DESIGN AND SYNTHESIS OF SPECIFIC INHIBITORS OF THE 3′-PROCESSING STEP OF HIV-1 INTEGRASE

The novel dinucleotide 5′-phosphate, [(L,D)-pIsodApdC], discovered in our laboratory, is a strong inhibitor of HIV-1 integrase for both the 3′-processing and the strand transfer steps. The rationale used in this molecular design was that residues immediately upstream of the dinucleotide cleavage site in the 3′-processing step might provide critical recognition/binding sites on integrase. The rationale for the second type of inhibitors was based on the elimination products (linear and cyclic dinucleotides) of 3′-processing. However, while the linear dinucleotide 5′-phosphate (pdGpdT) was active, its cyclic counterpart was inactive against both wild-type and mutant HIV integrase.     

Integration requires a specific interaction of the donor DNA terminal 5'-cytosine with glutamine 148 of the HIV-1 integrase flexible loop

Integration is essential for retroviral replication and gene therapy using retroviral vectors. Human immunodeficiency virus, type 1 (HIV-1), integrase specifically recognizes the terminal sequences of each long terminal repeat (LTR) and cleaves the 3'-end terminal dinucleotide 5'-GT. The exposed 3'-hydroxyl is then positioned for nucleophilic attack and subsequent strand transfer into another DNA duplex (target or chromosomal DNA). We report that both the terminal cytosine at the protruding 5'-end of the long terminal repeats (5'-C) and the integrase residue Gln-148 are critical for strand transfer. Proximity of the 5'-C and Gln-148 was demonstrated by disulfide cross-linking. Cross-linking is inhibited by the inhibitor 5CITEP 1-(5-chloroindol-3-yl)-3-hydroxy-3-(2H-tetrazol-5-yl)-propenone. We propose that strand transfer requires a conformational change of the integrase-viral (donor) DNA complex with formation of an H-bond between the N-3 of the 5'-C and the amine group of Gln-148. These findings have implications for the molecular mechanisms coupling 3'-processing and strand transfer as well as for the molecular pharmacology of integrase inhibitors.     

HIV-1 integrase can process a 3'-end crosslinked substrate.

Integrase of the human immunodeficiency virus type-1 (HIV-1) recognizes specific sequences located in the U3 and U5 regions at the ends of viral DNA. We synthesized DNA duplexes mimicking the U5 region and containing either 2'-aminonucleosides or non-nucleoside 1,3-propanediol insertions at the third and terminal positions and studied their interactions with HIV-1 integrase. Both modifications introduced a local structural distortion in the DNA double helix. Replacement of the terminal nucleosides by corresponding 2'-aminonucleosides had no significant effect on integrase activity. We used an integrase substrate bearing terminal 2'-aminonucleosides in both strands to synthesize a duplex with cross-linked strands. This duplex was then used to determine whether terminal base pair disruption is an obligatory step of retroviral DNA 3'-processing. Processing of the cross-linked analog of the integrase substrate yielded a product of the same length as 3'-processing of the wild-type substrate but the reaction efficiency was lower. Replacement of the third adenosine in the processed strand by a corresponding 2'-aminonucleoside did not affect integrase activity, whereas, its replacement by 1,3-propanediol completely inhibited 3'-processing. Both modifications of the complementary thymidine in the nonprocessed strand increased the initial rate of 3'-processing. The same effect was observed when both nucleosides, at the third position, were replaced by corresponding 2'-aminonucleosides. This indicates that the local duplex distortion facilitated the cleavage of the phosphodiester bond. Thus, a localized destabilization of the third A-T base pair is necessary for efficient 3'-processing, whereas 3'-end-fraying is important but not absolutely required.     

Processing of deoxyuridine mismatches and abasic sites by human immunodeficiency virus type-1 integrase.

