Preferential Inhibition of the Magnesium-Dependent Strand Transfer Reaction of HIV-1 Integrase by -Hydroxytropolones

Integration is a crucial step in the life cycle of human immunodeficiency virus type 1 (HIV-1); therefore, inhibitors of HIV-1 integrase are candidates for antiretroviral therapy. Two 7-hydroxytropolone derivatives ( -hydroxytropolones) were found to inhibit HIV-1 integrase. A structure-activity relationship investigation with several tropolone derivatives from The National Cancer Institute compound repository demonstrated that the 7-hydroxy group is essential for integrase inhibition. -Hydroxytropolones preferentially inhibit strand transfer and are inhibitory both in the presence of magnesium or manganese. Lack of inhibition of disintegration in the presence of magnesium coupled with results from different cross-linking assays suggests -hydroxytropolones as interfacial inhibitors. We propose that -hydroxytropolones chelate the divalent metal (Mg or Mn ) in the enzyme active site. The most active compound against HIV-1 integrase in biochemical assays [2,4,6-cycloheptatrien-1-one, 2,7-dihydroxy-4-isopropyl (NSC 18806) IC50 4.8 2.5 M] exhibits weak cytoprotective activity against HIV-1IIIB in a cell-based assay. -Hydroxytropolones represent a new family of inhibitors for the development of novel drugs against HIV infection. The screening and investigation of novel drugs against human immunodeficiency virus (HIV) remain critical because of the ongoing AIDS epidemics and because of the fast emergence of virus variants resistant to present antiviral therapy (Kellerman et al., 2005). The replication steps of HIV, a member of the retrovirus family, are well known and can therefore be targeted rationally (for general review, see De Clercq, 2005). After HIV binding to the host cell, viral single-stranded RNA genomes are released into the cell and serve as templates for the virus-encoded reverse transcriptase to synthesize double-stranded DNA copies bearing the long terminal repeats (LTRs) at both ends (Turner and Summers, 1999). The viral linear DNA is integrated into the host genome in a reaction catalyzed by the viral enzyme integrase (IN). Integration is essential for viral replication because integrated viral DNA (provirus) serves as a template for the synthesis of new viruses after processing by the host cell transcription-translation machines (Brown, 1990; Fesen et al., 1993; Asante-Appiah and Skalka, 1997; Van Maele and Debyser, 2005). Antiviral therapy currently uses a combination of reverse transcriptase and HIV protease inhibitors. Inhibitors of virus fusion to the host cells have recently been developed (Barbaro et al., 2005; De Clercq, 2005). Because HIV integrase is crucial for virus replication, the search for integrase inhibitors has been ongoing (Fesen et al., 1993; Hazuda et al., 2000; Debyser et al., 2002; Deprez et al., 2004; Johnson et al., 2004; Pommier et al., 2005). Integrase inserts the proviral DNA into host chromosomes in two steps: 3 processing (3 -P) and strand transfer (ST). 3 -P is an endonucleolytic cleavage reaction removing the 3 ends of the viral LTR DNA (generally a dinucleotide pGpT for HIV-1) immediately 3 from the conserved sequence (CA for HIV-1) (Fig. 1A). ST is the insertion of the processed 3 ends of the viral DNA into the cell genome (Asante-Appiah and Skalka, 1997). The HIV-1 integrase catalytic site contains three essential amino acids: Asp64, Asp116, and Glu152 (D,D-35 E-motif) that coordinate at least one and probably two divalent cations (Mg or Mn ) between the enzyme and its DNA substrates (Engelman and Craigie, 1992; Chiu and Davies, 2004). This research was supported by the Intramural Research Program of the National Institutes of Health, National Cancer Institute, Center for Cancer Research. Article, publication date, and citation information can be found at http://molpharm.aspetjournals.org. doi:10.1124/mol.105.020321. ABBREVIATIONS: HIV, human immunodeficiency virus; IN, integrase; 3 -P, 3 -processing; ST, strand transfer; LTR, long terminal repeat; DMSO, dimethyl sulfoxide; MOPS, 3-(N-morpholino)propanesulfonic acid; bp, base pair(s); PAGE, polyacrylamide gel electrophoresis; L-731,988, 4-(1-(4-fluoro-benzyl)-1Hpyrrol-2-yl)-2,4-dioxo-butyric acid; NSC 624404, 2-(4-{bis[2-hydroxy-5-(methylethyl)-3-oxocyclohepta-1,4,6-trienyl]methyl}phenoxy) acetic acid, sodium salt; NSC 310618, 1,2,3,4-tetrahydro-2-7-dihydroxy-9-methyl-2-(1-methylethenyl)-6H-benzocyclohepten-6-one. 0026-895X/06/6904-1454–1460 MOLECULAR PHARMACOLOGY Vol. 69, No. 4 U.S. Government work not protected by U.S. copyright 20321/3100083 Mol Pharmacol 69:1454–1460, 2006 Printed in U.S.A. 1454 at A PE T Jornals on N ovem er 7, 2017 m oharm .aspeurnals.org D ow nladed from The ST inhibitors 5-chloroindolyltetrazolylpropenone and L-731,988 have been proposed to chelate the divalent metal cations (Mg or Mn ) in the enzyme active site (Grobler et al., 2002; Marchand et al., 2003; Pommier et al., 2005). 