Publications

Treatment of 3',5'-di-O-acetyl-(E)-5-(2-bromovinyl)-2'-deoxyuridine (2) with p-chlorophenyl phosphorodichloridate and 1,2,4-triazole gave 1-(3,5-di-O-acetyl-2-deoxy-beta-D-erythro-pentofuranosyl)-(E)-5-(2-br o movinyl)- 4-(1,2,4-triazol-1-yl)pyrimidin-2(1H)-one (3). Reaction of 3 with ammonia gave (E)-5-(2-bromovinyl)-2'-deoxycytidine (1), the overall yield from 2 being 60%. A similar 4-(1,2,4-triazol-1-yl) derivative (4) was obtained from 3',5'-di-O-acetyl-thymidine by the use of phosphoryl chloride as the condensing agent. Treatment of thymidine with trimethylsilyl chloride and then with phosphoryl chloride and 1,2,4-triazole gave upon workup 1-(2-deoxy-beta-D-erythro-pentofuranosyl)-5-methyl-4(1,2,4-triazol -1-yl) pyrimidin-2(1H)-one (5). (E)-5-(2-Bromovinyl)-2'-deoxyuridine (BVDU) when similarly treated gave the corresponding (E)-5-(2-bromovinyl) compound 7. A minor product formed in both cases was a 4-(1,2,4-triazol-1-yl) derivative in which the nucleoside 5'-hydroxyl group had been replaced by chlorine (6 and 8). Whereas compounds 4-6 and 8 did not exhibit a selective antiviral effect, compounds 1-3 and 7 proved almost as active as the reference compound BVDU. In particular, compound 7, the 4-triazolyl derivative of BVDU, would seem worth pursuing for its potential as an inhibitor of herpes simplex virus type 1 and varicella-zoster virus.

New 5'-O-carbonate prodrugs of zidovudine (AZT) have been synthesized in order to enhance its uptake by HIV-1 infected cells, to improve its anti-HIV potency, and to optimize the intramolecular cyclic rearrangement process related to the 5'-O-(4-hydroxybutyl) carbonate moiety. Evidence of this prodrug rearrangement was confirmed by comparison of the serum half-lives of the 3'-azido-3'-deoxythymidin-5'-yl O-(4-hydroxyalkyl or -alkenyl or -alkylepoxide) carbonate prodrugs with our thermodynamic predictions. Interestingly, these 5'-O-carbonate prodrug series show increased anti-HIV potencies in conjunction with, or without, reduced cytotoxicity as compared to AZT that lead to a gain in selectivity indexes. The cytotoxicity of AZT could be reduced with these 5'-O-carbonate prodrug series by delaying the 5'-O-glucuronidation of AZT, which is one of the major limitations of AZT.

Human immunodeficiency virus (HIV) is the etiological agent of the acquired immune deficiency syndrome (AIDS). The current strategy for the treatment of HIV infection is called Highly Active Antiretroviral Therapy (HAART) and is based on cocktails of drugs that are currently approved by the Food and Drug Administration. These drugs include compounds that target the viral entry step and the enzymes reverse transcriptase or protease. The introduction of HAART has dramatically changed the landscape of HIV disease. Death from AIDS-related diseases has been reduced significantly since HAART came into use. Nevertheless it is not clear how long clinical benefit will last taking into account the emergence of multiple drug-resistant viral strains. Addition of new anti-HIV drugs targeting other steps of the viral replication cycle may increase the potency of inhibition and delay resistance development. HIV integrase is an essential enzyme in the HIV life cycle and is an attractive target for new drug development. Despite years of intensive research, only two classes of compounds that inhibit integration have been identified until now, namely the diketo acids and the pyranodipyrimidines. In this review we will point to new potential antiviral targets related to retroviral integration that are amenable to drug development. We will describe the pitfalls of currently used integrase assays and propose new strategies and technologies for the discovery of HIV integration inhibitors. Furthermore, we will describe the two classes of integrase inhibitors and discuss their antiviral activity, molecular mechanism of anti-HIV action and the selection of HIV resistance against these drugs.

(E)-3',5'-diamino-5-(2-bromovinyl)-2',3',5'-trideoxyuridine (5), the diamino analogue of BVDU (1), was synthesized from BVDU. In contrast with BVDU, compound 5 did not show activity against herpes simplex virus or varicella-zoster virus.

1-(3′-Fluoro-2′,3′,6′-trideoxy-β-D-allofuranosyl)thymine [7] and 1-(3′-fluoro-2′,3′,6′-trideoxy-α-L-talofuranosyl) thymine [8] were synthesized starting from the corresponding 2,3′-anhydro nucleoside derivatives. The fluorine was introduced stereoselectively by opening of the anhydro bridge in the presence of HF/AIF 3 . The 5′-C-methyl derivatives, [7] and [8], of 3′-fluoro-3′-deoxythymidine (FLT) were evaluated for their inhibitory effect against human immunodeficiency virus type 1 (HIV-1) and type 2 (HIV-2). The compounds [7] and [8] had antiviral activity which was three orders of magnitude lower than the reference compound 3′-fluoro-3′-deoxythymidine. None of the compounds showed appreciable activity against other RNA or DNA viruses at subtoxic concentrations.

Various Mannich bases of chalcones and related compounds displayed significant cytotoxicity toward murine P388 and L1210 leukemia cells as well as a number of human tumor cell lines. The most promising lead molecule was 21 that had the highest activity toward L1210 and human tumor cells. In addition, 21 exerted preferential toxicity to human tumor lines compared to transformed human T-lymphocytes. Other compounds of interest were 38, with a huge differential in cytotoxicity between P388 and L1210 cells, and 42, with a high therapeutic index when cytotoxicity to P388 cells and Molt 4/C8 T-lymphocytes were compared. In general, the Mannich bases were more cytotoxic than the corresponding chalcones toward L1210 but not P388 cells. A ClusCor analysis of the data obtained from the in vitro human tumor screen revealed that the mode of action of certain groups of compounds was similar. For some groups of compounds, cytotoxicity was correlated with the sigma, pi, or molar refractivity constants in the aryl ring attached to the olefinic group. In addition, the IC50 values in all three screens correlated with the redox potentials of a number of Mannich bases. X-ray crystallography and molecular modeling of representative compounds revealed various structural features which were considered to contribute to cytotoxicity. While a representative compound 15 was stable and unreactive toward glutathione (GSH) in buffer, the Mannich bases 15, 18, and 21 reacted with GSH in the presence of the pi isozyme of glutathione S-transferase, suggesting that thiol alkylation may be one mechanism by which cytotoxicity was exerted in vitro. Representative compounds were shown to be nonmutagenic in an intrachromosomal recombination assay in yeast, devoid of antimicrobial properties and possessing anticonvulsant and neurotoxic properties. Thus Mannich bases of chalcones represent a new group of cytotoxic agents of which 21 in particular serves as an useful prototypic molecule.