We have examined the activities of HIV-1 integrase on substrates containing mismatches, composed of deoxyuridine at different positions in either the processed or nonprocessed strand of viral DNA, within and near the conserved CA dinucleotide of the U5 end of the HIV-1 LTR. Substitution in the processed strand of either the C or A of the CA dinucleotide or of the G 5' to the CA reduced strand transfer six-, three- and seven-fold respectively. 3'-processing was also reduced by substitution at the GC but not at the A. Substitution in the nonprocessed strand of the G nucleotide at the processing site abolished strand transfer while substitution of the T had no effect. DNA binding of HIV-1 integrase was not affected by deoxyuridine substitutions. Deoxyuridine substitution outside the trinucleotide remained compatible with enzyme activity. Enzymatically generated abasic sites were created at each mismatch to determine the effect of a missing base on integrase activity. Consistent with the deoxyuridine mismatch observations, 3'-processing and strand transfer were abolished when the abasic site was substituted for either of the nucleotides of the GCA trinucleotide. Integrase was, however, able to tolerate mismatches within this trinucleotide during the disintegration reaction. Taken together, these results suggest that base-mismatched or base-deleted substrates, which can be created by the proofreading-deficient HIV-1 RT, can be tolerated by HIV-1 integrase when located outside of the GCA trinucleotide at the U5 end of the LTR.     

Position-specific suppression and enhancement of HIV-1 integrase reactions by minor groove benzo[a]pyrene diol epoxide deoxyguanine adducts: implications for molecular interactions between integrase and substrates

The viral protein HIV-1 integrase is required for insertion of the viral genome into human chromosomes and for viral replication. Integration proceeds in two consecutive integrase-mediated reactions: 3'-processing and strand transfer. To investigate the DNA minor groove interactions of integrase relative to known sites of integrase action, we synthesized oligodeoxynucleotides containing single covalent adducts of known absolute configuration derived from trans-opening of benzo-[a]pyrene 7,8-diol 9,10-epoxide by the exocyclic 2-amino group of deoxyguanosine at specific positions in a duplex sequence corresponding to the terminus of the viral U5 DNA. Because the orientations of the hydrocarbon in the minor groove are known from NMR solution structures of duplex oligonucleotides containing these deoxyguanosine adducts, a detailed analysis of the relationship between the position of minor groove ligands and integrase interactions is possible. Adducts placed in the DNA minor groove two or three nucleotides from the 3'-processing site inhibited both 3'-processing and strand transfer. Inosine substitution showed that the guanine 2-amino group is required for efficient 3'-processing at one of these positions and for efficient strand transfer at the other. Mapping of the integration sites on both strands of the DNA substrates indicated that the adducts both inhibit strand transfer specifically at the minor groove bound sites and enhance integration at sites up to six nucleotides away from the adducts. These experiments demonstrate the importance of position-specific minor groove contacts for both the integrase-mediated 3'-processing and strand transfer reactions.     

Integrase of Mason-Pfizer monkey virus

The gene encoding an integrase of Mason-Pfizer monkey virus (M-PMV) is located at the 3'-end of the pol open reading frame. The M-PMV integrase has not been previously isolated and characterized. We have now cloned, expressed, isolated, and characterized M-PMV integrase and compared its activities and primary structure with those of HIV-1 and other retroviral integrases. M-PMV integrase prefers untranslated 3'-region-derived long-terminal repeat sequences in both the 3'-processing and the strand transfer activity assays. While the 3'-processing reaction catalyzed by M-PMV integrase was significantly increased in the presence of Mn(2+) and Co(2+) and was readily detectable in the presence of Mg(2+) and Ni(2+) cations, the strand transfer activity was strictly dependent only on Mn(2+). M-PMV integrase displays more relaxed substrate specificity than HIV-1 integrase, catalyzing the cleavage and the strand transfer of M-PMV and HIV-1 long-terminal repeat-derived substrates with similar efficiency. The structure-based sequence alignment of M-PMV, HIV-1, SIV, and ASV integrases predicted critical amino acids and motifs of M-PMV integrase for metal binding, interaction with nucleic acids, dimerization, protein structure maintenance and function, as well as for binding of human immunodeficiency virus type 1 and Rous avian sarcoma virus integrase inhibitors 5-CI-TEP, DHPTPB and Y-3.     

