. Phurmacol. Note Added in Proof-After this paper was submitted, a commu- 31,1089-1095 nication with regard to the cellular pharmacology of D4T appeared 25. Popovic, M., Sarngadharan, M. G., Read, E., and Gallo, R. C. in Biochemical Pharmmology (35). (1984) Science (Wash. D.C.) 224,497-500 26. De Clercq, E., and Merigan, T. C. (1971) Proc. SOC. Exp. Biol. REFERENCES Med. 137,590-594 1. Mitsuva, H., and Broder, S. (1986) Proc. Natl. Acud. Sci. U. S. A. 27. Balzarini, J., Herdewijn, P., Pauwels, R., Broder, S., and De 83,-1911-.1915 Clercq, E. (1988) Biochem. Phurmacol. 37, 2395-2403 2. Balzarini. J.. Pauwels. R.. Herdewiin, P.. De Clercq, E., Cooney, 28. Hoard, D. E., and Ott, D. G. (1965) J. Am. Chem. SOC. 87,1785- D. A., Kang, G.-J.,’ Dalal, M., johns; D. G., and Broder, S. (1986) Biochem. Biophys. Res. Commun. 140,735-742 29. Balzarini, J., Pauwels, R., Baba, M., Robins, M. J., Zou, R., 3. Balzarini, J., Kang, G.-J., Dalal, M., Herdewijn, P., De Clercq, Herdewijn, P., and De Clercq, E. (1987) Biochem. Biophys. Res. E., Broder, S., and Johns, D. G (1987) Mol. Phurmacol. 32, Commun. 145,269-276 162-167 30. Balzarini, J., Baba, M., Pauwels, R., Herdewijn, P., Wood, S. G., 4. Lin, T.-S., Schinazi, R. F., Chen, M. S., Kinney-Thomas, E., and Robins, M. J., and De Clercq, E. (1988) Mol. Phurmmol. 33, Prusoff, W. H. (1987) Biochem. Pharmmol. 36, 311-316 243-249 Mitsuya, H., Weinhold, K. J., Furman, P. A., St. Clair, M. H., 31. Balzarini, J., De Clercq, E., Mertes, M. P., Shugar, D., and Nusinoff Lehrman, S., Gallo, R. C., Bolognesi, D., Barry, D. Torrence, P. F. (1982) Biochem. Phurmmol. 31, 3673-3682 W., and Broder, S. (1985) Proc. Natl. Acud. Sci. U. S. A. 82, 32. Yarchoan, R., Klecker, R. W., Weinhold, K. J., Markham, P. D., 7096-7100 Lyerly, H. K., Durack, D. T., Gelmann, E., Nusinoff Lehrman, 6. Herdewijn, P., Balzarini, J., De Clercq, E., Pauwels, R., Baba, S., Blum, R. M., Barry, D. W., Shearer, G. M., Fischl, M. A., M., Broder, S., and Vanderhaeghe, H. (1987) J. Med. Chem. Mitsuya, H., Gallo, R. C., Collins, J. M., Bolognesi, D. P., 30,1270-1278 Myers, C. E., and Broder, S. (1986) Lancet 1, 575-580 7. Balzarini, J., and Broder, S. (1988) in Clinical Use of Antiviral 33. Zaharko, D. S., Bolten, B. J., Chiuten, D., and Wiernik, P. H. Drugs (De Clercq, E., ed) pp. 361-385, Martinus Nijhoff Pub- (1979) Cancer Res. 39,4777-4781 lishing, Norwell, MA 34. Yarchoan, R., Perno, C. F., Thomas, R. V., Klecker, R. W., Allain, 8. Nakashima, H., Matsui, T., Harada, S., Kobayashi, N., Matsuda, J.-P., Wills, R. J., McAtee, N., Fischl, M. A,, Dubinsky, R., A., Ueda, T., and Yamamoto, N. (1986) Antimicrob. Agents McNeely, M. C., Mitsuya, H., Pluda, J. M., Lawley, T. J., Chemother. 30,933-937 Leuther, M., Safai, B., Collins, J. M., Myers, C. E., and Broder, 9. Baba, M., Pauwels, R., Herdewijn, P., De Clercq, E., Desmyter, S. (1988) Lancet 1, 76-81 J., and Vandeputte, M. (1987) Biochem. Biophys. Res. Commun. 35. August, E. M., Marongiu, M. E., Lin, T.-S., and Prusoff, W. H. 142, 128-134 (1988) Biochem. Phurmmol. 37, 4419-4422 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Biological Chemistry Unpaywall

Differential Patterns of Intracellular Metabolism of 2′,3′-Didehydro-2′,3′-dideoxythymidine and 3′-Azido-2′,3′-dideoxythymidine, Two Potent Anti-human Immunodeficiency Virus Compounds

Balzarini, J; Herdewijn, P; Clercq, E De
Journal of Biological Chemistry – Apr 1, 1989
Download PDF
7 pages

Loading next page...
