| |
HIV protein Vpr may lead to new therapies
|
| |
| |
PHILADELPHIA, March 31 (UPI) -- Researchers said Friday they have found an HIV protein that might provide a key in the fight against AIDS.
Scientists at the University of Pennsylvania School of Medicine said that an HIV-1 accessory protein called "Vpr" appears to destroy the host cell's ability to survive by binding to a host receptor.
This binding ability then keeps an important enzyme from activating the host cell's immune system.
This finding might not only eventually provide a new approach to treating AIDS, but may also help in the battle against inflammatory diseases like rheumatoid arthritis, and even sepsis, the researchers said.
Earlier research had already shown that the Vpr protein binds to the glucocorticoid receptor (GR) of the host cell, but the question of whether the GR pathway was required for Vpr to commandeer the host cell's machinery had still eluded scientists.
To answer this question, the researchers used an siRNA, a short sequence of RNA used to silence gene expression, to completely destroy expression of the glucocorticoid receptor protein.
When the researchers kept the glucocorticoid receptor protein from being made, Vpr did not kill host cells, they noted.
"This indicated that glucocorticoid receptor function is not what's really necessary for Vpr activity," the researchers said. "The glucocorticoid receptor-Vpr complex must be interacting with something else."
The findings appear in the February print issue of Nature Cell Biology.
The HIV-1 Vpr and glucocorticoid receptor complex is a gain-of-function interaction that prevents the nuclear localization of PARP-1
Letter
Nature Cell Biology 8, 170 - 179 (2006)
Feb 2006
Karuppiah Muthumani1, Andrew Y. Choo1, 2, Wei-Xing Zong3, Muniswamy Madesh3, Daniel S. Hwang1, Arumugam Premkumar4, Khanh P. Thieu1, Joann Emmanuel1, Sanjeev Kumar1, Craig B. Thompson3 & David B. Weiner1
1 Department of Pathology and Laboratory Medicine, University of Pennsylvania School of Medicine, Philadelphia, PA 19104, USA.
2 Present Address: Program for Biological and Biomedical Sciences (BBS), Department of Cell Biology, Harvard Medical School, Boston, MA 02115, USA.
3 Abramson Family Cancer Research Institute, Department of Cancer Biology, University of Pennsylvania School of Medicine, Philadelphia, PA 19104, USA.
4 Laboratory of Molecular Neuropharmacology, Memorial Sloan-Kettering Cancer Center, New York, NY 10021, USA.
ABSTRACT
The Vpr protein of HIV-1 functions as a vital accessory gene by regulating various cellular functions, including cell differentiation, apoptosis, nuclear factor of kB (NF-kB) suppression and cell-cycle arrest of the host cell. Several reports have indicated that Vpr complexes with the glucocorticoid receptor (GR), but it remains unclear whether the GR pathway is required for Vpr to function1. Here, we report that Vpr uses the GR pathway as a recruitment vehicle for the NF-kB co-activating protein, poly(ADP-ribose) polymerase-1 (PARP-1). The GR interaction with Vpr is both necessary and sufficient to facilitate this interaction by potentiating the formation of a Vpr-GR-PARP-1 complex. The recruitment of PARP-1 by the Vpr-GR complex prevents its nuclear localization, which is necessary for Vpr to suppress NF-kB. The association of GR with PARP-1 is not observed with steroid (glucocorticoid) treatment, indicating that the GR association with PARP-1 is a gain of function that is solely attributed to HIV-1 Vpr. These data provide important insights into Vpr biology and its role in HIV pathogenesis.
BACKGROUND. A trademark of HIV infection is the diminution of the CD4+ T-cell count of the host, which invariably leads to eventual immunodeficiency2. It is believed that various viral factors contribute to this effect by suppressing both immune activation and T-cell expansion3, 4, 5, 6. The 96-amino-acid viral protein R (Vpr), which has a relative molecular mass of 14,000, has been implicated in both the destruction and suppression of potential antigen-specific T cells through multiple mechanisms7. In fact, Vpr is sufficient to suppress mitogen or anti-CD3-dependent proliferation and activation of T cells. Additionally, Vpr is present in the serum of infected patients and can efficiently reactivate viruses from latency8, 9. Furthermore, Vpr possesses intrinsic transduction properties, which indicates that there are various viral-induced pathogenesis events that occur within a non-viral infection setting10. Other reported important activities include host cell-cycle arrest at the G2/M stage, nuclear transport of the pre-integration complex, host-cell apoptosis, nuclear herniations and the induction of immune suppression11, 12, 13, 14, 15, 16, 17, 18, 19, 20.
