Adenine, a Selective Anti-Human Immunodeficiency Virus Agent

Proc. Nati. Acad. Sci. USA Vol. 86, pp. 332-336, January 1989 Medical Sciences Marked in vivo antiretrovirus activity of 9-(2-phosphonylmethoxy- ethyl)adenine, a selective anti-human immunodeficiency virus agent (acquired immunodeficiency syndrome/antiviral chemotherapy) JAN BALZARINI*t, LIEVE NAESENS*, PIET HERDEWUN*, IVAN ROSENBERGt, ANTONIN HOLYt, RUDI PAUWELS*, MASANORI BABA*, DAVID G. JOHNS§, AND ERIK DE CLERCQ* *Rega Institute for Medical Research, Katholieke Universiteit Leuven, Minderbroedersstraat 10, B-3000 Leuven, Belgium; tInstitute of Organic Chemistry and Biochemistry, Czechoslovak Academy of Sciences, 16610 Prague, Czechoslovakia; and §Developmental Therapeutics Program, Division of Cancer Treatment, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892 Communicated by R. C. Gallo, October 6, 1988 (receivedfor review July 1, 1988)

ABSTRACT 9-(2Phophonylmethoxyethyl)adenne (PMEA) in vivo. No animal model is known in which HIV infection leads is a potent and selective inhibitor of the replication of human to severe immunodeficiency and animal death. Recently, a immunodeficiency virus (HIV) in vitro in human T-Iymphocyte variety of murine, feline, and simian retrovirus test systems MT4, H9, and ATH8 cells. PMEA also inhibits Moloney munne have been proposed for the in vivo evaluation of promising sarcoma virus (Mo-MSV)-induced transformation ofmurine C3H anti-HIV compounds (17-22); in several of these models the embryo fibroblasts. Moreover, PMEA causes a dose-dependent antiretrovirus activity ofAZT has been assessed. Tavares et al. suppression of tumor formation and assocated mortality in mice (17) found that AZT prevents Rickard feline leukemia virus inoculated with Mo-MSV. At a dose of 50 or 20 mg/kg per day infection in 6-week-old kittens ifadministered immediately after PMEA effected a 90-100% protection of the mice against Mo- infection. Ruprecht et al. (19) found that in mice infected with MSV-induced tumor formation and mortality. Even with a PMEA Rauscher murine leukemia virus AZT prevented development dose as low as 1 to 5 mg/kg per day, tumor formation was of splenomegaly and suppressed viremia when treatment was signifiantly delayed and the survival rate was siantly en- started soon after virus inoculation. In this study, AZT was hanced. In parallel experiments, azidothymidine exhibited a com- given i.p. at daily doses of60, 120, and 240 mg/kg (or 0.5 to 1.0 parable inhibitory effect on Mo-MSV-induced tumor formation mg/ml in drinking water). and associated death only at a 25-fold higher dose than PMEA. Our group reported that AZT was endowed with remarkable Because PMEA has stronger in vivo andretrovirus potency and antiretrovirus activity when given i.p. at daily doses of 125, 25, selectivity than azidothymidine and various other compounds and 5 mg/kg ofbody weight to newborn NMRI (Naval Medical currently being subjected to clinical trials, PMEA studies should be Research Institute) mice infected with Moloney murine sar- pursued to assess the potential of this compound in the treatment coma virus (Mo-MSV) (22). This murine retrovirus, when of acqured immunodeficiency syndrome (AIIDS) and other retro- injected s.c. in the left hind leg of 2- to 3-day-old NMRI mice, virus infections in humans. causes tumor formation within 4 to 5 days and death within 10 to 12 days after infection. Both tumor appearance and the Acquired immunodeficiency syndrome (AIDS) is caused by associated mortality are functions of the virus inoculum size. the retrovirus, human immunodeficiency virus (HIV) (1, 2). Both Mo-MSV-induced tumor formation in vivo and Mo- Because an urgent need exists to find an effective chemo- MSV-induced cell transformation in vitro depend on the viral therapeutic drug against AIDS, many substances have been replicative