Use of Patient-Derived Human Immunodeficiency Virus Type 1 Integrases To Identify a Protein Residue That Affects Target Site Selection

To identify parts of retroviral integrase that interact with cellular DNA, we tested patient-derived human immunodeficiency virus type 1 (HIV-1) integrases for alterations in the choice of nonviral target DNA sites. This strategy took advantage of the genetic diversity of HIV-1, which provided 75 integrase variants that differed by a small number of amino acids. Moreover, our hypothesis that biological pressures on the choice of nonviral sites would be minimal was validated when most of the proteins that catalyzed DNA joining exhibited altered target site preferences. Comparison of the sequences of proteins with the same preferences then guided mutagenesis of a laboratory integrase. The results showed that single amino acid substitutions at one particular residue yielded the same target site patterns as naturally occurring integrases that included these substitutions. Similar results were found with DNA joining reactions conducted with Mn(2+) or with Mg(2+) and were confirmed with a nonspecific alcoholysis assay. Other amino acid changes at this position also affected target site preferences. Thus, this novel approach has identified a residue in the central domain of HIV-1 integrase that interacts with or influences interactions with cellular DNA. The data also support a model in which integrase has distinct sites for viral and cellular DNA.     

Nonspecific alcoholysis, a novel endonuclease activity of human immunodeficiency virus type 1 and other retroviral integrases.

Retroviral integrase (IN) exhibits a previously unrecognized endonuclease activity which we have termed nonspecific alcoholysis. This action occurred at every position in nonviral DNA sequences except those near 5' ends and is clearly distinguished from, and was not predicted by, the site-specific alcoholysis activity previously described for IN at the processing site near viral DNA termini. The integrases of human immunodeficiency virus type 1, visna virus, and Rous sarcoma virus exhibited different target site preferences in this new assay. The isolated central domain of human immunodeficiency virus type 1 IN preferred the same sites as the full-length protein. Nonspecific alcoholysis may provide insights into the structure and function of IN and other endonucleases and suggests that stimulators of some activities possessed by retroviral enzymes should be sought as antiviral agents.     

An amino acid in the central catalytic domain of three retroviral integrases that affects target site selection in nonviral DNA

Integrase can insert retroviral DNA into almost any site in cellular DNA; however, target site preferences are noted in vitro and in vivo. We recently demonstrated that amino acid 119, in the alpha2 helix of the central domain of the human immunodeficiency virus type 1 integrase, affected the choice of nonviral target DNA sites. We have now extended these findings to the integrases of a nonprimate lentivirus and a more distantly related alpharetrovirus. We found that substitutions at the analogous positions in visna virus integrase and Rous sarcoma virus integrase changed the target site preferences in five assays that monitor insertion into nonviral DNA. Thus, the importance of this protein residue in the selection of nonviral target DNA sites is likely to be a general property of retroviral integrases. Moreover, this amino acid might be part of the cellular DNA binding site on integrase proteins.     

Juxtaposition of two viral DNA ends in a bimolecular disintegration reaction mediated by multimers of human immunodeficiency virus type 1 or murine leukemia virus integrase.

Integration of retroviral DNA involves a coordinated joining of the two ends of a viral DNA molecule into precisely spaced sites on target DNA. In this study, we designed an assay that requires two separate oligonucleotides to be brought together via interactions between integrase promoters to form a "crossbones" substrate that mimics the integration intermediate. The crossbones substrate contains two viral DNA ends, each joined to one strand of target DNA and separated by a defined length of target DNA. We showed that purified integrases of human immunodeficiency virus type 1 (HIV-1) and murine leukemia virus (MLV) could mediate a concerted strand cleavage-ligation between the two half-substrates at one or both viral DNA joining sites (trans disintegration). Another major product, termed fold-back, resulted from an intramolecular attack on the phosphodiester bond at the viral-target DNA junction by the 3'-OH group of the same DNA molecule (cis disintegration). The activity of integrase on the crossbones substrate depended on the presence of viral DNA sequences. For trans disintegration, the optimal length of target DNA between the viral DNA joining sites of the crossbones substrate corresponded to the spacing between the staggered joints formed on two opposite strands of target DNA during retroviral DNA integration in vivo. The activity of integrases on crossbones did not require complementary base pairing between the two half-substrates, indicating that the half-substrates were juxtaposed solely through protein-DNA interactions. The crossbones assay, therefore, measures the ability of integrase to juxtapose two viral DNA ends, an activity which heretofore has been difficult to detect by using purified integrase in conventional assays. Certain mutant integrases that were otherwise inactive with the crossbones substrate could complement one another, indicating that no single protomer in the integrase multimer requires a complete set of functional domains either for catalytic activity or for juxtaposition of the two viral DNA ends by the active multimer.     

