Restriction of Feline Immunodeficiency Virus by Re
http://www.100md.com
病菌学杂志 2005年第24期
Molecular Medicine Program, Mayo Clinic College of Medicine, Rochester, Minnesota
Cystic Fibrosis Research and Treatment Center, University of North Carolina, Chapel Hill, North Carolina
ABSTRACT
The Ref1 and Lv1 postentry restrictions in human and monkey cells have been analyzed for lentiviruses in the primate and ungulate groups, but no data exist for the third (feline) group. We compared feline immunodeficiency virus (FIV) to other restricted (human immunodeficiency virus type 1 [HIV-1], equine infectious anemia virus [EIAV]) and unrestricted (NB-tropic murine leukemia virus [NB-MLV]) retroviruses across wide ranges of viral inputs in cells from multiple primate and nonprimate species. We also characterized restrictions conferred to permissive feline and canine cells engineered to express rhesus and human TRIM5 proteins and performed RNA interference (RNAi) against endogenous TRIM5. We find that expression of rhesus or human TRIM5 proteins in feline cells restricts FIV, impairing pseudotyped vector transduction and viral replication, but rhesus TRIM5 is more restricting than human TRIM5. Notably, however, canine cells did not support restriction by human TRIM5 and supported minimal restriction by rhesus TRIM5, suggesting that these proteins may not function autonomously or that a canine factor interferes. Stable RNAi knockdown of endogenous rhesus TRIM5 resulted in marked increases in FIV and HIV-1 infectivities while having no effect on NB-MLV. A panel of nonprimate cell lines varied widely in susceptibility to lentiviral vector transduction, but normalized FIV and HIV-1 vectors varied concordantly. In contrast, in human and monkey cells, relative restriction of FIV compared to HIV-1 varied from none to substantial, with the greatest relative infectivity deficit for FIV vectors observed in human T-cell lines. Endogenous and introduced TRIM5 restrictions of FIV could be titrated by coinfections with FIV, HIV-1, or EIAV virus-like particles. Arsenic trioxide had complex and TRIM5-independent enhancing effects on lentiviral but not NB-MLV infection. Implications for human gene therapy are discussed.
INTRODUCTION
Innate cellular defenses against retroviruses include early postentry restrictions to infection that are dependent upon the identity of the viral capsid protein and are mediated by dominantly acting, species-specific host factors. Murine leukemia virus (MLV) variants provide the classic model of this kind of restriction. N-tropic MLV (N-MLV) is restricted by the Friend virus susceptibility factor 1 b (Fv1b) allele in BALB/c mice but not in NIH Swiss mice, whereas B-MLV is restricted by the Fv1n allele in NIH Swiss mice but not in BALB/c mice. NB-MLV is restricted by neither allele. The n and b Fv1 alleles encode retroviral Gag-like proteins similar to murine endogenous retroviral Gag (5). The critical viral determinant maps to capsid residue 110 (7, 23, 51).
Human and simian cells display similar activities that restrict N-MLV but not B-MLV or NB-MLV (3, 4, 8, 11-13, 16, 48). Although the viral determinant of N-MLV restriction in nonmurine cells also maps to capsid residue 110 (47, 51), the mechanisms appear to differ. Fv1 acts after reverse transcription, whereas the human activity Ref1 and simian activity Lv1 block infection prior to or concurrent with reverse transcription (6). In addition to restricting N-MLV, Lv1 has been shown to restrict lentiviruses: human immunodeficiency virus type 1 (HIV-1) and type 2 (HIV-2) and equine infectious anemia virus (EIAV) as well as certain simian lentiviruses (12, 15, 16, 29, 52). These data are consistent with the inability of HIV-1 to productively infect Old World Monkeys. In contrast, the human cell activity of Ref1 restricts EIAV and N-MLV but does not restrict HIV-1, HIV-2, or simian immunodeficiency viruses (SIVs) (8, 12, 16).
An informative characteristic of Fv1, Lv1, and Ref1 is their saturability when cells are infected with large amounts of virus or vector, or coinfected with other restricted retroviruses or virus-like particles (VLPs). Titration of the restricting activity in this way requires only that the coinfecting particles be entry competent and contain mature capsids of any retrovirus that is also restricted; they need not contain viral genomes or enter via the same receptor pathway (1, 3, 4, 12, 52). Together, these compelling infectivity and cross-titration phenotypes have supported a model in which cytoplasmic factors can block infection at low viral inputs by interacting directly or indirectly with the viral capsid protein (6).
