Intermolecular masking of the HIV-1 Rev NLS by the cellular protein HIC: Novel insights into the regulation of Rev nuclear import
- 6k Downloads
The HIV-1 regulatory protein Rev, which is essential for viral replication, mediates the nuclear export of unspliced viral transcripts. Rev nuclear function requires active nucleocytoplasmic shuttling, and Rev nuclear import is mediated by the recognition of its Nuclear Localisation Signal (NLS) by multiple import factors, which include transportin and importin β. However, it remains unclear which nuclear import pathway(s) predominate in vivo, and the cellular environment that modulates Rev nucleocytoplasmic shuttling remains to be characterised.
In our study, we have identified the cellular protein HIC (Human I-mfa domain-Containing protein) as a novel interactor of HIV-1 Rev. We demonstrate that HIC selectively interferes with Rev NLS interaction with importin β and impedes its nuclear import and function, but does not affect Rev nuclear import mediated by transportin. Hence, the molecular determinants mediating Rev-NLS recognition by importin β and transportin appear to be distinct. Furthermore, we have employed HIC and M9 M, a peptide specifically designed to inhibit the transportin-mediated nuclear import pathway, to characterise Rev nuclear import pathways within different cellular environments. Remarkably, we could show that in 293T, HeLa, COS7, Jurkat, U937, THP-1 and CEM cells, Rev nuclear import is cell type specific and alternatively mediated by transportin or importin β, in a mutually exclusive fashion.
Rev cytoplasmic sequestration by HIC may represent a novel mechanism for the control of Rev function. These studies highlight that the multivalent nature of the Rev NLS for different import receptors enables Rev to adapt its nuclear trafficking strategy.
KeywordsNuclear Localisation Signal Nuclear Import Cytosolic Extract Parent Plasmid Nuclear Localisation Signal Domain
The HIV-1 regulatory protein Rev (18 kDa) is essential for HIV-1 replication [1, 2]. Rev is predominantly localised in the nucleus/nucleolus , and its primary function is to mediate the nuclear export of partially spliced and unspliced viral transcripts. Rev has also been shown to modulate splicing and translation of viral transcripts, and their subsequent packaging, and to interfere with integration of the HIV-1 genome [4, 5, 6, 7]. Rev nuclear export of unspliced viral transcripts requires active shuttling of the protein between the nucleus and cytoplasm via nuclear pore complexes (NPCs) which is mediated by two major functional domains, the Nuclear Localisation Signal (NLS) and the Nuclear Export Signal (NES) [8, 9]. The leucine-rich Rev NES binds directly to CRM1, which in concert with DDX3, a DEAD box RNA helicase, facilitates Rev nuclear export of unspliced viral transcripts via the NPC [10, 11, 12, 13, 14]. Also, Rev-export function was shown to be inhibited by Nuclear Factor 90 (NF90). The basic arginine-rich Rev NLS mediates both Rev nuclear import and binding to the Rev Response Element (RRE), a cis-acting RNA element present in all unspliced viral transcripts [16, 17, 18]. The Rev NLS is recognized by at least 5 different importin β family members, including importin β, transportin, importin 5, importin 7 and importin 9, which facilitate its nuclear import [19, 20, 21, 22, 23]. Despite evidence showing the utilisation of multiple nuclear import receptors in vitro by Rev, it remains unclear if some are redundant and/or if, under specific conditions, one nuclear import pathway may predominate in vivo. Hutten et al. described transportin as a major Rev nuclear import receptor . However, this study was restricted to HeLa cells, and the molecular and cellular determinants governing the interaction of Rev with one or the other nuclear import receptors have not been investigated.
In this report, we identified a novel interaction between HIV-1 Rev and the cellular protein HIC (Human I-mfa domain-Containing protein). HIC is a 246 amino acid protein with a prominently cytoplasmic distribution. The cysteine-rich C-terminal domains of HIC and the Inhibitor of MyoD family a (I-mfa) share a high homology (74%) and are essential for their activities . HIC acts as a regulator of transcription and interacts with and/or modulates the activity of several cellular and viral transcription factors, including Axin, cyclin T1 and T2, TCF1, HIV-1 Tat, HTLV-1 Tax and KSHV LANA [24, 25, 26, 27, 28, 29]. We previously reported that the ectopic expression of HIC resulted in the mislocalisation of HIV-1 Tat to the cytoplasm. This contrasted with an earlier report which showed that Tat and HIC co-localised in the nucleolus . Nevertheless, these studies were descriptive, and the effects of HIC on the nuclear transport machinery were not investigated.
