Zinc chelation inhibits HIV Vif activity and liberates antiviral function of the cytidine deaminase APOBEC3G

Transcription

1 The FASEB Journal Research Communication Zinc chelation inhibits HIV Vif activity and liberates antiviral function of the cytidine deaminase APOBEC3G Zuoxiang Xiao,*, Elana Ehrlich,* Kun Luo,*, Yong Xiong,* and Xiao-Fang Yu*,,1 *Department of Molecular Microbiology and Immunology, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA; and Second Affiliated Hospital, School of Medicine, Zhejiang University, Zhejiang, China ABSTRACT APOBEC3 proteins are cellular antiviral proteins that are targeted for proteasomal degradation by primate lentiviral Vif proteins. Vif acts as a substrate receptor for the Cullin5 (Cul5) E3 ubiquitin ligase, specifically interacting with Cul5 through a novel H- x5 - C- x C- x3 5 -H zinc binding motif. Using the membrane-permeable zinc chelator, N,N,N,N -tetrakis- (2-pyridylmethyl) ethylenediamine (TPEN), we demonstrated a requirement for zinc for Vif function in vivo. Treatment with TPEN at an IC50 of 1.79 M inhibits Cul5 recruitment and APOBEC3G (A3G) degradation. Zinc chelation prevented Vif function in infectivity assays, allowing the virus to become sensitive to the antiviral activity of A3G. Zinc chelation had no effect on cellular Cul5-SOCS3 E3 ligase assembly, suggesting that zinc-dependent E3 ligase assembly may be unique to HIV-1 Vif, representing a new target for novel drug design. Xiao, X., Ehrlich, E., Luo, K., Xiong, Y., Yu, X-F. Zinc chelation inhibits HIV Vif activity and liberates antiviral function of the cytidine deaminase APOBEC3G. FASEB J. 21, (2007) Key Words: HIV-1 virion lentiviral Vif proteins TPEN zinc binding motif Apobec3g (a3g) is a cellular antiviral protein that, when incorporated into the budding HIV-1 virion, dramatically reduces infectivity in the subsequent cell (1 7). However, A3G incorporation into the virion is dramatically reduced due to the action of the HIV-1 Vif protein. HIV-1 Vif hijacks the cellular ubiquitin proteasome system and targets A3G for proteasomal degradation (8 17). The Cullin E3 ubiquitin ligases are a family of modular RING E3 ligases that consists of three main components: a Cullin (Cul1,2,3,4a,4b,5, or 7), an adaptor protein, and a substrate receptor (18). The E3 ligase is the third enzyme in the ubiquitination sequence and is responsible for substrate specificity. Cullin acts as scaffolding on which the adaptor protein and substrate receptor assemble in order to bring a specific substrate in close proximity to the E2 ubiquitin-conjugating enzyme. The substrate receptor determines the specificity of the protein to be degraded and binds to Cullin via an adaptor protein (19). The E2-conjugating enzyme transfers multiple ubiquitin molecules to the substrate, targeting it for proteasomal degradation. HIV-1 Vif co-opts the Cullin5 E3 ubiquitin ligase, acting as a substrate receptor, targeting A3G for proteasomal degradation (Fig. 1A) (8 17). Both Cul2 and Cul5 bind their substrate receptors through the ElonginB-ElonginC adaptor proteins (20). Cellular substrate receptors have an additional interface that determines Cul2 or Cul5 selection, termed the Cul2 or Cul5 box, respectively (15, 21). Primate lentiviral Vif proteins do not have a Cul5 box although they specifically select Cul5 (15). We previously identified a highly conserved HCCH zinc binding motif and demonstrated its requirement for Cul5 selection (15, 22, 23). We propose that this zinc binding domain is acting to stabilize a putative helix with a hydrophobic face that is required for Cul5 interaction (Fig. 1B) (23). Here we use the membrane-permeable zinc chelator N,N,N,N -tetrakis-(2-pyridylmethyl) ethylenediamine (TPEN) to evaluate the zinc requirement in HIV-1 Vif-mediated A3G degradation in vivo. Treatment with TPEN resulted in a defect in the ability of Vif to recruit Cul5 to the E3 ligase. This resulted in increased A3G stability, allowing it to be packaged within the virion, where it drastically inhibited virus infectivity. The TPEN concentrations used in these studies did not affect the function of a number of cellular zinc binding proteins. MATERIALS AND METHODS Plasmid construction Plasmids NL4 3, NL4 3 Vif, phiv-1vif-hemagglutinin, phiv-1vif-cmyc, SIVagmTanVif, SIVsyk173Vif phapo3g- 1 Correspondence: Department of Microbiology and Immunology, Johns Hopkins Bloomberg School of Public Health, 615 N. Wolfe St., Baltimore, MD 21205, USA. xfyu@jhsph.edu doi: /fj com /07/ FASEB 217