A substitution in rous sarcoma virus integrase that separates its two biologically relevant enzymatic activities

Retroviral integrase prepares viral DNA for integration by removing 2 nucleotides from each end of unintegrated DNA in a reaction referred to as processing. However, it has been known since the processing assay was first described that avian integrases frequently nick 3 nucleotides, as well as 2 nucleotides, from viral DNA ends when reaction mixtures contain Mn2+. We now report that specificity for the biologically relevant "-2" site is enhanced when the serine at amino acid 124 of Rous sarcoma virus (RSV) integrase is replaced by alanine, valine, glycine, lysine, or aspartate. The protein with a serine-to-aspartate substitution exhibited especially high fidelity for the correct site, as evidenced by a ratio of -2 nicks to -3 nicks that was more than 40-fold greater than that for the wild-type enzyme in reactions with Mn2+. Even with Mg2+, the substituted proteins exhibited greater specificity than the wild type, especially the S124D protein. Moreover, this protein was more efficient than the wild type at processing viral DNA ends. Unexpectedly, however, the S124D protein was significantly impaired at catalyzing the insertion of viral DNA ends in reactions with Mn2+ and joining was undetectable in reactions with Mg2+. Thus, the S124D protein has separated the processing and joining activities of integrase. Similar results were found for human immunodeficiency virus integrase with the analogous substitution. No proteins with comparable properties have been described. Moreover, RSV virions containing integrase with the S124D mutation were unable to replicate in cell cultures. Together, these data suggest that integrase has evolved to have submaximal processing activity so that it can also catalyze DNA joining.     

Role of the nonspecific DNA-binding region and alpha helices within the core domain of retroviral integrase in selecting target DNA sites for integration

Retroviral integrase plays an important role in choosing host chromosomal sites for integration of the cDNA copy of the viral genome. The domain responsible for target site selection has been previously mapped to the central core of the protein (amino acid residues 49-238). Chimeric integrases between human immunodeficiency virus type 1 (HIV-1) and feline immunodeficiency virus (FIV) were prepared to examine the involvement of a nonspecific DNA-binding region (residues 213-266) and certain alpha helices within the core domain in target site selection. Determination of the distribution and frequency of integration events of the chimeric integrases narrowed the target site-specifying motif to within residues 49-187 and showed that alpha 3 and alpha 4 helices (residues 123-166) were not involved in target site selection. Furthermore, the chimera with the alpha 2 helix (residues 118-121) of FIV identity displayed characteristic integration events from both HIV-1 and FIV integrases. The results indicate that the alpha 2 helix plays a role in target site preference as either part of a larger or multiple target site-specifying motif.     

Characterization of the functional domains of human foamy virus integrase using chimeric integrases

Retroviral integrases insert viral DNA into target DNA. In this process they recognize their own DNA specifically via functional domains. In order to analyze these functional domains, we constructed six chimeric integrases by swapping domains between HIV-1 and HFV integrases, and two point mutants of HFV integrase. Chimeric integrases with the central domain of HIV-1 integrase had strand transfer and disintegration activities, in agreement with the idea that the central domain determines viral DNA specificity and has catalytic activity. On the other hand, chimeric integrases with the central domain of HFV integrase did not have any enzymatic activity apart from FFH that had weak disintegration activity, suggesting that the central domain of HFV integrase was defective catalytically or structurally. However, these inactive chimeras were efficiently complemented by the point mutants (D164A and E200A) of HFV integrase, indicating that the central domain of HFV integrase possesses potential enzymatic activity but is not able to recognize viral or target DNA without the help of its homologous N-terminal and C-terminal domains.     