 
/lp/unpaywall/differential-patterns-of-intracellular-metabolism-of-2-3-didehydro-2-3-aAv4Jg0FVx

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
0021-9258
DOI
10.1016/s0021-9258(18)83322-1
Publisher site
See Article on Publisher Site

Abstract

Vol. 264, No. 11, Issue of ’ April pp. 61274133,1989 15, THE JOURNAL OF BIOLOGICAL CHEMISTRY Printed in U. S.A. 0 1989 by The American Society for Biochemistry and Molecular Biology, Inc. Differential Patterns of Intracellular Metabolism of 2’,3’-Didehydro- 2’,3‘-dideoxythymidine and 3’-Azido-2’,3’-dideoxythymidine, Two Potent Anti-human Immunodeficiency Virus Compounds* (Received for publication, October 21, 1988) Jan Balzarini, Piet Herdewijn, and Erik De Clercq From the Rega Institute for Medical Research, Katholieke Universiteit Leuven, B-3000 Leuuen, Belgium 3’-Azido-2‘,3’-dideoxythymidine (AZT) and 2’,3’- didehydro-2’,3’-dideoxycytidine (ddeCyd, ddddcyd, D4C) (2- didehydro-2‘,3’-dideoxythymidine (D4T) are potent 4), 3‘-azido-2‘,3‘-dideoxythymidine (AzddThd, AZT) (5-8), and selective inhibitors of human immunodeficiency 2’,3’-didehydro-2’,3’-dideoxythymidine (ddeThd, ddddThd, virus replication in MT-4 and ATH8 cells. They are D4T) (3,9, lo), 3’-azido-2’,3’-dideoxyuridine (AzddUrd) (11, also inhibitory to the replication of murine retrovi- 12), 3’-fluoro-2’,3‘-dideoxythymidine (FddThd) (6, 12-14), ruses, i.e. Moloney murine sarcoma virus-induced and 3’-fluoro-2’,3’-dideoxyuridine (FddUrd) (12)) have been transformation of C3H cells. In MT-4 cells AZT is recognized as potent and selective inhibitors of the replication readily phosphorylated to its 5’-monophosphate, while of human immunodeficiency virus (HIV), the etiologic agent the 5‘-di- and 5“triphosphates are generated to a 200- of the acquired immunodeficiency syndrome (AIDS). The 600-fold lower extent than the 5’-monophosphate. 2‘,3‘-dideoxyribonucleoside analogues are assumed to be tar- D4T is phosphorylated in MT-4 cells to its 5’-mono- geted at the virus-encoded RNA-directed DNA polymerase phosphate at a 300-600-fold lower extent than AZT. (reverse transcriptase). To exert their antiretroviral activity, The phosphorylation of AZT in the thymidine kinase- the ddNs should be phosphorylated intracellularly to their 5’- deficient cell line (Raji/TK-) is severely depressed, triphosphate (ddNTP) metabolites. The phosphorylation while D4T phosphorylation is only slightly diminished steps are catalyzed by cellular enzymes, as has been shown in Raji/TK- as compared to Raji/O cells. D4T has a 10- for AZT in H9, ATH8, and Molt/4F cells (7,15,16), for ddCyd fold lower affinity for phosphorylation by crude MT-4 in ATH8 and Molt/4F cells (16-19), and for FddThd in cell extracts than AZT (IC,,,, 142 and 14 PM, respec- Ehrlich ascites carcinoma, MT-4, CEM, and H9 cells (13, 14, tively), and the Vmax for phosphorylation of D4T is only 5% that of AZT. D4T is phosphorylated by MT-4 cell 20). The nucleoside kinases (ie. dThd kinase and dCyd ki- extracts about 180-fold less efficiently than AZT nase) convert the ddNs (i.e. AZT and ddCyd, respectively) to (VmaJKm, 0.06 for D4T, as compared to 11 for AZT), their 5’-monophosphates. However, the affinity of the ddNs and this is consistent with the differences found in the for their respective nucleoside kinases may differ considerably amounts of phosphorylated products of D4T and AZT from one compound to another (i.e. K, value of AZT for H9 formed in intact MT-4 cells. The 5’-triphosphates of cell dThd kinase, 3 FM (15); K, value of FddThd for Ehrlich AZT and D4T are equipotent in their inhibitory effects ascites carcinoma cell dThd kinase, 11.2 PM (20); K,,, value of on the reverse transcriptases from human immunode- ddCyd for Molt/4F cell and KB cell dCyd kinase, 200 and 220 ficiency virus and Moloney murine leukemia virus. PM, respectively (7, 19)) and determines the eventual intra- cellular levels of the ddNTPs. In an attempt to gain a better insight in the metabolism and mechanism of antiretrovirus action of the ddNs, we Several pyrimidine 2’,3’-dideoxynucleoside analogues investigated the intracellular phosphorylation of two struc- (ddNs)’ (i.e. 2’,3’-dideoxycytidine (ddCyd, D2C) (1, 2), 2‘,3‘- turally related and potent anti-HIV compounds (AZT and D4T) in MT-4 cells. This cell line is exquisitely sensitive to * These investigations were supported in part by the AIDS Basic the inhibitory effects of 2‘,3‘-dideoxythymidine analogues Research Programme of the European Community and by grants (ie. AZT and D4T) on HIV replication and would therefore from the Belgian Fonds voor Geneeskundig Wetenschappelijk On- seem useful to investigate in parallel the intracellular metab- derzoek (Projects 3.0040.83 and 3.0097.87) and the Belgian Gecon- certeerde Onderzoeksacties (Project 85/90-79). The costs of publica- olism of AZT and D4T. D4T has not previously been studied tion of this article were defrayed in part by the payment of page for its metabolic fate within the cell. charges. This article must therefore be hereby marked “aduertise- rnent” in accordance with 18 U.S.C. Section 1734 solely to indicate MATERIALS AND METHODS this fact. ’ The abbreviations used are: ddNs, pyrimidine 2’,3’-dideoxynu- Cells-The origin and cultivation of MT-4, ATH8, and C3H cells cleoside analogues; ddCyd, D2C, 2’,3’-dideoxycytidine; ddeCyd, have been described previously (21-23). MT-4 cells were a gift from ddddCyd, D4C, 2‘,3‘-didehydro-2‘,3’-dideoxycytidine; AzddThd, Dr. N. Yamamoto (Yamaguchi University, Yamaguchi, Japan). Both AZT, 3’-azido-2’,3’-dideoxythymidine; ddeThd, ddddThd, D4T, MT-4 and ATH8 cells represent immortalized helper/inducer T cell 2’,3’-didehydro-2’,3’-dideoxythymidine; AzddUrd, 3’-azido-2’,3’-di- clones, obtained by cloning a normal T4 cell line in the presence of deoxyuridine; FddThd, 3’-fluoro-2’,3’-dideoxythymidine; FddUrd, tumor cells derived from a patient with adult T cell leukemia or 3’-fluoro-2’,3’-dideoxyuridine; HIV, human immunodeficiency virus; lethally irradiated HTLV-I-producing MJ tumor cells, respectively. MSV, Moloney murine sarcoma virus; MLV, Moloney murine leu- C3H cells represent a continuously growing murine embryo fibroblast kemia virus; HPLC, high performance liquid chromatography; AZT- cell line. Cultivation and characterization of the Fhji/O cell line and TP, AZT 