Recently, the tripartite motif (39) protein TRIM5 was identified as the major factor restricting HIV-1 infection in rhesus cells (49). Its properties appear to account for the major phenotypic features of Lv1 restriction (15, 49). In addition, orthologs of TRIM5 have subsequently been shown to mediate Ref1 (human TRIM5 [TRIM5hu]) as well as the broad restriction to HIV-1, N-MLV, EIAV, and SIVmac in African green monkey (AGM) cells (TRIM5agm) (15, 22, 34, 46, 57). Rhesus TRIM5 (TRIM5rh) expression in normally HIV-1-permissive HeLa cells confers strong restriction of HIV-1 but not SIVmac (49). Introduction of TRIM5rh into Crandell feline kidney cells (CrFK), which lack restriction factors (7, 11, 15, 51), confers potent restriction of HIV-1, SIVagm, EIAV, and N-MLV (15, 22, 41, 47). Human TRIM5hu expression results in a weak effect on HIV-1 and strong restriction of EIAV and N-MLV (22, 49). B-MLV and NB-MLV are not restricted by any TRIM5 alleles so far described, which has led to their use as negative controls experimentally (12).
Three groups of lentiviruses infect primates (e.g., HIV-1), ungulates (e.g., EIAV), and felines (feline immunodeficiency virus [FIV]) respectively. While Lv1 and Ref1 restriction patterns and TRIM5-mediated blocks to viruses in the first two lentiviral groups have been well described, no data have been reported for the third. FIV is the cause of a pandemic in multiple feline species. The virus is highly divergent at the nucleotide and protein levels from primate and ungulate lentiviruses, including the capsid (CA) region implicated in mediating Lv1/Ref1 restriction. Uniquely among the nonprimate lentiviruses, FIV causes CD4+ T-cell depletion and AIDS. Here we report a detailed analysis of FIV postentry restriction, which we carried out by characterizing Lv1 and Ref1 restrictions compared to other retroviruses that served as positive (HIV-1 and EIAV) and negative (NB-MLV) controls and by examining specific effects of TRIM5 cDNA expression and TRIM5-directed RNA interference (RNAi). We also studied the effects of arsenic trioxide, a mitochondrial membrane potential-disrupting agent that enhances HIV-1 vector transduction in some cells (54), apparently by attenuating postentry restrictions (1, 2, 22). Finally, we consider implications for use of FIV vectors (36, 40) in human gene therapy.
MATERIALS AND MATHODS
Cells. Cell lines were cultured in RPMI or Dulbecco's modified Eagle medium with 10% fetal calf serum. 293T human embryonic kidney cells were used as lentivirus and subgenomic lentivector producer cells. For NB-MLV production, the Moloney murine leukemia virus-based retrovirus producer cell line Phoenix A, a gift of G. Nolan, was used. 293T, HeLa (human epithelial adenocarcinoma), Crandell feline kidney (CrFK), HT1080 (human epithelial fibrosarcoma), RD (human rhabdomyosarcoma), Vero (African green monkey kidney), FRhK4 (fetal rhesus kidney), 3T3 (NIH Swiss murine embryo fibroblast), human leukemic T-cell (H9, Jurkat clone E6-1, MT-4, Molt-4 clone 8, and SupT1), MDBK (bovine kidney), and canine kidney (DK), skin fibroma (A72), and osteosarcoma (D17a) cell lines were used in transduction experiments.
