Targeting the HIV integration process in antiviral therapy.

AuthorMcNeely, Melissa
PositionLEADING ARTICLE - Clinical report


In the global fight against HIV/AIDS, it is important to utilise every weapon available. Current therapy includes an array of drugs inhibiting many different viral processes; inhibitors of reverse transcription, proteolytic maturation, fusion and entry have all been very effective in reducing HIV replication enough to prevent disease progression, especially when used in conjunction with highly active antiretroviral therapy (HAART). However, owing to the high mutation rate of the HIV reverse transcriptase, multiple drug-resistant strains continue to emerge, necessitating continuing searches for new drug targets.

The recent developments in drug discovery targeting viral integrase (IN) have been very promising. HIV IN is a crucial protein for HIV infection and is responsible for the insertion of the provirus into the human genome, permanently linking the viral genome to the host genome, a key part of viral replication. Gaps in scientific knowledge and difficulties in getting around certain biological characteristics have hindered development of IN inhibitors until recently. The successful introduction of the IN inhibitors, raltegravir and elvitegravir, in the clinic has demonstrated the effectiveness of targeting the integration process in antiviral therapy. However, the success of both compounds is somewhat confounded by the emergence of resistant strains.


Protein-protein interactions (PPI) represent a relatively untapped and potentially rich source of therapeutic targets [1-3]. Recently, there have been significant advances in identification and validation of cellular co-factors involved in HIV infection, and cellular binding partners of HIV IN involved in both integration and nuclear import have been described [4-6]. It has become apparent that inhibition of IN function by interference with its cellular interaction partners would indeed represent a significant addition to the therapeutic arsenal.



The HIV-1 replication cycle is initiated by the attachment of the viral envelope glycoprotein gp120 to the cellular CD4 receptor (Figure 1). The interaction of gp120 with CD4 triggers a conformational change in the gp120 molecule, allowing binding to the CXCR4 or CCR5 chemokine coreceptor. Another conformational change in the gp41 viral protein brings about fusion of the viral particle with the cellular membrane, which is then followed by viral uncoating. In the next step, reverse transcription of the positive RNA strand is initiated, during which the single-stranded RNA is copied into double-stranded cDNA. This will then be imported into the nucleus as a pre-integration complex (PIC). As well as viral cDNA, this complex contains viral matrix (MA), viral protein R (Vpr), reverse transcriptase (RT), integrase and several cellular proteins.

In the nucleus, the proviral DNA associates with the host chromosomes, and integration of the viral cDNA occurs. This step is catalysed by the viral enzyme IN, and from this point on the provirus will forever form part of the cellular genome. During the lifespan of the infected cell, the virus may either remain present as a latent reservoir or transcription of viral genes may initiate the formation of new virions. Viral proteins and genomic RNA are assembled at the cellular membrane carrying envelope proteins. Immature viral particles bud from the membrane and mature into infectious virions through proteolytic processing of the viral polyproteins by the viral protease [7].

The standard therapy for HIV-1-infected patients, HAART, is based on combinations of potent drugs targeting different steps of this replication cycle, given to reduce viral loads to a minimum and to minimise the risk of development of resistance. There are three main classes of drugs: protease inhibitors (PIs), nucleoside or nucleotide reverse transcriptase inhibitors (NRTIs) and non-nucleoside reverse transcriptase inhibitors (NNRTIs). US FDA-approved antiviral drugs typically target viral proteins for specificity and low cytotoxicity. However, because of the high error rate of reverse transcriptase and HIV's short replication cycle, the HIV genome mutates very quickly resulting in rapid emergence of drug-resistant HIV strains.

Although HAART for HIV infection is based on combinations of truly effective antiretrovirals, this continuous emergence of drug-resistant virus strains spurs scientists to look for second-generation and novel inhibitors to outpace the development of resistance. Also, drug-associated side effects compromise the adherence of patients to the treatment, which in turn fosters the development of resistance. Therefore, efforts to discover new drugs and drug classes are highly important in the field of HIV research. To avoid cross-resistance, novel drugs should preferentially target as-yet unexploited steps of the viral replication cycle.

