No abstract

In an effort to further extend the number of targets for development of antiretroviral agents, we have used an in vitro integrase assay to investigate a variety of MATERIALS AND METHODS Materials. HIV-1 integrase protein (3.5 pmol per reaction), produced via an Escherichia coli expression vector as described (13), was obtained from R. Craigie (Laboratory of Molecular Biology, National Institute of Diabetes and Digestive and Kidney Diseases) and stored at -70°C in 1 M NaCl/20 mM Hepes, pH 7.6/1 mM EDTA/1 mM dithiothreitol/20% glycerol (wt/vol). Caffeic acid phenethyl ester (CAPE) was brought to our attention by Dezider Grunberger (Columbia University, New York), who supplied the compound. Doxorubicin, 5-iminodaunorubicin, mitoxantrone, ellipticine, ellipticinium, 9-aminoacridine, amsacrine, ditercalinium, ethidium, camptothecin, 9-aminocamptothecin, 10,11-methylenedioxycamptothecin, etoposide (VP-16), teniposide (VM-26), and quercetin were obtained through the Developmental Therapeutics Program, National Cancer Institute. Hydroxyrubicin and adriamycinone were obtained through Waldemar Priebe (M. D. Anderson Hospital, Houston). Naphthoquinone, 5,8-dihydroxy-1,4-naphthoquinone, 5-hydroxy-1,4-naphthoquinone, and dihydroxyanthraquinone were purchased from Aldrich. Chloroquine, primaquine, quinacrine, and amodiaquine were obtained through Sigma. Hydroxychloroquine was from SterlingWinthrop Research Institute. Mefloquine was from Hoffmann-La Roche.Oligonucleotide Substrate. Oligonucleotides were obtained from Midland Certified Reagent (Midland, TX), and were HPLC-purified before use. The following complementary oligonucleotides were used as substrates:AE118: 5'-GTGTGGAAAATCTCTAGCAGT-3' and AE117: 5'-ACTGCTAGAGATTTTCCACAC-3' (2). 2399The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Efficient replication of HIV-1 requires integration of a DNA copy of the viral genome into a chromosome of the host cell. Integration is catalyzed by the viral integrase, and we have previously reported that phenolic moieties in compounds such as flavones, caffeic acid phenethyl ester (CAPE, 2), and curcumin confer inhibitory activity against HIV-1 integrase. We now extend these findings by performing a comprehensive structure-activity relationship using CAPE analogues. Approximately 30 compounds have been prepared as HIV integrase inhibitors based on the structural lead provided by CAPE, which has previously been shown to exhibit an IC50 value of 7 microM in our integration assay. These analogues were designed to examine specific features of the parent CAPE structure which may be important for activity. Among the features examined for their effects on inhibitory potency were ring substitution, side chain length and composition, and phenyl ring conformational orientation. In an assay which measured the combined effect of two sequential steps, dinucleotide cleavage and strand transfer, several analogues have IC50 values for 3'-processing and strand transfer lower than those of CAPE. Inhibition of strand transfer was assayed using both blunt-ended and "precleaved" DNA substrates. Disintegration using an integrase mutant lacking the N-terminal zinc finger and C-terminal DNA-binding domains was also inhibited by these analogues, suggesting that the binding site for these compounds resides in the central catalytic core. Several CAPE analogues were also tested for selective activity against transformed cells. Taken together, these results suggest that the development of novel antiviral agents for the treatment of acquired immune deficiency syndrome can be based upon inhibition of HIV-1 integrase.

The natural dibenzylbutyrolactone type lignanolide (-)-arctigenin (2), an inhibitor of human immunodeficiency virus type-1 (HIV-1) replication in infected human cell systems, was found to suppress the integration of proviral DNA into the cellular DNA genome. In the present study 2 was tested with purified HIV-1 integrase and found to be inactive in the cleavage (3'-processing) and integration (strand transfer) assays. However, the semisynthetic 3-O-demethylated congener 9 characterized by a catechol substructure exhibited remarkable activities in both assays. Structure-activity relationship studies with 30 natural (1-6), semisynthetic (7-21), and synthetic (37-43, 45, 46) lignans revealed that (1) the lactone moiety is crucial since compounds with a butane-1,4-diol or tetrahydrofuran substructure and also lignanamide analogues lacked activity and (2) the number and arrangement of phenolic hydroxyl groups is important for the activity of lignanolides. The congener with two catechol substructures (7) was found to be the most active compound in this study. 7 was also a potent inhibitor of the "disintegration" reaction which models the reversal of the strand transfer reaction. The inhibitory activity of 7 with the core enzyme fragment consisting of amino acids 50-212 suggests that the binding site of 7 resides in the catalytic domain.