5”triphosphate; D4T-TP, D4T 5”triphosphate; ddThd, its dThd kinase-deficient derivative (Raji/TK-) has been described 2’,3’-dideoxythymidine; AZT-MP, AZT 5’-monophosphate; AZT- previously (24). DP, AZT 5’diphosphate; D4T-MP, D4T 5’-monophosphate; D4T- Viruses-Human immunodeficiency virus (HIV) was obtained DP, D4T 5“diphosphate. from the culture supernatant of H9 cells persistently infected with 6127 6128 Phosphorylation of D4T and AZT in MT-4 Cells HTLV-IIIB (derived from a pool of American patients with AIDS) pared from exponentially growing MT-4 cells, which were first washed and kindly provided by Dr. R. C. Gallo (National Cancer Institute, twice with phosphate-buffered saline at 4 “C and then homogenized Bethesda, MD) (25). Moloney murine sarcoma virus (MSV) was by sonication. The suspensions were clarified by centrifugation at prepared from tumors obtained in 3-day-old NMRI mice that were 100,000 X g for 60 min. In the enzyme experiments, [methyL3H] inoculated intramuscularly with the virus (26). Moloney murine leu- dThd, [methyL3H]AZT, or [meth~l-~H]D4T served as the radiola- kemia virus (MLV) was from Electronucleonics (Bethesda, MD). beled substrates. Substrate concentrations were as follows: for Compounds-3‘-Azido-2‘,3’-dideoxythymidine (AZT) and 2’,3’- [methyL3H]dThd, 5, 10, 25, and 50 p~; for [meth~l-~H]AZT, 2, 5, 10, didehydro-2’,3’-dideoxythymidine (D4T) were synthesized according 20, and 40 p~; and for [methyL3H]D4T, 20,40,100,200, and 250 p~. to previously published methods (6). Their formulae are shown in The K,,, and V,, values for the radiolabeled substrates were derived Fig. 1. The 5”triphosphate derivative of AzddThd was synthesized from Lineweaver-Burk plots, using a linear regression analysis pro- as described earlier (27,28). The 5”triphosphate of D4T was prepared gram. In the competition assays, [meth~l-~H]AZT and [methyL3H] following essentially the same synthetic procedure. The other re- D4T were incubated in the presence of 10 p~ unlabeled dThd or agents used were of the highest quality available. dTTP. The other experimental conditions used for the dThd kinase Radio~hemicals-[methyl-~H]AZT (specific radioactivity, 10 Ci/ assay were as previously described (31). mmol) and [rnethyL3H]D4T (specific radioactivity, 20 Ci/mmol) were Inhibition of the HIV- or MLV-associated reverse transcriptase by obtained from Moravek Biochemicals Inc., Brea, CA. [methyL3H] the 5”triphosphates of AZT (AZT-TP) and D4T (DIT-TP) was dThd (specific radioactivity, 40 Ci/mmol) was from the Radiochem- determined as previously described (27). Inhibition of reverse tran- ical Centre Amersham (Amersham, United Kingdom). scriptase by the test compounds was estimated during the period that Antiretrouiral Assays-The procedures for measuring anti-HIV the reverse transcriptase activity increased linearly. In the MLV activity in ATH8 and MT-4 cells and anti-MSV activity in C3H cells reverse transcriptase experiments, the endogenous viral RNA served have been described previously (2, 3, 22, 29, 30). as the template. In the HIV reverse transcriptase assays, exogenous Cytostatic Assays-Cytostatic effects of the compounds were as- poly(A) .oligo(dT)lz-la served as the template. DNA polymerase a was sessed by measuring inhibition of cell proliferation. The experimental obtained from calf thymus (Pharmacia, Uppsala, Sweden). The re- procedures have been described previously (3, 12). Briefly, exponen- action mixture (40 pl) contained 4 pl of activated DNA (0.5 mg/ml), tially growing MT-4, CEM, and H9 cells were seeded in 200-p1 4 p1 of bovine serum albumin (5 mg/ml), 2 pl of a mixture of dGTP, microplate wells and incubated in the presence of varying concentra- dCTP, and dATP, 2 mM), 4 pl of [meth~l-~H]dTTP (4 pCi; 30 p~), tions of the test compound (50,000 cells/well). After an incubation 6 p1 of distilled water, 4 pl of a mixture containing 200 mM Tris-C1 period of 3,4, and 5 days the cell number was determined in a Coulter (pH 7.5), 30 mM MgClZ, and 2 mM dithiothreitol, 8 pl of the appro- counter (12). ATH8 cells were seeded at lo5 cells/culture tube (2 ml/ priate inhibitor concentrations (final concentrations: 500, 100, 20, 4 tube) in the presence of the test compound, and after 7 days of pM) and 12 pl of enzyme (0.024 unit). Determination of the Lipophilicity of Test Compounds-To estimate incubation the number of living cells was determined in a blood cell the lipid solubility of 2‘-deoxythymidine (dThd), 2’,3’-dideoxythy- counting chamber using the trypan blue dye exclusion technique (3). Metabolism of [methyL3H1AZT and [meth~l-~H]D4T in Human midine (ddThd), D4T, and AZT, the partition of the compounds MT-4 and Raji Cells-The metabolism of the radiolabeled compounds between 1-octanol and 10 mM potassium phosphate buffer, pH 7.4, was monitored according to previously established procedures (16, was measured as previously described (30). Briefly, a 50 pM concen- 17). Briefly, the MT-4 cells were seeded at 2-4 X 10’ cells/ml and tration of the test compound in 10 mM potassium phosphate buffer incubated with varying concentrations of the radiolabeled compounds was thoroughly mixed with an equal volume of 1-octanol for 30 min. (see footnote to Table 111). At different time intervals (see footnote Then, the mixture was further equilibrated at room temperature for to Table IV), cells were centrifuged, washed, and precipitated with 60 min, UV absorption was measured for the aqueous and alcoholic cold trichloroacetic acid (10%). After centrifugation, the supernatants liquid phases, and the percentage of the test compound present in each liquid phase was calculated based on their absorption maxima. were neutralized with tri-n-octylamine in Freon. HPLC analysis of Alternatively, the RF values of dThd, ddThd, D4T, and AZT were the neutralized cell extracts was carried out using a Partisil-SAX determined by thin layer chromatography on Silica Gel M5735 with radial compression column. A linear gradient of 0.007 M ammonium a mixture of ch1oroform:methanol (92.5:7.5). dihydrogen phosphate (pH 3.80) to 0.25 M ammonium dihydrogen + 0.50 M KC1 (pH 4.5) was used. The different fractions Tumor Formation in NMRI Mice Inoculated Intramuscularly with phosphate Moloney Murine Sarcoma Virus (MSV)-NMRI mice were inoculated of the eluate were assayed for