Virus and Host Factors in HIV Pathogenesis
Jay A. Levy, MD
Professor of Medicine; Director, Laboratory for Tumor and AIDS Virus Research, University of California, San Francisco
Basic science highlights from the 2nd IAS Conference on HIV Pathogenesis and Treatment dealt with further insights into HIV:cell interactions -- processes both at the cell membrane level and those involving intracellular proteins that lead to virus integration and viral production. Many of these new observations offer hope for novel antiviral therapies. For example, several presentations covered the newly discovered innate intracellular control of HIV via the cytosine deaminase protein APOBEC3G, the effect of which can be countered through the action of the Vif protein (see below). Attention given to recent information dealing with several other viral accessory genes promises, after several years of research, to provide some explanation for why these accessory gene are present in lentiviruses and absent in the simple retroviruses, such as the oncoviruses.
An excellent summary of HIV interactions at the cell surface was presented by Robert Doms of the University of Pennsylvania, Philadelphia. He described the initial interaction of the viral gp120 with the CD4 binding site on the cell surface which leads to conformational changes in the envelope, allowing a second interaction of the virus envelope with chemokine coreceptors. This latter event produces a further change in gp41 before it fuses and penetrates the cell surface.[2,3] That event involves helical regions (HRs) 1 and 2, and a process by which HR-2 folds back on itself to bind to HR-1. This action permits virus fusion and entry of the HIV capsid into the cell. The new drug, enfuvirtide (T-20), which is a mimic of HR-2, can block the binding to HR-1 and prevent HIV infection.
Features that can influence resistance to enfuvirtide include a rapid triggering of the HR-2, an increase in the number of viral envelope "knobs" that interact with the cell, the efficiency of conformational changes to the envelope, and a high affinity of the viral envelope for the chemokine coreceptor. Of note, Dr. Doms remarked that primary viruses can vary by 2-4 logs in their sensitivity to entry inhibitors while exhibiting more modest variability in their sensitivity to other antiviral drugs.
In this regard, expression of more copies of the CCR5 coreceptor on the cell (varying from 10,000 to 90,000 copies per cell) can also affect the net activity of enfuvirtide. Thus, infected subjects who express low levels of CCR5 may respond better to entry inhibitors than would individuals who express higher levels of this molecule. For this reason the use of a coreceptor antagonist in combination with enfuvirtide could increase the effectiveness of this antiviral approach by providing synergistic inhibition of the virus. In general, a slow rate of viral fusion, as occurs when CCR5 levels are low, would permit the best antiviral activity of enfuvirtide.
It is believed that understanding further the virus:cell membrane interactions will lead to the development of other inhibitors that can act on conformational structures of the viral envelope formed during the attachment and entry processes. In this author's opinion, a glycolipid is most likely the location for the HIV envelope fusion with the cell, and approaches to target this part of the cell membrane may bring universal antiviral activity.
Intracellular Events: Vif:APOBEC3G Interaction
Certainly the most exciting new observation on host antiviral activities during HIV infection are the studies on Vif reported at the meeting and recently published by 4 separate groups.[4-7] These findings deal with the cellular enzyme, APOBEC3G, that, in the absence of the viral Vif protein, can block HIV infection.
Ned Landau of the Salk Institute for Biological Studies, San Diego, California, was the first speaker to describe this activity, which was also addressed by Brian Cullen, of Duke University Medical Center, Durham, North Carolina, and Didier Trono of the University of Geneva, Switzerland, in a Forum session entitled "Viral Molecular Pathogenesis." The novel observations on APOBEC3G have their origin in Michael Malim's initial finding that the viral accessory gene, vif, was important for the virus to avoid innate or natural intracellular antiviral activity. Vif permitted the infection of a wide variety of human cells that were not susceptible to infection by HIV variants lacking vif. HIV contains 5 other accessory genes, besides vif, they are tat, nef, vpr, vpu, nef, and rev. While tat and rev have transcriptional and splicing activities, the roles of the other accessory genes have not been consistently defined. Several appear to have pleiotropic activities (see below). The role of vif now appears further elucidated.