power, as does the cytopathogenicity of HIV for evaluated for antiretrovirus activity in vitro, and some human T4 lymphocytes in vitro. We therefore considered compounds reported to be effective against HIV replication Mo-MSV-induced tumor formation in vivo and associated in vitro are currently being clinically investigated for AIDS mortality an appropriate model to assess the in vivo antiretro- treatment (3-15). So far, only 3'-azido-2',3'-dideoxythy- virus activity of selected anti-HIV agents. midine (N3ddThd, AZT, Zidovudine) has unambiguously Recently, we described a series ofacyclic purine derivatives demonstrated clinical benefit (7-9). In a double-blind, place- endowed with antiretrovirus and broad-spectrum anti-DNA bo-controlled trial, 282 patients with AIDS or AIDS-related virus activity (23, 24). Among them, we found that the acyclic complex (ARC) were randomly assigned to receive either adenosine derivative PMEA [9-(2-phosphonylmethoxyeth- AZT (250 mg) or placebo by mouth every 4 hr for a total yl)adenine] (Fig. 1) inhibits HIV-induced cytopathogenicity in period of 24 weeks. When the study was completed, 19 MT4 cells and HIV antigen expression in H9 cells at concen- placebo recipients (of 137) and one AZT recipient (of 145) had trations ('1.6-2 ,uM) significantly below the toxicity threshold died. A decreased frequency of opportunistic infections, for the host cells ("40-67 AM) (Table 1 and data not shown) increased base-line Karnofsky performance score, increased (24). PMEA also proved effective in inhibiting Mo-MSV- weight, and a significant increase in the number of CD4 cells induced transformation of murine C3H embryo fibroblasts in were noted in the AIDS or ARC subjects receiving AZT (9). vitro; it did so at a concentration similar to that found inhibitory However, serious adverse reactions-particularly bone mar- to HIV replication in MT4 and H9 cells. row suppression (megaloblastic anemia)-but including also We then evaluated PMEA for its antiretrovirus effect in headache, confusion, anxiety, nausea, and insomnia were Mo-MSV-infected newborn NMRI mice and included in this observed, indicating that AZT is not an innocuous drug (16). study a number of anti-HIV agents that are currently in These results further emphasize the need for more efficacious and selective antiretrovirus agents. Abbreviations: HIV, human immunodeficiency virus; Mo-MSV, A major problem in screening compounds proven active Moloney murine sarcoma virus; AIDS, acquired immunodeficiency against HIV in vitro is the lack ofany suitable HIV test system syndrome; PMEA, 9-(2-phosphonylmethoxyethyl)adenine; AZT, 3'- azido-2' ,3'-dideoxythymidine; ddCyd, 2' ,3'-dideoxycytidine; ddAdo, 2',3'-dideoxyadenosine; ATA, aurintricarboxylic acid; The publication costs of this article were defrayed in part by page charge CCID50, cell culture 50%6 infective dose; ED50, 50%o effective dose; payment. This article must therefore be hereby marked "advertisement" CD50, 50%o cytotoxic dose. in accordance with 18 U.S.C. §1734 solely to indicate this fact. tTo whom reprint requests should be addressed. 332 Downloaded by guest on September 29, 2021 Medical Sciences: Balzarini et al. Proc. Natl. Acad. Sci. USA 86 (1989) 333

NH2 Inhibitory Effects of Test Compounds on the Initiation of Mo-MSV-Induced Tumor Formation in NMRI Mice and on the Survival of Mo-MSV-Inoculated NMRI Mice. Two- to three- day-old NMRI mice (weighing -2 g) were inoculated s.c. in the as N left hind leg with 50 ,ul of Mo-MSV (100 focus-forming units measured by in vitro determination of the virus-induced trans- formation ofmurine C3H embryo fibroblast cells). At 4 to 5 days LM2I 0I l 11 after infection, tumors develop and rapidly increase in volume CH2OCH2- P- OH upon further aging of the mice. Within 10 to 12 days after I infection, mice (then weighing -5 to 6 g) die