2 Figure 1. Intracellular zinc chelation inhibits Vif interaction with Cul5. A) HIV Vif acts as a substrate receptor, assembling with Cul5, ElonginB, and ElonginC to form an E3 ubiquitin ligase. The Cul5-Vif E3 ubiquitin ligase binds A3G and brings it in close proximity to the E2 ubiquitin-conjugating enzyme. Polyubiquitinated A3G is recognized by the proteasome and subsequently degraded. B) Model of the zinc-stabilized motif in HIV Vif that is required for Cul5-Vif interaction. C) Increasing concentrations of TPEN decrease Cul5 interaction in a dose-dependent manner. 293T cells were transfected with Vif-hemagglutinin expression vectors and treated with the indicated amounts of TPEN. Vif-hemagglutinin was immunoprecipitated and eluted materials were then analyzed by SDS-PAGE and immunoblotting for Cul5, Vif-hemagglutinin, ElonginB, and ElonginC. D) TPEN treatment has no effect on transcription and translation of Cul5 and Vif. 293T cells were transfected with Vif-hemagglutinin and treated with the indicated amounts of TPEN. Cell lysates were harvested and analyzed by SDS-PAGE and immunoblotting for Cul5 and Vif-hemagglutinin. hemagglutinin, and phapo3g-myc have been described (8, 22, 24). VR1012 is the parental vector of phiv-1 Vif and was used as a control vector in the absence of Vif. The adenovirus E4orf6 and human p53 expression vectors (pcmv6.9 and pc53sn3, respectively) were gifts from Gary Ketner (25). Hemagglutinin (HA) -tagged SOCS3 was a generous gift from Nicholas A. Cacalano (UCLA, Los Angeles, CA, USA). Cell culture, transfection, MAGI assay, and antibodies 293T and MAGI-CCR5 cells (26) were maintained and transfected or infected as described previously (8). The MAGI assay was performed as described (8). The antibodies used in this study have been described (8): anti-hemagglutinin antibody (Ab) -agarose conjugate, antimyc Ab-agarose conjugate, anti-vif, anti-elongin B, anti-elongin C, anti-myc, anti-hemagglutinin, and anti-human ribosomal P antigens. The antip53 mouse monoclonal antibody (mab) was obtained from Oncogene Research Products (Cat. #OP03; San Diego, CA, USA). In vivo zinc chelation Medium was removed 12 h prior to harvesting cells and replaced with either fresh medium alone or medium containing the indicated concentration of TPEN (Sigma, St. Louis, MO, USA). After a 12 h incubation, the cells were harvested for immunoprecipitation and immunoblot analysis or the medium was collected for MAGI assay. Immunoprecipitation and immunoblot analysis Transfected 293T cells were harvested, washed twice with cold PBS, and lysed in lysis buffer (50 mm Tris, ph 7.5, with 150 mm NaCl, 1% Triton X-100, and complete protease inhibitor cocktail tablets) at 4 C for 1 h, then centrifuged at 10,000 g for 30 min. For myc tag immunoprecipitation, precleared cell lysates were mixed with antimyc Ab (Upstate Biotechnology, Lake Placid, NY, USA) and incubated with protein G at 4 C for 3 h. For HA tag immunoprecipitation, precleared cell lysates were mixed with anti-hemagglutinin Ab-conjugated agarose beads (Roche, Nutley, NJ, USA) and incubated at 4 C for 3 h. Samples were then washed three times with washing buffer (20 mm Tris, ph 7.5, with 100 mm NaCl, 0.1 mm EDTA, and 0.05% Tween-20). Beads were eluted with elution buffer (0.1 M glycine-hcl, ph 2.0) or 2 loading buffer. The eluted materials were then analyzed by SDS-PAGE and immunoblotting as described (8). 218 Vol. 21 January 2007 The FASEB Journal XIAO ET AL.

