HIV-1 has three enzymes: protease, reverse transcriptase (RT) and integrase (IN). The third, IN, is being vigorously pursued as a novel therapeutic target [1–4]. Efficacy of IN inhibitors will ultimately depend on the ability of such agents to prevent or to slow replication of viruses found in infected individuals. The error rate of reverse transcription, coupled with the high rate of viral turnover in infected individuals, makes HIV sequence variability within infected individuals inevitable. Indeed, genetic variability has been documented for protease [5,6], RT [5,6] and the envelope glycoproteins . Some of this genetic variability can result in phenotypic changes that are critical to drug development. For example, mutations that lead to protease resistance have been found even among untreated individuals [8,9]. Therefore, IN sequence variability from clinical isolates is likely and the effects of such mutations on IN function may aid in our understanding of integration.
Upon viral entry into the cell and after reverse transcription of the viral genome, IN catalyzes the integration of a cDNA copy of the genome into the host chromosome (for review see Asante-Appiah and Skalka ). IN can perform three enzymatic functions: 3′ end processing, strand transfer and disintegration. All three reactions can be performed in vitro using oligonucleotides that resemble the viral long terminal repeat (LTR) ends and recombinant IN purified from Escherichia coli[11,12]. The 3′ end-processing reaction catalyzes the removal of two nucleotides from each 3′ end of the viral DNA, thus exposing an absolutely conserved CpA dinucleotide. Additionally, IN catalyzes a concerted cleavage–ligation reaction resulting in viral DNA that is covalently attached to target DNA, a process called strand transfer. Studies have shown that both 3′ end processing and strand transfer result from a direct transesterification reaction . Disintegration is the reverse of the strand-transfer reaction in which IN can resolve a partially integrated piece of DNA into two separate entities ; this reaction has only been observed in vitro. Disintegration is especially useful for mechanistic studies since the core region of the protein, amino acid residues 50–212, is sufficient to catalyze the disintegration reaction whereas both N- and C-termini are required for 3′ end processing and strand transfer [14,15].
IN is a 32 kDa protein with three distinct regions: the N-terminus, the core and the C-terminus. The N-terminus contains a HHCC (HX3−−7HX23−−32CX2C) motif where H is histidine, C is cysteine and X is any amino acid. It is believed to be important in protein–protein interactions and aids in the the formation of multimers of IN . The core contains three absolutely conserved residues that make up the catalytic triad, an aspartic acid residue at position 64, another aspartic acid at position 116, and a glutamic acid at position 152. This DD35E motif is highly conserved throughout retroviral IN as well as in some bacterial transposases  and has been shown to be absolutely necessary for enzyme activity [18,19]. The C-terminus shares some homology to the Src homology region 3 (SH3), common to some signal transduction molecules [20,21] and is believed to play a role in binding to host DNA.
To date, there has been limited IN sequence analysis from clinical isolates of HIV. Mutations within IN have been shown to have dramatic effects on enzyme function in vitro and in vivo[22–26]. In this study, clinical isolates of HIV RNA were sequenced for IN and the effects of mutations in IN on function and viral replication were compared with IN from a previously unstudied laboratory adapted strain of HIV, HIVIIIRF, and the prototypic HIV IN derived from HIVNL4−−3.
Materials and methods
HIV was isolated from four patients and used for RNA isolation and assessment.
The following oligonucleotides were synthesized and desalted by SigmaGenosys (The Woodlands, Tx). The numbers in parentheses are the corresponding nucleotide numbers in the HIV genome based on the sequence of HIVNL4−−3.