Vector stocks. Transfections were done with the calcium phosphate transient transfection method in two-chamber cell factories (CF2) (26). Medium was changed 12 to 16 h later, and supernatants were collected 48 h thereafter, filtered through a 0.2-μm-pore-size filter, concentrated by ultracentrifugation over a sucrose cushion in a swinging bucket SW28 rotor at 25,000 rpm for 2 h, and aliquoted and frozen at –80°C. FIV-based lentiviral vectors were produced in 293T cells using a 5' U3-substituted system (26, 35, 36). Enhanced green fluorescent protein (eGFP) transfer vector GiNW was cotransfected with the pFP93 packaging construct (26) and vesicular stomatitis virus G protein (VSV-G) envelop construct pMD.G (60). The HIV-1 GFP vectors, which have the same internal promoter (cytomegalovirus [CMV]), were produced by transfection of 293T cells with HIV-1 transfer construct pHR · -hCMV-eGFP (59) containing gfp, HIV-1 packaging construct pCMVR8.91 (60), and pMD.G. For NB-MLV GFP vector, MLV transfer construct pLEGFP (Clontech, Palo Alto, CA) and pMD.G were cotransfected into Phoenix A cells. The EIAV GFP vector SIN6.1eGFPW and the EIAV packaging plasmid pEV53B have been described previously (31, 32). EIAV vectors were produced by transfection of pEV53B, SIN6.1eGFPW, and pCI-VSV-G (31) into 293T cells at a ratio of 1:1:0.6. The day following transfection, the medium was changed to include 10 mM sodium butyrate, and viral supernatants were collected 24 h later.
VLPs. To generate FIV VLPs, packaging construct pFP93 was cotransfected by calcium phosphate coprecipitation with pMD.G into 293T cells in CF2 chambers. For HIV-1 VLPs, HIV-1 packaging construct pCMVR8.91 was cotransfected with pMD.G. For NB-MLV VLPs, pMD.G was transfected into Phoenix A cells. Supernatants were collected 48 h after changing the medium, filtered through a 0.2-μm-pore-size filter, concentrated by ultracentrifugation over a sucrose cushion in a swinging bucket SW28 rotor at 25,000 rpm for 2 h, and then aliquoted and frozen at –80°C. EIAV VLPs were harvested from 293T cells in a manner similar to EIAV GFP vector following transfection of pEV53B and pCI-VSV-G.
Replicating virus stocks. Wild-type, replication-competent FIV was produced in 293T cells by transfection of pCT5, in which replacement of the 5' FIV U3 element by the human CMV immediate-early promoter enables human cell expression (35, 36). The medium was replaced 14 h after plasmid transfection, and supernatants were collected 32 h later. Supernatants were filtered (0.45 μM) and frozen in 500-μl aliquots. Cotransfection of pCT5 with pMD.G was used in the preparation of some stocks to boost the initial round infectivity for CrFK cell replication studies. Titers were determined for virus stocks on CrFK cells using a focal infectivity assay (38) as described previously (35).
Generation of stable TRIM5 allele-expressing cell lines. Stable cell lines expressing TRIM5 alleles were generated with MLV-based retroviral vectors kindly provided by J. Sodroski (49); 7.5 μg of each pLPCX vector encoding hemagglutinin (HA) epitope-tagged or untagged human or rhesus TRIM5 and puromycin resistance was transfected by calcium phosphate coprecipitation with 2.5 μg pMD.G into 1 x 106 Phoenix A cells in 6-cm-diameter dishes. Medium was changed 16 h later, and supernatants were collected 48 h thereafter and filtered. CrFK, D17, 293T, or HT1080 cells were seeded, 2.5 x 105 per well, into three wells of a six-well plate 24 h before transduction. The cells were transduced a second time 12 h later. Selection in 5 μg/ml (for CrFK), 0.3 μg/ml (for D17), 3 μg/ml (for 293T), or 0.5 μg/ml (for HT1080) puromycin (Sigma) was begun 48 h after the second transduction. Lines were robustly polyclonal, with outgrowth of more than 105 colonies from initial selection of each line.
Normalization of vector and virus stocks. Reverse transcriptase (RT) activities were used to normalize viral preparations. Undiluted or diluted supernatants (10 μl) were assayed by mixing with 50 μl of RT cocktail consisting of poly(A) template (5 μg/ml) and pd(t)12-18 oligonucleotide (1.57 μg/ml) in 50 mM Tris (pH 7.8), 7.5 mM KCl, 2 mM dithiothreitol, 5 mM MgCl2, 0.05% NP-40, and 1 μCi [32P]TTP. Samples were incubated in a Falcon flexible plastic 96-well plate for 1.5 h at 37°C. Five microliters of the reaction was blotted onto a Whatman DE-81 ion-exchange filter paper 96-well plate. After drying in a 37°C oven for 15 min, the plate was washed with 2x SSC (1x SSC is 0.15 M NaCl plus 0.015 M sodium citrate) and ethanol using a Filtermate Harvester. The plate was dried in a 37°C oven for 30 min. The bottom of the plate was sealed, and 35 μl of Microscint-O scintillation fluor was pipetted into each well. A TopCount NXT microplate luminescence-scintillation counter (Packard) was used to measure the radioactivity of each sample.
