|Home | About | Journals | Submit | Contact Us | Français|
Recent studies have identified host cell factors that regulate early stages of HIV-1 infection including viral cDNA synthesis and orientation of the HIV-1 capsid (CA) core toward the nuclear envelope, but it remains unclear how viral DNA is imported through the nuclear pore and guided to the host chromosomal DNA. Here, we demonstrate that N-terminally truncated POM121C, a component of the nuclear pore complex, blocks HIV-1 infection. This truncated protein is predominantly localized in the cytoplasm, does not bind to CA, does not affect viral cDNA synthesis, reduces the formation of 2-LTR and diminished the amount of integrated proviral DNA. Studies with an HIV-1-murine leukemia virus (MLV) chimeric virus carrying the MLV-derived Gag revealed that Gag is a determinant of this inhibition. Intriguingly, mutational studies have revealed that the blockade by N-terminally-truncated POM121C is closely linked to its binding to importin-β/karyopherin subunit beta 1 (KPNB1). These results indicate that N-terminally-truncated POM121C inhibits HIV-1 infection after completion of reverse transcription and before integration, and suggest an important role for KPNB1 in HIV-1 replication.
Recent genetic approaches have identified a considerable number of host cell factors regulating the replication of human immunodeficiency virus type 1 (HIV-1), including APOBEC3G, TNPO3, NUP358, NUP153, LEDGF/p75, TSG101 and BST-2/Tetherin [1–6]. Like the other lentiviruses, HIV-1 evolved to be able to infect non-dividing cells by delivering its reverse-transcribed DNA into the host cell nucleus through nuclear pore complexes (NPC) [7–13]. A large number of studies has indicated that the HIV-1 capsid (CA) is an essential determinant of HIV-1 infection of non-dividing cells. HIV-1 CA was necessary for infection of non-dividing cells as demonstrated by using chimeric viruses in which the Gag region of HIV-1 had been replaced by that of murine leukemia virus (MLV) . A carboxy-terminally truncated form of mouse cleavage and polyadenylation specificity factor subunit 6 (mCPSF6-358) inhibited HIV-1 infection by preventing nuclear import, and HIV-1 harboring the N74D mutation of CA evaded mCPSF6-358-mediated inhibition, taking an alternative route for nuclear entry . The CA determines which NPC proteins, such as nucleoporin NUP358 and NUP153 or the nuclear transport factor, Transportin 3 (TNPO3), will be employed for crossing the nuclear envelope [16–18].
Nuclear import of host cell proteins is generally mediated by importin family members. Importin-α family proteins mainly recognize nuclear-localizing signals (NLSs) of cargo proteins and importin-βfamily proteins are usually necessary for docking to the NPC and translocation through the nuclear pore . Importin-β/karyopherin subunit beta 1 (KPNB1) protein was shown to bind to and import the HIV-1 proteins Tat and Rev independently of Importin-α . Recent reports showed the TNPO3, a member of the importin-β protein family , prevents cleavage and polyadenylation specificity factor subunit 6 (CPSF6)-mediated capsid stabilization in the cytoplasm and contributes to efficient nuclear entry of the CPSF6-associated HIV-1 preintegration complex (PIC) [12, 21]. There have been conflicting reports about the role of an importin-β family member importin-7 in HIV-1 nuclear import [22, 23]. Thus, the roles of multiple host cell factors in HIV-1 nuclear import remain to be further studied.
In the present study, we employed cDNA expression screening to identify an N-terminally truncated form of POM121C which strongly inhibited HIV-1 replication. POM121C is a nucleoporin located in the middle of the nuclear pores which plays an essential role in the formation of NPC . Here, we describe how this truncated protein interferes with HIV-1 replication.
HEK293 (Invitrogen Corp., Carlsbad, CA), HEK293T (Invitrogen), HeLa (ATCC), and Plat-E packaging cells  were propagated in Dulbecco’s modified Eagle’s medium containing 10% fetal bovine serum (FBS) and penicillin-streptomycin. The human T cell lines MT4C5 (kindly provided by Dr. Tetsuro Matano, National Institute of Infectious Diseases, Japan) and Jurkat (kindly provided by Dr. Klaus Strebel, National Institutes of Health, USA) were maintained in complete RPMI 1640 medium supplemented with 10% FBS and penicillin-streptomycin. Phytohemagglutinin (PHA)-activated PBMCs (PHA-PBMCs) (kindly provided by Dr. Rika Ann Furuta, Japanese Red Cross Osaka Blood Center) were cultured in RPMI 1640 containing 10% FBS, penicillin/ streptomycin, and 100 U IL-2 per ml.