Irreversible inhibition of human immunodeficiency virus type 1 integrase by dicaffeoylquinic acids

Human immunodeficiency virus type 1 (HIV-1) and other retroviruses require integration of a double-stranded DNA copy of the RNA genome into the host cell chromosome for productive infection. The viral enzyme, integrase, catalyzes the integration of retroviral DNA and represents an attractive target for developing antiretroviral agents. We identified several derivatives of dicaffeoylquinic acids (DCQAs) that inhibit HIV-1 replication in tissue culture and catalytic activities of HIV-1 integrase in vitro. The specific step at which DCQAs inhibit the integration in vitro and the mechanism of inhibition were examined in the present study. Titration experiments with different concentrations of HIV-1 integrase or DNA substrate found that the effect of DCQAs was exerted on the enzyme and not the DNA. In addition to HIV-1, DCQAs also inhibited the in vitro activities of MLV integrase and truncated variants of feline immunodeficiency virus integrase, suggesting that these compounds interacted with the central core domain of integrase. The inhibition on retroviral integrases was relatively specific, and DCQAs had no effect on several other DNA-modifying enzymes and phosphoryltransferases. Kinetic analysis and dialysis experiments showed that the inhibition of integrase by DCQAs was irreversible. The inhibition did not require the presence of a divalent cation and was unaffected by preassembling integrase onto viral DNA. The results suggest that the irreversible inhibition by DCQAs on integrase is directed toward conserved amino acid residues in the central core domain during catalysis.     

Weak palindromic consensus sequences are a common feature found at the integration target sites of many retroviruses

Integration into the host genome is one of the hallmarks of the retroviral life cycle and is catalyzed by virus-encoded integrases. While integrase has strict sequence requirements for the viral DNA ends, target site sequences have been shown to be very diverse. We carefully examined a large number of integration target site sequences from several retroviruses, including human immunodeficiency virus type 1, simian immunodeficiency virus, murine leukemia virus, and avian sarcoma-leukosis virus, and found that a statistical palindromic consensus, centered on the virus-specific duplicated target site sequence, was a common feature at integration target sites for these retroviruses.     

Role of the His-Cys finger of Moloney murine leukemia virus integrase protein in integration and disintegration.

Retroviral integrases mediate site-specific endonuclease and transesterification reactions in the absence of exogenous energy. The basis for the sequence specificity in these integrase-viral DNA recognition processes is unknown. Structural analogs of the disintegration substrate were made to analyze the disintegration reaction mechanism for the Moloney murine leukemia virus (M-MuLV) integrase (IN). Modifications in the target DNA portion of the disintegration substrate decreased enzymatic activity, while substitution of the highly conserved CA in the viral long terminal repeat portion had no effect on activity. The role of the His-Cys finger region in catalysis was addressed by N-ethylmaleimide (NEM) modification of the cysteine residues of M-MuLV IN as well as by mutations. Both integration activities, 3' processing, and strand transfer, were completely inhibited by NEM modification of M-MuLV IN, while disintegration activity was only partially sensitive. However, structural analogs of the disintegration substrates that were modified in the target DNA and had the conserved CA removed were not active with NEM-treated M-MuLV IN. In addition, mutants made in the His-Cys region of M-MuLV IN were examined and found to also be completely blocked in integration but not disintegration activity. These data suggest that the domains of M-MuLV IN that are required for the forward integration reaction substrate differ from those required for the reverse disintegration reaction substrate.     

Critical contacts between HIV-1 integrase and viral DNA identified by structure-based analysis and photo-crosslinking.