Stable TRIM5 RNAi. Two 63-nucleotide (nt) oligonucleotides encoding a short hairpin RNA (shRNA1) predicted to be processed by Dicer to the small interfering RNA (siRNA1) of Stremlau et al. (49) were synthesized such that annealing resulted in BamHI and HindIII adaptor termini (sense, GATCCGCTCAGGGAGGTCAAGTTGTTCAAGAGACAACTTGACCTCCCTGAGCT TTTTTGGAAA; antisense, AGCTTTTCCAAAAAAGCTCAGGGAGGTCAAGTTGTCTCTTGAACAACTTGACCTCCCTGAGCG. This adaptor was inserted into pSilencer (Ambion). Stable knockdown cell lines were generated as previously described for LEDGF/p75 (24) by stable transfection of plasmid linearized in prokaryotic sequences. A control cell line was developed using the same construct with a control 21-nt scrambled sequence shRNA in place of the 21-nt anti-TRIM5 sequence. Selection was in 0.125 mg hygromycin (Sigma) per ml, starting 48 h after transfection.
Vector transduction of different cell lines. A72, D17a, DK, CrFK, RD, 293T, HT1080, HeLa, H9, MT-4, Molt-4, Jurkat, Vero, FRhK4, 3T3, and MDBK cells, cell lines expressing HA-tagged human and rhesus TRIM5, rhesus TRIM5 shRNA knockdown FRhK4 cells, and scramble shRNA FRhK4 cell lines were seeded in 24-well plates at 45,000 cells/well. Cells were transduced with twofold or fourfold serial dilutions of FIV, HIV-1, EIAV, and NB-MLV GFP vectors. Cells were fixed in 1% formaldehyde and analyzed by fluorescence-activated cell sorting 48 h posttransduction. Mean titers per 1,000 RT units were calculated from the linear ranges of infectivity curves (25-27). In some cases, cells were treated with different doses of As2O3 (0.1 to 12 μM; Sigma) at the time of transduction. As2O3 was prepared as described by Berthoux et al. (2).
Assay for FIV replication. CrFK-derived cell lines that express human or rhesus TRIM5 were infected overnight at multiplicities of infection (MOI) of 0.01, 0.1, and 1.0. Cells were washed three times with phosphate-buffered saline and new medium was added. Supernatants were collected and frozen periodically over 3 weeks and RT activities were determined.
Immunoblotting. Cells were lysed in 150 mM NaCl, 1% NP-40, 0.5% deoxycholate, 0.1% sodium dodecyl sulfate, 50 mM Tris, pH 8.0, containing protease inhibitors (Complete Mini; Roche). After normalization for protein concentration, lysates were combined with ?-mercaptoethanol, boiled, and separated on a 10% sodium dodecyl sulfate-polyacrylamide gel. Proteins were transferred to a polyvinylidene difluoride membrane (Immobilon-P; Millipore) using a Bio-Rad Trans-Blot SD semi-dry electrophoretic transfer cell. The transferred blot was blocked in 10% milk before probing with anti-HA antibody (1:2,000; Novus Biologicals) in 5% milk-0.1% Tween. Peroxidase-conjugated goat anti-mouse immunoglobulin G (1:2,000; Calbiochem) was used as the secondary antibody. Blots were developed with a chemiluminescence kit (Amersham Pharmacia) and exposed to radiographic film.
Confocal immunofluorescence microscopy. CrFK or D17 cell lines (2 x 105) (parental, or lines stably expressing HA-tagged TRIM5hu or TRIM5rh) were seeded into Lab-Tek II chamber slides (Nalge Nunc International). Cells were fixed in 4% paraformaldehyde (Tousimis Research Corp.) then methanol before blocking in 10% fetal bovine serum (Gibco) and 20 mM ammonium chloride (Sigma). Cells were probed with rat anti-HA primary antibody (1:500; Roche) and goat anti-rabbit immunoglobulin G conjugated to Cy3 (1:500; Chemicon International). Slides were sealed with ProLong Gold mounting medium (Molecular Probes) and imaged on a Zeiss LSM 510 confocal laser scanning microscope mounted on an Axiovert 100 M (Zeiss, Thornwood, NY).