HEK293T cells cultured in a 10-cm dish were cotransfected with 8 μg of pNL4-3luc (env-/nef-) , pNL4-3tk , pNL4-3luc CA-N74D , pNL4-3luc IN-D116G , pLai3ΔenvLuc2 , or pMHIV-mMA12CA-Luc  plus 2 μg of pHCMV-G (VSV-G) using FuGENE 6 (Promega Corp, Madison, WI) according to the manufacturer’s instructions. Virus supernatant was harvested 48 h post-transfection and filtered through a 0.45 μm-pore syringe filter. Replication-competent HIV-1NL4-3 or NL4-3luc-RC virus stocks were prepared by transfection of HeLa cells as previously reported . Viruses were harvested 48 h post-transfection and filtered through a 0.45 μm-pore syringe filter. Titers of the virus stocks were quantified by HIV-1 CA (p24) enzyme-linked immunosorbent assay (ELISA) (ZeptMetrix Corporation, Buffalo, NY) and by determining their reverse transcriptase (RT) activity. For construction of the cDNA expression library, cDNAs were synthesized from poly(A)-RNA of Jurkat cells with random hexamer primers using the SuperScript Choice system (Invitrogen Corp., Calsbad, CA) according to the manufacturer’s instructions. The synthesized cDNAs were inserted between the BstXI sites of pMRX  using BstXI adaptors (Invitrogen Corp., Carlsbad, CA), generating a retroviral cDNA expression library. For production of the retrovirus reporter vector, Plat-E cells were cotransfected with a retrovirus transfer plasmid: pMX-luc  and VSV-G using FuGENE 6 transfection reagent. Culture supernatant of Plat-E cells was collected 60 h after transfection and filtered through a 0.45 μm-pore syringe filter.
POM121C (614–987) was amplified by PCR using total DNA isolated from surviving cell pools. The sequences of the primers used for amplification are shown in Table 1. The amplified cDNA was cloned into the pCR2.1-TOPO vector (Life Technologies Corp., Carlsbad, CA), and sequenced. The cDNA was then subcloned into the pMRX-HA-ires-puro vector . Full-length POM121C (2–987) (GenBank accession number NM_001099415) was amplified by RT-PCR using total RNA isolated from HEK293 cells as the template. RT-PCR was performed with the PrimeScript II high-fidelity one-step RT-PCR kit (TaKaRa Bio., Inc., Shiga, Japan). The amplified RT-PCR products were inserted into pCR2.1 and sequenced. The cDNA was inserted into the BamH I and Not I sites of pMRX-HA-ires-puro. To generate the POM121C-mutant expression constructs, the cDNAs of POM121C mutants were amplified by PCR using pMRX-HA-POM121C (614–987) as the template. The amplified PCR products were inserted into the pCR2.1-TOPO vector and sequenced. The cDNAs were inserted into the BamH I and Not I sites of pMRX-HA-ires-puro. The resultant plasmids are referred to as pMRX-HA-POM121C (2–987), pMRX-HA-POM121C (614–987), pMRX-HA-POM121C (614–873), pMRX-HA-POM121C (686–987), pMRX-HA-POM121C (686–873), pMRX-HA-POM121C (801–987), pMRX-HA-POM121C (614–973) and pMRX-HA-POM121C (614–987: Δhelix). For construction of Strep-tag II fusion-GFP, -Cyclophilin A (CYPA) and -POM121C (614–987) protein expression vectors, the cDNAs of GFP and POM121C (614–987) were amplified by PCR using pMax-GFP (Lonza Japan Ltd., Tokyo, Japan), pcDNA-HA-CypA  or pMEX-HA-POM121C (614–987) as template. The amplified products were inserted into the Kpn I and Xho I sites of pCAG-One-STrEP-FLAG-tag (OSF) , and sequenced. The resultant plasmids are referred to as pCAG-OSF-GFP, pCAG-OSF-CypA and pCAG-OSF-POM121C (614–987). The GST gene was amplified by PCR using pGEX-4T-3 (GE Healthcare Life Sciences, Pittsburgh, PA) as the template. For generating expression vectors for GST-tagged proteins, a two-step PCR was carried out. The GST gene fragment and POM121C-mutant fragments were linked and amplified by fusion PCR. After cloning and confirmation of the nucleotide sequence, fused DNA fragments were excised by digestion with Bgl II and Not I and then inserted into the BamH I and Not I sites of pMRX-HA-ires-puro. The resultant plasmids are referred to as pMRX-HA-GST, pMRX-HA-GST-POM121C (614–987), pMRX-HA-GST-POM121C (801–987), pMRX-HA-GST-POM121C (801–973) and pMRX-HA-GST-POM121C (801–987: Δhelix). pMX-luc was constructed by inserting an Sma I-Hind III DNA fragment carrying the luciferase gene of Igκcona-luc  into the Hpa I and Hind III sites of the pMX vector . To generate the HIV-1NL4-3 vector carrying the firefly luciferase gene (pNL4-3luc-RC), fragments of the env gene and luciferase gene were fused by fusion PCR. The amplified fragments were extracted after digestion with Hpa I and Not I and inserted into the Hpa I and Not I sites of pNL-E , generating pNL4-3luc-RC. The coding sequence of HA-tagged POM121C (614–987) was amplified by PCR using pMRX-HA-POM121C (614–973) as the template. The PCR product was cloned into the pENTR/D-TOPO (Life Technologies Corp., Carlsbad, CA) vector. The coding sequence of HA-tagged POM121C (614–987) was transferred to the lentivirus destination vector CSII-EF-IB-RfA  using Gateway LR Clonase II Enzyme Mix (Life Technologies Corp., Carlsbad, CA). The resultant plasmid is referred to as pCSII-EF-IB-POM121C (614–987).