3 RESULTS Although we previously demonstrated the requirement of the HCCH motif for zinc coordination (23), we wanted to assess the requirement of zinc for the interaction of HIV-1 Vif with the Cul5 E3 ligase and degradation of A3G in vivo. To this end, we used the membrane-permeable zinc chelator TPEN. Our goal was to identify a TPEN concentration that would interfere with Vif selection of Cul5 while leaving the function of cellular zinc binding proteins such as A3G and Rbx, a zinc-dependent component of the E3 ligase, intact. Specifically interfering with the Vif-Cul5 interaction would allow HIV-1 to efficiently package A3G, allowing HIV to become sensitive to the activities of the antiviral protein even in the presence of Vif. When Vif-expressing 293T cells were treated with increasing concentrations of TPEN, Cul5-Vif interaction decreased in a dose-dependent manner while interaction with ElonginB and ElonginC remained constant (Fig. 1C). From these data we determined TPEN to have an IC50 of 1.79 M. TPEN treatment had no effect on cellular Cul5 protein levels, suggesting that this effect was due to the requirement of zinc for Vif-Cul5 interaction and not to a global effect on translation (Fig. 1D). These data demonstrate the requirement of zinc in vivo for recruitment of Cul5 by HIV-1 Vif. The H- x5 -C- x C- x3 5 -H motif is highly conserved in primate lentiviral Vif proteins. The spacing between the first His and the first Cys is invariably five amino acids. The spacing between the first and second Cys is also highly conserved, with the exception of SIVagm, which has 17 rather than 18 amino acids between these positions. In addition, while most HIV/SIV Vif proteins have five amino acids between the second Cys and the second His, SIVsyk Vif has only three amino acids (Fig. 2A). To verify that zinc is required for Cul5 selection and E3 ligase complex formation in Vif proteins with divergent zinc binding motif spacing, we again used TPEN. 293T cells were transfected with SIVagm or SIVsyk Vif expression vectors and treated with 4 M TPEN or control vehicle DMSO (Fig. 2B). Upon immunoprecipitation of Vif, we detected equal amounts of ElonginB and ElonginC independent of TPEN treatment; however, for all immunoprecipitated Vif proteins we observed a marked decrease in Cul5 coimmunoprecipitation in TPEN-treated cells when compared with vehicle-treated controls (Fig. 2B). These data taken together suggest that the HCCH motif in primate lentiviral Vif proteins represents a zinc binding region that is required for interaction with Cul5 in vivo, irrespective of minor spacing divergence. Although few well-characterized cellular functions of the Cul5 E3 ligases are known, some Cul5-specific substrate receptors have been identified (20, 21, 27, 28). To determine whether zinc-mediated Cul5 recruitment was unique to primate lentiviral Vif proteins, we assessed the effect of TPEN on the cellular Cul5- SOCS-3 E3 ligase. Upon immunoprecipitation of SOCS-3 in the presence of 4 M TPEN or control DMSO, we detected equal interaction with Cul5, suggesting that zinc is not required for interaction of the cellular substrate receptor SOCS-3 with Cul5 (Fig. 2C). We observed similar results when we evaluated the effect of TPEN on the interaction of the adenovirus protein E4orf6 with Cul5 (data not shown). To determine the functional relevance of the zinc binding motif in HIV-1 Vif, we studied the effect of zinc chelation on A3G degradation. In the absence of TPEN, HIV-1 Vif induces the degradation of A3G, but upon addition of TPEN, A3G degradation was abolished (Fig. 3A, compare lanes 1 and lane 3). TPEN had no effect on A3G expression or stability in the absence of HIV-1 Vif (Fig. 3A, lane 2). The observed effect of zinc chelation on A3G degradation was presumably due to the loss of