Interactions of HIV-1 DNA heterocyclic bases with viral DNA

Human immunodeficiency virus type 1 integrase is one of three viral enzymes, and it realizes a key process of the viral replication cycle, i.e. viral DNA integration into infected cell genome. Integrase recognizes nucleotide sequences located at the ends of the viral DNA U3 and U5 LTRs and catalyzes 3'-processing and strand transfer reactions. To study the interactions between integrase and viral DNA at present work, we used modified integrase substrates mimicking the terminal U5 LTR sequence and containing non-nucleoside insertions in one or/and both strands. It is shown that the substrate modifications have no influence on the integrase binding rate, while the heterocyclic bases removal in the 5th and 6th substrate positions and in the 3rd position of the substrate processed strand distinctly inhibits the integrase catalytic activity. This fact demonstrates these bases significance for the active enzyme/substrate complex formation. On the contrary, modification of the 3rd position within substrate non-processed strand stimulates 3'-processing. Since heterocyclic base elimination results in disruption of the DNA complementary and staking interactions, this result shows that DNA double helix destabilization close to the cleaved bond promotes the 3'-processing.     

Structural determinants for HIV-1 integrase inhibition by beta-diketo acids.

Among all the HIV-1 integrase inhibitors, the beta-diketo acids (DKAs) represent a major lead in anti-HIV-1 integrase drug design. These derivatives inhibit the integration reaction in vitro with a strong specificity for the 3'-end joining step. They are also antiviral and inhibit integration in vivo. The aim of the present study has been to investigate the molecular interactions between DKAs and HIV-1 integrase. We have compared 5CITEP with one of the most potent DKAs reported by the Merck group (L-708,906) and found that 5CITEP inhibits 3'-processing at concentrations where L-708,906 is only active on strand transfer. We also report a novel bifunctional DKA derivative that inhibits 3'-processing even more effectively than 5CITEP. The interactions of these inhibitors with the viral DNA donor ends have been studied by performing experiments with oligonucleotides containing defined modifications. We propose that the bifunctional DKA derivative binds to both the acceptor and donor sites of HIV-1 integrase, whereas the monofunctional L-708,906 derivative binds selectively to the acceptor site.     

Methylphosphonodiester substitution near the conserved CA dinucleotide in the HIV LTR alters both extent of 3'-processing and choice of nucleophile by HIV-1 integrase.

We present evidence suggesting that the 3'-processing activity of HIV-1 integrase is dramatically affected by electrostatic and/or steric perturbations 3' to the conserved CA dinucleotide. When the phosphodiester bond 3' to the scissile phosphodiester is replaced by a methylphosphonodiester linkage, 3'-processing decreases by two orders of magnitude. This block of cleavage can be somewhat overcome by increasing the pH of the reaction. Labeling of the substrates at the 3'-end revealed blockage of water and glycerol, but stimulation of the viral DNA 3'-hydroxyl, acting as the nucleophile with the methylphosphonodiester substrate. Interestingly, a circular trinucleotide was formed using the phosphodiester and methylphosphonodiester substrates when the terminal nucleotide was 3'-deoxyadenosine but not 2'-deoxyadenosine. Mutagenesis of the enzyme active site has previously been shown to alter the choice of nucleophile in the 3'-processing reaction. Taken together, the results in this study suggest that 'mutagenesis' of the DNA backbone can also alter the choice of nucleophile.     

Relationship between the oligomeric status of HIV-1 integrase on DNA and enzymatic activity

The 3'-processing of the extremities of viral DNA is the first of two reactions catalyzed by HIV-1 integrase (IN). High order IN multimers (tetramers) are required for complete integration, but it remains unclear which oligomer is responsible for the 3'-processing reaction. Moreover, IN tends to aggregate, and it is unknown whether the polymerization or aggregation of this enzyme on DNA is detrimental or beneficial for activity. We have developed a fluorescence assay based on anisotropy for monitoring release of the terminal dinucleotide product in real-time. Because the initial anisotropy value obtained after DNA binding and before catalysis depends on the fractional saturation of DNA sites and the size of IN.DNA complexes, this approach can be used to study the relationship between activity and binding/multimerization parameters in the same assay. By increasing the IN:DNA ratio, we found that the anisotropy increased but the 3'-processing activity displayed a characteristic bell-shaped behavior. The anisotropy values obtained in the first phase were predictive of subsequent activity and accounted for the number of complexes. Interestingly, activity peaked and then decreased in the second phase, whereas anisotropy continued to increase. Time-resolved fluorescence anisotropy studies showed that the most competent form for catalysis corresponds to a dimer bound to one viral DNA end, whereas higher order complexes such as aggregates predominate during the second phase when activity drops off. We conclude that a single IN dimer at each extremity of viral DNA molecules is required for 3'-processing, with a dimer of dimers responsible for the subsequent full integration.     