from the viral OH infection. Drug treatment was started 2 to 3 hr before injection of the and further administration was FIG. 1. Formula of PMEA. virus, compound given daily i.p. for an additional 9 days. The mean day of tumor clinical trial or considered for clinical studies-i.e., AZT, initiation (± SD) and the mean day of survival of the mice (± 2' ,3 '-dideoxycytidine (ddCyd), 2' ,3 '-dideoxyadenosine SD) were calculated, and the statistical significance of the (ddAdo), ribavirin, fusidic acid, and suramin. average delay oftumor formation and the mean day of survival in the treated groups versus the untreated (control) group was assessed the two-tailed Student's t test. MATERIALS AND METHODS by Inhibitory Effects of Test Compounds on HIV-Induced RESULTS Cytopathogenicity in MT-4 Cells. The procedure to determine Antiretrovirus Effect of PMEA and Other Test Compounds the anti-HIV activity in MT-4 cells has been described (25). in Vitro. Of the compounds listed in Table 1, AZT emerged Briefly, HIV-infected MT-4 cells were seeded at 5 x 104 cells as the most potent and selective inhibitor of HIV replication into 0.32-cm2 wells ofa 96-well microplate containing various in MT-4 cells (ED50, 0.004,.M), followed by ddCyd (ED50, concentrations of the test compounds. Virus input was 100 0.6 1LM) and Evans blue (ED50, 1.5 MtM). With an ED50 of 2 cell culture 50% infective doses (CCID50) per well. After a MtM, PMEA was slightly superior to ddAdo and aurintricar- 5-day incubation at 370C, the number of viable cells was boxylic acid (ATA) (ED50, 5-10 MM), whereas suramin determined in parallel for both HIV- and mock-infected cells showed a 50% inhibitory effect at 32 MLM. Ribavirin as well as in a blood cell-counting chamber using the trypan blue fusidic acid were inactive as anti-HIV agents at subtoxic dye-exclusion method. The 50% effective dose (ED50) was concentrations. These data agree with previously reported defined as the concentration of compound that protected values for ribavirin in HIV-infected ATH8 cells (26) and for HIV-infected cells by 50%, whereas the 50% cytotoxic dose fusidic acid in HIV-infected H9 cells (unpublished data). AZT (CD50) corresponded to the concentration of compound that and PMEA were equally potent in HIV-infected MT-4 cells reduced the number of viable mock-infected cells by 50%. and MoSV-infected C3H cells. However, ddCyd, ddAdo, Inhibitory Effects of Test Compounds on Mo-MSV-Induced and the anionic substances suramin, ATA, and Evans blue Transformation of Murine C3H Embryo Fibroblasts. The were invariably more active in the human HIV/MT-4 than anti-Mo-MSV assay was done as described (24). Briefly, the murine Mo-MSV/C3H assay system. murine C3H-3T3 embryo fibroblast cells were seeded at 5 x Effect of PMEA and Other Test Compounds on Tumor 105 cells per ml into 1-cm2 wells of a 48-well microplate. Formation in Mo-MSV-Infected Newborn NMRI Mice. Treat- Twenty-four hours later, the cell cultures were infected with ment of the Mo-MSV-infected mice with PMEA at a dose 80 focus-forming units of Mo-MSV for 60-90 min at 37°C. ranging from 1 to 50 mg/kg per day significantly delayed Then the medium was replaced by 1 ml of fresh culture tumor formation (P < 0.0005) (Table 2, Fig. 2A). At a PMEA medium containing various concentrations of the test com- dose of 20 or 50 mg/kg per day, no tumor development (till pounds. After 6 days, transformation of the cell cultures was 25 days after infection) was seen in 83% and 95% of mice, examined microscopically. The ED50 was defined as the respectively. In parallel with PMEA AZT was also investi- concentration of compound that inhibited Mo-MSV-induced gated at different daily doses (625, 125, 25, and 5 mg/kg). In cell transformation by 50%. contrast to PMEA, which increased the mean tumor initiation time by more than 2-fold at a dose as low as 5 