Cul5, as zinc chelation had no effect on Vif-A3G interaction (Fig. 3B). In the absence of TPEN and in the presence of Vif, A3G levels were reduced in both the cell and virion, as would be expected (Fig. 3C, Figure 2. Zinc is required for Cul5 selection by diverse primate lentiviral Vif proteins. A) Alignment illustrating divergence in the spacing of the SIVagm and SIVsyk zinc binding motif. Hydrophobic residues are represented by a phi ( ). B) SIVagm and SIVsyk require zinc for Cul5 selection despite divergence in the spacing from the zinc binding consensus of HIV Vif. 293T cells were transfected with the indicated SIV Vif expression vectors and treated with 4 M TPEN. Cell lysates were immunoprecipitated and analyzed by SDS-PAGE, followed by immunoblotting for Cul5, HA, ElonginB, and ElonginC. C) The cellular substrate receptor SOCS-3 does not require zinc for Cul5 selection. 293T cells were transfected with SOCS-3-hemagglutinin and treated with 4 M TPEN. Cell lysates were immunoprecipitated and analyzed by SDS-PAGE and immunoblotting. ZINC CHELATION INHIBITS VIF FUNCTION 219

4 Figure 3. Intracellular zinc chelation inhibits Vif-mediated A3G degradation. A) TPEN treatment inhibits A3G degradation. 293T cells were transfected with the indicated plasmids in the presence of 4 M TPEN or control vehicle. Cell lysates were analyzed for A3G degradation by SDS-PAGE followed by immunoblotting for A3G-hemagglutinin, Vif-myc, and ribosomal p19 as a loading control. B) Zinc chelation has no effect on Vif-A3G interaction. 293T cells were transfected with the indicated plasmids and treated with 4 M TPEN or control vehicle. Vif-hemagglutinin was immunoprecipitated from prepared lysates. Lysates were analyzed by SDS-PAGE followed by immunoblotting for Vif-hemagglutinin and A3G-myc. C) TPEN treatment increases virion incorporation of A3G. Cells were infected with WT or Vif NL4 3 virus in the presence of A3G-myc and treated with 4 M TPEN or DMSO. Cells were harvested and virus was isolated from the supernatant by ultracentrifugation and analyzed by SDS-PAGE and immunoblotting against Vif, myc, and p24. D) TPEN does not interfere with Rbx function. Adenovirus E4orf6-mediated p53 degradation is not affected by 4 M TPEN. 293T cells were transfected with p53 and E4orf6-myc expression vectors in the presence of 4 M TPEN or control vehicle. Cells were harvested and analyzed for p53 degradation by SDS-PAGE followed by immunoblotting for p53, E4orf6-myc, and ribosomal p19 as a loading control. compare lanes 2 and 3). Upon addition of TPEN, however, cellular A3G protein levels were increased, enhancing incorporation into virions even in the presence of Vif (Fig. 3C, lanes 1 and 3). These data demonstrate the requirement of zinc for Vif degradation of A3G and subsequent exclusion from the virus. Rbx is required for recruitment of the E2 ubiquitinconjugating enzyme and neddylation of Cullin, two functions that are required for E3 activation and subsequent activity. Since zinc is required for Rbx function, we wanted to confirm that the effects of TPEN treatment we had observed were not due to disruption of Rbx function. To evaluate the effect of TPEN on Rbx function, we used one of the other well-characterized Cul5 E3 ligase systems. The adenovirus protein E4orf6 acts as a Cul5 substrate receptor, functioning to degrade cellular p53 (29). In the absence of E4orf6, p53 levels are stable (Fig. 3D, lanes 3 and 4). However, in the presence of the viral substrate receptor, p53 levels decline (Fig. 3D, lane 2). p53 levels also declined in the presence of 4 M TPEN, indicating that the effect of zinc chelation on A3G degradation is not due to a defect in Rbx function (Fig. 3D, lane 1). To assess the biological relevance of the zinc require- 220 Vol. 21 January 2007 The FASEB Journal XIAO ET AL.