The viral protein, Vif, is a 192 amino acid, 23 Kd protein that is a late gene product in all lentiviruses except EIAV and is not found in the more simple viruses. Without the vif gene (eg, the delta vif virus), many cells are nonpermissive; they have a dominant negative factor that blocks virus replication. The "permissive cells," such as CEM-SS, 293T cells, Jurkat, and Sup T1 cells, do not express this protein and thus can be infected by viruses lacking Vif (delta vif). These delta vif viruses, however, are not infectious for primary CD4 lymphocytes and macrophages. In the latter cells, wild-type viruses replicate because the Vif protein inhibits the activity of the intracellular protein, APOBEC3G (apolipoprotein B mRNA-editing enzyme, catalytic polypeptide-like 3G complex), a member of a family of cytidine deaminases.
APOBEC3G (or CEM15), is a 384 amino acid protein that converts cytosine to uracil, generally during the editing of mRNA. The recent work of several groups[4-7,9] now indicates that this enzymatic effect is not targeting the viral RNA but rather the single-stranded DNA. In the absence of the viral Vif protein, mutant viruses contain several G-to-A mutations resulting from activity of this cytidine deaminase. Essentially, the reverse transcription forming the negative-strand DNA causes the insertion of a uracil instead of cytosine, leading to the inclusion of an adenosine instead of guanosine in the positive-strand DNA. That process leads to degradation of some of viral DNA, and any integrated virus cannot produce very high levels of infectious virus. With passage, more G-to-A mutations in the viral DNA occur and infectivity is eliminated. Some activation steps can increase the APOBEC3G concentration and it is surmised that polymorphisms in the gene can influence APOBEC3G levels in cells. Thus, variations in susceptibility of human cells to HIV infection can result.
Another important new finding is that APOBEC3G can be found in the virion. In wild-type viruses, only a small quantity of this enzyme can be found, whereas the delta vif virus contains large amounts. It now appears (although it is not yet vigorously proven) that Vif binds APOBEC3G in the cytoplasm and prevents its encapsulation into the virion. Thus, when the wild-type virus infects cells, reverse transcription can readily take place. The delta vif virus with APOBEC3G in the virion cannot complete normal reverse transcription, and the mutations lead to DNA degradation or production of noninfectious particles.
This Vif:APOBEC3G interaction could be a drug target. However, because APOBEC3G is an important mRNA editing enzyme in the cell, it will be necessary to find a method of preventing the attachment of Vif to APOBEC3G without decreasing the activity of this cytidine deaminase within the cell.
In this context, it is noteworthy that other accessory proteins interact with host factors. For instance, Tat interacts with cyclin T1 and Rev with CERM-1. Moreover there are other intracellular blocking activities, one of which, LV-1 (or PML), appears to affect early steps in reverse transcription similar to the Fv-1 gene in the murine system. This latter process appears to be associated with expression of an endogenous retrovirus that interacts with the viral capsid. In the discussion period of the Forum session, it was noted that Vpr can bring the uracil DNA glycosylase hUNG-2 enzyme into a virion that can repair the uracil base change. This interaction may be a viral means of rescuing HIV from the effect of APOBEC3G. It merits further study. A suggestion given by Dr. Trono was that APOBEC3G probably is present in cells as an innate response against transposons.[10,11]
Other Intracellular Activities
Discussions on other cellular proteins that interact with viral genes centered on the ERK-2 protein that phosphorylates the viral Gag-derived protein p6. For example, B. Hemonnot, Institute Biologie, Montpellier, France, demonstrated that if this MAP kinase is not active in cells, a good separation of the viral particle from the cell membrane does not take place; HIV therefore becomes localized to the cell surface. Thus, targeting ERK-2 present in virions can block virus replication. Perhaps noteworthy, in this author's opinion, is the concept that via an accumulation of viral antigens on the cell surface, a block of ERK-2 activity might provide an effective way of inducing immunologic reactions against HIV.