- INS: 5′-GGTCTCCGCGGGAATCAGGAAAGTAC-3′ (3579–3605)
- INX: 5′-GCTTTTCTAGAAATATACATATGGTG-3′ (4497–4522)
- Core1: 5′-CAGCTGTGATAAATGTCAGCTA-3′ (3721–3742)
- Core2: 5′-CCATTTGTACTGCTGTCTTAA-3′ (4122–4142)
- INSpf: 5′-GCAATTTCACCAGTACTACAGT-3′ (3963–3984)
- IN3'end: 5′-GTGACATAAAAGTAGTGCCAAG-3′ (4371–4392)
- IN5'end: 5′-GCTGGCTACTATTTCTTTTG-3′ (3705–3724)
- IN Nde: 5′-GCAATCAGGAAACATATGTTTTTA GATGGA-3′
- IN Hind: 5′-TGGTGTTTTACTAAGCTTTTCCAT CTGTTA-3'
- V1: 5′-ATGTGGAAAATCTCTAGCAGT-3′
- V2: 5′-ACTGCTAGAGATTTTCCACAT-3′
- db-Y1: 5′-TGCTAGTTCTAGCAGGCCCTTGGGC CGGCGCTTGCGCC-3′
Cells and viruses
MT2 cells, a CD4 human T-cell leukemia cell line, HeLa–CD4 cells, a cervical carcinoma cell line that expresses CD4 owing to stable transfection, H9 cells, a human CD4 T lymphoblastoid cell line, and CEM-SS cells, a human CD4 T-lymphoblastoid cell line that is extremely sensitive to HIV infection, were cultured in RPMI-1640 supplemented with 11.5% fetal bovine serum and 2 mmol/l l-glutamine. Clinical isolates of HIV were obtained according to institutional review board approved protocols from the University of California, Irvine Medical Center (UCIMC) (isolate HIVR19) and Vanderbilt University (isolates HIVR101, HIVR104 and HIVR106). All infectious molecular clones were derived from the wild-type HIVNL4−−3 plasmid, which contains silent mutations that allow the replacement of wild-type IN by variant IN, as described previously . HIV7−−3is a reference strain containing the silent mutations used for cloning and an IN gene from HIVNL4−−3 that was multiply passaged in tissue culture . This virus has been shown to be identical to HIVNL4−−3. The molecular clones were transfected into HeLa–CD4 cells and replicated in H9 cells, as described previously .
HIV-positive serum was overlayed onto MT2 cells and virus replication was measured by antigen production using an indirect immunofluorescence assay (IFA). When cultures were 100% positive by IFA, cells were removed from the culture supernatants by low-speed centrifugation followed by filtration through 0.45 μm nitrocellulose filters. HIV-positive culture supernatants were then back-passaged onto peripheral blood mononuclear cells. Cells were removed from the supernatant fluids and the virions were amplified one time on CEM-SS. Total passage number was three, including the original isolation from the patient. Virions were pelleted by low-speed centrifugation as described previously .
Cloning of integrase sequences
Cloning was performed as described . Briefly, pelleted virions were lysed and the RNA isolated (Gentra Systems, Minneapolis, MN). This RNA was subjected to a reverse transcription reaction using Moloney murine leukemia virus (Mo-MLV) RT (Superscript II, Life Technologies, Rockville, MD) and an IN-specific primer, INX. The cDNA product was used as the template in a polymerase chain reaction (PCR) utilizing a high-fidelity polymerase, Pfu (Stratagene La Jolla, CA), and primers that amplify the IN sequence (INX and INS). The PCR products were then ligated into pPCR-Script (Stratagene la Jolla, CA) and positive colonies were identified by blue/white screening using 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside. Plasmids from positive colonies were sequenced using the Sequenase 2.0 kit (USB, Cleveland, OH). Sequencing primers used were INS, INX, INSpf, Core1, Core2, IN 3′ end and IN 5′ end. Re-isolation and re-sequencing of several different viruses, including two clinical isolates and HIVNL4−−3, has indicated that both tissue culture adaptations and PCR-induced mutations are rare to non-existent using this protocol (unpublished results).
The INS and INX primers used to amplify the original sequence have silent mutations that introduce unique restriction sites, Sac II and Xba I, respectively; this allows only the IN sequence to be cloned into an HIV infectious molecular clone (pNL4–3), which also contains these unique sites . IN from pPCR-Script clones were amplified using Taq polymerase and IN Nde I and IN Hin dIII primers; this allowed the IN gene to be cloned into a bacterial expression vector, pT7.7, using Nde I and Hin dIII restriction endonucleases as described previously  The resulting vector contains a 6-histidine tag fused to IN under the control of a T7 RNA polymerase promoter. All clones were re-sequenced to ensure that neither the amplification of IN for insertion into HIVNL4−−3 nor the amplification of IN for insertion into pT7.7 resulted in additional mutations.