HEK293, HeLa, Jurkat and MT4C5 cells were transduced with retrovirus vectors for POM121C mutants and puromycin resistance and express POM121C-mutants. Cells stably expressing POM121C-mutants were obtained after selection with 4 μg/ml of puromycin for HEK293 and HeLa or 2 μg/ml of puromycin for Jurkat and MT4C5. PHA-stimulated PBMCs were transduced with lentivirus vectors for POM121C mutants and Blasticidin S resistance, and were cultured in the presence of 6 μg/ml of Blasticidin S.
HEK293 and HeLa cells (3 × 105) were infected for 24 h with 10 ng (p24) of VSV-G-pseudotyped NL4-3luc (VSV-G/NL4-3luc) or VSV-G-pseudotyped MLV-luc (VSV-G/MLV-luc) normalized by reverse transcriptase (RT) counts corresponding to 10 ng (p24) of VSV-G/NL4-3luc in 12-well plates. Cells were then harvested and lysed 24 h post-infection. MT4C5, Jurkat and PHA-PBMCs (5 × 105) were infected for 24 h with 10 ng (p24) of VSV-G/NL4-3luc in 24-well plates, harvested and lysed 24 h post-infection. MT4C5 cells (2 × 105) were infected for 72 h with 10 ng (p24) of NL4-3luc-RC in 24-well plates. Cells were then harvested and lysed 72 h post-infection. To determine luciferase activity, cell lysates were mixed with luciferase substrate and light emission was measured using a GloMax-Multi Detection System (Promega Corp., Madison, WI).
MT4C5 cells (1 × 105) were exposed to HIV-1NL4-3 stock containing 10 pg of p24. Virus production was monitored for 8 days post-infection by measuring RT activity in the culture supernatants. Mean values from three independent experiments are shown.
Whole-cell lysates were prepared as follows: cells were washed twice with phosphate-buffered saline (PBS) (-), suspended in PBS(-) (500 μl per 1 × 107 cells) and mixed with an equal volume of 2 × sample buffer (4% sodium dodecyl sulfate, 125 mM Tris-HCl, pH 6.8, 10% 2-mercaptoethanol, 10% glycerol, and 0.002% bromphenol blue). Proteins were solubilized by heating for 5 min at 95°C. Samples were subjected to SDS-PAGE, transferred to PVDF membranes, and reacted with rat monoclonal antibody to HA (#11867423001, Sigma-Aldrich Co., St. Louis, MO), mouse monoclonal antibody to FLAG [29–22381, Wako Pure Chemical Industries, Ltd., Osaka, Japan], mouse monoclonal antibody to HIV-1 p24 (#ab9071, Abcam, Inc., Cambridge, MA), goat polyclonal antibody to HIV-1 gp120 (#ab21179, Abcam, Inc., Cambridge, MA), HIV-1-positive pooled serum from infected individuals (subtype B) , mouse monoclonal antibody to α-tubulin (#T6199, Sigma-Aldrich Co, St. Louis, MO), rabbit polyclonal antibody to CYPA (#BML-SA296, Enzo Life Sciences, Inc., Farmingdale, NY), or mouse monoclonal antibody to KPNB1 (#GTX22811, GeneTex, Inc., Irvine, CA). Membranes were then incubated with horseradish peroxidase-conjugated secondary antibody (#NA934 for anti-rabbit IgG, #NA931 for ant-mouse IgG, #NA933 for anti-human IgG, Amersham Biosciences, Piscataway, NJ), and proteins were visualized by Western Lightning Plus-ECL (PerkinElmer, Waltham, MA) or enhanced chemiluminescence (Pierce Biotechnology, Rockford, IL).
HeLa cells (1 × 104) were cultured for 48 h at 37°C in Nunc Lab-Tek II 8-well glass Chamber Slides (Thermo Fisher Scientific, Waltham, MA). Cells were then fixed with 4% paraformaldehyde for 10 min at room temperature followed by two washes in PBS(-). Fixed cells were permeabilized for 20 min at room temperature with 0.5% Triton X-100 in PBS(-) prior to incubation with antibodies. For staining, cells were incubated at 4°C overnight with primary antibodies diluted appropriately in 20% ImmunoBlock (DS Pharma Biomedical Co., Ltd., Osaka, Japan)/PBS(-). Cells were washed twice with PBS(-) and incubated with Alexa Fluor 594-conjugated secondary antibodies (Thermo Fisher Scientific, Waltham, MA) and Hoechst33342 (Thermo Fisher Scientific, Waltham, MA) for 2 h at room temperature. Samples were then washed twice with PBS(-) and mounted with VECTASHIELD (Vector Laboratories, inc., Burlingame, CA). Images were acquired by FV10i Confocal Laser Scanning Microscopy (OLYMPUS Corp., Tokyo, Japan)
HEK293, control vector-expressing HEK293  or POM121C (614–987)-expressing HEK293 (POM121C [614–987]) cells (2 × 105) in a 12-well plate were co-transfected with 0.45 μg of pNL4-3luc together with 0.05 μg of pGL4.84-EF1α-hRlucCP (Renilla-luc)  using FuGENE6 according to the manufacturer’s instructions. Cells were harvested and lysed 48 h post-transfection. The luciferase activity was measured using the GloMax-Multi Detection System (Promega Corp, Madison, WI).