radioactivity in a toluene-based scin- tillant. at 2 days after birth with MSV (day 0). AZT and D4T were admin- Phosphorylation of [methyL3H]D4T and [methyL3H]AZT was also istered daily by the intraperitoneal route at a dose of 125 or 25 mg/ examined in Raji/O cells and its dThd kinase-deficient Raji/TK- kg/day, starting 2 h before virus inoculation (day 0) and continued until day 5. Tumor appearance and mortality associated herewith counterpart, and H9 cells. The procedure was essentially the same as that described for the MT-4 cells. were recorded daily. There were 10 mice per group. Effect of AZT and D4T on the Intracellular Radiolabeled [methyl- 3H]dTMP, [meth~l-~HIdTDP, and [meth~l-~HIdTTP Pools in MT-4 RESULTS Cells-MT-4 cells were seeded at 400,000 cells/ml in 5-ml tissue Antiretroviral and Cytotoxic Effects of AZT and D4T“AZT culture bottles for 24 h in the presence of different concentrations of and D4T were evaluated for their antiretroviral effects in AZT or D4T (0, 50, 500 p~). Then, 0.04 p~ [methyL3H]dThd (10 human MT-4 and ATH8 cells infected with HIV and murine pCi/5-ml tissue culture bottle) was added to the cell cultures, and 2 h, the trichloroacetic acid-soluble intracellular material was after C3H cells infected with MSV (Table I). Both AZT and D4T analyzed by thin layer chromatography in a mixture of isopropyl proved to be potent and highly selective inhibitors of HIV alcohol:NH3:HzO (63:l) and by HPLC using the same separation replication in MT-4 cells. AZT showed an ED50 (50% effective method as described above. dose) of 0.005 PM, whereas D4T was 10-fold less active. Since Enzyme Assays-Cell extracts containing dThd kinase were pre- the CDS0 (50% cytotoxic dose) of AZT and D4T was 8.1 and 19 PM, respectively, AZT could be considered as 3-fold more selective as an anti-HIV agent than D4T in MT-4 cells. In 0 A ..JJ.. ATH8 cells, however, AZT and D4T were about 480- and 80- fold less effective against HIV than in MT-4 cells, while their cytotoxic effects were diminished only by a factor of 5. In murine C3H cells AZT was 100-fold more effective in inhib- iting MSV cell transformation than was D4T (EDso, 0.023 and 2.1 PM, respectively), while neither compound was toxic for C3H cells at 200 PM (Table I). When the cytostatic effects of AZT and D4T were evaluated against a number of human T cell lines (i.e. MT-4, H9, CEM, FIG. 1. Structural formulae of 3’-azido-2’,3‘-dideoxythy- ATHS), remarkable differences were noted (Table 11). AZT midine (AxddThd, AZT) and 2’,3’-didehydro-2’,3’-dideoxy- was considerably more inhibitory to the growth of MT-4 and thymidine (ddeThd, Dan. Phosphorylation of D4T and AZT in MT-4 Cells 6129 TABLE I Inhibitory effects of AZT and D4T on HIV-induced cytopathogenicity in human MT-4 and ATH8 cells and MSV-induced transformation of murine C3H cells MT-4 cells C3H cells Compound EDm’ EDmc MTCd PM PM PM 0.023 f 0.01 >200 AZT 0.005 f 0.001 8.1 f 1.8 >200 19 f 3.6 2.1 f 0.7 D4T 0.05 f 0.001 a 50% effective dose, or dose required to inhibit HIV-induced destruction of MT-4 or ATH8 cells by 50%. 50% cytotoxic dose, or dose required to inhibit MT-4 or ATH8 cell proliferation by 50%. 50% effective dose, or dose required to inhibit MSV-induced transformation of C3H cells by 50%. Minimal toxic concentration, or lowest dose required to cause a microscopical alteration of normal C3H cell morphology. e Data taken from Ref. 3. TABLE I11 TABLE I1 Phosphorylation of 1 [methyl-3H]AZT and 1 p~ Imeth~l-~HI Cytostatic effects of AZT and D4T on MT-4, ATH8, H9, D4T in MT-4 cells as a function of incubation time and CEM cells ~~ ~ Phosphorylated products” 50% cytostatic dose Time of Compound Cell line incubation 5’-Monophosphate 5”Diphosphate 5’-Triphosphate nmlllOg celb PM PM AZT MT-4 53 f9 42 f 7 36f 10 AZT H9 3.5 0.68 CEM 582 f 73 >lo00 >lo00 ATH8 40 24 0.80 48 0.72 D4T MT-4 112 f 13 84 & 12 41 f 18 72 95 0.43 0.15 H9 264 f 70 243 f 19 228 f 13 CEM 138f 56 143 f 15 158f 29 D4T ATH8 110 0.06 0.23 7 0.29 a Data taken from Ref. 3. Due to the long time that the ATH8 cells required for proliferation, the inhibitory effects of the compounds on 48 0.24 ATH8 cell growth were measured only after 7 days of incubation. Data represent average values for two separate experiments. ATH8 cells than for CEM or H9 cells. The antiproliferative effect of AZT on MT-4 cells slightly increased upon longer incubation times (Table 11). Less striking differences were noted in the cytostatic effects of D4T on the different T cell lines. The CD, values of D4T for H9, CEM, and ATH8 cells ranged from 112 to 264 PM. The antiproliferative effect of D4T on MT-4 cells significantly increased upon longer incu- bation times (Table 11). Phosphorylation of [methyl-3HJAZT and [methyL3H]D4T in MT-4 Cells following Different Incubation Times-The metabolism of radiolabeled AZT and D4T by MT-4 cells was followed upon incubation of the cells with 1 PM radiolabeled compounds for 3.5, 7, 24, 48, and 72 h (Table 111). A typical HPLC chromatogram for the intracellular phosphorylation of [methyC3H]AZT and [methyL3H]D4T is depicted in Fig. 2. AZT was extensively metabolized to its 5”monophosphate derivative (AZT-MP) in MT-4 cells. Upon incubation with 1 PM AZT, the intracellular AZT-MP levels measured after 3.5- F r a c t i on nun brr 48 h ranged from 100 to 164 nmol/lOg cells and slightly decreased to 95 nmol/lOg cells after 72 h. In contrast, rela- FIG. 2. Ion exchange (Partisil-SAX) HPLC elution profile of a 10% trichloroacetic acid extract of MT-4 cells incubated tively small levels of the 5’-diphosphate (AZT-DP) (1.3-1.9 for 24 h with 1 p~ [methyZ-’H]AZT (M) or 1 phi [methyl- nmol/lOg cells) and 5’-triphosphate (AZT-TP) (0.68-0.80 *H]D4T (.