Glucocorticoid receptor II (GR-II) has been identified as an in vivo target for Vpr12, 20, 21, 22. The Vpr-GR interaction is dependent on the signature LXXLL motif, the abrogation of which attenuates the GR-dependent co-activation and transcription that is induced by Vpr. In addition, co-treatment with the GR antagonist mifpristone (Mif) blocks several pathogenic functions of Vpr, including apoptosis and viral transcription12, 17, 19. However, the mechanism behind nuclear factor of kB (NF-kB) suppression by Vpr currently remains unresolved. Furthermore, the functional deviations between Vpr and glucocorticoid treatments indicates that different mechanisms may occur.
In an effort to understand the role of the GR in Vpr-mediated NF-B suppression, we compared NF-kB-dependent transcriptional activation in cells with a functional GR and in CV-1 cells, a monkey kidney cell line that expresses an endogenous GR, but is refractory of function23. As shown in Fig. 1a, co-transfection of Vpr but not of a control vector into HeLa cells, is sufficient to inhibit tumour necrosis factor-ƒÊ (TNF-)-induced NF-kB transcription. The inhibition was also observed in cells prone to HIV-1 infection, including Jurkat T cells, U937 monocytes and primary peripheral blood leukocyte (PBL) cells and macrophages (Fig. 1c-f). More interestingly, the same inhibitory effect was also observed in CV-1 cells that possess a non-functional GR (Fig. 1b), indicating that GR-mediated transcription is not required for NF-B suppression, contrary to previous reports that suggested that GR activation leads to an upregulation of inhibitory I-B12. This was further verified, as shown by the fact that inhibition of de novo protein synthesis via cycloheximide treatment did not attenuate Vpr-mediated NF-B-dependent transcription (Fig. 1g). Vpr treatment was also accompanied by a reduced nuclear duration of RelA (p65) in both functional GR and non-functional GR cells (Fig. 1h). This result could be due to a failure of the formation of transcriptional complexes, which prevents acetylation and extended presence of RelA within the nucleus, as Vpr did not significantly affect its initial nuclear localization24. As upstream kinase inhibition could manifest the same effect, we next examined the activity of I-B kinase-b (IKKb). Vpr treatment did not affect the kinase activity of IKK (Fig. 1i) nor did it affect in vivo phosphorylation and turnover of I-kBƒÊ (Fig. 1j). However, Vpr potently attenuated the DNA-binding activity of NF-kB (RelA) at both the initial (Fig. 1k) and the later time points, and this effect was specific to RelA and not to other transcriptional factors. Last, co-transfection or Vpr treatment directly attenuated RelA-mediated transcriptional activation (see Supplementary Information, Fig. S1a-d). Taken together, these results indicate that Vpr suppresses NF-kB at the transcriptional level independently of concomitant de novo transcription induced by GR activation.
A hypothesis for this inhibition is that Vpr may destabilize the NF-kB transcriptional complex through a direct and/or indirect cross-talk mechanism. Interestingly, recent studies have indicated a role for PARP-1 in co-activating NF-kB-dependent gene expression25, 26, 27. Evidence indicates that PARP-1 is linked to the transcriptional potential of NF-B through formation of a critical transcriptional complex within the nucleus26, 27 and, consequently, Parp-1-deficient mice are refractory to toxin-induced sepsis. In view of this, we noticed that treatment of both HeLa and Jurkat cells leads to a dosage-dependent reduction in nuclear localization of PARP-1, as measured by biochemical and immunofluoresence methods (Fig. 2a, b). This effect was specific to PARP-1 and was not associated with other trafficking factors (such as eIF4E) and constitutive nuclear factors (such as PCNA) (Fig. 2c-e). The decrease in nuclear PARP-1 was concomitant with an increase in cytoplasmic PARP-1, indicating a defect in the nuclear or cytoplasmic shuttling of PARP-1 (Fig. 2d). Co-staining of Vpr-GFP (green fluorescent protein) with both PARP-1 and GR revealed significant colocalization within the cytoplasmic and perinuclear regions (see Supplementary Information, Fig. S2a), whereas GFP alone failed to affect PARP-1 nuclear localization.