Measurement of reverse-transcribed viral cDNA in HIV-1-infected cells was performed as described previously . Prior to infection, virus solution was treated with 100 U of RNase-free DNase I (Sigma-Aldrich Co, St. Louis, MO) in the presence of 10 mM MgCl2 for 30 min at 37°C. For infection, 7 × 105 target EV-control HEK293 or POM121C (614–987)-expressing HEK293 cells were incubated for 12 or 24 h with VSV-G/NL4-3luc virus stock containing 23 ng of p24. Total cellular DNA was extracted using the DNeasy Blood & Tissue kit (Qiagen, Inc., Valencia, CA). VSV-G/NL4-3luc inactivated by incubation at 65°C for 30 min was used as a negative control. Primers U5-gag/F2 (5´-GTAGTGTGTGCCCGTCTGTTG-3´) (nt 553 to 573) and U5-gag/R2 (5´-CAAGCCGAGTCCTGCGT-3´) (nt 689 to 705) and probe U5-gag/probe2 (5´-FAM-TGGCGCCCGAACAGGGACTT-TAMRA-3´) (nt 636 to 655) were used for amplification and quantification of the U5-gag region of the HIV-1 late reverse transcription product. Primers 5’-CCCTCAGACC CTTTTAGTCAGTG-3’ (nt 9668–9690) and 5’-TGGTGTGTAGTTCTGCCAATCA-3’ (nt 77–98) and probe 5’-FAM-TGTGGATCTACCACACACAAGGCTACTTCC-TAMRA-3’ (nt 46–75) were used for amplification of the 2-LTR circle from the HIV-1NL4-3 cDNA. The integrated viral DNA was measured as described previously . Briefly, the first PCR was carried out with an Alu-sequence-specific sense primer (Alu-HIV) and an antisense primer recognizing the LTR/gag region of the HIV sequence (M661). The first-round PCR samples were diluted with deionized distilled water by a factor of 10, and 5 μL was used as the template for the real-time PCR assay for measuring R/U5 DNA using M667, AA55 and HIV-FAM. For standardization, β-globin cDNA was quantified as described previously . Real-time PCR was carried out with the StepOnePlus Real-Time PCR system (Applied Biosystems, Carlsbad, CA). The ratios of viral cDNA toβ-globin cDNA levels were calculated.
Envelope-stripped HIV-1 virions (naked capsid cores) were prepared as described previously . Briefly, HIV-1-containing culture supernatants were prepared by transiently transfecting HeLa cells with pNL4-3 using LipofectAMINE LTX PLUS (Invitrogen Corp., Carlsbad, CA). Two ml of 20% sucrose solution was placed at the bottom of SW55 centrifuge tubes and overlaid with 3 ml of the HIV-1-containing culture supernatant described above. Samples were then centrifuged for 60 min at 35,000 rpm at 4°C. Particulate HIV-1 was resuspended with PBS (-) containing a protease inhibitor cocktail (Nacalai Tesque, Inc., Kyoto, Japan). This suspension was loaded onto the top of a discontinuous sucrose density gradient composed of 1.0 ml 30% sucrose solution at the bottom of SW55 centrifuge tubes covered or not covered by 1.0 ml 0.1% Triton X-100 in 10% sucrose solution and then centrifuged in an SW55Ti rotor for 120 min at 35,000 rpm at 4°C. Particulate CA proteins were subjected to Western blotting with anti-HIV-1 env (gp120) or HIV-1-positive pooled serum from infected individuals (subtype B) to verify removal of HIV-1 env and MA proteins.
Naked HIV-1 capsid cores isolated by ultracentrifugation were absorbed onto Formvar-coated copper grids, and stained with 2% phosphotungstic acid solution. The images were recorded with a Tecnai F20 transmission electron microscope (FEI company, Hillsboro, OR) at 200kV.
HEK293 cells were transfected with pCAG-OSF, pCAG-OSF-GFP, pCAG-OSF-POM121C (614–987) or pCAG-OSF-CypA, harvested 48 hr post-transfection and lysed in a 7 ml-Dounce homogenizer. Cell extracts were incubated with Strep-Tactin Sepharose for 2 h at 4°C. Purified Strep-tagged protein complexes were incubated with naked HIV-1 capsid cores for 2 h at 4°C. After extensive washing, Strep-tagged protein complexes were released by boiling in SDS-PAGE loading buffer and the proteins were analyzed by 12% SDS-PAGE by Western blotting using anti-FLAG or anti-p24 antibody.