5 ment in Vif-Cul5 interaction and assess whether A3G retained antiviral function in the presence of 7 M TPEN, we studied the effect of TPEN on virus infectivity. MAGI cells were infected with wild-type (WT) or Vif-deficient virus produced in 293T or 293T-A3G cells in the presence of 7 M TPEN or control DMSO. We set the infectivity of WT virus in the absence of A3G to 100% and calculated the relative infectivity under all treated conditions (Fig. 4). As expected, in the presence of A3G and in the absence of Vif, there was a drastic decrease of infectivity to 8% (Fig. 4, lane 4). However, in the presence of Vif, A3G was neutralized, resulting in 91% infectivity (Fig. 4, lane 3). One would expect zinc chelation to have an effect on virus infectivity independent of Vif function. We observed a modest decrease in infectivity of 15% even in the absence of A3G (Fig. 4, lanes 5 and 6). However, when we treated producer cells with TPEN in the presence of A3G, Vif was no longer able to neutralize A3G, resulting in an infectivity loss of 87% compared with an infectivity of 91% in the absence of TPEN (Fig. 4, compare lanes 3 and 7). These data demonstrate two important points: 1) zinc is required for Vif function in the neutralization of A3G, and 2) TPEN, at the concentrations used in this assay, has no effect on A3G function (Fig. 4, lanes 7 and 8). DISCUSSION Viruses routinely use cellular machinery for replication and protection from the host defenses. Humans have evolved an innate defense to retroviruses, possibly through coevolution with endogenous retroelements, in the form of the APOBEC3 proteins. However, the virus maintains the upper hand in the form of the Vif protein, capable of hijacking the cellular ubiquitin Figure 4. Zinc chelation drastically reduces virus infectivity in the presence of A3G. MAGI cells were infected with WT or Vif-deficient NL4 3 produced in the absence or presence of A3G and 7 M TPEN as indicated. Relative infectivity was calculated by setting WT virus infectivity to 100% (lane 1). proteasome system and targeting APOBEC3 for degradation. Both Cul2 and Cul5 E3 ligases recruit their substrate receptors through the ElonginB/ElonginC module. It has been demonstrated in cellular E3 ligases that Cul2/Cul5 selection is determined via the presence of a Cul2 or Cul5 box directly downstream of the BC box. Careful inspection of the primate lentiviral Vif proteins did not reveal the presence of a conserved Cul5 box; however, Vif clearly selects Cul5 and not Cul2 (15, 16). We did in fact identify a highly conserved HCCH motif that we previously characterized as a novel zinc binding domain required for Cul5 selection (15, 22, 23). The data presented here demonstrate in vivo the requirement of zinc for Vif function. We used the membrane-permeable zinc chelator TPEN to demonstrate a zinc requirement for recruitment of Cul5 to the E3 ligase (Fig. 1C). TPEN treatment specifically diminished Cul5 interaction with primate lentiviral Vif proteins, rendering the Vif-Cul5 E3 ligase ineffective (Fig. 2B). Zinc chelation had no effect on ElonginB/ElonginC and A3G interaction with Vif, demonstrating that zinc is not required for global Vif structure (Fig. 1C and Fig. 2B). It is interesting that zinc is not required for SOCS3 interaction with Cul5 or E4orf6 degradation of p53 via Cul5 (Fig. 2C and Fig. 3D). These data are important in demonstrating that the HCCH zinc binding motif is unique to the Vif-Cul5 interaction. The ability of E4orf6 to efficiently degrade p53 in the presence of TPEN addresses the effect of zinc chelation on Rbx. Presumably Vif binds zinc weakly relative to cellular proteins such as A3G and Rbx at the concentrations used in our studies. TPEN treatment inhibited Vif-mediated A3G degradation, allowing HIV to become sensitive to the activity of A3G (Fig. 3A and Fig. 4, lane 7). Inhibition of A3G degradation via TPEN treatment may have occurred for the following reasons: 1) Rbx is a zinc binding protein that is required for E3 function, and was inhibited by TPEN; 2) global Vif and A3G structure was disrupted by zinc chelation; and 