Next, we hypothesized whether Vpr may be interacting with PARP-1 and, therefore, facilitating its failed nuclear localization. Therefore, plasmid (p) Vpr was transfected into CV-1, PBL cells and macrophages with appropriate controls; they were immunoprecipitated with either anti-Vpr or PARP-1 antibodies and were blotted for the other. As shown in Fig. 2f-h, Vpr was sufficient to co-immunoprecipitate with PARP-1, indicating that the two reside together in a complex in CV-1 and primary HIV-1 target cells. We then asked whether this requires a Vpr-GR interaction, which we have shown previously to be a cellular receptor for Vpr. As shown in Fig. 2i and Supplementary Information, Fig. S2b, the GR antagonist mifpristone is sufficient to reverse the cellular accumulation of PARP-1, indicating that the interaction between Vpr and GR is necessary for Vpr to affect PARP-1 localization. Not surprisingly, the interaction between Vpr and PARP-1 was also sensitive to mifpristone in PBL and HeLa cells (Fig. 2j, k), which indicates that the Vpr-GR complex may facilitate the interaction with PARP-1, and therefore regulate its localization. Paradoxically, steroid treatment (dexamethasone) alone is insufficient to relocate PARP-1 from the nucleus to the cytoplasm (Fig. 2i; and see Supplementary Information, Fig. S2b). In fact, recombinant protein (r) Vpr and dexamethasone exhibit competing effects on PARP-1 localization (see Supplementary Information, Fig. S2c). Therefore, although the Vpr interaction with GR was required for PARP-1 interaction, GR activation by steroids alone is insufficient, implying a molecular distinction between Vpr and steroid-receptor activation.
Next, we examined the molecular nature of the Vpr-PARP-1 complex. We in vitro translated PARP-1, Vpr and GR with 35S-Met and performed co-precipitation experiments. As shown in Fig. 3a-c, although Vpr directly interacts with GR as previously shown20, Vpr was unable to directly interact with PARP-1. In addition, as is consistent with dexamethasone experiments, GR is unable to interact with PARP-1. PARP-1 was only sufficient to complex with Vpr when it was complexed with GR (Fig. 3c). This was also observed in glutathione S-transferase (GST) pull-down assays and confirmed by western blot analysis of the pull-down (Fig. 3d, e). Taken together, these results indicate that the Vpr-GR interaction is necessary and sufficient to recruit PARP-1 in vitro.
To determine whether this can be extrapolated in vivo, we reduced the endogenous levels of GR via small interfering RNA (siRNA). Of the specific clones tested, GR-21 specifically and efficiently knocked down GR protein levels (Fig. 3f). Subsequently, transfection of Vpr into these cells failed to co-immunoprecipitate PARP-1 compared with siRNA control cells, indicating that Vpr requires GR to interact with PARP-1 in vivo (Fig. 3g). Biochemical fractionation of His-tagged Vpr through a nickel column and subsequent co-immunoprecipitation of PARP-1 or GR reveals that the three proteins exist in one complex rather than mutually exclusive complexes in vivo (Fig. 3h).
If Vpr recruitment of PARP-1 via GR interaction is, in fact, regulating transcriptional suppression, then prevention of either the Vpr interaction with GR or with PARP-1 should be sufficient to reverse the NF-kB suppression that is induced by Vpr. Accordingly, co-treatment of mifpristone, which abrogates the Vpr-GR interaction, was sufficient to significantly (P < 0.001) recover the transcriptional suppression of co-transfected RelA in HeLa and CV-1 cells, respectively (Fig. 4a, b). Similar results were also obtained in HIV-1 target cells, including Jurkat T cells and U937 monocytes (Fig. 4c, d). In a similar fashion, loss of GR function via siRNA also exhibited an inability for Vpr to suppress NF-B, again indicating that GR is necessary for Vpr to inhibit NF-kB (Fig. 4e).
The role of PARP-1 in regulating NF-B suppression seems to be entirely structural and not associated with its enzymatic properties, as DPQ (a nicotinic acid analogue) and a PARP-1 enzymatic inhibitor, did not significantly repress TNF--induced NF-B transcription28. Meanwhile, small interfering RNA (siRNA)-mediated knockdown of PARP-1 had a more pronounced effect on NF-B-dependent transcription relative to knockdown efficiency (see Supplementary Information, Fig. S3a-c). To determine whether recruitment of PARP-1 to the Vpr-GR complex was required for Vpr to suppress NF-B, we screened previously described mutants29 for their ability to interact with GR and/or PARP-1. As shown in Fig. 4f, a point mutation of Vpr at H71 to tyrosine (Y) exhibited binding to GR but, importantly, failed to coprecipitate PARP-1. Consistently, transfection of this mutant into HeLa cells exhibited an obvious attenuation of cytoplasmic localization of PARP-1 compared with wild-type Vpr (Fig. 4g). Last, transfection of this mutant, but not others, exhibited a significant reduction in its ability to repress NF-B-dependent transcription (Fig. 4h). Taken together, these results indicate that Vpr recruitment to the GR complex is required for efficient NF-B transcription. It also indicates that, to a large extent, the Vpr-GR interaction is merely a priming effect to potentiate the recruitment of PARP-1, which is necessary for transcriptional suppression.