Parental, EV-control or POM121C mutant-expressing HEK293 cells (7 × 107) were harvested, resuspended in 1 ml of IP lysis buffer (25 mM Tris-HCl [pH 7.4], 150 mM NaCl, 1% NP40, 5% glycerol and protease inhibitor cocktail [Nacalai Tesque, Inc., Kyoto, Japan]) and incubated on ice for 30 min. The cell lysates were cleared by centrifugation at 8,000 × g for 3 minutes at 4°C. Part of the cell extract was reserved as input. Glutathione-Sepharose 4B (GE Healthcare Life Sciences, Pittsburgh, PA) was added to each cell extract and the mixtures were rotated for 2 h at 4°C. The beads were then washed five times with IP wash buffer (20 mM Tris [pH 7.4], 150 mM NaCl, 0.1% Triton X-100) and bound proteins were eluted with SDS sample buffer.
Proteins were identified by liquid chromatography nano-electron spray ionization tandem mass spectrometry (LC-nESI-MS/MS) (QTRAP 5500 LC/MS/MS System, AB SCIEX, Concord, ON, Canada). Proteins excised from the gels were analyzed by LC-nESIMS/MS after separation on a HiQ sil C18W-3P column (0.1 mmΦ × 100 mm, KYA TECH Corporation, Tokyo, Japan). The proteins were eluted at a flow rate of 300 nL/min using water containing 0.1% (v/v) formic acid as the eluent A and acetonitrile containing 0.1% (v/v) formic acid as the eluent B with a linear gradient from 5% B to 45% B in 70 min. The data were analyzed using ProteinPilot™ software (Version 3, Applied Biosystems, Warrington, UK). Protein candidates were selected from those that had more than one distinct peptides with at least 95% confidence.
All data were obtained from at least three independent experiments. The average values are presented with error bars indicating the standard deviation (SD) and the statistical significance was analyzed using one-way analysis of variance (ANOVA) with Dunnett’s multiple comparison test, or Student’s t-test. All the statistical analyses were performed using Prism 6 software (GraphPad Software, Inc). P values below 0.05 (P<0.05, *; P<0.01, **; P<0.001,***) were considered significant. Unpaired two-tailed Student’s t-test was used for the data shown in Figs 1C (panel PBMCs) and and7C7C to test whether the means of the two groups were significantly different. One-way analysis of variance (ANOVA) with Dunnett’s multiple comparison test was used for the data shown in Figs 1C (panels HEK293 and Jurkat), 1D, 1E, ,2A,2A, 3B, 3D, 4C, 4E, 5B, 5C, 6D, 6G and and7A7A to determine whether the means of multiple groups were significantly different from a single group.
To identify proteins that interfere with HIV-1 infection, we employed a functional screen using a cDNA expression library. The details of the method of screening were essentially the same as those previously reported  except that complementary DNAs were synthesized from poly(A)-RNA of the Jurkat human T-cell line with random hexamer primers. In brief, HEK293 cells transduced with a cDNA expression library in a murine leukemia virus (MLV) vector were challenged with vesicular stomatitis virus glycoprotein (VSV-G)-pseudotyped recombinant HIV-1 capable of expressing herpes simplex virus thymidine kinase (HIV-1-TK)  followed by selection of uninfected cells in the presence of ganciclovir (GCV). After several rounds of HIV-1-TK infection and subsequent GCV selection, we characterized surviving cell clones resistant to HIV-1 infection and obtained a cDNA that encodes an N-terminally truncated form of POM121C (POM121C [614–987]) (Fig 1A and 1B). Stable expression of POM121C (614–987) in HEK293 cells with an influenza virus hemagglutinin (HA) tag, resulted in potent suppression of single-round infection of these cells with VSVG-pseudotyped NL4-3luc reporter virus (Fig 1C, left panel). Similar inhibition of HIV-1 infection was observed in the Jurkat cell line stably expressing POM121C (614–987) (Fig 1C, middle panel). Transduction of phytohemagglutinin/interleukin-2-stimulated peripheral blood mononuclear cells (PBMCs) with a lentivirus vector expressing HA-tagged POM121C (614–987) also resulted in a marked resistance to VSV-G-pseudotyped NL4-3luc infection (Fig 1C, right panel). Stable expression of POM121C (614–987) in MT4C5 cells markedly reduced the infectivity of VSV-G-pseudotyped HIV-1 (Fig 1D, left panel), replication-competent HIV-1 reporter virus, NL4-3luc-RC (Fig 1D, middle panel) and NL4-3 in spreading infection assays (Fig 1D, right panel). Neither the surface expression of CD4 and CXCR4 nor cell growth was affected by stable POM121C (614–987) expression (data not shown). However, stable expression of HA-tagged full-length POM121C in HeLa cells did not significantly influence VSV-G-pseudotyped HIV-1 infection (Fig 1E, right panel). A weaker expression of full-length POM121C (Fig 1E, left panel) may partly explain this lack of inhibition, but immuno-fluorescence studies revealed that the POM121C (614–987) was localized predominantly in the cytoplasm of HeLa cells, whereas full-length POM121C was found in the nucleus (Fig 1F). These results suggest that POM121C (614–987) expressed in the cytoplasm reduced the infectivity of HIV-1 in human cells.