In this report, we explored the mechanisms whereby HIC could regulate Rev nuclear import and demonstrated that HIC selectively blocks importin β- but not transportin-mediated Rev nuclear import via a mechanism involving the intermolecular masking of Rev NLS by HIC. In addition, we employed HIC, as an inhibitor of importin β mediated Rev nuclear import, and M9 M, a peptide which specifically inhibits the transportin pathway, as tools to further characterise Rev nuclear import pathway(s) in HeLa, 293T, COS7, Jurkat, CEM, THP-1 and U937 cells. While we confirmed that transportin is the major import receptor for Rev in HeLa, THP-1 and U937 cells, we showed that in 293T, COS7, CEM and Jurkat cells, importin β-mediated Rev nuclear import is dominant. Subsequently, reporter gene assays revealed that HIC contributes to the control of Rev function in 293T Jurkat and CEM but not in HeLa, U937 and THP-1 cells. Collectively these results demonstrate that Rev nuclear import is tightly regulated and suggest that the molecular determinants mediating Rev transport by importin β and transportin are distinct, and that the Rev dominant nuclear import pathway is cell type specific.
HIC sequesters Rev in the cytoplasm by inhibiting its nuclear import in vivo
To determine whether HIC inhibits Rev nuclear import or promotes Rev nuclear export, we repeated the colocalisation studies, in the presence of Leptomycin B, which specifically inhibits nuclear export mediated by CRM1 . Since Rev localisation is the result of a net balance between Rev import and export, treatment with LMB interrupted the nuclear export and resulted in an overall increase in Rev nuclear localisation, as determined by an increase in both the number of cells displaying Rev exclusively in the nucleus and Rev nuclear signal intensity (Figure 1D). However, LMB did not prevent the overall effects of HIC on Rev cytoplasmic redistribution. Indeed when Rev was co-expressed with HIC, the percentage of cells displaying Rev in the nucleus decreased from 83% (Rev alone+LMB) to 43% (Rev+HIC+LMB) (Figure 1D upper panel). In addition, the Rev nuclear signal intensity also decreased from 72% (Rev alone+LMB) to 40% (Rev+HIC+LMB) (Figure 1D, lower panel). These observations support the view that HIC most likely sequesters Rev in the cytoplasm by inhibiting its nuclear import rather than promoting its nuclear export.
HIV-1 Rev nuclear import mediated by importin β is selectively blocked in vitro by a competitive excess of HIC
To further dissect the mechanisms underlying HIC inhibition of Rev nuclear import, we reduced the complexity of this system, by employing 2 μg of recombinant importin β or transportin, as sole import factors. Remarkably, HIC efficiently and distinctively abolished Rev nuclear import mediated by importin β but not by transportin (Figure 2B). As previously shown, HIC did not block transportin mediated nuclear import of M9. To further examine the effect of HIC on importin β-mediated Rev import, we performed similar experiments with Rev mutants (GST-YFP-RevΔN1/-RevΔN2/-RevNLS) with or without deletions of the Rev NLS domain (Figure 2C). Strong fluorescent signals were observed in the nucleus for both RevΔN1 and RevNLS, both of which encompass the functional NLS sequence, while RevΔN2 lost its nuclear localisation due to the absence of the NLS domain (Figure 2C). The addition of increasing amounts of HIC (0.5-2 μg) correlated with a decreasing nuclear signal intensity for both RevΔN1 and RevNLS (Figure 2C), demonstrating that the RevNLS is sufficient and necessary for both Rev nuclear import and its inhibition by HIC.
HIV-1 Rev and HIC interact directly in vitro and form a complex in vivo
Subsequently, co-immunoprecipitation assays were employed to investigate potential interactions between HIV-1 Rev and HIC in transfected 293T cells. HA-Rev was immunoprecipitated and HIC was specifically detected in the eluted fraction (Figure 3B). Similarly, the HIC mutant (144-246), which contains the I-mfa domain, but not the HIC mutant (2-144), interacted with Rev (Figure 3B). Thus, HIC and Rev interact in vivo and in vitro, and this interaction is mediated by the I-mfa domain of HIC and Rev NLS domain.