Purification of integrase
IN protein was induced using isopropyl-β-d-thiogalactopyranoside, expressed in E. coli and purified as described previously by Ni2+ affinity chromatography (Pharmacia, Peapack, NJ) . Protein concentrations were determined using a commercially available Coomassie brilliant blue assay (Pierce, Rockford, IL). Protein purity was assessed by sodium dodecylsulfate polyacrylamide gel electrophoresis (SDS–PAGE).
Integration assays utilizing recombinant IN were performed with two oligonucleotide substrates. The DNA substrate oligonucleotides (db-Y1 or ‘dumbbell’ and V1/V2) were purified using 15% PAGE. Purified db-Y1 and V1 were 5′ end-labeled with [γ-32P]-ATP and purified over a p6 micro spin column (Biorad, Hercules, CA) as described previously . The purified substrates were annealed by heating to 95°C for 5 min, followed by slow cooling to room temperature.
Processing of 3′ end and strand-transfer activity
The 3′ end-processing and strand-transfer reactions were performed as described previously [11,12]. IN from clinical isolates of HIV-1 were diluted in storage buffer (20 mmol/l HEPES pH 7.5, 10 mmol/l dithiothreitol (DTT), 0.3 mol/l NaCl, 10 mmol/l 3-[3-(cholamidopropyl)-dimethyl-ammonio]-1-propanesulfonate, 20% glycerol). Next, 1.0, 2.0, 3.0, 4.0, 5.0 or 8.0 pmol IN was added to reaction buffer (20 mmol/l HEPES pH 7.5, 10 mmol/l MgCl2, 35 mmol/l NaCl, 10 mmol/l dithiothreitol, 0.5% NP-40, 15% dimethyl sulfoxide). The reactions were started by the addition of 0.2 pmol V1/V2 substrate. Each concentration was assessed in quadruplicate apart from the 8 pmol concentration, which was in duplicate. Reactions proceeded for 1 h at 37°C and then were stopped by the addition of ethylenediaminetetraacetic acid (EDTA) pH 8.0 to a final concentration of 18 mmol/l. Gel loading buffer (98% formamide, 10 mmol/l EDTA pH 8.0, 0.05% bromophenol blue and 0.05% xylene cyanol) was then added and the 3′ end-processing and strand-transfer products were separated by 15% denaturing PAGE. Product formation was quantified using a Phosphorimager and ImageQuant software (Molecular Dynamics, Sunnyvale, CA).
Disintegration assays were performed by a protocol modified from Chow et al.. Purified IN was diluted from the original stocks in storage buffer to the appropriate concentrations. The concentration range chosen for each of the proteins was that in which there was a linear relationship between product formation and enzyme concentration. For HIV7−−3, HIVR101 and HIVR104IN, reaction concentrations of 0.5, 1.0, 1.5, 2.0 and 3.0 pmol protein were used; for HIVR19, HIVR106 and HIVRF IN, 2.0, 2.5, 3.0, 3.5 and 4.0 pmol protein were used. Each reaction contained reaction buffer with 30 mmol/l NaCl and 4.5 pmol labeled db-Y1 substrate in a final volume of 20 μl. Reactions were started by the addition of substrate and incubated for 1 h at 37°C. The reactions were stopped with EDTA. Gel loading buffer was then added and reaction products were separated using 15% denaturing PAGE. Quantification was performed on a Phosphorimager using ImageQuant software.
Viral clones were transfected into HeLa–CD4 cells with Lipofectin (Stratagene). After 2 days, H9 cells were added and the virus was allowed to replicate until the culture was 100% positive by IFA. Virus supernatant was separated and stored at −70°C. Virus supernatants were thawed and the 50% tissue culture infectious dose (TCID50) was determined by endpoint dilution, as described previously . In separate experiments in triplicate wells of 24-well plates, 10 000 cpm RT activity, 10 000 TCID50 or 25 000 TCID50 of each clone was added to MT2 cells. Each culture was monitored by IFA and RT assay every other day for 8 days. Only the data for the 10 000 cpm RT are reported but similar results were found for the other inoculations (10 000 TCID50 and 25 000 TCID50).