Having shown that POM121C (614–987) restricts single-round and spreading HIV-1 infection, we next addressed which stage of HIV-1 replication is targeted by this protein. Viral cDNA synthesis was assessed by quantitative polymerase chain reaction (qPCR) studies in HEK293 cells expressing or not expressing POM121C (614–987) (Fig 2A). Following infection with VSV-G-pseudotyped NL4-3luc, the late reverse transcription (RT) product (Fig 2A, top panel) in cells expressing POM121C (614–987) was produced similarly to control cells up to 24 hr after infection. Notably, expression of POM121C (614–987) resulted in a marked decrease of the amount of integrated provirus, consistent with the results obtained with reporter virus infection (Fig 2A, middle panel). POM121C (614–987) also moderately reduced production of the 2-LTR form that accumulates exclusively in the nucleus and is therefore recognized as a marker of nuclear import of viral cDNA (Fig 2A, bottom panel).
The moderate and marked reduction in the 2-LTR and integrated provirus forms, respectively, in the presence of POM121C (614–987) implies possible blockade by POM121C (614–987) at the nuclear import, nuclear trafficking, or integration stage. Recent studies indicate that HIV-1 nuclear entry is regulated by viral capsid (CA) [14, 42]. To investigate potential interactions between POMC121C (614–987) and the viral core in HIV-1-infected cells, we performed HIV-1 CA core binding assays with envelope-stripped HIV-1 virions. The integrity of the prepared naked capsid core was verified biochemically by Western blotting (Fig 2B) showing loss of the envelope and matrix proteins after stripping, and confirmed morphologically by electron microscopy (Fig 2C) showing the cone-shaped viral core. Pull-down assays with naked HIV-1 cores and purified OSF-tagged proteins revealed that POM121C (614–987) did not interact significantly with the capsid core (Fig 2D, panel pull down). In contrast, the association of the HIV-1 core with Cyclophilin A (CYPA), known to be an HIV-1 core-binding protein , was clearly demonstrated (Fig 2D, panel pull down). These results indicate that POM121C (614–987) inhibits HIV-1 infection after reverse transcription without binding to HIV-1 core.
To gain more insight into blockade of infectivity after completion of RT, HEK293 cells expressing POM121C (614–987) or mCPSF6-358, which had been reported to block nuclear import of viral cDNA , were infected with VSV-G-pseudotyped NL4-3luc carrying the N74D mutation in the CA or the D116G mutation in the integrase (IN). The N74D mutation significantly reduces the binding affinity of mCPSF6-358 to HIV-1 CA, resulting in viral escape from restriction by mCPSF6-358 , while the D116G mutation disables HIV-1 for integration within host chromatin without affecting RT or nuclear entry . We stably expressed POM121C (614–987) or mCPSF6-358 in HEK293 cells for single-round reporter virus infection assays. Expression of POM121C (614–987) or mCPSF6-358 was verified by Western blotting (Fig 3A). As shown in Fig 3B, mCPSF6-358 weakly affected N74D HIV-1 infection, consistent with a previous report . POM121C (614–987) potently or moderately inhibited infection with N74D or D116G mutant viruses, respectively (Fig 3B, panels CA N74D and IN D116G). Kinetic studies of viral cDNAs following infection of cells expressing POM121C (614–987) with the N74D or D116G viruses showed that the late RT form was produced at levels similar to those of control cells (Fig 3C, upper panels), while accumulation of the 2-LTR form was decreased (Fig 3C, lower panels). This suggests that the action of POM121C (614–987) does not depend on the route of nuclear import of HIV-1 cDNA. The moderate inhibition of D116G virus infection by POM121C (614–987) suggests that POM121C (614–987) not only blocks nuclear import but also affects integration of HIV-1 providing that the D116G virus lacks the integrase activity. In this case, the overall inhibitory effects of POM121C (614–987) on this mutant virus may be proportionally alleviated. Additionally, overexpression of POM121C (614–987) did not inhibit the infectivity of the VSV-G-pseudotyped murine leukemia virus (MLV)-based vector, suggesting that POM121C (614–987) expressed in the cytoplasm has no effect on gamma retrovirus infection (Fig 3D, middle panel). A previous study had shown that CA is a pivotal determinant of retrovirus infection in non-dividing cells, using an HIV-1-based chimeric virus MHIV-mMA12CA, in which the entire Matrix (MA) and CA proteins were replaced by those of MLV . Having shown that POM121C (614–987) does not inhibit MLV infection in dividing cells, we next investigated whether it inhibits infection with MHIV-mMA12CA, which should not encounter inhibition of nuclear entry in dividing cells. Indeed, POM121C (614–987) did not significantly inhibit MHIV-mMA12CA infection in dividing cells (Fig 3D, right panel). These results indicate that Gag is a genetic determinant of POM121C (614–987)-mediated HIV-1 restriction although POM121C (614–987) did not seem to interact with the capsid core (Fig 2D).