Integration is one step that has, thus far, remained relatively untouched. The HIV genome does not replicate episomally, but depends on integration into the host genome for replication. Without this process, unintegrated viral DNA quickly degrades [8]. The central role of IN in HIV infection coupled with the absence of a human homologue has rendered IN a very attractive subject for inhibitor development.


Viral integration is essential for HIV replication. Transcription of the viral genome and production of the viral proteins requires that the viral cDNA be fully inserted into the host genome. IN does this by catalysing two different reactions (see Figure 2): 3' processing and strand transfer. After completion of DNA synthesis, the blunt-ended double-stranded DNA must be prepared for integration. During 3' processing, endonucleolytic cleavage of the 3' ends of the viral DNA results in CA-3'-hydroxyl DNA ends. These recessed 3'-OH groups become the reactive intermediates for strand transfer, and by default are also the viral DNA attachment sites to the host genome [9].


After the 3' processing in the cytoplasm of the cell, IN associated with the viral cDNA within the PIC is transported into the nucleus where the strand transfer reaction occurs. The PIC associates with the host chromosome, and IN catalyses a transesterification-ligation reaction in which the reactive 3'-OH groups of the processed viral cDNA are covalently bonded to the 5' phosphate of the host DNA, resulting in a 5-bp single-stranded DNA gap at both viral-host DNA junctions. Once the last two nucleotides from the 5' ends of the proviral DNA are cleaved off and the gap is filled between the host and proviral DNA by host DNA repair enzymes, integration is complete.



As integration is such a critical step, it stands to reason that inhibitors of the process might meet with success in the clinic. Nevertheless, IN inhibitors are only recent, and rather tardy, additions to the HIV antiviral drug array. Certain biological characteristics of HIV-1 integration have hindered the search for selective IN inhibitors. Viral IN has no cellular counterpart, so unlike reverse transcriptase and protease inhibitors, for which analogue inhibitors targeting cellular DNA polymerases and proteases were already known, IN inhibitors had to be developed from scratch. In addition, the reactions catalysed by IN are single events, and not polymerisation reactions like those catalysed by reverse transcriptase (RT). The 3' processing and strand transfer reactions are carried out once at each long-terminal repeat (LTR) end during retroviral replication. Integration is therefore an all-or-nothing event, and so opportunities to inhibit it are relatively limited.

Another drawback in development of IN inhibitors has been the lack of a complete three-dimensional structure of full-length IN. This impedes rational structure-based design of inhibitors, so libraries of compounds must be screened for effectiveness in in vitro assays. In these assays, not all steps of the integration process can be accounted for, and discovered drugs might target biochemical interactions important for other enzymes. The assays share the drawback of only looking at one biochemical process at a time, and so simply cannot provide a comprehensive look at entire cellular processes. Of course, this makes it difficult initially to assess activity versus toxicity, once compounds are processed from pure in vitro screening assays into HIV replication assays in cell culture. New assays should be developed that allow direct identification of authentic IN inhibitors in cell culture, and as all steps of the integration process represent potential therapeutic targets, to allow identification of additional subtargets during the integration process.

During the decade of research leading to the introduction of IN inhibitors in the clinic (1990-2000), different mechanistic classes of compounds were identified as effective inhibitors of viral IN [10-12]: nucleosides and analogues; hydroxylated aromatic compounds; agents interacting with DNA; and peptides and antibodies. These inhibitors of recombinant IN can also be classified according to their antiviral and cytotoxic activity in cell culture, and to whether they target other viral processes besides integration. Unfortunately, most compounds that inhibit recombinant IN do not exhibit significant antiviral activity, even if they are not toxic in cell culture. This could be due to a number of causes, including inability of the compound to enter the cell. Another potential complication is the different conformational state of IN in each assay. Whereas the in vitro screening tools use isolated IN, in an actual infection of living cells this enzyme is embedded in the multiprotein PIC. Within the PIC, steric hindrance may limit accessibility of drugs to catalytic sites.



To continue reading

Request your trial