Susceptibility to l-chicoric acid
Recombinant HIVNL4−−3 containing IN genes from HIV7−−3, HIVR19, HIVR101, HIVR104, HIVR106 or HIVIIIRF was isolated from H9 cultures and frozen at −70°C. Susceptibility of each recombinant virus to l-chicoric acid was determined using a cytopathic effect assay, as described previously . Briefly, MT-2 cells were infected in triplicate at a multiplicity of infection of < 1 in the presence of serial dilutions of l-chicoric acid. After 6 or 7 days, the cells were harvested and viability was determined using Finter's Neutral red dye . This assay is highly correlated with direct measures of viral replication such as IFA, RT assay and viral RNA synthesis . The ability of l-chicoric acid to inhibit IN7−−3, INIIIRF, INR19, INR101, INR104 and INR106 was determined essentially as described, previously [1,27,30]. The only changes made were the IN proteins used for the assays.
Sequence analysis of clinical isolates of HIV integrase
RNA was harvested from minimally passaged HIV obtained from North American HIV-positive individuals. Each cDNA was then cloned and three to nine clones for each virus were sequenced. The complete IN sequence for the four unique isolates (HIVR101, HIVR104, HIVR106−−1, HIVR19), one laboratory adapted strain (HIVIIIRF) and one reference strain (HIV7−−3) have been deposited in GenBank (Accession numbers: HIVR101 AF203329; HIVR104 AF203330; HIVR106 AF203331; HIVR19 AF203332). The IN sequences contain between 3 and 5% variability at the amino acid level when compared with HIV7−−3. Conservative, as well as non-conservative, changes were found in all of the IN sequences, although there were more conservative than non-conservative changes for each protein (Table 1). None of these IN contained premature stop codons or changes to essential amino acids  in the N-terminus (HHCC) or core regions (DD35E) of the protein. Such results suggested that the IN genes encoded active IN proteins. Although tissue culture, PCR and cloning may select for specific variants, we have repeatedly passaged viruses in several cell lines and have amplified several isolates using PCR on more than one occasion and seen no genotypic changes, suggesting that neither the isolation nor amplification protocols used were likely causes of significant genotypic variation.
To determine phylogenetic relationships between these IN and several other known HIV IN, maximum likelihood analysis was performed. This enabled the construction of a phylogenetic tree based solely on IN base composition (Fig. 1). The HIV isolates described herein group relatively close together but are distinct from each other and other previously described HIV IN genes. Each of these IN sequences contained a unique array of amino acid changes that have not been identified among other HIV isolates. The reference sequences for comparison are from Skinner et al..
Expression of integrase
IN sequences derived from HIV7−−3, HIVR19, HIVR101, HIVR104, HIVR106 and HIVIIIRF were cloned into a bacterial expression vector and expressed in E. coli. The IN proteins (designated IN7−−3, INR19, INR101, INR104, INR106, and INIIIRF, respectively) were subsequently purified using Ni2+ affinity chromatography. The final enzyme preparations resulted in proteins that were of the predicted size (approximately 32 kDa) and were > 95% pure (not shown).
Specific activity of 3′ end processing for variant integrases
The 3′ end-processing specific activities were determined for each variant IN using in vitro assays of integration. The specific activity was derived by plotting the amount of 3′ end-processing product, including the strand-transfer product, formed within 1 h against the IN present in each reaction. A representative experiment is shown in Fig. 2a. Calculating the slope of the plot gave the activity for each of the proteins. Averaging the slopes of a minimum of three experiments determined the 3′ end-processing activities. The specific activities of INR101, INR104 and INR106were statistically the same as IN7−−3. In contrast, the INIIIRFand INR19showed statistically relevant attenuation when compared to IN7−−3.
Disintegration molecular activities of integrases from clinical isolates
The disintegration molecular activity of each of the variant HIV IN proteins was determined to examine the nature of IN attenuation observed for HIVIIIRF and HIVR19. The assays were performed in substrate excess so that the reaction approached zero order in respect to substrate concentration. The activity values were derived by calculating the slope for each protein within the linear range of the activity curve. A typical experiment is shown in Fig. 2b. For each protein, at least three separate experiments were performed and the slopes were averaged. The proteins fall into two distinct groups based on disintegration activity compared with wild-type IN. Three proteins had statistically significant differences in disintegration activity, INR19, INR106, and INIIIRF, and two proteins, INR101, and INR104, did not differ statistically from the reference IN7−−3.