POM121C (614–987) contains FG repeat motifs like other nucleoporins and a putative short α–helix motif at the extreme C-terminus. To determine the functional importance of these motifs, a series of truncation mutants was prepared and tested for their anti-HIV-1 activity in HEK293 cells (Fig 4A). Single-round infection studies revealed that mutants containing the α-helix structure (POM121C [686–987] and POM121C [801–987]) efficiently inhibited HIV-1 infection, although POM121C (801–987) was poorly expressed (Fig 4B and 4C). To assess the contribution of the C-terminal α-helix structure in POM121C (614–987) to the prevention of HIV-1 infection, we tested mutants with a small truncation or deletion of the α-helix motif of POM121C (614–987), POM121C (614–973) and POM121C (614–987: Δhelix). The deletion of the α–helix motif did not essentially alter subcellular localization (data not shown). Both α-helix motif mutants were expressed at similar levels to POM121C (614–987), but had lost inhibitory activity (Fig 4D and 4E). These results indicate that the α-helix motif plays an essential role in POM121C (614–987)-mediated HIV-1 restriction.
We next sought to determine how POM121C (614–987) interferes with HIV-1 infection. Because we failed to co-immunoprecipitate any viral protein with POM121C (614–987), a mammalian expression vector for POM121C (801–987) fused to the GST protein (GST-POM121C [801–987]) was constructed (Fig 5A and 5B). This fusion protein efficiently inhibited HIV-1 infection in HEK293 cells (Fig 5B, lower panel), did not affect the synthesis of the late RT product and reduced the amount of the 2-LTR form of viral cDNA in the same manner as POM121C (614–987) (Fig 5C). GST pull-down experiments using HEK293 cells stably expressing GST-POM121C (801–987) reproducibly demonstrated that several proteins specifically co-precipitated with GST-POM121C (801–987) (Fig 5D). Analyses of the isolated protein bands by mass spectrometry revealed that the most prominent, band 3, contains the karyopherin, importin subunit beta-1 (KPNB1) (Table 2).
Immunoblotting studies revealed that KPNB1 specifically interacts with GST-POM121C (801–987) (Fig 6A, panel GST pull-down). To investigate the functional consequences of POM121C (801–987) binding to KPNB1, truncation or deletion mutants of GST-POM121C (614–987) were stably expressed in HEK293 cells (Fig 6B and 6C) and tested for HIV-1 inhibition and KPNB1 binding (Fig 6D and 6E). As expected from the mutational studies of POM121C (614–987), two GST fusion proteins lacking the C-terminal α-helix motif were unable to inhibit HIV-1 infection, whereas those containing the C-terminal part of the FG repeat motifs and the α-helix motif did block HIV-1 in the single-round infection assay (Fig 6D). The binding of these mutants to KPNB1 correlated with the inhibition of HIV-1 infection (Fig 6E, panel GST pull-down). GST proteins linked to the extreme C-terminus of POM121C containing the α-helix motif but not the FG repeat motifs or simply to the α-helix motif failed to inhibit HIV-1 infection or to interact with KPNB1 (Fig 6F, 6G and 6H), suggesting that the α-helix motif alone is not sufficient for HIV-1 blockade or binding to KPNB1. These results suggest that POM121C (614–987)-mediated HIV-1 inhibition is closely linked to its binding to KPNB1.
Transient transfection with the proviral plasmid pNL4-3luc allows for bypassing the replication processes up to the integration stage and mimics the expression of viral genes from the integrated HIV-1 provirus. EV-control- or POM121C (614–987)-expressing HEK293 cells were transfected with the NL4-3luc provirus plasmid which has the luciferase gene in the nef position. This resulted in similar levels of reporter gene activity and newly-synthesized viral protein (Fig 7A and 7B). Likewise, POM121C (614–987) did not affect the release of HIV-1 CA protein from cells transfected with the pNL4-3 provirus plasmid (Fig 7C). These results suggest that POM121C (814–987) does not inhibit the late phase of the HIV-1 replicative cycle.