HIC selectively interferes with the Rev NLS interaction with importin β
The dominant Rev nuclear import pathway is cell type dependent
Remarkably, HIC selectively blocked Rev nuclear import in the presence of 293T, COS7, CEM or Jurkat cytosolic extracts (Figure 6 and Figure S1; Additional File 1). However, its nuclear import remained unaffected by HIC when HeLa, THP-1 or U937 cytosolic extracts were employed (Figure 6 and Figure S1; Additional File 1). In contrast, M9 M inhibited Rev nuclear import when HeLa, THP-1 or U937 cytosolic extracts were employed but did not affect Rev nuclear import mediated by 293T, COS7, CEM or Jurkat cytosolic extracts (Figure 6 and Figure S1; Additional File 1). These results are consistent with Hutten et al.  and support their finding that transportin is the major nuclear import receptor for Rev in HeLa cells. Additionally, our results demonstrated that transportin is also the primary import factor in U937 and THP-1 cells. In 293T, COS7, CEM and Jurkat however, the Rev dominant nuclear import pathway appears to involve importin β, which is consistent with our co-localisation assay performed in COS7 cells.
HIC inhibits Rev function in a cell-specific fashion
Collectively, these functional results correlate with HIC selective inhibition of Rev nuclear import mediated by 293T, Jurkat or CEM cytosol or importin β, but not by HeLa, U937 or THP-1 cytosol and transportin and strongly suggest that HIC contributes to the spatial control of Rev function in 293T, Jurkat and CEM cells.
This study identifies the cellular protein HIC as a novel interactor and regulator of HIV-1 Rev nuclear import and function. First, using in vitro nuclear import assays, we analysed the role of individual nuclear transport machinery components in mediating Rev nuclear import and employed importin β or transportin as the transport receptors in the presence of a competitive amount of recombinant HIC. We demonstrated that HIC selectively inhibited Rev nuclear import mediated by importin β but not by transportin and that the Rev NLS domain was sufficient and necessary for Rev nuclear import inhibition by HIC. Additional controls and complementary experiments demonstrated that HIC selectively and physically targeted the Rev NLS domain and not the import factors themselves. Furthermore, the observed inhibition of Rev nuclear import did not result from a general block of import pathways or the physical obstruction of the NPC since M9 or Rev nuclear import mediated by transportin remained unaffected by HIC. The molecular recognition of NLSs by import receptors in the cytoplasm determines their nuclear import rate [33, 34]. Based on the evidence herein, we propose that the HIC I-mfa domain binds to Rev NLS and selectively prevents its recognition by importin β and subsequent nuclear import proteins. This is similar to I-κB, which interacts with and masks the NF-κB NLS domain, also preventing its nuclear import [35, 36]. To extend our study to an in vivo cellular context, we next demonstrated that co-expression of HIC and Rev in COS7 cells resulted in the cytoplasmic sequestration of Rev with a concomitant reduction in its nuclear accumulation and this was dependent on the HIC I-mfa domain. Then, using leptomycin B, we further indicated that HIC most likely inhibits Rev nuclear import rather than promoting its nuclear export. This is also consistent with our competitive nuclear import assay performed with COS7 cytosol.
It should be noted that we have employed models where HIC was overexpressed or knocked-down in vivo, or added in excess amounts when used with in vitro nuclear import assays, and that HIC inhibitory effects on Rev function or Rev nuclear import were dose dependent. Collectively, these observations suggest that HIC acts as a cellular competitor with importin β for binding the Rev NLS, and that the balance between HIC and importin β could influence the rate of Rev nuclear import. Interestingly, we and others have recently described that the expression of HIC is tightly regulated at the transcriptional level [37, 38]. Furthermore, compared to 293T where endogenous levels of HIC expression are sufficient to down-modulate HIV-1 Rev activity, HIC is highly expressed in PBMCs and more specifically in HIV-1 target cells, including CD4+ T-cells and monocytes . Additionally, it is also possible that post-transcriptional modification(s) could modulate the relative affinity of Rev NLS for HIC and importin β, and mediate Rev release from sites of sequestration in the cytoplasm. In this regard, Rev is a target of the protein kinase CK2, which phosphorylates Rev at Ser5, Ser8 and results in the conformational change of a region encompassing the NLS domain [39, 40]. In addition, Rev Lys33 is also a target of ubiquitination .