Growth kinetics of HIVNL4−−3 containing variant integrases
Using the infectious molecular clone HIVNL4−−3, IN was replaced with each of the mutant IN forms and the resulting plasmid transfected into HeLa–CD4 cells and amplified on H9 cells. First, MT2 cells were infected with equal amounts of RT activity (10 000 cpm per 5 × 105 cells). The culture supernatants and cells, respectively, were assayed every day for RT activity and for HIV antigen synthesis by IFA (Fig. 3). Mutations within IN have been shown to affect RT activity ; therefore, to control for possible IN-induced RT defects, equal amounts of infectious virus were used to infect the target cells (10 000 infectious particles per 5 × 105 cells and 25 000 infectious particles per 5 × 105 cells) and infections monitored for RT activity and HIV antigen synthesis. The growth kinetics for all sets of experiments, performed at least twice, showed no significant effects on virus growth compared with HIV7−−3, indicating that the mutations had only a slight effect on viral replication.
Susceptibility of variant integrase proteins to l-chicoric acid
l-Chicoric acid is a potent and selective inhibitor of IN [1,30,34] that is amenable to derivatization [27,37]. Resistance to l-chicoric acid maps to the IN gene  and the compound acts in synergy with existing antiretroviral agents [35,36] these properties suggest that l-chicoric acid is an excellent lead candidate for an IN inhibitor. All of the IN proteins showed identical susceptibilities to l-chicoric acid (Table 2). Therefore, naturally occurring mutations in IN do not appear to be a hindrance to inhibitor design.
Mutations in non-conserved regions of retrotransposon IN have been shown to have dramatic affects on the life cycle of these transposable elements . The relative similarity of HIV IN to certain retrotransposon IN genes suggests the same may be true for HIV. Consequently, naturally occurring mutations within the HIV-1 IN gene could lead to differences in protein activity and to alterations in the viral life cycle.
Sequence analysis suggests that IN is relatively well conserved but can tolerate mutations at certain amino acid positions. Comparison of the proteins sequenced here with sequences identified by others [31,32] has identified six highly conserved regions within HIV IN (amino acid residues 1–10, 50–100, 139–152, 175–185, 235–251 and 260–265). Skinner et al. had previously published nucleotide sequence data for 100 IN from 10 HIV-1-infected individuals. These clones were derived from integrated cellular DNA amplified using Taq polymerase. Variable regions they reported  correspond well with the variable regions we identified here. However, the effects of the mutations they observed on the functions of IN were not reported . In contrast, the sequences described here are of complete IN genes that all resulted in active proteins and were derived from infectious virions not integrated sequences.
Sequence analysis of the variant IN has identified conservative mutations in some highly conserved residues such as V150A (HIVR101), F185Y (HIVR19) and M154L (HIVR104). Previous studies on the F185 residue have shown that the protein can tolerate certain changes at this position that dramatically increase protein solubility (F185K and F185H) but both F185K and F185A significantly impaired viral replication [39,40]. IN from HIVR19 containing the F185Y mutation showed some degree of attenuation for both 3′ end processing and disintegration but was not inhibited for growth in tissue culture. An M154I mutation has been reported to cause partial resistance to another class of IN inhibitor, the diketoacids ; the M154L mutation found in HIVR104 could be important in resistance to these compounds as well.
For each clinical isolate, three to nine clones were sequenced from the original RT-PCR reaction. With one exception, HIVR106, all clones from the same individual were identical. This may mean that IN sequences are highly conserved within an individual. Alternatively, the isolation methods for these studies may have selected for rapidly growing isolates of HIV that replicate in established T cell lines. Therefore, the lack of variability in clones within an individual may be caused by selection bias secondary to biological cloning. This interpretation is supported by the data of Skinner et al., who observed sequence variability in multiple clones from the same individual . However, the small degree of IN sequence variability from multiple studies, including Overbaugh et al. (unpublished results), Skinner et al. and our results, would suggest that IN variability is biologically constrained. Protein sequences differed from the HIV7−−3 IN reference sequence by about 5%, with INR104 being the most divergent. While the HHCC and DD35E motifs were conserved, there was some variability around the second and third core catalytic residues (D116 and E152) and near the two histidine residues of the HHCC domain in HIVR101, HIVR104, HIVR106 and HIVIIIRF.