David Rekosh of the University of Virginia, Charlottesville, Virginia, presented new observations on Nef, which has many described functions. He noted that about 10% of the Nef protein in cells targets the cell membrane via myristoylation. He reported that enhancing targeting of the cell membrane by Nef, using a Nef fusion protein (LAT-Nef) containing the N-terminal region of LAT that localizes to lipid rafts, produced differences in Nef function in terms of MHC downregulation, with a slightly lesser effect being observed on virus infectivity and CD4 molecule downmodulation. He raised the question of whether, in wild-type virus infection, the non-membrane-associated Nef contributed to these other roles of Nef in the cell.
A report presented by C. Berlioz-Torrent, Institut Cochin, Paris, France, demonstrated the binding of the viral envelope within the cell to a protein TIP 47 that is required for the transport of mannose-6-phosphate receptors to the trans-Golgi network (TGN). The envelope gp41 interacts with TIP 47 and other components (AP-1, AP-2), which allow retrograde transport of the envelope to the cell surface. Primarily, the tyrosine tail of gp41 is involved and is crucial for infectivity.
G. Blot, Institut Cochin, discussed 2-hybrid screen approaches that identified luman, an ATF/CREB transcription factor that also binds to the transmembrane portion of gp41. Its overproduction can decrease production of HIV. Inhibitory RNA (RNAi) approaches directed against luman mRNA demonstrated that reduced expression of luman led to increased production of HIV in HeLa P4-2 cells. This interaction therefore appears important in trafficking of the virus envelope within the cell.
In other studies, S. Emiliani, Institut Cochin, used yeast 2-hybrid screening to show that up to 9 cellular proteins bind to HIV integrase. Some involve nuclear transfer and others involve chromatin binding proteins. These investigators concentrated on LEDGF/p75 (lens epithelium-derived growth factor), a cellular factor that protects against stress-induced apoptosis. This protein has 2 isoforms, p52 and p75, of which p75 appears to be the most important for HIV infection. By using RNAi approaches, they found a block in integration (ie, proviral DNA formation) when p75 was inhibited. Further investigations will show where p75 binds to the integrase and whether it retains the integrase and/or helps its transport to the nucleus.
Furthermore, E. Le Rouzic, Institut Cochin, noted that Vpr helps in the nuclear translocation of the viral preintegration complex containing viral DNA. In this regard, investigators have recently studied the interaction of Vpr with nucleoporin (hCG1), a component of the nuclear pore. By selective mutation, they have shown the importance of this binding of Vpr to hCG1 on the transport of the preintegration complex to the nucleus. This process could represent another target for anti-HIV therapy.
Certainly, the years of research on accessory proteins have come to the forefront with the identification of the APOBEC:Vif interaction and the identification of other cellular proteins that could be targeted in therapy. Essentially, natural, innate factors can be discovered that have an antiviral activity similar to the secreted cytokines of interferon, chemokines, and the CD8+ cell antiviral factor. Cellular factors can be effective targets for therapy because their ability to mutate may be much more limited than that of viral proteins. The new observations summarized and reported at the Second IAS Conference underlined future work in which basic science can reveal novel avenues for effective anti-HIV therapies.
1. Doms R. HIV entry: insights into viral tropism, pathogenesis, and antiviral therapy. Posters and abstracts of the 2nd IAS Conference on HIV Pathogenesis and Treatment; July 13-16, 2003; Paris, France. Plenary Session.
2. Pierson TC, Doms RW. HIV-1 entry inhibitors: new targets, novel therapies. Immnol Lett. 2003;85:113-118.
3. Moore JP, Doms RW. The entry of entry inhibitors: a fusion of science and medicine. Proc Natl Acad Sci USA. In press.
4. Lecossier D, Bouchonnet F, Clavel F, Hance AJ. Hypermutation of HIV-1 DNA in the absence of the Vif protein. Science. 2003;300:1112.