mg/kg per day, Table 1. Inhibitory effects of test compounds on HIV-induced AZT had this effect only at a dose of 125 mg/kg per day-that cytopathogenicity in MT-4 cells and Mo-MSV-induced is, at a 25-fold higher dose than PMEA. Moreover, at an AZT transformation of murine C3H embryo fibroblasts dose of 25 or 125 mg/kg per day, no Mo-MSV-infected mice Mo-MSV- were protected against tumor initiation (mean day of tumor infected C3H initiation, 8.1 and 10.2 days, respectively). These data clearly HIV-infected MT-4 cells cells indicate that PMEA was superior to AZT as an antiretrovirus agent in this in vivo model. Compound ED50a, ,uM CD50 , AM ED50a, M As compared with PMEA and AZT, the other test com- PMEA 2.0 ± 0.4 67 ± 4 2.70 ± 0.45 pounds proved much less active, if not inactive: ddAdo and AZT 0.004 ± 0.001 8.0 ± 1.8 0.036 ± 0.013 ddCyd effected a 40-55% increase in the mean day of tumor ddCyd 0.60 ± 0.07 27 ± 4.4 29.0 ± 8.8 initiation when administered at a dose of 625 mg/kg per day ddAdo 9.2 ± 4.9 >500 16.8 ± 2.85 and 250 mg/kg per day, doses that were very close to the Ribavirin >20 26 ± 13 20.4 ± 7.2 toxicity threshold. Lower doses (i.e., 125 mg/kg per day for Fusidic acid >20 46 ± 12 >200 ddAdo and 62.5 mg/kg per day for ddCyd) were without a Suramin 32 ± 16 >500 150 ± 4 marked antiviral effect. Ribavirin had a modest inhibitory ATA 5.6 ± 2.4 >500 134 ± 5 effect on Mo-MSV-induced tumor formation (40% increase in Evans blue 1.5 ± 0.27 196 ± 28 >200 the mean day of tumor initiation) at 50 mg/kg per day (P < aED50 was defined as the concentration of compound that protected 0.0005). A slightly lower dose (40 mg/kg per day) was without HIV-infected cells by 50%o or inhibited Mo-MSV-induced C3H cell any significant effect (P < 0.10), whereas slightly higher transformation by S50o. doses (75 and 100 mg/kg per day) proved lethal within 9 days bCDSO corresponded to the concentration of compound that reduced for 73% and 89% of the mice, respectively. These data the number of viable mock-infected cells by 50%. indicate that ribavirin has only little activity as an antiretro- Downloaded by guest on September 29, 2021 334 Medical Sciences: Balzarini et al. Proc. Natl. Acad. Sci. USA 86 (1989)

Table 2. Inhibitory effects of test compounds on the initiation of tumor formation and survival of NMRI mice inoculated with Mo-MSV Compound dose, Mice at start of Day of tumor initiation, Day of animal death,a mg/kg per day experiments, no. mean ± SD mean ± SD PMEA 50 21 18 0b (P < 0.0005) c 20 23 18.2 ± 3.2b (p < 0.0005) d 5 20 11.83 ± 2.4 (P < 0.0005) 16.4 + 1.4e (P < 0.0005) 2.5 21 7.1 ± 1.5 (P < 0.0005) 14.0 + 1.4 (P < 0.0005) 1 21 6.0) ± 0.6 (P < 0.0005) 12.7 + 1.5 (P < 0.0005) 0 (control) 23 4.55 ± 0.6 10.5 + 0.7 AZT 625 10 14.5 ± 3.5' (P < 0.0005) 4.9 ± 3.3' (P < 0.0005) 125 21 10.2 ± 2.2 (P < 0.0005) 16.5 ± 1.6g (P < 0.0005) 25 23 8.1 ± 2.0 (P < 0.0005) 13.2 ± 2.3h (p < 0.0005) 5 12 6.2 ± 0.8 (P < 0.0005) 12.6 ± 1.3 (P < 0.0005) 0 (control) 43 4.5 ± 0.6 10.4 ± 0.8 ddCyd 250 10 7.8 ± 1.0 (P < 0.0005) 11.5 ± 2.4 (P < 0.025) 125 19 6.5 ± 0.5 (P < 0.0005) 12.3 ± 1.2 (P < 0.0005) 62.5 10 5.1 ± 0.3 (P < 0.25) 10.6 ± 1.2 (P < 0.25) 0 (control) 23 5.0 ± 0.2 10.2 ± 0.7 ddAdo 625 9 6.2 ± 1.1 (P < 0.0005) 13.4 ± 0.7 (P < 0.0005) 125 23 4.9 ± 0.7 (P < 0.010) 11.0 ± 0.8 (P < 0.0025) 0 (control) 43 4.5 ± 0.5 10.4 ± 0.8 Ribavirin 100 9 9.0 ± 0.0i (P < 0.0005) 7.7 ± 1.7' (P < 0.0005) 75 11 8.3 ± 0.6' (P < 0.0005) 8.0 ± 3.3k (P < 0.0010) 50 8 6.5 ± 1.1 (P < 0.0005) 14.0 ± 1.9 (P < 0.0005) 40 12 4.8 ± 0.6 (P < 0.10) 10.5 ± 1.2 (P > 0.25) 20 12 4.8 ± 0.6 (P < 0.10) 10.8 ± 0.9 (P < 0.25) 10 11 4.7 ± 0.5 (P < 0.25) 10.5 ± 0.7 (P > 0.25) 0 (control) 23 4.5 ± 0.6 10.5 ± 0.7 Fusidic acid 25 11 4.7 ± 0.8 (P > 0.25) 10.1 1.1 (P > 0.25) 0 (control) 19 4.6 ± 0.8 10.2 ± 0.7 Suramin 62.5 23 6.8 ± 2.22 (P < 0.05) 13.0 ± 1.44 (P < 0.25) 0 (control) 21 6.1 ± 0.7 12.1 ± 1.3 ATA 25 22 8.5 ± 1.5 (P < 0.0025) 14.7 ± 1.8 (P < 0.010) 0 (control) 18 7.5 ± 0.6 13.4 ± 1.4