RESULTS
Comparative restriction of FIV vectors in cells from primate and nonprimate species. We initially carried out single-round VSV-G pseudotyped FIV vector (26, 27, 36) infections of human (293T, RD, HeLa, HT1080, Jurkat, Molt-4, MT-4, SupT1, and H9), simian (FRhK4 and Vero), feline (CrFK), canine (D17a, A72, and DK), bovine (MDBK), and murine (3T3) cell lines. Each cell line was transduced with eGFP vector over a wide range of inputs in two- or fourfold increments. We observed wide variations in susceptibility, both between and within individual species (Fig. 1A). Some of the variation could be grouped as cell type dependent. For example, human epithelial-derived cell lines (HT1080 and HeLa) were significantly more transducible by FIV vectors than transformed human T-cell lines (Jurkat, H9, MT-4, Molt-4, and SupT1). However, considerable variation was also seen between fibroeptihelial cell lines from the same species. For example, FIV vector transduction efficiencies differed 18-fold between canine DK and D17a cells. Some cell lines were notably permissive (293T, CrFK, RD, D17a, and A72). HeLa, HT1080, MT-4, and 3T3 cells exhibited intermediate resistance to FIV vectors. The simian cell lines (FRhK4 and Vero), the canine DK cell line, and human leukemic T-cell lines (H9, Jurkat, and Molt-4) were significantly more resistant to FIV vector transduction. The most resistant was the bovine MDBK cell line.
The same cell lines were tested in parallel for HIV-1 vector transduction efficiency (Fig. 1B). The HIV-1 vector has the same genetic structure, i.e., with an internal CMV promoter driving expression of an eGFP reporter placed upstream of the 3' long terminal repeat, thus removing promoter activity as a variable. Those cells found to be highly permissive for FIV (293T, CrFK, RD, and A72) were also highly permissive for HIV-1. Most of the intermediate-susceptibility cell lines (HeLa, HT1080, and 3T3) retained the same relative position. However, infectivity curves in certain cell lines shifted markedly to the left compared to FIV (notably, the human T-cell lines H9, Jurkat, MT-4, Molt-4, and SupT1). The simian cell lines (FRhK4 and Vero) were also comparatively resistant to HIV-1, consistent with their previously described Lv1 phenotype (1), as were canine DK cells. Again, the most resistant cell line was MDBK, with FIV and HIV-1 having equivalent infectivity on an RT-normalized basis.
We then compared FIV and HIV-1 vectors as a function of infectivity (transducing units per 103 RT activity units). They were equivalent (no significant difference, P > 0.05) in permissive feline CrFK and human RD cells as well as the more-resistant murine 3T3, canine DK, and bovine MDBK cells (Fig. 1C). Both vectors were comparatively less efficient in the monkey cell lines (FRhK4 and Vero cells), with FIV displaying twofold less infectivity than HIV-1. This approximate twofold difference was also seen in 293T, A72, HeLa, and HT1080 cells. Both vectors were markedly impaired in H9 cells, consistent with prior observations that H9 cells are relatively resistant to HIV-1 infection but allow chronic viral production once infected (59). Interestingly, we documented a strong bias for HIV-1 vectors over FIV in all of the human T-cell lines tested. HIV-1 vectors were 5-, 5-, 9.5-, 12.4-, and 26-fold more efficient than FIV in H9, MT-4, Jurkat, SupT1, and Molt-4 cells respectively.