To determine whether the interaction between PARP-1 and Vpr is relevant in vivo, we next challenged mice with lipopolysaccharide (LPS) or staphyloccal enterotoxin B (SEB) and measured the ability of Vpr to suppress immune activation and PARP-1 nuclear localization. SEB and LPS have been extensively used as toxin models to study the effects on immune cells19, 30. Knockout mice that are deficient in the Parp-1 gene are refractory to the toxic effects of LPS challenge25. Using the D-galactosmine-sensitized mouse model30, we investigated the protective activity of recombinant Vpr that was delivered intravenously (Fig. 5a) against SEB and LPS challenge. Injection of Vpr into mice effectively repressed the hyperinflammation (TNF-a, interleukin (IL)-12 and IL-1) that was induced by SEB (Fig. 5c). The splenocytes from mice injected with mock, SEB, SEB + Vpr, SEB + Dex or SEB + Vpr + mifpristone were then extracted and lysed. The proteins were fractionated between cytoplasmic and nuclear extracts and blotted for PARP-1 expression. Mice that were injected with SEB alone demonstrated significant localization of PARP-1 to the nucleus, which supports its role as a co-activator for NF-kB. However, mice that were co-injected with SEB and Vpr failed to localize PARP-1 into the nucleus (Fig. 5d). This effect could be reversed by co-injection with mifpristone, indicating that the Vpr-GR interaction is necessary in vivo for preventing nuclear localization of PARP-1. By contrast, SEB- and dexamethasone-injected mice failed to prevent the localization of Parp-1 into the nucleus. These results support the in vitro finding that failed PARP-1 nuclear localization is specific to Vpr and is not manifested by a traditional steroid effect. Furthermore, immunoprecipitation of Vpr within the splenic lysates reveals its presence within this lymphoid organ post-intravenous injection (Fig. 5b). Taken together, these results indicate that the inhibition of nuclear localization of PARP-1 by Vpr is a relevant phenomenon both in vitro and in vivo.
In this study, we have elucidated the mechanistic role for GR in mediating NF-B suppression by Vpr. Although the presence of the structural GR protein is required, its transcriptional property is not. The structural GR, following interaction with Vpr, induces the recruitment of PARP-1, and the GR-Vpr interaction alone is sufficient to complex with PARP-1 in vitro. Accordingly, we propose that the 'gain of function' of the Vpr-GR interaction is a priming event for PARP-1 recruitment and hence prevents its nuclear accumulation. Although the molecular nature of the Vpr-Gr-PARP-1 complex remains unknown, the interaction of Vpr with GR may induce conformational changes within the complex that expose binding sites for the PARP-1 interaction. Consistently, steroid treatment is insufficient to recruit PARP-1 and interact with GR. Future work will identify the details of this regulation, such as the specific importin that may be involved or the role of the nuclear-cytoplasmic shuttling properties of the Vpr.
Although our results reveal a mechanism by which Vpr can suppress NF-B, we propose that prevention of PARP-1 nuclear localization may have greater consequences for HIV infection. First, PARP-1 has recently been shown to be a modulator of chromatin structure by binding to nucleosomes. In vivo, PARP-1 binding to nucleosomes has been correlated with transcriptionally repressed chromatin domains31. This indicates that Vpr has a potent ability to transcriptionally alter gene expression, which may also be associated with failed PARP-1 nuclear localization. Second, PARP-1 has been shown to be a negative regulator of TAT-mediated HIV transcription by directly competing for the TAT RNA32. Therefore, retaining PARP-1 within the cytoplasm could potentially augment TAT-mediated transcription and viral infectivity. Consistently, it has been observed that Vpr can augment TAT-mediated viral production33, 34.
In conclusion, the present results help to clarify a controversy in the Vpr literature, but at the same time raise several important questions regarding Vpr biology. We observe that the GR complex is a central target of Vpr in vitro and in vivo. The Vpr-GR complex is a gain-of-function interaction that represses NF-B transcription through the additional recruitment of PARP-1, a crucial regulator of NF-kB. It is likely that Vpr targeting of PARP-1 could add to the T-cell depletion and immunosuppressive effects that are observed during HIV-1 infection, illustrating the complexity of HIV-1 infection. These studies indicate that understanding the relationship of Vpr to PARP-1 activation may reveal new insights into the important roles of Vpr, such as cell-cycle arrest and apoptosis, in host-cell pathogenesis.
|
|
| |
| |
|
|
|