The disintegration reaction has been shown to require only the core portion of the protein, demonstrating that the core mediates catalysis. Three of the proteins described here (INR19, INIIIRF and INR106) showed a slight, but statistically relevant, attenuation in disintegration activity when compared with IN7−−3. This suggests that the mutations found within the core region of these proteins may directly affect the catalytic process of integration. The 3′ end-processing and strand-transfer activities for the clinical isolates correlated with disintegration except for INR106. INR106 contained a non-conservative mutation inside the core region that is not found in any of the other proteins tested. Therefore, it is possible the K111T mutation compensates for a slightly impaired core. Both INIIIRF and INR19 showed a slight but statistically relevant attenuation in 3′ end processing whereas the other proteins did not differ statistically from IN7−−3. Examination of both disintegration and end-processing results indicate that the in vitro activities of the mutant proteins were all relatively similar, with minor differences observed in three of the proteins.
The influences of the mutant IN on viral life cycle were tested by replication kinetic assays in which the wild-type IN gene from HIVNL4−−3 was replaced with each mutant IN. There were no statistically relevant differences in the replication kinetics of the viruses, indicating that minor changes in activity do not grossly affect viral replication. Indeed, it has been shown by several investigators that a greater than fourfold difference in IN activity is needed to produce a noticeable effect in tissue culture replication of HIV viruses (P. J. King and W. E. Robinson, Jr, unpublished data [26,39,41,42]). The results of each of these experiments indicated no discernable differences in replication whether based on RT or numbers of infectious particles, indicating that these mutations also have no gross affect on virus replication or RT activity.
The continued development of IN inhibitors will eventually depend on their activity against clinical isolates of HIV. The finding that all five variant IN proteins have similar susceptibility to an inhibitor of IN, l-chicoric acid, demonstrated that natural variation within the IN gene may not be a barrier to inhibitor design and synthesis. Furthermore, since resistance to l-chicoric acid resulted in slightly impaired viral replication  it may be that resistance to an inhibitor of IN will be less likely to develop than resistance to current inhibitors of HIV targeted to RT and protease. These data, combined with those of others, suggest that IN may be the least mutable of the three HIV enzymes (RT, protease, and IN). At the very least, there are regions of IN, which constitute over half of the protein, that are absolutely conserved; this may prove to be an important virologic constraint if IN inhibitors are developed and used clinically.
The authors would like to thank Brenda McDougall for excellent technical assistance and Samson A. Chow for his gift of the pT7.7 expression vector and for advice on integrase assays.
1. Robinson WE Jr, Cordeiro M, Abdel-Malek S. et al
. Dicaffeoylquinic acid inhibitors of human immunodeficiency virus integrase: inhibition of the core catalytic domain of human immunodeficiency virus integrase. Mol Pharmacol 1996, 50: 846 –855.
2. Hazuda DJ, Pelock P, Witmer M. et al
. Inhibitors of strand transfer that prevent integration and inhibit HIV-1 replication in cells. Science 2000, 287: 646 –650.
3. Pommier Y, Neamati N. Inhibitors of human immunodeficiency virus integrase. Adv Virus Res 1999, 52: 427 –458.
4. Robinson WE Jr. HIV integrase: the next target? Infect Med 1998, 15: 129 –137.
5. Birk M, Sonnerborg A. Variations in HIV-1 pol
gene associated with reduced sensitivity to antiretroviral drugs in treatment-naive patients. AIDS 1998, 12: 2369 –2375.
6. Shafer R, Stevenson D, Chan B. Human immunodeficiency virus reverse transcriptase and protease sequence database. Nucl Acids Res 1999, 27: 348 –352.