5. Harris RS, Bishop KN, Sheehy AM, et al. DNA deamination mediates innate immunity to retroviral infection. Cell. 2003;113:803-809.
6. Mangeat B, Turelli P, Caron G, Friedli M, Perrin L, Trono D. Broad antiretroviral defence by human APOBEC3G through lethal editing of nascent reverse transcripts. Nature. 2003;424:99-103.
7. Zhang H, Yang B, Pomerantz RL, Zhang C, Arunachalam SC, Gao L. The cytidine deaminase CEM15 induces hypermutation in newly synthesized HIV-1 DNA. Nature. 2003;403:94-98.
8. Landau N. Host/virus mechanisms in the molecular pathogenesis of HIV. Posters and abstracts of the 2nd IAS Conference on HIV Pathogenesis and Treatment; July 13-16, 2003; Paris, France. Plenary Session.
9. Cullen BR, Bogerd HP, Doehle B. Mechanism of action of the HIV-1 VIF accessory. The 2nd IAS Conference on HIV Pathogenesis and Treatment. Paris, France: International Medical Press; 2003;8(suppl 1):S213.
10. Trono D. Innate intracellular antiretroviral defenses. Posters and abstracts of the 2nd IAS Conference on HIV Pathogenesis and Treatment; July 13-16, 2003; Paris, France. Abstract 121.
11. Sheehy AM, Gaddis NC, Chol JD, Malim MH. Isolation of a human gene that inhibits HIV-1 infection and is suppressed by the viral Vif protein. Nature. 2002; 418: 646-650.
12. Hemonnot B, Cartier C, Gay B, Bardy M, Devaux C, Briant L. ERK-2 dependent phosphorylation of Pg GAG protein is required for HIV-1 replicative cycle. The 2nd IAS Conference on HIV Pathogenesis and Treatment. Paris, France: International Medical Press; 2003;8(suppl 1):S232-S233.
13. Rekosh D, Alexander M, Ravichandran K, Hammarskjold ML, Myles H. HIV-1 Nef associates with lipid rafts to down-modulate cell surface CD4 and class 1 MHC expression and to increase viral infectivity. The 2nd IAS Conference on HIV Pathogenesis and Treatment. Paris, France: International Medical Press; 2003;8(suppl 1):S233.
14. Blot G, Janvier K, Le Panse S, Benarous R, Berlioz-Torrent C. Identification of TIP47 as the first cellular co-factor involved in the retrograde transport of the HIV-1 envelope to the trans-golgi network: effect on the env incorporation and on HIV-1 infectivity. The 2nd IAS Conference on HIV Pathogenesis and Treatment. Paris, France: International Medical Press; 2003;8(suppl 1).
15. Blot G, Lopez S, Benarous R, Berlioz-Torrent S. Luman, a cyclic amp response element (CRE)-binding protein/activating transcription factor, a new partner of the cytoplasmic domain of HIV-1 TMgp41. The 2nd IAS Conference on HIV Pathogenesis and Treatment. Paris, France: International Medical Press; 2003;8(suppl 1).
16. Emiliani S, Rain JC, Maroun M, et al. A cellular co-factor of HIV-1 integrase essential for HIV-1 replication: a novel target for anti-HIV-1 drug discovery. The 2nd IAS Conference on HIV Pathogenesis and Treatment. Paris, France: International Medical Press; 2003;8(suppl 1):S214.
17. Le Rouzic E, Villet R, Mousnier A, Dargemont C, Benichou S. Functional analysis of the interaction between Vpr HIB-1 and the nucleoporin hCG1. The 2nd IAS Conference on HIV Pathogenesis and Treatment. Paris, France: International Medical Press; 2003;8(suppl 1):S270.
18. Le Rouzic E, Mousnier A, Rustum C, et al. Docking of HIV-1 Vpr to the nuclear envelope is mediated by the interaction with the nucleoporin hCG1. J Biol Chem. 2002;277:45091-45098.