---). The column was equilibrated and developed nmol/lOg cells) were detected within the first 48 h of incuba- with 0.007 M ammonium phosphate (pH 3.80) for 6 min, followed by tion; and after 72 h, the AZT 5‘-di- and 5”triphosphate levels a linear gradient to 0.25 M ammonium phosphate + 0.50 M KC1 (pH decreased to 0.43 and 0.15 nmol/lOg cells, respectively. The 4.5) over the next 20 min, and finally by 20-min isocratic elution with accumulation of AZT 5‘-monophosphate in MT-4 cells con- the latter buffer. trasts sharply with the low levels found for the D4T 5’- monophosphate (D4T-MP) (Table 111). The D4T 5”triphos- following 24-h incubation with MT-4 cells is presented. No phate (D4T-TP) levels were similar to those detected for metabolites other than the 5’-mono-, 5’-di-, and 5”triphos- D4T-MP. The intracellular DIT-TP levels were only 2-3-fold phates of AZT and D4T were detected. Identification of the lower than the AZT-TP levels after 24-48 h incubation, and 5’-mono-, 5‘-di-, and 5”triphosphate metabolites of AZT and the 72-h levels of DIT-TP were similar to the AZT-TP levels D4T was ascertained by spiking the cell extracts with the (Table 111). chemically synthesized 5’-mono- and 5”triphosphates of AZT In Fig. 2, the phosphorylation pattern of AZT and D4T and D4T. The radiolabeled peaks that were tentatively iden- 0.03 0.14 72 0.15 0.07 0.35 24 0.21 0.10 0.23 0.09 0.24 3.5 0.47 147 1.34 100 1.96 164 1.53 0.71 155 1.30 >loo0 >loo0 >loo0 Day3 Day4 Day5 Dayl“ 4.1‘ 110‘ 2.4‘ 40‘ CDmb EDw” CDmb cells ATHB 6130 Phosphorylation of D4T and AZT in MT-4 Cells TABLE IV Phosphorylation of Im~thyl-~HlAZT and [methyL3H1D4T in MT-4 cells upon 24-h incubation as a function of the input concentrations of the nucleosides Initial Compound concentration Nucleoside 5'-Monophosphate 5"Diphosphate 5"Triphosphate P'U nm01/10~ cells" AZT 50 5.6 1.6 10 35 377 4.0 1.1 1 .o 0.53 1.7 26 1.6 0.54 0.37 1.6 21 1.4 0.45 D4T 50 93 2.4 10 19 0.77 1.0 1.0 1.1 2.0 0.13 0.16 0.22 0.25 Data represent average values for two separate experiments. TABLE V TABLE VI Phosphorylation of 1 p~ [meth~l-~H]AZT and 1 p~ [rneth~l-~H] Effect of dThd and dCyd on the phosphorylation of 1 pM [methyL3H] D4T in human lymphoblast Raji/O and Raji/TK- cells upon 24-h AZT and 1 pM [rnethyL3H]D4T in MT-4 cells incubation 5'-Phosphorylated metabolites" Concentration of 5'- phosphorylated metabolites" Compound Upon addition of Compound 250~~ dThd+ 25aM 1OOOpM Raji/O RaiiiTK- dThd dCyd 1000 p~ dCyd nm01/10~ cells nmol/109 cells AZT 2.52 0.704 AZT 4.25 2.88 2.15 1.62 AZT-MP 31.7 0.053 AZT-MP 100 42.0 AZT-DP 1.01 C0.005 AZT-DP 1.96 1.12 AZT-TP AZT-TP D4T 1.91 1.77 D4T D4T-MP 0.038 0.025 D4T-MP 0.007 DIT-DP 0.011 0.006 D4T-DP 0.098 C0.005 0.005 0.006 D4T-TP DIT-TP C0.005 Data represent average values for two separate experiments. Values are the mean of two separate experiments. added to the cell extracts, the D4T 5'-tri- and 5"diphosphate peaks significantly diminished in size, whereas the D4T 5'- monophosphate peak increased (data not shown). Phosphorylation of [meth~l-~H/AZT and [methyL3H/D4T in MT-4 Cells as a Function of Different Input Concentra- tions-MT-4 cells were incubated with [ methyL3H]AZT and [methyL3H]D4T at concentrations ranging from 0.25 to 50 pM. Formation of the 5'-mono-, 5'-di-, and 5"triphosphate metabolites of AZT and D4T increased with higher initial concentrations of the nucleosides. While the intracellular levels of AZT-MP increased pro- portionally with increasing concentrations of the input AZT, or about 60-fold when the initial AZT concentration increased from 0.37 to 50 p~, AZT 5'-diphosphate and 5"triphosphate levels increased only by a factor of 4-5-fold under the same .4 I experimental conditions. In contrast, when the initial concen- QE Ql 0.2 f 1 ldThd 1-1 or l/AZT~--l(phf'Ij tration of D4T was raised from 0.25 to 50 p~, phosphorylated products (5'-mOnO-, 5'-di-, and 5"triphosphate) were formed " " "" + "" + "" ~ "" ~ "" dsam 0.02 a03 QW (105 at a 30-80-fold greater extent (Table IV). The D4T-TP levels I/ 0CTl~-*l(pM) were inferior to the AZT-TP levels if the initial concentration of the nucleosides was lower than 1 p~, equal to the AZT-TP FIG. 3. Double-reciprocal (Lineweaver-Burk) plots for levels if the initial nucleoside concentrations were 1-10 p~ dThd (o"o), AZT (W), and D4T (A---A) phospho- rylation by MT-4 cell extracts. and superior to the AZT-TP levels if the initial nucleoside concentrations were 50 pM. Essentially the same results were obtained in H9 cells (data not shown). tified as the 5'-mono- and 5"triphosphate of AZT and D4T Metabolism of [meth~l-~H/AZT and [methyL3H1D4T in coincided with the authentic 5'-mono- and 5'-triphosphates Raji/O and dThd Kinase-deficient Raji/TK- Cells-Phospho- of AZT and D4T on the chromatogram. Upon treatment of rylation of [ methyk3H]AZT and [ methyL3H]D4T was exam- the cell extracts with alkaline phosphatase, the radiolabeled ined in the human B-lymphoblast Raji/O cell line and its peaks identified as the phosphorylated derivatives of AZT dThd kinase-deficient counterpart Raji/TK- (Table V). and D4T disappeared while the nucleoside forms accumulated AZT was phosphorylated in Raji/O cells about 2-3-fold less (data not shown). Similarly, after phosphodiesterase had been 0.231 0.019 0.014 0.017 0.015 0.201 0.041 0.025 1.35 0.621 0.762 0.658 0.795 0.427 0.313 0.022 0.721 0.007 0.717 0.036 18.6 1.19 0.44 0.03 0.06 0.06 1.7 2.3 70 3.1 1.3 4.0 219 1212 Phosphorylation of D4T and AZT in MT-4 Cells 6131 TABLE VI1 Zntracellular [methyL3H]dTTP levels and ratios of intracellular [methyl-3HldTMP/([methyl-3HldTDP + [methyl- 3H]dTTP) levels in MT-4 cells preincubated with AZT and D4T prior to pulse labeling with 0.04 pM [methyl-3Hl dThd for 2 h Data represent average values for two to three separate experiments. [meth~l-~H)dTMP Ratio Intracellular levels of [methyL3H]dTTP Compound [methyL3H]dTDP + [methyL3H]dTTP concentration D4T AZT D4T AZT PM pmolllOQ cells 160 0.71 0.71 0 160 164 8.54 0.96 50 24.5 273 17.0 1.77 500 3.6 TABLE VI11 and D4T, it appeared that AZT was phosphorylated to a Inhibitory effects of AZT-TP and DIT-TP on the actiuity of MLV- similar extent as dThd, while D4T was phosphorylated about and HZV-associated reverse transcriptase and DNA polymerase a 200-fold less efficiently. 