Five milligrams of Evans blue/kg per day did not significantly increase the mean day of tumor initiation (day 7.7 ± 0.84 versus day 7.4 ± 0.60 for control). Nor did it result in an increase of the mean day of animal death (day 11.6 ± 0.65 versus day 12.7 ± 1.15 for control). A higher dose of Evans blue was toxic to the mice. aMice died from drug toxicity before tumor formation for the following drug treatment regimens (mean day ± SD of animal death between parentheses): PMEA at 125 mg/kg per day (day 5.3 ± 2.0), ddCyd at 625 mg/kg per day (day 8.5 ± 1.9), fusidic acid at 625 mg/kg per day (day 1.0 ± 0.0) and 125 mg/kg per day (day 1.0 ± 0.0), suramin at 125 mg/kg per day (day 3.2 ± 1.6), and ATA at 62.5 mg/kg per day (day 3.5 ± 2.3). bThe mean days of tumor initiation in the groups treated with PMEA at 50 mg/kg per day (day 18) or 20 mg/kg per day (day 18.2) were calculated from 1 mouse of21 and 4 mice of23, respectively; the remaining mice did not develop any tumor during the 25-day observation after virus inoculation. cAt 50 mg of PMEA/kg per day, 95% of mice were alive on day 25 of the experiment. dAt 20 mg of PMEA/kg per day, all mice (100%1) were alive on day 25 of the experiment. 'At S mg of PMEA/kg per day, 65% of mice were alive on day 25 of the experiment. fMean day of tumor initiation in the group of mice treated with 625 mg of AZT/kg per day was calculated for 2 of 10 mice. Eight mice died from toxicity of the compound before any tumor appeared, which resulted in a shorter survival time of the mice than that of control. gAt 125 mg of AZT/kg per day, 71% of mice were alive on day 25 of the experiment. hAt 25 mg of AZT/kg per day, 26% of mice were alive on day 25 of the experiment. iThe mean day of tumor initiation in the group of mice treated with ribavirin at 100 mg/kg per day (day 9.0) or 75 mg/kg per day (day 8.3) was calculated from 1 mouse of 9 and 3 mice of 11, respectively; the other mice died from toxicity of the compound before any tumor developed, which resulted in a shorter survival time of the mice than that of control.

virus agent and a very narrow therapeutic index. Also, fusidic mice, whereas at 25 mg/kg per day it had no antiretrovirus acid, a compound recently claimed to be effective against activity (P > 0.25). HIV in vitro at concentrations easily attained in vivo (14, 15) We also included suramin and the related anionic sub- proved ineffective in both our in vitro human and murine stances Evans blue and ATA in our in vivo retrovirus model. assay systems and in vivo. At doses of 625 or 125 mg/kg per Although suramin delayed the tumor initiation time when day fusidic acid caused almost instantaneous death of the administered at 62.5 mg/kg per day, at a dose of 125 mg/kg Downloaded by guest on September 29, 2021 Medical Sciences: Balzarini et al. Proc. Natl. Acad. Sci. USA 86 (1989) 335

vivors when the compound was administered at 25 or 125 mg/ kg per day, respectively (Table 2). A dose of 5 mg of PMEA/ kg perday (65% long-term survivors and a mean of16.4 days for animal death) and a dose of 125 mg of AZT/kg per day (71.4% 0-10 100 long-term survivors and a mean of 16.5 days for animal death) could be considered as equipotent, which means that PMEA is 25-fold more potent an antiretrovirus agent than AZT. C 80. Ofthe other compounds, only ddAdo, ribavirin, and ddCyd Go significantly increased the life-span of the mice (P < .0 0.0005)-this increase being 30%o, 33%, and 20%o ofcontrol at 6060 doses of625, 50, and 125 mg/kg per day, respectively. These doses were close to the a - toxicity threshold, suggesting 0 narrow therapeutic index for these compounds. Neither 00 fusidic acid, suramin, Evans blue, nor ATA had any protec- 0 tive effect against mortality (Table 2). We also administered PMEA at 75 and 50 mg/kg per day by .0 2 0- the i.p. route to adult mice to investigate whether the milk- suckling baby mice, which had been inoculated with Mo- MSV, could be protected by the compound via the mother's 5 10 15 20 25 milk. Under these experimental conditions, however, no T i m e (days) protection against tumor formation and associated mortality 1015 2 2 was seen, indicating that PMEA is either not secreted in the milk or PMEA is not well absorbed when given orally, or both. 1 00_ Clearly PMEA was toxic to the mice at a 5-fold lower dose than AZT (Table 2). At 125 mg/kg per day PMEA reduced the average survival time of the mice to 5.3 days. To achieve a 8 I, 0 similar reduction in the survival time, AZT had to be administered at a daily dose of 625 mg/kg. Mortality rates caused by AZT at 625 mg/kg per day and PMEA at 125 mg/kg 2 60- per day did not differ statistically. Antitumor Effect of PMEA. To ascertain that the inhibitory 0 effects of PMEA on Mo-MSV-induced tumor development 40. and associated mortality were not due to a direct antitumor .0 action of the compound, additional experiments were done 2 20- with PMEA (20 mg/kg per day). Treatment started 7 days after the mice had been inoculated with Mo-MSV. The mice were then bearing clearly visible tumors. Under these conditions, PMEA had no substantial effect on tumor development or 0 T 5 10 15 20 25 life-span of the mice (data not shown). A similar observation was T m e (days) made for AZT, indicating that the beneficial effects of PMEA and AZT on the course ofthe retrovirus infection were FIG. 2. Tumor formation (Upper) and survival (Lower) ofNMRI not mediated by a direct cytostatic action but should be mice inoculated with Mo-MSV after treatment with indicated doses attributed to the antiviral activity of the compounds. of PMEA. PMEA (in a 25-/II vol) was injected i.p. at doses of either 1 (0), 2.5 (A), 5 (n), 20 (V), 50 (o), or 125 (x) mg/kg per day starting DISCUSSION on the day of infection and continuing for an additional 9 days. In our HIV Control mice received vehicle only (e). Treated groups numbered assay system, MT-4 cells were infected with HIV, 21-23 mice, except for the group treated with 125 mg of PMEA per and virus-induced cytopathogenicity was recorded at the fifth kg per day, which numbered 12 mice. day after infection based on cell viability determination by trypan blue dye exclusion (25). The cytotoxicity of the test per day it was clearly toxic, thus indicating a very narrow compounds for mock-infected MT-4 cells was determined in therapeutic window. Evans blue and ATA had no appreciable parallel. In the Mo-MSV assay system, murine C3H embryo antiretroviral activity at doses of 5 and 25 mg/kg per day, fibroblasts were infected with Mo-MSV, and virus-induced respectively (Table 2). transformation of the cells was recorded microscopically at Effect of PMEA and Other Test Compounds on Mortality of the sixth day after infection (27). Mo-MSV-Infected Newborn NMRI Mice. Another important AZT is exquisitely active as an antiretrovirus agent in vitro variable that could be followed to assess antiretrovirus in murine systems-presumably because it is rapidly metab- activity in our in vivo model is mortality accompanying olized to its 5'-triphosphate in murine cells (22). In human Mo-MSV-induced tumor cells it is less efficiently phosphorylated to the 5'- progression (Table 2). PMEA triphosphate stage. This may explain why AZT is more proved particularly effective in reducing the mortality rate inhibitory to Mo-MSV in murine C3H cells than to HIV in when administered at a dose of 50, 20, or 5 mg/kg per day: human ATH8, H9, or peripheral blood mononuclear cells (6). at these doses (given on the day of infection and for the MT-4 cells are highly sensitive to the anti-HIV effects ofAZT following 