Analysis of the crystal structure of HIV-1 integrase reveals a cluster of lysine residues near the active site. Using site-directed mutagenesis and photo-crosslinking we find that Lys156 and Lys159 are critical for the functional interaction of integrase with viral DNA. Mutation of Lys156 or Lys159 to glutamate led to a loss of both 3' processing and strand transfer activities in vitro while maintaining the ability to interact with nonspecific DNA and support disintegration. However, mutation of both residues to glutamate produced a synergistic effect eliminating nearly all nonspecific DNA interaction and disintegration activity. In addition, virus containing either of these changes was replication-defective at the step of integration. Photo-crosslinking, using 5-iododeoxyuracil-substituted oligonucleotides, suggests that Lys159 interacts at the N7 position of the conserved deoxyadenosine adjacent to the scissile phosphodiester bond of viral DNA. Sequence conservation throughout retroviral integrases and certain bacterial transposases (e.g. Tn10/IS10) supports the premise that within those families of polynucleotidyl transferases, these residues are strategic for DNA interaction.     

Activities of the feline immunodeficiency virus integrase protein produced in Escherichia coli.

Retroviral DNA integration requires the activity of at least one viral protein, the integrase (IN) protein. We cloned and expressed the integrase gene of feline immunodeficiency virus (FIV) in Escherichia coli as a fusion to the malE gene and purified the IN fusion protein by affinity chromatography. The protein is active in site-specific cleavage of the viral DNA ends, DNA strand transfer, and disintegration. FIV IN has a relaxed viral DNA substrate requirement: it cleaves and integrates FIV DNA termini, human immunodeficiency virus DNA ends, and Moloney murine leukemia virus DNA ends with high efficiencies. In the cleavage reaction, IN exposes a specific phosphodiester bond near the viral DNA end to nucleophilic attack. In vitro, either H2O, glycerol, or the 3' OH group of the viral DNA terminus can serve as nucleophile in this reaction. We found that FIV IN preferentially uses the 3' OH ends of the viral DNA as nucleophile, whereas HIV IN protein preferentially uses H2O and glycerol as nucleophiles.     

Targeted modification of mammalian genomes

The stable and site-specific modification of mammalian genomes has a variety of applications in biomedicine and biotechnology. Here we outline two alternative approaches that can be employed to achieve this goal: homologous recombination (HR) or site-specific recombination. Homologous recombination relies on sequence similarity (or rather identity) of a piece of DNA that is introduced into a host cell and the host genome. In most cell types, the frequency of homologous recombination is markedly lower than the frequency of random integration. Especially in somatic cells, homologous recombination is an extremely rare event. However, recent strategies involving the introduction of DNA double-strand breaks, triplex forming oligonucleotides or adeno-associated virus can increase the frequency of homologous recombination.

Site-specific recombination makes use of enzymes (recombinases, transposases, integrases), which catalyse DNA strand exchange between DNA molecules that have only limited sequence homology. The recognition sites of site-specific recombinases (e.g. Cre, Flp or ΦC31 integrase) are usually 30–50 bp. In contrast, retroviral integrases only require a specific dinucleotide sequence to insert the viral cDNA into the host genome. Depending on the individual enzyme, there are either innumerable or very few potential target sites for a particular integrase/recombinase in a mammalian genome. A number of strategies have been utilised successfully to alter the site-specificity of recombinases. Therefore, site-specific recombinases provide an attractive tool for the targeted modification of mammalian genomes.     

The role of manganese in promoting multimerization and assembly of human immunodeficiency virus type 1 integrase as a catalytically active complex on immobilized long terminal repeat substrates.