Effect of DNA modifications on DNA processing by HIV-1 integrase and inhibitor binding: role of DNA backbone flexibility and an open catalytic site

Integration of the viral cDNA into host chromosomes is required for viral replication. Human immunodeficiency virus integrase catalyzes two sequential reactions, 3'-processing (3'-P) and strand transfer (ST). The first integrase inhibitors are undergoing clinical trial, but interactions of inhibitors with integrase and DNA are not well understood in the absence of a co-crystal structure. To increase our understanding of integrase interactions with DNA, we examined integrase catalysis with oligonucleotides containing DNA backbone, base, and groove modifications placed at unique positions surrounding the 3'-processing site. 3'-Processing was blocked with substrates containing constrained sugars and alpha-anomeric residues, suggesting that integrase requires flexibility of the phosphodiester backbone at the 3'-P site. Of several benzo[a]pyrene 7,8-diol 9,10-epoxide (BaP DE) adducts tested, only the adduct in the minor groove at the 3'-P site inhibited 3'-P, suggesting the importance of the minor groove contacts for 3'-P. ST occurred in the presence of bulky BaP DE DNA adducts attached to the end of the viral DNA suggesting opening of the active site for ST. Position-specific effects of these BaP DE DNA adducts were found for inhibition of integrase by diketo acids. Together, these results demonstrate the importance of DNA structure and specific contacts with the viral DNA processing site for inhibition by integrase inhibitors.     

2' Derivatives of guanosine and inosine cyclic 3',5'-phosphates. Synthesis, enzymic activity, and the effect of 8-substituents.

A series of representative derivatives of guanosine cyclic 3',5'-phosphate (cGMP) and inosine cyclic 3',5'-phosphate (cIMP) which contained modifications in either the 2' position or the 8 and 2' positions were synthesized. Three types of derivatives were investigated: (1) derivatives in which the 2' position has been altered to produce a 2'-deoxynucleoside cyclic 3',5'-phosphate or a 9-beta-D-arabinofuranosylpurine cyclic 3',5'-phosphate; (2) 2'-omicron-acyl derivatives; and (3) doubly modified derivatives containing a 2' modification [as in (1) and (2)] and an 8-substitution. 2'-Deoxyinosine cyclic 3',5'-phosphate and 9-beta-D-arabinofuranosylhypoxanthine cyclic 3',5'-phosphate were obtained by HNO2 deamination of 2'-deoxyadenosine cyclic 3',5'-phosphate and 9-beta-D-arabinofuranosyladenine cyclic 3',5'-phosphate (ara-cAMP), respectively. Treatment of 8-bromo-2'-omicron-(p-toluenesulfonyl) adenosine cyclic 3',5'-phosphate with NaSH yielded the intermediate 8,2'-anhydro-9-beta-D-arabinofuranosyl-8-mercaptoadenine cyclic 3',5-phosphate, which was converted directly to 2'-deoxyadenosine cyclic 3',5'-phosphate (dcAMP) by treatment with Raney nickel. 8-Bromo-2'-omicron-(p-toluenesulfonyl) guanosine cyclic 3',5'-phosphate was converted to 8,2'-anhydro-9-beta-D-arabinofuranosyl-8-mercaptoguanine cyclic 3',5'-phosphate, and the latter was desulfurized with Raney nickel to give 2-deoxyguanosine cyclic 3',5'-phosphate. Ara-cAMP, 9-beta-D-arabinofuranosylguanine cyclic 3',5'-phosphate, and 9-beta-D-arabinofuranosyl-8-mercaptoguanine cyclic 3',5'-phosphate have been previously reported (Mian et al. (1974), J. Med. Chem. 17, 259). 8-Bromo-2'-omicron-acetylinosine cyclic 3',5'-phosphate and 8-[(p-chlorophenyl)thio]-2'-omicron-acetylinosine cyclic 3',5'-phosphate were produced by acylation of 8-bromoinosine cyclic 3',5'-phosphate and 8-[(p-chlorophenyl)thio]inosine cyclic 3',5'-phosphate, respectively; while 8-bromo-2'-omicron-butyrylguanosine cyclic 3',5'-phosphate was synthesized by bromination of 2'-omicron-butyrylguanosine cyclic 3',5'-phosphate.     