50% inhibitory concentration Effect of dThd and dCyd on the Phosphorylation of [methyl- 3H]AZT and [methyL3H]D4T in MT-4 Cells-As shown in DNA Compound Table VI, addition of 250 p~ dThd (in the presence of 1000 30-min polymerase a 30-min p~ dCyd to avoid cytotoxicity of dThd) resulted in a dramatic assay assay 30-min assay assay decrease of the levels of phosphorylated metabolites of AZT and D4T. Under our experimental conditions, the AZT 5‘- AZT-TP” 1.15 f 0.4 0.51 f 0.3 0.43 >500 mono- and 5”triphosphate levels decreased by 80- and 35- 175 f 117 D4T-TP 1.63 f 0.4 1.18 f 0.2 0.84 fold, respectively; the D4T 5’-monophosphate levels de- Data taken from Ref. 27. creased by 30-fold; and the D4T-TP levels even fell below the detection limit (0.005 nmol/lOg cells) (Table VI). Addition of TABLE IX a lower concentration of exogenous dThd (25 p~) or addition Partition coefficients (P) between 1-octanol and 10 mMpotassium of 1000 p~ dCyd also resulted in a significant but less marked Rf values in chloroform:methanol(92.5:7.5) on phosphate buffer and diminution of the phosphorylated AZT and D4T pools than silica gel TLC for dThd, ddThd, D4T. and AZT that observed with 250 p~ dThd. Compound RF* P” Effect of Preincubation of MT-4 Cells with AZT and D4T dThd 0.124 f 0.059 0.067 f 0.005 on the Intracellular Phosphorylation of 0.04 phi [methyL3HI ddThd 0.339 f 0.088 0.233 f 0.005 dThd-MT-4 cells were incubated for 24 h with different D4T 0.325 f 0.091 0.154 f 0.008 concentrations of AZT or D4T prior to a pulse labeling of the AZT 0.431 & 0.082 0.964 f 0.038 cells with 0.04 p~ [meth~l-~H]dThd (10 pCi/4 X lo6 cells/ “Data represent average values (fS.D.) for at least two to three culture) for 2 h. Then, the intracellular ratios of [methyL3H] separate experiments. dTMP/( [ methyL3H]dTDP + [ methyL3H]dTTP) were deter- mined. Without inhibitor, the ratio of radiolabeled dTMP/ efficiently than in MT-4 cells, and D4T was phosphorylated (dTDP + dTTP) was 0.71 (Table VII). This corresponds to 10-fold less efficiently in Raji/O than MT-4 cells (compare 160, 32, and 137 pmol of [meth~l-~HIdTTP, [methyL3H] the data in Tables IV and V). In the dThd kinase-deficient dTDP, and [methyL3H]dTMP per lo9 MT-4 cells, respec- cell line Raji/TK-, AZT was 500-1000-fold less extensively tively. However, increasing initial concentrations of AZT phosphorylated to its 5’-mOnO-, 5‘-di-, and 5”triphosphate resulted in a dramatic increase of the [meth~l-~H]dTMP/ than in Raji/O cells. AZT-DP levels were even under the ( [methyb3H]dTDP + [ methyL3H]dTTP) ratio (17 in the pres- detection limit (0.005 nmol/lOg cells). In contrast, D4T phos- ence of 500 p~ AZT). In contrast, 500 p~ D4T afforded a phorylation proceeded at about the same rate in Raji/TK- as slight increase of the ratios of radiolabeled dTMP/(dTDP + in Raji/O cells (Table V). These data suggest a crucial role for dTTP) (Table VII). cellular dThd kinase in the phosphorylation of AZT but not Inhibitory Effects of AZT-TP and DIT-TP on HIV-associ- D4T in Raji cells. ated Reverse Transcriptase and DNA Polymerase a-AZT- Phosphorylation of Lmeth~l-~HJdThd, [methyL3HJAZT, and TP and D4T-TP were evaluated for their inhibitory effects [methyL3H]D4T by MT-4 Cell Extracts-Different concentra- on the reverse transcriptases of MLV and HIV and DNA tions of [ methyL3H]dThd, [ methyL3H]AZT, and [ meth~l-~H] polymerase a. The initial concentration of [ methyL3H]dTTP D4T were incubated with MT-4 cell extracts under assay in the reaction mixture was 1 p~. AZT-TP and D4T-TP were conditions stable for optimal dThd kinase activity. Michaelis- strongly inhibitory to the reverse transcriptases. Their 50% Menten constants (K,) and maximal velocity ( Vmax) values inhibitory concentrations were in the range of 0.5-1.5 pM for were calculated from Lineweaver-Burk diagrams (Fig. 3). the MLV reverse transcriptase and in the range of 0.02-0.1 dThd and AZT were phosphorylated to a comparable extent p~ range for HIV reverse transcriptase (Table VIII). AZT- by the MT-4 cell extracts, while D4T was phosphorylated much less efficiently. The V,,, values for phosphorylation of MP had no effect on the enzyme, even at a concentration of 1000 p~ (27). AZT-TP and D4T-TP were considerably less dThd, AZT, and D4T were 190, 153, and 8.2 nmol/mg of protein/h, respectively. The K, values were 13.5, 13.8, and inhibitory to DNA polymerase a than MLV and HIV reverse 142 p~, respectively. Addition of 10 p~ dThd or 10 ~LM dTTP transcriptase. Their 50% inhibitory concentrations were >500 considerably decreased the phosphorylation of radiolabeled and 175 pM, respectively. AZT or D4T by MT-4 cell extracts (data not shown). When Lipophilicity of AZT and D4T-To estimate the lipid solu- the V,,./K, ratio (14, 11, and 0.06, respectively) was taken bility of AZT, D4T, dThd, and ddThd, their RF values were as a parameter for the phosphorylation rate of dThd, AZT, determined by thin layer chromatography in a mixture of PM PM PM PM MLV HIV 6132 Phosphorylation of D4T and AZT in MT-4 Cells ch1oroform:methanol (Table IX). The RF values of AZT, D4T, D4T and AZT phosphorylation by MT-4 cell extracts. This dThd, and ddThd were 0.43,0.33,0.12, and 0.34, respectively. suggests that dThd kinase is responsible for the phosphoryl- The partition of the test compounds between 1-octanol and ation of AZT and D4T. However, the phosphorylation pat- potassium phosphate buffer was also measured (Table IX). terns obtained for D4T and AZT in Raji/O and Raji/TK- cells AZT was far more lipophilic than D4T (partition coefficient implicate dThd kinase in the activation of AZT but not D4T. (P):0.964 and 0.154, respectively). The lipophilicity of D4T Furthermore, D4T is equally cytostatic for Raji/O and Raji/ was higher than that of dThd but lower than that of ddThd TK- cells and inhibit Simian AIDS-related retrovirus (SRV)- (Table IX). induced syncytium formation almost to a similar extent in Raji/O and Raji/TK- cells (data not shown). These observa- DISCUSSION tions, again, argue against the role of dThd kinase in the The 2’,3’-dideoxythymidine analogues AZT and D4T are phosphorylation of D4T. Thus, it is unclear which enzyme(s) potent inhibitors of HIV and MSV replication in vitro and are responsible for the phosphorylation of D4T. Whatever show a marked “therapeutic index” in MT-4 cells (ratio of enzyme is responsible, it is highly susceptible to inhibition by the compound concentration required to inhibit cell growth dThd as well as dTTP. by 50% to the concentration required to inhibit virus repli- The phosphorylation data obtained in MT-4 extracts are cation by 50%). Although structurally related, AZT and D4T in agreement with our observation that (i) phosphorylation show a totally different pattern of intracellular phosphoryla- of D4T and AZT in intact MT-4 cells is severely suppressed tion. In contrast to AZT that accumulates mainly as its 5’- by dThd, and (ii) both the cytostatic and antiretroviral activ- monophosphate