7. Willey R, Rutledge R, Dias S. et al
. Identification of conserved and divergent domains within the envelope gene of the acquired immunodeficiency syndrome retrovirus. Proc Natl Acad Sci USA 1986, 83: 5038 –5042.
8. Lech W, Wang G, Tang Y. et al
. In vivo
sequence diversity of the protease of human immunodeficiency virus type-1: presence of protease inhibitor resistant variants in untreated subjects. J Virol 1996, 70: 2038 –2043.
9. Cornelissen M, van den Burg R, Zorgdrager R, Lukashov F, Goudsmit J. polgene diversity of five human immumodeficiency virus type-1 subtypes: evidence for naturally occurring mutations that contribute to drug resistance, limited recombination patterns, and common ancestry for subtypes B and D.
J Virol 1997, 71: 6348 –6358.
10. Asante-Appiah E, Skalka AM. Molecular mechanisms in retrovirus DNA integration. Antiviral Res 1997, 36: 139 –156.
11. Vincent KA, Ellison V, Chow SA, Brown PO. Characterization of human immunodeficiency virus type 1 integrase expressed in Escherichia coli
and analysis of variants with amino-terminal mutations. J Virol 1993, 67: 425 –437.
12. Chow SA, Vincent KA, Ellison V, Brown PO. Reversal of integration and DNA splicing mediated by integrase of human immunodeficiency virus. Science 1992, 255: 723 –726.
13. Engelman A, Mizuuchi K, Craigie R. HIV-1 DNA integration: mechanism of viral DNA cleavage and DNA strand transfer. Cell 1991, 67: 1211 –1221.
14. Bushman FD, Engelman A, Palmer I, Wingfield P, Craigie R. Domains of the integrase protein of human immunodeficiency virus type 1 responsible for polynucleotidyl transfer and zinc binding. Proc Natl Acad Sci USA 1993, 90: 3428 –3432.
15. Engelman A, Bushman FD, Craigie R. Identification of discrete functional domains of HIV-1 integrase and their organization within an active multimeric complex. EMBO J 1993, 12: 3269 –3275.
16. Zheng R, Jenkins TM, Craigie R. Zinc folds the N-terminal domain of HIV-1 integrase, promotes multimerization, and enhances catalytic activity. Proc Natl Acad Sci USA 1996, 93: 13659 –13664.
17. Polard P, Chandler M. Bacterial transposases and retroviral integrases. Mol Microbiol 1995, 15: 13 –23.
18. Engelman A, Craigie R. Identification of conserved amino acid residues critical for human immunodeficiency virus type 1 integrase function in vitro. J Virol 1992, 66: 6361 –6369.
19. Leavitt AD, Shiue L, Varmus HE. Site-directed mutagenesis of HIV-1 integrase demonstrates differential effects on integrase functions in vitro
. J Biol Chem 1993, 268: 2113 –2119.
20. Lodi PJ, Ernst JA, Kuszewski J. et al
. Solution structure of the DNA binding domain of HIV-1 integrase. Biochemistry 1995, 34: 9826 –9833.
21. Eijkelenboom A, Lutzke R, Boelens R, Plasterk R, Kaptein R, Heard K. The DNA-binding domain of HIV-1 integrase has an SH3-like fold. Nat Struc Biol 1995, 2: 807 –810.
22. King PJ, Robinson WE Jr. Resistance to the anti-human immunodeficiency virus type 1 compound l- chicoric acid results from a single mutation at amino acid 140 of integrase. J Virol 1998, 72: 8420 –8424.
23. Cannon PM, Wilson W, Byles E, Kingsman SM, Kingsman AJ. Human immunodeficiency virus type 1 integrase: effect on viral replication of mutations at highly conserved residues. J Virol 1994, 68: 4768 –4775.
24. Lutzke R, Plasterk R. Structure based mutational analysis of the C-terminal DNA-binding domain of human immunodeficiency virus type-1 integrase: critical residues for protein oligomerization and DNA binding. J Virol 1998, 72: 4841 –4848.
25. Engelman A, Craigie R. Efficient magnesium-dependent human immunodeficiency virus type 1 integrase activity. J Virol 1995, 69: 5908 –5911.