HIV-1 integrase preassembled on donor DNA is refractory to activity stimulation by LEDGF/p75

LEDGF/p75 is known to enhance the integrase strand transfer activity in vitro, but the underlying mechanism is unclear. Using an integrase assay with a chemiluminescent readout adapted to a 96-well plate format, the effect of LEDGF/p75 on both the 3'-processing and strand transfer steps was analyzed. Integrase inhibitors of the strand transfer reaction remained active in the presence of LEDGF/p75, but displayed 3- to 7-fold higher IC50 values. Our analyses indicate that, in the presence of 150 nM LEDGF/p75, active integrase/donor DNA complexes were increased by 5.3-fold during the 3'-processing step. In addition, these integrase/donor DNA complexes showed a 4.5-fold greater affinity for the target DNA during the subsequent strand transfer step. We also observed a 3.7-fold increase in the rate constant of catalysis of the strand transfer step when 150 nM LEDGF/p75 was present during the 3'-processing step. In contrast, when LEDGF/p75 was added at the beginning of the strand transfer step, no increase in either the concentration of active integrase/donor DNA complex or its rate constant of strand transfer catalysis was observed. This observation suggested that the integrase/donor DNA formed in the absence of LEDGF/p75 became refractory to the stimulatory effect of LEDGF/p75. Instead, this LEDGF/p75 added at the start of the strand transfer step was able to promote the formation of a new cohort of active integrase/donor DNA complexes which became functional with a delay of 45 min after LEDGF/p75 addition. We propose a model whereby LEDGF/p75 can only bind integrase before the latter binds donor DNA whereas donor DNA can engage either free or LEDGF/p75-bound integrase.     

Major and minor groove contacts in retroviral integrase-LTR interactions

The 3'-processing activities of HIV-1, HTLV-2, and M-MuLV integrases (INs) with their corresponding U5 end of the viral DNA molecule were examined to define functional group determinants of U5 terminus recognition and catalysis. Nucleotide analogues were incorporated into the U5 terminus to produce conservative modifications in the surface of the major and/or minor grooves to map the hydrogen-bonding contacts required for LTR-IN interaction. Specifically, the phylogenetically conserved CA (positions 4 and 3, respectively) and the 5'-proximal nucleotide (position 5) were replaced with base analogues in plus and/or minus strands. For each integrase, similar major and minor groove contacts were identified in the guanine and adenine of the conserved CA/GT. Overall, perturbances in the minor groove resulted in a greater decrease in 3'-processing activity than the major groove substitutions. Additionally for HIV-1 and HTLV-2 INs, we observed an increase in the 3'-processing activity with an O4-MeThy substitution at position 3 of the minus strand. O4-MeThy may act to destabilize Watson-Crick base pairing and in doing so provide these INs with a more favorable interaction with the adjacent scissile bond. At position 5, a substantial divergence among the three INs was noted in the functional groups required for 3'-processing activity, thereby supporting the role of this position in providing some level of substrate specificity.     

Pre-organized structure of viral DNA at the binding-processing site of HIV-1 integrase

The integration of the human immunodeficiency virus type 1 DNA into the host cell genome is catalysed by the viral integrase (IN). The reaction consists of a 3'-processing [dinucleotide released from each 3' end of the viral long terminal repeat (LTR)] followed by a strand transfer (insertion of the viral genome into the human chromosome). A 17 base pair oligonucleotide d(GGAAAATCTCTAGCAGT), d(ACTGCTAGAGATTTTCC) reproducing the U5-LTR extremity of viral DNA that contains the IN attachment site was analysed by NMR using the classical NOEs and scalar coupling constants in conjunction with a small set of residual dipolar coupling constants (RDCs) measured at the 13C/15N natural abundance. The combination of these two types of parameters in calculations significantly improved the DNA structure determination. The well-known features of A-tracts were clearly identified by RDCs in the first part of the molecule. The binding/cleavage site at the viral DNA end is distinguishable by a loss of regular base stacking and a distorted minor groove that can aid its specific recognition by IN.     

