HIV-1 NUCLEAR IMPORT: IN SEARCH OF A LEADER

Michael I. Bukrinsky¹ and Omar K. Haffar²

1The Picower Institute for Medical Research, Manhasset, NY 11030, ²Cytokine Networks Inc., 101 Elliott Ave West, Suite 428, Seattle, WA 98119

Received 11/7/97 Accepted 11/14/97

5. NULEAR IMPORT OF HIV-1 AS A DRUG TARGET

The protein-protein interaction between the NLSs of the HIV-1 PIC and karyopherinalphapresents an attractive new target for drug development. Interrupting the process of viral replication at the step of nuclear importation of the HIV-1 PIC may be accomplished by developing inhibitors to either of the interacting proteins. In contrast to inhibitors that target viral NLS-proteins, such as MA, inhibitors that interact with karyopherin alpha may affect normal cell function and may be less specific.

5.1. Compounds which bind to the MA NLS

Arylene bis(methyl ketone) compounds modified with a pyrimidine side chain were the first small molecules shown to associate with MA, inhibit binding of the HIV-1 PIC to karyopherin alpha, and block viral replication (85,86). These small molecules are represented by the prototypic compound CNI-H0294. CNI-H0294 forms Schiff-base adducts via its carbonyl moieties with the lysine residues in the MA NLSs (85). This interaction inactivates the MA NLS and represents the primary mechanism of action of CNI-H0294. The specificity of the compound for HIV appears to be derived from the poly-lysine nature of the MA NLS which allows the formation of Schiff base adducts with two neighboring residues. This property of the compound is dictated by the spatial separation of the carbonyl moieties. Additionally, CNI-H0294 associates with the viral RT via the pyrimidine side chain (86). RT is a component of the PIC and may be located in proximity to MA in the HIV-1 PIC (14,15). The exact site on RT where CNI-H0294 binds remains to be identified, however, the presence of fully reverse transcribed nascent cDNA in treated infected cells suggests that it is outside the active site of RT. Consistent with this assumption, CNI-H0294 and its analogues are not antagonistic to AZT or 3TC (see below). Binding to RT appears to stabilize the otherwise reversible Schiff base adducts between CNI-H0294 and lysines in the MA NLS. The mechanism of interaction between CNI-H0294 and the PIC proteins was confirmed using analogues of CNI-H0294 with modifications in either the bis(methyl ketone) group or the pyrimidine ring (86). For instance, removing either the pyrimidine ring or one of the carbonyl moieties reduced the activity of the inhibitor by approximately 200-fold.

CNI-H0294 and its functional analogues specifically inhibited nuclear importation of the HIV-1 PIC as determined by reduction in the levels of 2-LTR circles, a nuclear form of the HIV-1 DNA (85), while binding and entry of the virus into target cells and reverse transcription of the viral RNA were not affected. These results supported the hypothesis that the compounds specifically target the nuclear import step of HIV-1 replication. The CNI compounds inhibited infection of primary macrophage cultures with clinical or lab-adapted isolates of HIV-1 (85). Similarly, these compounds inhibited acute infection of activated PBMC cultures, as well as virus replication in endogenously infected PBMCs collected from HIV-1 seropositive individuals (O.K.H. et al., submitted). Since these individuals commonly carry a swarm of virus quasispecies, this result suggests that the inhibitory effect of the CNI compounds is virus strain-independent. In addition, this latter finding provided an independent confirmation for the role of active nuclear import mechanism in HIV-1 replication in activated T lymphocytes (see above). CNI-H0294 and analogues were also evaluated in vitro in combination with AZT and 3TC. The nuclear importation inhibitors had at least an additive effect when combined with these RT inhibitors (M.I.B. and O.K.H., manuscript in preparation). Furthermore, an analogue of CNI-H0294 inhibited virus replication in activated PBMCs collected from an HIV-1 infected individual resistant to AZT and ddC therapies (O.K.H. and M.I.B., unpublished). Although this represents a single observation, it does suggest that these novel compounds may prove useful for treating HIV-infected cohorts where resistance to established therapies is on the rise.

A distinct NLS in IN has been recently identified, and proposed to play a role in nuclear importation of the HIV-1 PIC (13). Given the poly-lysine nature of this new NLS, and the anti-HIV-1 properties of CNI-H0294 and its analogues, it is plausible that these compounds may also bind to the IN NLS. It would be necessary to address this question to fully delineate the mechanism of action of this class of inhibitors.

5.2. Compounds which bind karyopherin alpha

Inhibition of nuclear importation can also be accomplished by compounds which associate with karyopherin alpha, the cellular NLS receptor. Gulizia et al. (87) and Gallay et al. (88) utilized the prototypic NLS peptide of the SV40 large T antigen as an inhibitor of karyopherin alpha-PIC binding, and demonstrated the critical importance of this protein-protein interaction in nuclear importation of the HIV genome and HIV-1 infection of target cells. The inhibitory effects of the SV40 NLS peptide were attributed to its ability to compete with the HIV-1 PIC for binding to karyopherin alpha. This was confirmed by Gallay et al. (88) who showed by biochemical analysis that binding of Rch1, one form of the human karyopherin alpha, to the MA NLS was competitively inhibited by the SV40 NLS peptide. One limitation however, for the application of NLS peptides as therapeutic compounds for HIV-1 infection is the high levels of peptides necessary to achieve the inhibitory effect. For instance, approximately 100 µM of SV40 NLS was required to inhibit HIV-1 infection, and no inhibition was observed at concentrations below 20 µM (87). Similarly, a very high concentration of the NLS peptide (500 µM) was required to inhibit binding of recombinant karyopherin alpha and PIC in cell free assays (88). These observations may be a result of two distinct events. First, the NLS peptide of SV40 is rich in lysine residues and is therefore highly charged, which limits its uptake into cells across the plasma membrane. Second, it is estimated that each karyopherin alpha protein contains eight NLS binding sites (89). Although, it is not yet clear whether all of these binding sites can be occupied at the same time, this observation suggests that efficient inhibition of all NLS-binding sites on karyopherin alpha may be difficult to achieve. In addition, given the central role that this class of proteins plays in cell activation and transport of transcriptional factors to the nucleus, it remains to be determined whether a significant therapeutic index can be achieved using this approach.

5.3. Perspectives for development of nuclear import inhibitors as anti-HIV drugs

Unlike the classical approach to anti-viral therapeutic drug development which targets the viral enzymes, such as RT, protease, or integrase, targeting nuclear importation represents a novel paradigm for therapeutic intervention in HIV-1 infection. The availability of new inhibitory compounds validates the potential of nuclear importation as a target for drug development. Importantly, these novel compounds inhibit HIV-1 infection in both primary macrophages and activated primary T lymphocytes, the principle target cells of the virus in vivo. Given the demonstrated efficacy of a combination therapy approach to treatment of HIV infection, it is expected that these and other new compounds will be administered together with the available approved therapies. The ability of the nuclear importation inhibitors to synergize with nucleoside analogues in vitro, suggests a promising potential for success of such therapy in vivo.