resulting in relatively small levels of the 5’- ity of D4T and AZT in MT-4 cells are reversed by the addition di- and 5’-triphosphate derivatives, D4T does not accumulate of dThd. The latter findings also suggest that a phosphoryl- as it 5’-monophosphate. Due to the low extent by which D4T ated product of D4T and AZT, presumably their 5”triphos- is initially phosphorylated to its 5’-monophosphate, the D4T- phate derivatives, accounts for the biological activity of the TP levels that are eventually achieved are lower than the compounds. One of the factors that may contribute to the AZT-TP levels, at least when starting with an input nucleo- decreased intracellular formation of phosphorylated D4T side concentration of 1 p~ or less. However, when the initial products in the presence of dThd (and dCyd) may be the dose of AZT and D4T was increased, anabolism of D4T to competition of dThd with D4T for the catalytic site of D4T D4T-TP was facilitated to a relatively greater extent over kinase. Additional inhibition may occur after the conversion AZT-TP formation from AZT, so that at input nucleoside of dThd to dTTP, because dTTP could act as a potent concentrations greater than 10 PM the levels obtained for (allosteric) inhibitor of the putative D4T kinase. Also, inter- D4T-TP exceeded those of AZT-TP. ference with the cellular transport (uptake) of D4T may be Furman et al. (15) demonstrated that AZT-MP accumula- taken in consideration to explain the decreased anabolism of tion causes a severe inhibition of dTMP kinase resulting in D4T in the presence of dThd. Should this be the case, how- an efficient blockage of its further phosphorylation to AZT- ever, it is surprising that 25 p~ dThd would afford a similar DP. From our findings, one may assume that when the initial inhibitory effect on the D4T transport system as 1000 PM AZT concentrations are higher than 1 p~, dTMP kinase of dCyd. MT-4 cells must be blocked to a significant extent. Conse- AZT is active as an inhibitor of HIV replication in MT-4 quently, increase of the initial AZT concentration from 1 to cells and of C3H cell transformation by MSV at lower con- 50 p~ would not lead to a proportional increase in the centrations than is D4T. Since AZT-TP and D4T-TP are formation of AZT-DP and AZT-TP. D4T is apparently unable equally inhibitory to the HIV- and MLV-associated reverse to block its own phosphorylation, and, hence, increase in the transcriptases, this greater potency of AZT against HIV rep- initial D4T concentration results in a fairly proportional lication can most probably be attributed to its more efficient increase in the levels of all phosphorylated products of D4T phosphorylation within the cells to the 5”triphosphate. In- (Table IV). Additional evidence that D4T-MP unlike AZT- deed, higher 5’4riphosphate levels are achieved within MT-4 MP does not significantly inhibit dTMP kinase is obtained cells for AZT than for D4T if the nucleosides are added to from our pulse-labeling experiments in which MT-4 cells were the cells at a concentration of 0.25-0.37 p~. Due to the preincubated with AZT or D4T and then labeled with [methyl- relatively low specific radioactivity of both compounds, phos- 3H]dThd. A block at the level of dTMP kinase should result phorylation of the compounds at lower (antivirally active) in an increase of intracellular radiolabeled dTMP and a concentrations could not be accurately determined. concomitant decrease of the radiolabeled dTTP pools com- The observation that HIV reverse transcriptase has a very pared to control. In contrast to AZT, D4T did not lead to a high affinity for AZT-TP and D4T-TP, this affinity being significant accumulation of [ methyL3H]dTMP and depletion several orders of magnitude higher than the affinity for cel- of [methyG3H]dTTP pools. The fact that D4T, unlike AZT, lular DNA polymerase a, may explain the selectivity and does not interfere with dTMP kinase and thus fails to affect potency of both AZT and D4T as an antiretroviral chemo- the salvage pathway of dThd, may be clinically advantageous therapeutic agent in vitro. In this respect, D4T-TP proved 3- in that a shut-off of this pathway, because of the ensuing fold more inhibitory to DNA polymerase a than AZT-TP. reduction in dTTP pool levels, may lead to substantial cyto- In view of the propensity of HIV to infect and damage the toxicity. central nervous system, potential treatment schedules for It has been established that AZT and dThd are phospho- AIDS patients should address the ability of candidate anti- rylated equally well by the cytosol dThd kinase (15, 16). The AIDS drugs to cross the blood brain barrier. Preliminary K, and Vmax values of dThd kinase are very similar for both evidence suggests that AZT favorably influences the course compounds. We now find that D4T is phosphorylated by MT- of neurological disorders associated with AIDS (32). Com- 4 cell extracts at a much higher K,,, and with a significantly pounds with high lipophilicity may be expected to cross the lower Vmax than AZT and dThd. The phosphorylation of D4T blood brain barrier more easily than polar compounds. AZT in MT-4 cells is sensitive to inhibition by dThd, and so is the proved clearly more lipophilic than dThd (P, 0.964 and 0.067, phosphorylation of AZT. dTTP, a well known allosteric in- respectively). D4T showed a higher partition coefficient than hibitor of dThd phosphorylation by dThd kinase, also inhibits dThd, but its lipophilicity was inferior to that of AZT. Also, Phosphorylation of D4T and AZT in MT-4 Cells 6133 10. Lin, T.-S., Schinazi, R. F., and Prusoff, W. H. (1987) Biochem. D4T had a higher partition coefficient value than ddCyd (P, Phurmncol. 36,2713-2718 0.050). Since it has been established that dThd as well as 11. Schinazi, R. F., Chu, C. K., Ahn, M.-K., and Sommadossi, J.-P. ddCyd cross the blood brain barrier (33, 34), one may specu- (1987) J. Cell. Biochem. Suppl. 11D, 74 late from our lipophilicity data that D4T also does so. 12. Balzarini, J., Baba, M., Pauwels, R., Herdewijn, P., and De Clercq, Finally, we found that D4T is considerably less active than E. (1988) Biochem. Phurmacol. 37, 2847-2856 13. Matthes, E., Lehmann, Ch., Scholz, D., Rosenthal, H. A., and AZT as an antiretroviral agent in mice. Treatment of the Langen, P. (1988) Biochim. Biophys. Res. Commun. 