26. Wiskerchen M, Muesing MA. Human immunodeficiency virus type 1 integrase: effects of mutations on viral ability to integrate, direct viral gene expression from unintegrated viral DNA templates, and sustain viral propagation in primary cells. J Virol 1995, 69: 376 –386.
27. King PJ, Ma G, Miao W. et al
. Structure–activity relationships: analogues of the dicaffeoylquinic and dicaffeoyltartaric acids as potent inhibitors of human immunodeficiency virus type 1 integrase and replication. J Med Chem 1999, 42: 497 –509.
28. Robinson WE Jr, Montefiori DC, Gillespie DH, Mitchell WM. Complement-mediated, antibody-dependent enhancement of human immunodeficiency virus type 1 (HIV-1) infection in vitro increases HIV-1 RNA and protein synthesis and infectious virus production. J Acquir Immune Defic Syndr 1989, 2: 33 –42.
29. Montefiori DC, Robinson WE Jr, Schuffman SS, Mitchell WM. Evaluation of antiviral drugs and neutralizing antibodies against human immunodeficiency virus by a rapid and sensitive microtiter infection assay. J Clin Microbiol 1988, 26: 231 –235.
30. Robinson WE Jr, Reinecke MG, Abdel-Malek S, Jia Q, Chow SA. Inhibitors of HIV-1 replication that inhibit HIV integrase. Proc Natl Acad Sci USA 1996, 93: 6326 –6331.
31. Cannon PM, Byles ED, Kingsman SM, Kingsman AJ. Conserved sequences in the carboxyl terminus of integrase that are essential for human immunodeficiency virus type 1 replication. J Virol 1996, 70: 651 –657.
32. Skinner LM, Lamers SL, Sanders JC, Eyster ME, Goodenow MM, Katzman M. Analysis of a large collection of natural HIV-1 integrase sequences, including those from long-term nonprogressors. J Acquir Immune Defic Syndr Hum Retrovirol 1998, 19: 99 –110.
33. Shin CG, Taddeo B, Haseltine WA, Farnet CM. Genetic analysis of the human immunodeficiency virus type 1 integrase protein. J Virol 1994, 68: 1633 –1642.
34. McDougall B, King PJ, Wu BW, Hostomsky Z, Reinecke MG, Robinson WE Jr. Dicaffeoylquinic and dicaffeoyltartaric acids are selective inhibitors of human immunodeficiency virus type 1 integrase. Antimicrob Agents Chemother 1998, 42: 140 –146.
35. Beale K, Robinson WE Jr. Combinations of reverse transcriptase, protease, and integrase inhibitors can be synergistic in vitro against drug-sensitive and RT inhibitor-resistant molecular clones of HIV-1. Antiviral Res 2000, 46: 223 –232.
36. Robinson WE Jr. l- Chicoric acid, an inhibitor of human immunodeficiency virus type 1 (HIV-1) integrase, improves on the in vitro anti-HIV-1 effect of zidovudine plus a protease inhibitor (AG1350). Antiviral Res 1998, 39: 101 –111.
37. Lin Z, Neamati N, Zhao H. et al
. Chicoric acid analogues as HIV-1 integrase inhibitors. J Med Chem 1999, 42: 1401 –1414.
38. Nymark-McHahon M, Sandmeyer S. Mutations in nonconserved domains of Ty3 integrase affect mutiple stages of the Ty3 life cycle. J Virol 1999, 73: 453 –465.
39. Engelman A, Liu Y, Chen H, Farzan M, Dyda F. Structure-based mutagenesis of the catalytic domain of human immunodeficiency virus type 1 integrase. J Virol 1997, 71: 3507 –3514.
40. Jenkins T, Engelamn A, Ghirlando R, Craigie R. A soluble active mutant of HIV-1 integrase: involvement of both the core and carboxyl-terminal domains in multimerization. J Biol Chem 1996, 271: 7712 –7718.
41. Taddeo B, Haseltine WA, Farnet CM. Integrase mutants of human immunodeficiency virus type 1 with a specific defect in integration. J Virol 1994, 68: 8401 –8405.
42. van den Ent F, Vos A, Plasterk R. Mutational scan of the human immunodeficiency virus type 2 integrase protein. J Virol 1998, 72: 3916 –3924.