In the present study, we also attempted to characterise Rev nuclear import pathways within different cellular environments and employed the M9 M peptide, which was specifically designed to inhibit the transportin pathway. Remarkably, HIC and M9 M had opposite effects on Rev nuclear import in a cell type-specific fashion. As previously described by Hutten et al., M9 M inhibited Rev nuclear import in the presence of HeLa cytosolic extracts . Similarly, M9 M but not HIC, inhibited Rev nuclear import using U937 and THP-1 cellular extracts. These results further substantiate the view that transportin and not importin β acts as the major import receptor for Rev in HeLa, U937 and THP-1 cells. Nevertheless, when 293T, Jurkat or CEM cytosolic extracts were employed, HIC but not M9 M inhibited Rev nuclear import, revealing that importin β but not transportin is the dominant nuclear import pathway in these cells. Finally, using a CAT reporter gene assay, overexpression of HIC was shown to reduce Rev activity in 293T, Jurkat and CEM cells. These results were substantiated by siRNA knockdown of endogenous HIC, which remarkably increased Rev activities in 293T cells. In contrast, HIC had no significant effect on Rev function in HeLa, U937 and THP-1 cells. Collectively, these findings support the hypothesis that Rev nuclear import pathways is determined by the cellular context and that importin β and transportin alternate as major import receptors for Rev in a cell specific and mutually exclusive fashion. Remarkably, we also revealed here that while HIC displaces importin β from Rev NLS, Rev could bind simultaneously to both HIC and transportin, strongly suggesting that the molecular determinants for Rev binding to importin β and transportin are different. The multivalent nature of the Rev NLS for multiple import factors would enable Rev to exploit multiple import pathways and to adapt its nuclear trafficking strategy to different cellular environments.
Given the observed effects of HIC on both Tat and Rev localisation and functions, it would be of interest to correlate their respective use of specific nuclear import pathways and the cell specific HIC endogenous level of expression, with HIV-1 replication and to distinguish the effects of HIC-Tat and HIC-Rev interactions on HIV-1 life cycle by employing Tat- or Rev-independent viruses.
Finally, we also describe how HIC interacts and interferes with SV40T NLS nuclear import, which constitutes the archetype of import mediated by the importin α/β pathway. Interestingly, other reports describe interactions of HIC and I-mfa with basic regions. These include the Axin GSK-3 binding domain, and Cyclin T1 KRM, both of which have a high K/R residue content [26, 27]. It would be of interest to investigate if HIC could also modulate their nuclear import.
We have identified HIC as a novel cellular cofactor for the HIV-1 regulatory protein Rev. We propose that the intermolecular masking of Rev NLS by HIC by which HIC control of Rev nuclear import can contribute to the spatial control of its activity. We also show that Rev nuclear import is cell specific and alternatively mediated by transportin or importin β.
DNA constructs and plasmids
pCAGGS-HA-Rev was created by cloning Rev (aa 1-116) sequence into pCAGGS . pFLAG-HIC, pFLAG-HIC(2-144), pFLAG-HIC(144-246) and pDM128-RRE were described previously [25, 43]. pGEX-GST-SV40TNLS-GFP, pGEX-GST-importin β, pQE80-RanQ69L, pGEX-GST-YFP, GST-M9 M and His-tagged transportin and importin β vectors were described previously [32, 44, 45, 46, 47]. pGEX-GST-YFP-M9: M9 sequence encoding hnRNP A1 (aa268-305) was cloned into pGEX-GST-YFP . pGEX-GST-YFP-Rev (aa1-116), pGEX-GST-YFP-RevΔN1 (aa35-116), pGEX-GST-YFP-RevΔN2 (aa46-116) and pGEX-GST-YFP-RevNLS (aa35-46) were generated by cloning Rev relevant sequences into pGEX-GST-YFP (Figure 7A). pBAD/6×His-HIC was described previously .