3) Vif zinc binding domain was disrupted interfering with Cul5 interaction, resulting in inefficient degradation. We addressed the role of Rbx by using the E4orf6-p53 degradation system. In this system, TPEN treatment has no effect on p53 degradation (Fig. 3D). These data suggest that compared to Rbx, which has eight Cys and His residues that coordinate multiple zinc ions, Vif has a relatively weak zinc binding capability and is susceptible to TPEN concentrations that are too low to disrupt Rbx function. If global Vif protein structure had been affected, one would presume that Vif would no longer be able to form stable protein-protein interactions; however, when Vif is immunoprecipitated, comparable amounts of ElonginB and ElonginC coprecipitated regardless of TPEN treatment. As for the effect of TPEN on A3G function, one would assume that zinc chelation would inhibit the two active sites of A3G and therefore inhibit A3G antiviral function. However, A3G was demonstrated to be functional in the infectivity assay, where in ZINC CHELATION INHIBITS VIF FUNCTION 221

6 the absence of A3G and TPEN, infectivity is equal to 100%; upon addition of A3G in the presence of TPEN, however, infectivity was drastically reduced even in the presence of Vif, due to the sensitivity of the virus to A3G activity (Fig. 4, lanes 7 and 8). Therefore, the likely explanation for the TPEN-induced defect in A3G degradation is the effect of TPEN on Vif-Cul5 E3 ligase assembly. The data presented here demonstrate the requirement for zinc in the degradation of A3G by Vif via the Cul5 E3 ligase. Our data also suggest that the zinc requirement for Cullin-substrate receptor interaction may be unique to the Vif-Cul5 E3 ligase and that this interaction may be disrupted using TPEN concentrations that do not affect stronger zinc binding cellular proteins such as A3G and Rbx. Further work characterizing this zinc binding motif and more detailed mapping of the Cul5 interface will facilitate the development of novel antiviral therapeutics. This work was supported by a grant from the NIH (A ) and funding from the National Science Foundation of China (NSFC ) and Cheung Kong Scholars Program Foundation of the Chinese Ministry of Education to X-F.Y. We thank Nicholas A. Cacalano (UCLA) for the SOCS3 expression vector and D. McClellan for editorial assistance. The following reagents were obtained through the AIDS Research Reagents Program, Division of AIDS, NIAID, NIH: SIVagmTan, SIVsyk173, MAGI-CCR5 cells and anti-vif antibody. REFERENCES 1. Harris, R. S., and Liddament, M. T. (2004) Retroviral restriction by APOBEC proteins. Nat. Rev. Immunol. 4, Navarro, F., and Landau, N. R. (2004) Recent insights into HIV-1 Vif. Curr. Opin. Immunol. 16, Turelli, P., and Trono, D. (2005) Editing at the crossroad of innate and adaptive immunity. Science 307, Goff, S. P. (2003) Death by deamination: a novel host restriction system for HIV-1. Cell 114, Rose, K. M., Marin, M., Kozak, S. L., and Kabat, D. (2004) The viral infectivity factor (Vif) of HIV-1 unveiled. Trends Mol. Med. 10, Bieniasz, P. D. (2004) Intrinsic immunity: a front-line defense against viral attack. Nat. Immunol. 5, Sheehy, A. M., Gaddis, N. C., Choi, J. D., and Malim, M. H. (2002) Isolation of a human gene that inhibits HIV-1 infection and is suppressed by the viral Vif protein. Nature 418, Yu, X., Yu, Y., Liu, B., Luo, K., Kong, W., Mao, P., and Yu, X. F. (2003) Induction of APOBEC3G ubiquitination and degradation by an HIV-1 Vif-Cul5-SCF complex. Science 302, Mehle, A., Strack, B., Ancuta, P., Zhang, C., McPike, M., and Gabuzda, D. (2004) Vif overcomes the innate antiviral activity of APOBEC3G by promoting its degradation in the ubiquitinproteasome pathway. J. Biol. Chem. 279, Stopak, K., de Noronha, C., Yonemoto, W., and Greene, W. C. (2003) HIV-1 Vif blocks the antiviral activity of APOBEC3G by impairing both its translation and intracellular stability. Mol. Cell. 12, Marin, M., Rose, K. M., Kozak, S. L., and Kabat, D. (2003) HIV-1 Vif protein binds the editing enzyme APOBEC3G and induces its degradation. Nat. Med. 9, Conticello, S. G., Harris, R. S., and Neuberger, M. S. (2003) The Vif protein of HIV triggers degradation of the human antiretroviral DNA deaminase APOBEC3G. Curr. Biol. 13, Sheehy, A. M., Gaddis, N. C., and Malim, M. H. (2003) The antiretroviral enzyme APOBEC3G is degraded by the proteasome in response to HIV-1 Vif. Nat. Med. 9, Liu, B., Sarkis, P. T., Luo, K., Yu, Y., and Yu, X. F. (2005) Regulation of Apobec3F and human immunodeficiency virus type 1 Vif by Vif-Cul5-ElonB/C E3 ubiquitin ligase. J. Virol. 