HIV-1

A great obstacle to anti-HIV therapy is given by the progressive reduction in drug effectiveness after few month of treatment. In fact, the virus mutates rapidly its genome and, over typical time-scales of few months, becomes capable of eluding the action of drugs targeting the current viral protein targets. Conventional therapies, based on combinations of drugs, target different steps of the viral life cycle. These include: the viral entry (coreceptor antagonists and fusion inhibitors), the reverse transcription (nucleoside and non-nucleoside inhibitors of the viral reverse transcriptase),the integration (integrase inhibitors) and the viral maturation (protease inhibitors) (Fig. 1). All of these may unfortunately become ineffective in the long run.

HIV-1 attack must rely on cellular partners (only 15 proteins are encoded by the virus!). Since human proteins are highly conserved compared to hypervariable viral ones they provide a much stiffer template for rational drug design without the well-known drug resistance problems related with retroviral infection. Moreover, an indirect antiviral agent, might be combined with the existing direct antiviral drugs resulting in additive or even synergistic treatment options with the potential of eliminating the HIV reservoir. Hence, the structural determination of the macromolecular aggregates formed in infected cells is highly desirable.
Here we are using and developing a variety of computational approaches for the structural prediction of macromolecular aggregates complexes involving HIV-1 biomolecules.

Transcription is an essential step in the replication cycle of HIV that involves a complex interplay of viral and multiple cellular proteins. Control of HIV transcription by host factors is at the basis of post-integrative latency. The discovery of transcription inhibitors is an innovative line of investigation in medicinal chemistry. This is an excellent target that would lead to control of HIV-replication not only in acutely infected cells but also in chronically infected ones possibly limiting the maintenance of a reservoir of producing cells. To date no selective inhibitor of HIV gene expression and transcription has led to a successful clinical development. 

Gene expression from proviral DNA is regulated by two viral proteins, the Transactivator of Transcription (Tat) and Rev, as well as several known and unknown host-cellular factors. In particular, Tat is crucial for viral replication and it is considered one of the best targets for inhibition of HIV-1 transcription. Tat protein binds to HIV-1 TAR RNA and allows the replication of the entire viral genome. The process of transactivation is complex and require the participation of other molecules (e.g.: CyclinT1, p/CAF). The lack of detailed atomistic structural information of Tat-host protein complexes has required a molecular modeling strategy, validated against molecular biology experiments (Collaboration with Prof. M. Giacca at ICGEB-Trieste).

We have identified conserved structural and energetic features among different protein isolates and predicted the structural determinants of Tat in complex with one of the host cell cognate proteins, p/CAF. These findings may help the design of ligands interfering with Tat function. Work is in progress to further refine our models (Collaboration with Prof. M. Giacca and A. Marcello at ICGEB-Trieste).

Fig. 2. Contact surface between p/CAF and Tat proteins

Also TAR has been long considered as an attractive new chemotherapeutic target. Ligands binding at the interface between TAR and its cognate virally-enclosed protein Tat would block the production of full-length transcripts and reduce HIV-1 replication rates. Importantly, the therapeutic benefit would be realized without interference with host cell function, as there is no human counterpart to Tat or TAR. This contrasts with current anti-HIV drugs, which may interfere with the infected cell’s function as they target enzymes.

We are currently investigating structural, NMR and energetic properties of TAR-ligands complexes (Collaboration with Prof. M. Parrinello (ETH Zürich, Lugano, Switzerland) and Prof. G. Varani). This project has been also funded by European DECI initiative retaining it for operation in the DEISA supercomputing infrastructure and benefiting of the resource allocation of 400.000 cpu-hours.  

The protease, which cleaves HIV polypeptide chain at specific location is a major target for anti HIV therapy. We have obtained insights on drug resistance mutations by investigating the binding mechanism of a peptide substrate (Thr-Ile-Met-Met-Gln-Arg, cleavage site p2-NC of the viral polyprotein) to wild-type HIV-1 protease. The calculated free energy of binding and the kinetic constants for association and dissociation are consistently with experiments (collaboration with A. Laio, SISSA).

Early QM and QM/MM calculations, validated against experimental data offered an explanation of the observed changes in catalytic rate of compensatory mutations in drug resistance variants. Recent calculations confirmed our proposal (Carnevale et al., Journal of Molecular Structure: THEOCHEM 89897-105 (2009), special issue on QM/MM methods).

Uncoating and recruitment of the virion. The uncoating and recruitment of the human HIV-1 capsid (CA) protein in the virions is triggered by proline trans/cis isomerization catalyzed by the human protein cyclophilin A isomerase (CypA). 

We are using metadynamics simulations to investigate the isomerization of CA’s model substrate HAGPIA in water and in its target protein CypA. Our results allow us to propose a novel mechanistic hypothesis, which is finally consistent with all of the available molecular biology data. 

Interaction between HIV-1 integrase and its cellular partners . HIV-1 integrase (IN) is one of the major targets for anti AIDS therapies. The enzyme inserts the viral DNA into the host cell genome in an essential step of HIV-1 life cycle. This multi-step process starts in the cytosol and culminates in the nucleus of the infected cell. Along this pathway, besides its DNA substrate, IN interacts with a wide range of different host-cell proteins. These cellular cofactors are exploited for various functions, including nuclear import DNA target-site selection and virion assembly. Most antiretroviral agents may affect IN interactions with the viral and/or host DNA. In the first case, inhibitors may hamper DNA-breaking and joining reactions required for the integration of the viral DNA into the host chromosome. These processes involve two different catalytic reactions, i.e. 3’-processing and strand transfer, occurring at the same active site. Indeed, a drug which works in this way has just been approved by FDA for use in the clinics. 

Inhibition of specific IN-host cofactor interactions may be an alternative promising approach, as viral strains resistant to IN-cofactor-disrupting drugs might emerge at a considerably slower rate than those targeting viral proteins subjected to a high rate of variability (such as IN, Reverse Transcriptase and PR). A promising target is the prolyl-isomerase Pin1, whose binding to IN has been recently shown to affect IN function. Unfortunately, at present, the rational design of ligands affecting the affinity for this or any other specific IN/target complex is hampered by the lack of structural information. We are using computational tools to provide a structural model of the Pin1-IN complex. The resulting model has been validated against in vivo experiments performed by the group of Prof. M. Giacca at ICGEB-Trieste.