153,825- MSV-infected newborn mice for 5 subsequent days with AZT at 125 mg/kg/day increased the mean tumor initiation time 14. Balzarini, J., Matthes, E., Meeus, P., Johns, D. G., and De Clercq, by 2-fold, and resulted in a 71% survival of the mice at day (1988) Proceedings of the VI International Symposium on E. 25. In contrast, D4T treatment at 125 and 25 mg/kg/day only Human Purine & Pyrimidine Metabolism, Hakone, Japan, July resulted in a modest delay of tumor formation, without a 17-21, 1988, in press 15. Furman, P. A,, Fyfe, J. A., St. Clair, M. H., Weinhold, K., Rideout, dramatic effect on the increase of the survival rate of the mice J. L., Freeman, G. A., Nusinoff Lehrman, S., Bolognesi, D. P., (data not shown). This decreased efficiency may, at least in Broder, S., Mitsuya, H., and Barry, D. W. (1986) Proc. Natl. part, be related to the low efficiency by which D4T is con- Acad. Sci. U. S. A. 83,8333-8337 verted by the cellular phosphorylating enzymes to its 5’- 16. Balzarini, J., Pauwels, R., Baba, M., Herdewijn, P., De Clercq, monophosphate and eventually 5’-triphosphate in the murine E., Broder, S., and Johns, D. G. (1988) Biochem. Phurmacol. model. The in vivo data are consistent with the in vitro data 37,897-903 obtained in MSV-infected murine C3H cells. D4T was 100- 17. Cooney, D. A., Dalal, M., Mitsuya, H., McMahon, J. B., Nadkarni, M., Balzarini, J., Broder, S., and Johns, D. G. (1986) Biochem. fold less efficient in the latter system than AZT. Pharmncol. 35,2065-2068 In conclusion, this is the first report in which the cellular 18. Balzarini, J., Cooney, D. A,, Dalal, M., Kang, G-J., Cupp, J. E., metabolism and kinetic properties of D4T were investigated S., and Johns, D. G. (1987) Mol. Phur- De Clercq, E., Broder, and compared to those of AZT. A close correlation was found mmol. 32,798-806 between the antiviral and cytostatic activity of the compounds 19. Starnes, M. C., and Cheng, Y. (1987) J. Biol. Chem. 262, 988- and their metabolism to their 5”triphosphate form. D4T has 991 20. Langen, P., Kowollik, G., Etzold, G., Venner, H., and Reinert, H. unique metabolic features in that it does not accumulate as (1972) Acta Biol. Med. Ger. 29,483-494 its 5’-monophosphate and generates similar levels of 5‘- 21. Mitsuya, H., Guo, H.-E., Cowman, L., Megson, M., Reitz, M. S., mono-, 5’-di-, and 5”triphosphate. The fact that D4T does Jr., and Broder, S. (1984) Science (Wmh. D.C.) 225, 1484- not seem to block dTMP kinase may give D4T a potential edge over AZT. 22. Pauwels, R., De Clercq, E., Desmyter, J., Balzarini, J., Goubau, P., Herdewijn, P., Vanderhaeghe, H., and Vandeputte, M. Acknowledgments-We thank Ann Absillis, Lizette van Berckelaer, (1987) J. Virol. Methods 16, 171-185 23. Harada, S., Koyanagi, Y., and Yamamoto, N. (1985) Science 229, Miette Stuyck, Ria Van Berwaer, and Luk Kerremans for excellent technical assistance. The dedicated editorial help of Christiane Cal- 563-566 lebaut is highly appreciated. 24. Balzarini, J., De Clercq, E., Torrence, P. F., Mertes, M. P., Park, J. S., Schmidt, C. L., Shugar, D., Barr, P. J., Jones, A. S., Verhelst, G., and Walker, R. T. (1982) Biochem. Phurmacol. Note Added in Proof-After this paper was submitted, a commu- 31,1089-1095 nication with regard to the cellular pharmacology of D4T appeared 25. Popovic, M., Sarngadharan, M. G., Read, E., and Gallo, R. C. in Biochemical Pharmmology (35). (1984) Science (Wash. D.C.) 224,497-500 26. De Clercq, E., and Merigan, T. C. (1971) Proc. SOC. Exp. Biol. REFERENCES Med. 137,590-594 1. Mitsuva, H., and Broder, S. (1986) Proc. Natl. Acud. Sci. U. S. A. 27. Balzarini, J., Herdewijn, P., Pauwels, R., Broder, S., and De 83,-1911-.1915 Clercq, E. (1988) Biochem. Phurmacol. 37, 2395-2403 2. Balzarini. J.. Pauwels. R.. Herdewiin, P.. De Clercq, E., Cooney, 28. Hoard, D. E., and Ott, D. G. (1965) J. Am. Chem. SOC. 87,1785- D. A., Kang, G.-J.,’ Dalal, M., johns; D. G., and Broder, S. (1986) Biochem. Biophys. Res. Commun. 140,735-742 29. Balzarini, J., Pauwels, R., Baba, M., Robins, M. J., Zou, R., 3. Balzarini, J., Kang, G.-J., Dalal, M., Herdewijn, P., De Clercq, Herdewijn, P., and De Clercq, E. (1987) Biochem. Biophys. Res. E., Broder, S., and Johns, D. G (1987) Mol. Phurmacol. 32, Commun. 145,269-276 162-167 30. Balzarini, J., Baba, M., Pauwels, R., Herdewijn, P., Wood, S. G., 4. Lin, T.-S., Schinazi, R. F., Chen, M. S., Kinney-Thomas, E., and Robins, M. J., and De Clercq, E. (1988) Mol. Phurmmol. 33, Prusoff, W. H. (1987) Biochem. Pharmmol. 36, 311-316 243-249 Mitsuya, H., Weinhold, K. J., Furman, P. A., St. Clair, M. H., 31. Balzarini, J., De Clercq, E., Mertes, M. P., Shugar, D., and Nusinoff Lehrman, S., Gallo, R. C., Bolognesi, D., Barry, D. Torrence, P. F. (1982) Biochem. Phurmmol. 31, 3673-3682 W., and Broder, S. (1985) Proc. Natl. Acud. Sci. U. S. A. 82, 32. Yarchoan, R., Klecker, R. W., Weinhold, K. J., Markham, P. D., 7096-7100 Lyerly, H. K., Durack, D. T., Gelmann, E., Nusinoff Lehrman, 6. Herdewijn, P., Balzarini, J., De Clercq, E., Pauwels, R., Baba, S., Blum, R. M., Barry, D. W., Shearer, G. M., Fischl, M. A., M., Broder, S., and Vanderhaeghe, H. (1987) J. Med. Chem. Mitsuya, H., Gallo, R. C., Collins, J. M., Bolognesi, D. P., 30,1270-1278 Myers, C. E., and Broder, S. (1986) Lancet 1, 575-580 7. Balzarini, J., and Broder, S. (1988) in Clinical Use of Antiviral 33. Zaharko, D. S., Bolten, B. J., Chiuten, D., and Wiernik, P. H. Drugs (De Clercq, E., ed) pp. 361-385, Martinus Nijhoff Pub- (1979) Cancer Res. 39,4777-4781 lishing, Norwell, MA 34. Yarchoan, R., Perno, C. F., Thomas, R. V., Klecker, R. W., Allain, 8. Nakashima, H., Matsui, T., Harada, S., Kobayashi, N., Matsuda, J.-P., Wills, R. J., McAtee, N., Fischl, M. A,, Dubinsky, R., A., Ueda, T., and Yamamoto, N. (1986) Antimicrob. Agents McNeely, M. C., Mitsuya, H., Pluda, J. M., Lawley, T. J., Chemother. 30,933-937 Leuther, M., Safai, B., Collins, J. M., Myers, C. E., and Broder, 9. Baba, M., Pauwels, R., Herdewijn, P., De Clercq, E., Desmyter, S. (1988) Lancet 1, 76-81 J., and Vandeputte, M. (1987) Biochem. Biophys. Res. Commun. 35. August, E. M., Marongiu, M. E., Lin, T.-S., and Prusoff, W. H. 142, 128-134 (1988) Biochem. Phurmmol. 37, 4419-4422

Journal

Journal of Biological Chemistry – Unpaywall

Published: Apr 1, 1989

There are no references for this article.