Western-Blotting analysis was performed using BioTrace™PVDF (Pall Corporation) and the SNAP i.d. Protein Detection System (Millipore) according to the manufacturer's instructions. The following primary antibodies were employed: ANTI-MDFIC AB2 and ANTI-FLAG M2 (Sigma); anti_HA High Affinity 3F10 (Roche); Transportin 1 (D45), NTF97/Importin beta (3E9) and Tubulin antibodies (Abcam). The following secondary antibodies (GE Healthcare) were employed: ECL™Anti-mouse IgG and ECL™Anti-rabbit IgG.
Cell culture and Transfection
293T, HeLa and COS7 cell lines were maintained in Dulbecco's modified Eagle's medium (DMEM) with 0.3 gm/L of L-Glutamine (GIBCO) supplemented with 10% fetal calf serum and antibiotics. Jurkat, CEM, THP-1 and U937 cell lines were cultured in RPMI 1640 medium containing 10% fetal calf serum and supplemented with 0.3mg/L of L-Glutamine (GIBCO) and antibiotics. Transient DNA transfections were performed using FuGENE6 (Roche Diagnostics, Mannheim, Germany) according to the manufacturer's protocol. The total amount of DNA was equilibrated by addition of parent plasmid. Approximately 30,000 HeLa cells and 60,000 293T cells were reverse-transfected with siRNA (30 pmol) in individual wells of a 24 well plate using Lipofectamine™ RNAiMAX (Invitrogen) according to manufacturer's instructions. siRNAs were obtained from QIAGEN: MDFIC_3 (5'-GGAUUGUAGGAGUGGAAGATT-3'), MDFIC_5 (5'GGAGUGAGCUGGCUGGAAATT-3'), MDFIC_7 (5'-CAUGAGAUUUAGCAGACUATT-3') and luciferase GL2 siRNA (negative control). Quantitative real time RT-PCR analysis of HIC mRNA expression was performed 72 hours post-transfection as described before (Gu et al., 2009).
Jurkat, CEM, U937 and THP-1 cells were transfected by nucleofection with the Nucleofector device I from Amaxa Biosystems and using the Nucleofector Kit V or Nucleofector Kit C, according the manufacturor's instructions. Total amount of DNA was equilibrated by addition of parent plasmid.
Purification of Recombinant Proteins
Recombinant GST fusion proteins (GST-YFP Rev/RevΔN1/RevΔN2/RevNLS/M9, GST-SV40TNLS-GFP, GST- importin β, GST-M9M) and His-tagged proteins (RanQ69L, importin β, transportin, HIC) were produced as described previously [25, 32, 44, 45, 46, 47].
293T cells were transfected with 4 μg of pCAGGS-HA-Rev and/or 1 μg of pFLAG-HIC. 48 hours post-transfection, cells were lysed in 200 μl Native lysis buffer (TBS with 5 mM EDTA PH 8.0, 10 mM NaF, 1% Triton X-100, 10 mM DTT, protein inhibitors). Lysates were incubated with 20 μl of HA Matrix affinity (Roche) at 4°C, overnight and bound complexes were eluted in Leammli sample buffer. Western-Blotting was carried out using Anti-HA High Affinity 3F10 (Roche) or Anti-FLAG M2 monoclonal antibody (Sigma).
COS7 cells were transfected with 0.5 μg of pCAGGS-HA Rev, and/or 0.5 μg of pFLAG HIC, pFLAG-HIC(2-144), and/or pFLAG-HIC(144-246). 48 hours post-transfection, cells were fixed in 4% paraformaldehyde and permeabilised with 0.2% Triton X-100. Indirect immnunofluorescence was performed using Anti-HA High Affinity 3F10 (Roche), Anti-FLAG M2 monoclonal antibody (Sigma) as primary antibodies, and anti-rat Ig-Biotin (DAKO), anti-rabbit Ig (Molecular Probes), Streptavidin Texas Red 594 (Molecular Probes) and anti Mouse IgG1 FITC (Molecular Probes) as secondary antibodies. Stained cells were visualized using an Olympus BX50 fluorescent microscope and DP70 camera.