Based on similar late RT product synthesis 24 h post infection, we have shown here that POM121C (614–987) does not bind to the capsid (CA) core nor interfere with reverse transcription (Fig 2). After completing reverse transcription, the viral pre-integration complex (PIC) enters the nucleus by passing through the nuclear pore and approaches the host genome. One intriguing finding in the current study was that POM121C (614–987) markedly reduces the amount of the integrated provirus form, but the 2-LTR form is reduced to a lesser extent (Fig 2A). This is reminiscent of the difference in 2-LTR production levels and integrated forms previously shown in Nup153-depleted cells, in which HIV-1 nuclear entry was blocked . Perhaps POM121C (614–987) localized in the cytoplasm primarily impairs nuclear entry of HIV-1, and this may indirectly affect the proper intra-nuclear migration process and eventually limit integration. HIV-1 CA plays a pivotal role in nuclear import of viral DNA and this may potentially be a primary cause of POM121C (614–987)-mediated inhibition. Indeed, POM121C (614–987) inhibited HIV-1 but not MHIV-mMA12CA in dividing cells, suggesting that viral capsid defines the mode of nuclear entry and therefore determines whether POM121C (614–987) blocks infection or not. Accumulating evidence indicates that HIV-1 utilizes a particular pathway for nuclear entry that depends on CYPA, cleavage and polyadenylation-specific factor 6 (CPSF6) and NUP358, thereby evading innate immune sensors in the cytoplasm . The HIV-1 capsid mutant N74D, which is not targeted by CPSF6, can use a different pathway for nuclear entry that does not involve CPSF6 nor NUP358, but is recognized by cytoplasmic innate immune sensors . The present results that POM121C (614–987) inhibited infection with wild-type or N74D HIV-1 in a similar manner indicate that POM121C (614–987) suppresses both NUP358-dependent and -independent pathways (Fig 3B).
The production of similar amounts of the late RT product but marked decrease of the 2-LTR form of viral DNA in cells infected with the D116G mutant HIV-1 further support the idea that blockade by POM121C (614–987) occurs before integration. Taking these results together, we conclude that POM121C (614–987) inhibits HIV-1 replication after completion of reverse transcription and before the integration event.
To investigate how POM121C (614–987) blocks HIV-1 replication during nuclear import and intra-nuclear trafficking of HIV-1 DNA, we sought proteins binding POM121C (614–987) involved in HIV-1 inhibition. Binding of POM121C to viral components remains controversial. Fouchier et al.  reported yeast two-hybrid interaction of POM121 with Vpr, while Le Rouzic et al.  did not detect this interaction. Using immunoprecipitation and GST pull-down assays, we investigated whether POM121C interacts with Vpr but failed to detect any positive signal. Attention was therefore directed towards host cell factors that are targeted by POM121C (614–987). Because POM121C (801–987) proved to be sufficient to inhibit HIV-1 infection when fused with GST (Fig 4C), this fusion protein was used as bait. In this way, we have identified a number of candidates by MS at a good confidence level (Table 2). We selected KPNB1 for further study because this protein was found in the most prominent band 3 and had the highest number of matches. Importantly, the KPNB1 binding properties of POM121C mutants closely correlated with their level of HIV-1 blockade (Fig 6). The binding of POM121C mutants to KPNB1 is, however, not completely unexpected given that several previous studies demonstrated interactions between FG sequence repeats found in many nucleoporins and transport factors including KPNB1 [10, 47]. POM121C may belong to those nucleoporins that sequentially dock and undock with KPNB1/PIC complexes when HIV-1 PIC transits through nuclear pore complexes. Our attempts to determine whether endogenous full-length POM121C interacts with KPNB1 by means of immunoprecipitation experiments with two different antibodies specifically recognizing POM121C or KPNB1 were unsuccessful due in part to the limited expression of full-length POM121C in the nucleus. Thus, the possibility remains that endogenous POM121C physically interacts with KPNB1 under physiological conditions. To confirm the importance of endogenous POM121C or KPNB1 for HIV-1 replication, we had repeatedly attempted to knock-down POM121C or KPNB1 using shRNA, but unfortunately, depletion of these essential protein seriously affected the viability of cells, which made it impossible to assess HIV-1 infection. In future studies, it will be important to determine the exact role of POM121C and KPNB1 in the nuclear import and integration of HIV-1 DNA.
Accumulating evidence indicates that the HIV-1 capsid is an essential determinant of nuclear translocation of viral cDNA and that many host cell factors including karyopherins and nuclear pore complex components mediate this process. We have shown in this article that an N-terminally truncated form of POM121C potently inhibits HIV-1 infection after completion of reverse transcription and before integration. This inhibition closely correlates with its binding to KPNB1, suggesting an important role for this karyopherin in HIV-1 replication.
We thank the following researchers for providing invaluable materials: Dr. I.S.Y. Chen for pNL4-3luc (env -/ nef -), pHCMV-VSV-G and pCMV-ΔR8.2 packaging plasmids, Dr. H. Miyoshi (RIKEN, Japan) for CS-CDF-CG-PRE plasmid, Drs. M. Yamashita and M. Emerman for pLai3ΔenvLuc2 and pMHIV-mMA12CA-Luc. We thank Makiko Nawa (Laboratory of Cytometry and Proteome Research, Tokyo Medical and Dental University) for her help with TOF-Mass analysis.
Funded by grant 26460550 from the Ministry of Education, Culture, Sports, Science, and Technology to H.T., http://www.jsps.go.jp/j-grantsinaid/index.html, grant 21390136 from the Ministry of Education, Culture, Sports, Science, and Technology to S.Y., http://www.jsps.go.jp/j-grantsinaid/index.html, grant H22-AIDS-007 for Young Scientists of HIV/AIDS research from the Ministry of Health, Labor, and Welfare in Japan to H.T., http://www.mhlw.go.jp.
All relevant data are within the paper and its Supporting Information files.