5-Chloroindolyltetrazolylpropenone has been cocrystallized in the catalytic domain of HIV integrase and shown to bind in the DDE motif (Goldgur et al., 1999). The diketo acid derivative L-731,988 was shown to block binding of target DNA in the integrase active site (Espeseth et al., 2000). The selective inhibition of the strand transfer reaction by diketo acids has been proposed to be due to their interfacial inhibition on preformed integrase-viral DNA complexes (Pommier and Marchand, 2005). We previously reported one derivative (NSC 624404) containing the characteristic seven-membered tropolone ring as novel inhibitor of HIV-1 integrase in a four-point pharmacophore analysis of the National Cancer Institute drug database (Neamati et al., 1997). While screening the National Cancer Institute chemical library for HIV-1 integrase inhibitors, we recently found additional positive hits with tropolone derivatives. We report here the structure-activity relationship of tropolones available in the National Cancer Institute compound repository on HIV integrase activities. Tropolone derivatives are present in cupressaceous trees from genus Thuja and are probably responsible for resistance of fungal and insect attack on the heartwood (Baya et al., 2001; Diouf et al., 2002; Lim et al., 2005). Our experiments demonstrate the ability of the monomer 7-hydroxytropolones ( -hydroxytropolones) to preferentially inhibit the ST reaction by interfering with the enzyme catalytic site. -Hydroxytropolone derivatives are new lead inhibitors for HIV-1 integrase. Materials and Methods Compounds. All drugs were obtained from the National Cancer Institute chemical repository from the Developmental Therapeutics Program (National Institutes of Health, Bethesda, MD). Compounds were dissolved in 100% DMSO. Stock solutions (10 mM) were stored at 20°C. Recombinant HIV Integrase and Oligonucleotide Substrates. Expression and purification of the recombinant HIV-1 integrase in Escherichia coli were performed according to Leh et al. (2000) and Marchand et al. (2001) with addition of 10% glycerol to all buffers. The preparation of the Q148C/SSS-mutant integrase is described in Johnson et al. (2006). The oligonucleotide substrates, except those used for the disulfide cross-linking (Fig. 5A), were purchased from Integrated DNA Technologies, Inc. (Coraville, IA) and purified by polyacrylamide gel. The sequences of DNA substrates are shown in Figs. 1A, 2A, 3A, and 6A. The single-stranded oligonucleotides were 5 end-labeled with [ -P]ATP (PerkinElmer Life and Analytical Sciences, Boston, MA) and T4 polynucleotide kinase (New England BioLabs, Ipswich, MA). Unincorporated nucleotide was removed using mini Quick Spin Oligo columns (Roche Diagnostics, Indianapolis, IN). Substrates were obtained after annealing with complementary nonlabeled oligonucleotides. The thiolmodified substrate (Fig. 5A) for disulfide cross-linking assay was synthesized by W. Santos and G. Verdine (Department of Chemistry and Chemical Biology, Harvard University, Cambridge, MA) as described previously (He and Verdine, 2002). Integrase Catalytic Assays. Reactions were performed in 10 l with 300 nM recombinant IN, 20 nM 5 end P-labeled oligonucleotide substrates, and inhibitors at the indicated concentrations. We included 10% DMSO in controls. Reactions were incubated for 40 min at 37°C in a buffer containing a final concentration of 25 mM MOPS, pH 7.2, 25 mM NaCl, 14.3 mM -mercaptoethanol, and 7.5 mM divalent cations (MgCl2 or MnCl2 as indicated). Reactions were stopped by addition of 20 l of loading dye (10 mM EDTA, 98% deionized formamide, 0.025% xylene cyanol, and 0.025% bromphenol blue). Reactions were heated at 95°C for 1 min before electrophoresis in 20% polyacrylamide-7 M urea gels. Gels were dried, and reaction products were visualized and quantitated with a PhosphorImager (GE Healthcare, Little Chalfont, Buckinghamshire, UK). Densitometric analyses were performed using ImageQuant from the Molecular Dynamics software. The concentrations at which enzyme activity was reduced by 50% (IC50) was determined using Prism software (GraphPad Software Inc., San Diego, CA) for nonlinear regression to fit dose-response data to logistic curve models. Integrase Binding to HIV DNA Using the Disulfide-CrossLinking Assay. The disulfide cross-linking assay was described in detail previously (Johnson et al., 2006). In brief, 10 M recombinant Q148C/SSS-mutant integrase was incubated with 10 M DNA substrate (Fig. 5A) containing tethered thiols in the presence of 20 mM Tris, pH 7.4, 10% glycerol, and 7.5 mM divalent cations (MgCl2 or MnCl2 as indicated) for 20 min at 37°C. Reactions were stopped by the addition of 20 mM methylmethanethiosulfonate (capping reagent). Nonreducing gel loading buffer (100 mM Tris-HCl, pH 6.8, 4% SDS, 0.2% bromphenol blue, and 20% glycerol) was added, and samples were heated at 95°C before loading onto 16% Tricine gels (Invitrogen, Carlsbad, CA). Gels were stained with Microwave Blue according to manufacturer’s recommendations (Protiga, Frederick,