9 days) 95%, 100%, and 65% of the mice were (Table 1) and several other 2',3'-dideoxythymidine analogues recorded as long-term survivors (Table 2, Fig. 2B). At a (data not shown). So far the molecular basis of this phenom- PMEA dose of 5 mg/kg per day, the life-span of the enon is unclear.O The rapid conversion of AZT to its active Mo-MSV-infected mice was increased by 60%o, and even at a PMEA dose as low as 1 survival time mg/kg per day 1Balzarini, J., Matthes, E., Meeus, P., Johns, D. G. & De Clercq, E. increased significantly (20%6) (P < 0.005). (1988) Proceedings of the VI International Symposium on Human In parallel experiments, AZT treatment of Mo-MSV-in- Purine and Pyrimidine Metabolism, Hakone, Japan, July 17-21, fected mice also resulted in 26.1% and 71.4% long-term sur- 1988. Downloaded by guest on September 29, 2021 336 Medical Sciences: Balzarini et al. Proc. Natl. Acad. Sci. USA 86 (1989) form in murine cells may explain why AZT is so efficient in Nusinoff-Lehrman, S., Gallo, R. C., Bolognesi, D., Barry, inhibiting Rauscher murine leukemia virus-induced spleno- D. W. & Broder, S. (1985) Proc. Natl. Acad. Sci. USA 82, megaly, as shown by Ruprecht et al. (19), and Mo-MSV- 7096-7100. induced tumor formation, as shown here. The fact that 7. Yarchoan, R., Klecker, R. W., Weinhold, K. J., Markham, P. D., Lyerly, H.K., Durack, D. T., Gelmann, E., Nusinoff- PMEA was compared with, and found more efficacious than, Lehrman, S., Blum, R. M., Barry, D. W., Shearer, G. M., AZT under conditions where the latter showed optimal Fischl, M. A., Mitsuya, H., Gallo, R. C., Collins, J. M., antiviral activity adds more weight to the potential of PMEA Bolognesi, D. P., Myers, C. E. & Broder, S. (1986) Lancet i, as an antiretrovirus agent. 575-580. Our observations justify further studies on the pharmaco- 8. Yarchoan, R., Berg, G., Brouwers, P., Fischl, M. A., Spitzer, kinetic behavior of PMEA and its disposition in vivo (i.e., in A. R., Wichman, A., Grafman, J., Thomas, R. V., Safai, B., mice and monkeys). Preliminary data indicate that PMEA is Brunetti, A., Perno, C. F., Schmidt, P. J., Larson, S. M., rapidly cleared from the bloodstream (ti/2, 6-10 min) in NMRI Myers, C. E. & Broder, S. (1987) Lancet i, 132-135. mice given an i.v. bolus of 100 mg/kg. In view of the 9. Fischl, M. A., Richman, D. D., Grieco, M. H., Gottlieb, M. S., Volberding, P. A., Laskin, 0. L., Leedom, J. M., propensity of HIV to infect and damage the central nervous Groopman, J. E., Mildvan, D., Schooley, R. T., Jackson, system, potential treatment schedules for AIDS patients G. G., Durack, D. T., Phil, D., King, D. & The AZT Collab- should address the ability of a candidate drug to cross the orative Working Group (1987) N. Engl. J. Med. 317, 185-191. blood-brain barrier. Our initial studies with radio-labeled 10. McCormick, J. B., Getchell, J. P., Mitchell, S. W. & Hicks, PMEA in mice have indicated that PMEA predominantly D. R. (1984) Lancet ii, 1367-1369. accumulates in the kidneys and liver, while somewhat lower 11. Crumpacker, C., Heagy, W., Bubley, G., Monroe, J. E., levels are achieved in the brain. Also, the finding that PMEA Finberg, R., Hussey, S., Schnipper, L., Lucey, D., Lee, T. H., when given systemically (for example, i.p.) effects a signif- McLane, M. F., Essex, M. & Mulder, C. (1987) Ann. Intern. icant reduction in the mortality rate of mice inoculated Med. 107, 664-674. 12. Mitsuya, H. & Broder, S. (1986) Proc. Natl. Acad. Sci. USA 83, intracerebrally with herpes simplex virus type 1 (28) points to 1911-1915. the ability ofPMEA to cross the blood-brain barrier. Another 13. Yarchoan, R., Perno, C. F., Thomas, R. V., Klecker, R. W., important requirement any nucleoside analog aimed for oral Allain, J.