Synthesis and triplex-forming properties of cyclic oligonucleotides with (G,A)-antiparallel strands

Cyclic oligonucleotides carrying an oligopurine Watson-Crick sequence linked to the corresponding (G,A)- and (G,T)-antiparallel strands were prepared by nonenzymatic template-assisted cyclization of phosphorylated precursors. Cyclization was attempted using 3'-phosphate and 5'-phosphate linear precursors with carbodiimide or BrCN activation. The best results were obtained with the 5'-phosphorylated precursors and carbodiimide activation. Cyclic oligonucleotides bind polypyrimidine target sequence by formation of antiparallel triplexes. We have used UV and circular dichroism (CD) spectroscopy to analyze triplexes formed by cyclic oligonucleotides carrying G and A in the reverse-Hoogsteen strand. The relative stability of the triplexes formed by cyclic and linear oligonucleotides with a common polypyrimidine target was determined by melting experiments. The most-stable triplexes were formed by the cyclic oligonucleotide, followed by the unphosphorylated and phosphorylated oligonucleotide precursors, and, finally, the corresponding hairpin. Although the differences in binding affinity between cyclic oligonucleotides and their corresponding linear precursors are small, the use of cyclic oligonucleotides offers a clear advantage over conventional duplex recognition.     

Single amino acid substitution in HIV-1 integrase catalytic core causes a dramatic shift in inhibitor selectivity

HIV-1 integrase (IN) mediates the insertion of viral cDNA into the cell genome, a vital process for replication. This step is catalyzed by two separate DNA reaction events, termed 3'-processing and strand transfer. Here, we show that six inhibitors from five structurally different classes of compounds display a selectivity shift towards preferential strand transfer inhibition over the 3'-processing activity of IN when a single serine is substituted at position C130. Even though IN utilizes the same active site for both reactions, this finding suggests a distinct conformational dissimilarity in the mechanistic details of each IN catalytic event.     

Interaction of "aza" and "deaza" analogs of adenosine cyclic 3', 5'-phosphate with some enzymes of adenosine cyclic 3', 5'-phosphate metabolism: evidence that the lone pair electrons of N-3 are involved in the binding of adenosine cyclic 3', 5'-phosphate to type II adenosine cyclic 3', 5'-phosphate-dependent protein kinase.

Five hetercyclic analogs of adenosine cyclic 3',5'-phosphate (cyclic AMP) were examined for their ability (1) to stimulate type II cyclic AMP-dependent kinases from bovine brain, bovine heart, and rat liver; (2) to serve as substrates for "high Km" (Km for cyclic AMP = 0.13-0.43 mM) cyclic nucleotide phosphodiesterases from bovine heart, rabbit kidney, and rat liver; and (3) to inhibit the hydrolysis of cyclic AMP catalyzed by "low Km" (Km for cAMP = 0.32-1.5 muM) cyclic nucleotide phosphodiesterases from bovine brain, bovine heart, dog heart, rabbit liver, rat brain and rat liver. The analogs all had a purine ring system which had been modified by replacement of a ring carbon with nitrogen or vice versa to yield 2-aza-cAMP (7-amino-4-beta-D-ribofuranosylimidazo [4,5-d] -v-triazine cyclic 3',5'-phosphate); 8-aza-cAMP (7-amino-3-beta-D-ribofuranosyl-v-triazolo-[4,5-d]-pyrimidine cyclic 3',5'-phosphate); 1 deaza-cAMP (7-amino-3-beta-D-ribofuranosylimidazo [4,5-b[pyridine cyclic 3',5'-phosphate); 3-deaza-cAMP (4-amino-1-beta-D-ribofuranosylimidazo[4,5-c]pyridine cyclic 3',5'-phosphate) and 7-deaza-cAMP (7-amino-4-beta-D-ribofuranosylpyrrolo[2,3-d]pyrimidine cyclic 3',5'-phosphate).