The integration of a DNA copy of the viral genome into the genome of the host cell is an essential step in the replication of all retroviruses. Integration requires two discrete biochemical reactions; specific processing of each viral long terminal repeat terminus or donor substrate, and a DNA strand transfer step wherein the processed donor substrate is joined to a nonspecific target DNA. Both reactions are catalyzed by a virally encoded enzyme, integrase. A microtiter assay for the strand transfer activity of human immunodeficiency virus type 1 integrase which uses an immobilized oligonucleotide as the donor substrate was previously published (D. J. Hazuda, J. C. Hastings, A. L. Wolfe, and E. A. Emini, Nucleic Acids Res. 22;1121-1122, 1994). We now describe a series of modifications to the method which facilitate study of both the nature and the dynamics of the interaction between integrase and the donor DNA. The enzyme which binds to the immobilized donor is shown to be sufficient to catalyze strand transfer with target DNA substrates added subsequent to assembly; in the absence of the target substrate, the complex was retained on the donor in an enzymatically competent state. Assembly required high concentrations of divalent cation, with optimal activity achieved at 25 mM MnCl2. In contrast, preassembled complexes catalyzed strand transfer equally efficiently in either 1 or 25 mM MnCl2, indicating mechanistically distinct functions for the divalent cation in assembly and catalysis, respectively. Prior incubation of the enzyme in 25 mM MnCl2 was shown to promote the multimerization of integrase in the absence of a DNA substrate and alleviate the requirement for high concentrations of divalent cation during assembly. The superphysiological requirement for MnCl2 may, therefore, reflect an insufficiency for functional self-assembly in vitro. Subunits were observed to exchange during the assembly reaction, suggesting that multimerization can occur either before or coincident with but not after donor binding. These studies both validate and illustrate the utility of this novel methodology and suggest that the approach may be generally useful in characterizing other details of this biochemical reaction.     

Influence of subterminal viral DNA nucleotides on differential susceptibility to cleavage by human immunodeficiency virus type 1 and visna virus integrases.

A comparison of the extents of site-specific cleavage of U5 and U3 viral DNA termini by the integrases of human immunodeficiency virus type 1 and visna virus guided the quantitative testing of oligonucleotide substrates containing specific base substitutions. The simultaneous exchange of positions 5 and 6 between U3 substrates switched the patterns of differential susceptibility to the two integrases. The activity of visna virus integrase was more dependent on the identity of position 5 adjacent to the invariant CA bases than on position 6, whereas human immunodeficiency virus type 1 integrase appeared to interact even more critically with position 6. Although the paired natural substrates of most lentiviral integrases match at positions 7 and 8, these bases were not important for susceptibility of U5 substrates. In fact, the final six U5 positions contained all of the sequence information necessary for susceptibility. These results suggest that constraints other than integration influence the terminal inverted repeats of retroviral DNA.     

In vitro activities of purified visna virus integrase.

Although integration generally is considered a critical step in the retrovirus life cycle, it has been reported that visna virus, which causes degenerative neurologic disease in sheep, can productively infect sheep choroid plexus cells without detectable integration. To ascertain whether the integrase (IN) of visna virus is an inherently defective enzyme and to create tools for further study of integration of the phylogenetically related human immunodeficiency virus type 1 (HIV-1), we purified visna virus IN by using a bacterial expression system and applied various in vitro oligonucleotide-based assays to studying this protein. We found that visna virus IN demonstrates the full repertoire of in vitro functions characteristic of retroviral integrases. In particular, visna virus IN exhibits site-specific endonuclease activity following the invariant CA found two nucleotides from the 3' ends of viral DNA (processing activity), joins processed oligonucleotides to various sites on other oligonucleotides (strand transfer or integration activity), and reverses the integration reaction by resolving a complex that mimics one end of viral DNA integrated into host DNA (disintegration activity). In addition, although it has been reported that purified HIV-1 IN cannot specifically nick visna virus DNA ends, purified visna virus IN does specifically process and integrate HIV-1 DNA ends.     

Retroviral integrase superfamily: the structural perspective

The retroviral integrase superfamily (RISF) comprises numerous important nucleic acid‐processing enzymes, including transposases, integrases and various nucleases. These enzymes are involved in a wide range of processes such as transposition, replication and repair of DNA, homologous recombination, and RNA‐mediated gene silencing. Two out of the four enzymes that are encoded by the human immunodeficiency virus—RNase H1 and integrase—are members of this superfamily. RISF enzymes act on various substrates, and yet show remarkable mechanistic and structural similarities. All share a common fold of the catalytic core and the active site, which is composed primarily of carboxylate residues. Here, I present RISF proteins from a structural perspective, describing the individual members and the common and divergent elements of their structures, as well as the mechanistic insights gained from the structures of RNase H1 enzyme complexes with RNA/DNA hybrids.