-P., Wills, R. J., McAtee, N., Fischl, M. A., Dubin- use should fulfill is resistance to hydrolytic cleavage in acidic sky, R., McNeely, M. C., Mitsuya, H., Pluda, J. M., Lawley, conditions (for example, pH 1). We found that PMEA, in T. J., Leuther, M., Safai, B., Collins, J. M., Myers, C. E. & contrast to ddAdo, remained stable in 10% trichloroacetic Broder, S. (1988) Lancet i, 76-81. acid (pH < 1) for at least 60 min. 14. Faber, V., Dalgleish, A. G., Newell, A. & Malkovsky, M. In conclusion, because ofthe marked efficacy ofPMEA as (1987) Lancet ii, 827-828. an antiretrovirus agent and its superiority over AZT in the 15. Barnes, D. M. (1987) Science 238, 276. 16. Richman, D. D., Fischl, M. A., Grieco, M. H., Gottlieb, murine retrovirus model, PMEA merits further investigation M. S., Volberding, P. A., Laskin, 0. L., Leedom, J. M., as a promising candidate drug for the treatment of AIDS and Groopman, J. E., Mildvan, D., Hirsch, M. S., Jackson, G. G., other retroviral infections in humans. Durack, D. T., Nusinoff-Lehrman, S. & The AZT Collabora- tive Working Group (1987) N. Engl. J. Med. 317, 192-197. We thank Ann Absillis and Lizette van Berckelaer for their 17. Tavares, L., Roneker, C., Johnston, K., Nusinoff-Lehrman, S. excellent technical assistance and Christiane Callebaut for her & de Noronha, F. (1987) Cancer Res. 47, 3190-3194. dedicated editorial help. This research was supported in part by the 18. King, N. W. (1986) Vet. Pathol. 23, 345-353. AIDS Basic Research Programme of the European Community and 19. Ruprecht, R. M., O'Brien, L. G., Rossoni, L. D. & Nusinoff- by grants from the Belgian Fonds voor Geneeskundig Wetenschap- Lehrman, S. (1986) Nature (London) 323, 467-469. pelijk Onderzoek (projects 3.0040.83 and 3.0097.87) and the Belgian 20. Ruprecht, R. M., Rossoni, L. D., Haseltine, W. A. & Broder, Geconcerteerde Onderzoeksacties (project 85/90-79). S. (1985) Proc. Nati. Acad. Sci. USA 82, 7733-7737. 21. Sharpe, A. H., Jaenisch, R. & Ruprecht, R. M. (1987) Science 1. Barrd-Sinoussi, F., Chermann, J. C., Rey, R., Nugeyre, M. T., 236, 1671-1674. Chamaret, S., Gruest, J., Dauguet, C., Axler-Blin, C., Vdzinet- 22. Balzarini, J., Pauwels, R., Baba, M., Herdewijn, P., De Clercq, Brun, F., Rouzioux, C., Rozenbaum, W. & Montagnier, L. E., Broder, S. & Johns, D. G. (1988) Biochem. Pharmacol. 37, (1983) Science 220, 868-871. 897-903. 2. Gallo, R. C., Salahuddin, S. Z., Popovic, M., Sherer, G., 23. De Clercq, E., Holy, A., Rosenberg, I., Sakuma, T., Balzarini, Kaplan, M., Haynes, B. F., Parker, T. J., Redfield, R., Oleske, J. & Maudgal, P. C. (1986) Nature (London) 323, 464-467. J., Safai, B., White, G., Foster, P. & Markham, P. D. (1984) 24. Pauwels, R., Balzarini, J., Schols, D., Baba, M., Desmyter, J., Science 224, 500-503. Rosenberg, I., Holy, A. & De Clercq, E. (1988) Antimicrob. 3. Mitsuya, H., Popovic, M., Yarchoan, R., Matsushita, S., Agents Chemother. 32, 1025-1030. Gallo, R. C. & Broder, S. (1984) Science 226, 172-174. 25. Pauwels, R., De Clercq, E., Desmyter, J., Balzarini, J., 4. Broder, S., Yarchoan, R., Collins, J. M., Lane, H. C., Mark- Goubau, P., Herdewijn, P., Vanderhaeghe, H. & Vandeputte, ham, P. D., Klecker, R. W., Redfield, R. R., Mitsuya, H., M. (1987) J. Virol. Methods 16, 171-185. Hoth, D. F., Gelmann, E., Groopman, J. E., Resnick, L., 26. Balzarini, J., Mitsuya, H., De Clercq, E. & Broder, S. (1986) Gallo, R. C., Myers, C. E. & Fauci, A. S. (1985) Lancet ii, 627- Intern. J. Cancer 37, 451-457. 630. 27. Balzarini, J., Pauwels, R., Baba, M., Robins, M. J., Zou, R., 5. Collins, J. M., Klecker, R. W., Jr., Yarchoan, R., Lane, H. C., Herdewijn, P. & De Clercq, E. (1987) Biochem. Biophys. Res. Fauci, A. S., Redfield, R. R., Broder, S. & Myers, C. E. (1986) Commun. 145, 269-276. J. Clin. Pharmacol. 26, 22-26. 28. De Clercq, E., Holy, A. & Rosenberg, I. (1989) Antimicrob. 6. Mitsuya, H., Weinhold, K. J., Furman, P. A., St. Clair, M. H., Agents Chemother., in press. Downloaded by guest on September 29, 2021