79, Yu, Y., Xiao, Z., Ehrlich, E. S., Yu, X., and Yu, X. F. (2004) Selective assembly of HIV-1 Vif-Cul5-ElonginB-ElonginC E3 ubiquitin ligase complex through a novel SOCS box and upstream cysteines. Genes Dev. 18, Mehle, A., Goncalves, J., Santa-Marta, M., McPike, M., and Gabuzda, D. (2004) Phosphorylation of a novel SOCS-box regulates assembly of the HIV-1 Vif-Cul5 complex that promotes APOBEC3G degradation. Genes Dev. 18, Kobayashi, M., Takaori-Kondo, A., Miyauchi, Y., Iwai, K., and Uchiyama, T. (2005) Ubiquitination of APOBEC3G by an HIV-1 Vif-Cullin5-Elongin B-Elongin C complex is essential for Vif function. J. Biol. Chem. 280, Skowyra, D., Craig, K. L., Tyers, M., Elledge, S. J., and Harper, J. W. (1997) F-box proteins are receptors that recruit phosphorylated substrates to the SCF ubiquitin-ligase complex. Cell 91, Petroski, M. D., and Deshaies, R. J. (2005) Function and regulation of cullin-ring ubiquitin ligases. Nat. Rev. Mol. Cell. Biol. 6, Kamura, T., Sato, S., Haque, D., Liu, L., Kaelin, W. G., Jr., Conaway, R. C., and Conaway, J. W. (1998) The Elongin BC complex interacts with the conserved SOCS-box motif present in members of the SOCS, ras, WD-40 repeat, and ankyrin repeat families. Genes Dev. 12, Kamura, T., Maenaka, K., Kotoshiba, S., Matsumoto, M., Kohda, D., Conaway, R. C., Conaway, J. W., and Nakayama, K. I. (2004) VHL-box and SOCS-box domains determine binding specificity for Cul2-Rbx1 and Cul5-Rbx2 modules of ubiquitin ligases. Genes Dev. 18, Luo, K., Xiao, Z., Ehrlich, E., Yu, Y., Liu, B., Zheng, S., and Yu, X. F. (2005) Primate lentiviral virion infectivity factors are substrate receptors that assemble with cullin 5-E3 ligase through a HCCH motif to suppress APOBEC3G. Proc. Natl. Acad. Sci. U. S. A. 102, Xiao, Z., Ehrlich, E., Yu, Y., Luo, K., Wang, T., Tian, C., and Yu, X. F. (2006) Assembly of HIV-1 Vif-Cul5 E3 ubiquitin ligase through a novel zinc-binding domain-stabilized hydrophobic interface in Vif. Virology 349, Liu, B., Yu, X., Luo, K., Yu, Y., and Yu, X. F. (2004) Influence of primate lentiviral Vif and proteasome inhibitors on human immunodeficiency virus type 1 virion packaging of APOBEC3G. J. Virol. 78, Boyer, J. L., and Ketner, G. (2000) Genetic analysis of a potential zinc-binding domain of the adenovirus E4 34k protein. J. Biol. Chem. 275, Chackerian, B., Long, E. M., Luciw, P. A., and Overbaugh, J. (1997) Human immunodeficiency virus type 1 coreceptors participate in postentry stages in the virus replication cycle and function in simian immunodeficiency virus infection. J. Virol. 71, Brower, C. S., Sato, S., Tomomori-Sato, C., Kamura, T., Pause, A., Stearman, R., Klausner, R. D., Malik, S., Lane, W. S., Sorokina, I., et al. (2002) Mammalian mediator subunit mmed8 is an Elongin BC-interacting protein that can assemble with Cul2 and Rbx1 to reconstitute a ubiquitin ligase. Proc. Natl. Acad. Sci. U. S. A. 99, Kamura, T., Burian, D., Yan, Q., Schmidt, S. L., Lane, W. S., Querido, E., Branton, P. E., Shilatifard, A., Conaway, R. C., and Conaway, J. W. (2001) Muf1, a novel Elongin BC-interacting leucine-rich repeat protein that can assemble with Cul5 and Rbx1 to reconstitute a ubiquitin ligase. J. Biol. Chem. 276, Querido, E., Blanchette, P., Yan, Q., Kamura, T., Morrison, M., Boivin, D., Kaelin, W. G., Conaway, R. C., Conaway, J. W., and Branton, P. E. (2001) Degradation of p53 by adenovirus E4orf6 and E1B55K proteins occurs via a novel mechanism involving a Cullin-containing complex. Genes Dev. 15, Received for publication June 28, Accepted for publication August 14, Vol. 21 January 2007 The FASEB Journal XIAO ET AL.