In vitro nuclear import assay
293T, HeLa, COS7, THP-1, U937, CEM and Jurkat cytosolic extractions were performed as described previously . In vitro nuclear import assays and control experiments were performed with HeLa cells as described previously . Competitive In vitro nuclear import assays were performed by using GST-YFP-Rev/Rev NLS/RevΔN1/RevΔN2, GST-YFP-M9, GST-SV40TNLS-GFP as nuclear import substrate. RRL, 293T, HeLa, COS7, U937, Jurkat, CEM and THP-1 cytosolic extracts, recombinant N-His importin β or transportin C-His were employed as source of nuclear import factors. The addition of competitive and increasing amount of recombinant 6ΧHis HIC (0.5 to 4 μg) or GST-M9 M (4 μg), were employed for the competition assay. Images were captured using using an Olympus BX50 fluorescent microscope and DP70 camera. Cytosolic extracts (10 μg) were also subjected to Western Blot analysis to monitor the endogenous expression levels of importin β and transportin.
In vitro "pull down" assay
GST pull down assays were performed as previously described . Competitive in vitro GST pull down experiments were performed by adding 4 μg of recombinant N-His importin β or transportin C-His and increasing amount (0.5 to 4 μg) of recombinant 6ΧHis HIC to 0.5 μg of GST-YFP-Rev/Rev NLS immobilized on glutathione sepharose (Amersham).
Chloramphenicol acetyltransferase (CAT)-ELISA assay
293T and HeLa cells were transfected with 0.1 μg of CAT reporter gene DM128-RRE combined with pCAGGS-HA-Rev (0.05 μg), pFLAG-HIC (1-4 μg) or parent plasmid. Jurkat, CEM, U937 and THP-1 cells were transfected with 1 μg of CAT reporter gene DM128-RRE combined with pCAGGS-HA-Rev (0.5 μg), pFLAG-HIC (4 μg) or parent plasmid. For all cell lines, pRL-TK (Promega) was also transfected to control for transfection efficiency. 24 h post-transfection, the CAT-ELISA assays (Roche) were performed according to the manufacturer's protocols. Alternatively, 24 hours following reverse-transfection of HIC siRNAs (30pmoles) or control Luciferase GL2 (30pmoles) siRNA, cells were transfected with DM128-RRE (0.2 μg), pRL-TK (0.02 μg) and pCAGGS-HA-Rev (0.02 μg) or its parent plasmid and CAT assay was performed 48 hours post-transfection. Cell lysates were also subjected to Western Blot analysis to monitor HIC and Rev expression levels.
We wish to thank M. Angela Nieto and José-Manuel Mingot for the His-tagged transportin and importin β expression vectors and Yuh Min Chook for the GST-M9 M expression vector. pDM128-RRE, from Bryan Cullen, were obtained through the AIDS Research and Reference Reagent Program, Division of AIDS, National Institute of Allergy and Infectious Diseases, National Institutes of Health. pGEX-SV40TNLS-GFP, pGEX-importin β, and pQE80-RanQ69L from S. Kose and N. Imamoto were obtained from RIKEN, Saitama, Japan. This work was supported by the National Virus Reference Laboratory (NVRL), University College Dublin, Dublin, Ireland.
- 7.Levin A, Rosenbluh J, Hayouka Z, Friedler A, Loyter A: Integration of HIV-1 DNA is regulated by interplay between viral rev and cellular LEDGF/p75 proteins. Mol Med. 16: 34-44.Google Scholar
- 25.Gautier VW, Sheehy N, Duffy M, Hashimoto K, Hall WW: Direct interaction of the human I-mfa domain-containing protein, HIC, with HIV-1 Tat results in cytoplasmic sequestration and control of Tat activity. Proc Natl Acad Sci USA. 2005, 102: 16362-16367. 10.1073/pnas.0503519102.PubMedCentralCrossRefPubMedGoogle Scholar
- 29.Kusano S, Eizuru Y: Human I-mfa domain proteins specifically interact with KSHV LANA and affect its regulation of Wnt signaling-dependent transcription. Biochem Biophys Res Commun. 2010Google Scholar
- 39.D'Agostino DM, Ferro T, Zotti L, Meggio F, Pinna LA, Chieco-Bianchi L, Ciminale V: Identification of a domain in human immunodeficiency virus type 1 rev that is required for functional activity and modulates association with subnuclear compartments containing splicing factor SC35. J Virol. 2000, 74: 11899-11910.PubMedCentralCrossRefPubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.