Human and rhesus TRIM5 restrict FIV. After characterizing these differential restriction phenotypes for FIV versus HIV-1, we asked whether TRIM5 proteins could account for them. Since CrFK cells are widely reported to lack postentry restriction factors (7, 11, 15, 51), which is consistent with the data in Fig. 1, we stably expressed HA epitope-tagged and untagged TRIM5hu and TRIM5rh in this cell line, using MLV-based retroviral vectors prepared with the plasmids of Stremlau et al. (49). The four cell lines were selected in puromycin, and HA-tagged TRIM5 expression was verified by immunoblotting (Fig. 2A). Expression of the human and primate proteins was equivalent (Fig. 2A; corroborated in immunofluorescence analyses shown subsequently in Fig. 3). In addition, these cell lines were robustly polyclonal with respect to MLV vector insertion, representing outgrowth of more than 105 independent colonies from each transduction. FIV, HIV-1, EIAV, and NB-MLV GFP vector transductions were carried out, with twofold dilutions over a wide range. FIV infectivity shifted markedly to the right in CrFK cells expressing TRIM5hu or TRIM5rh (Fig. 2B). Rhesus TRIM5 caused a greater shift (64.6- ± 6.5-fold) than human TRIM5 (8.72- ± 2.4-fold). Similar to the findings of Hatziioannou et al. (15), analogously constructed vectors derived from the ungulate lentivirus EIAV (31) were shifted 25.9-fold and 11.6-fold in rhesus and human TRIM5-expressing CrFK cells, respectively, compared to parental CrFK cells. While HIV-1 was also strongly shifted (59.96- ± 16.6-fold to the right) by rhesus TRIM5, it was only weakly (approximately 2.4- ± 1.3-fold) blocked by human TRIM5, which is in agreement with prior results in HeLa (49) and CrFK (15) cells. In contrast, the negative control vector NB-MLV was unaffected by either rhesus or human TRIM5, establishing specificity (Fig. 2B). Furthermore, the same restricting activities were seen in CrFK cells expressing TRIM5 proteins lacking the HA tag, although lack of a good antibody to the native proteins prevented verification of equal expression (data not shown).
Subcellular distributions of human and rhesus TRIM5 proteins. Immunofluorescence analyses of the stable CrFK cell lines using anti-HA antibody showed similar, abundant expression and cytoplasmic distributions of human and rhesus TRIM5 (Fig. 3, top panels). The finely punctate foci we observed in stable cell lines are somewhat more diffusely distributed and smaller than the large and more-discrete structures, termed "cytoplasmic bodies," that were identified after transient transfection of a TRIM5-GFP fusion cDNA (39). In addition, we noted some variation on this point. For example, rhesus TRIM5 had a more-diffuse (although still heterogeneous) distribution in CrFK cells than did the human protein (Fig. 3); moreover, a minority of cells lacking discernible large foci in favor of more diffuse patterns could be found in each cell line and in the canine and human cell lines described below.
Canine D17 cells do not support restriction by introduced human TRIM5. Like feline CrFK cells, canine D17 cells are also reported to lack restriction factors and are highly permissive for HIV-1, FIV, and EIAV transduction (reference 17 and Fig. 1). Examinations of current drafts of the dog genome have not revealed any TRIM5 ortholog (45). We therefore derived stable D17 cell lines with the same TRIM5 retroviral vectors used for CrFK cells. Intracellular immunoreactivities and localizations of TRIM5hu and TRIM5rh were indistinguishable from CrFK cells (Fig. 3, middle panels). However, clearly different viral infectivity phenotypes were seen in these cell lines (Fig. 4A). Human TRIM5 did not restrict FIV or HIV-1 vectors (1.37- and 0.99-fold change in titer for FIV and HIV-1, respectively). Rhesus TRIM5 also had lesser effects in the canine cells than in the feline cells (2.21- and 2.10-fold decrease in titer for FIV and HIV-1, respectively). We additionally tested for restriction by TRIM5rh when retrovirally expressed in the relatively permissive human cell lines 293T and HT1080. TRIM5rh was similarly distributed in 293T cells (Fig. 3, bottom panels) and was able to restrict appreciably (Fig. 4B; 9.54- and 8.64-fold decrease in titer for FIV and HIV-1, respectively). TRIM5rh also restricted significantly in HT1080 cells (Fig. 4C; 35.01- and 35.64-fold decrease in titer for FIV and HIV-1, respectively). Again, NB-MLV titer was unaffected in any cell line.
Human and rhesus TRIM5 block FIV replication. We infected CrFK cells, which are permissive for FIV replication, and the CrFK-derived lines expressing TRIM5 proteins with replication-competent FIV at three MOI. Consistent with the single-round vector analyses, spreading infection was severely impaired in CrFK cells expressing either tagged or untagged TRIM5hu or TRIM5rh at MOI 0.01, 0.1, and 1.0 compared to parental CrFK cells (Fig. 5). Some RT activity accumulated in supernatants collected at 3 weeks from CrFK cells expressing untagged TRIM5rh at MOI 0.01 and from CrFK cells expressing HA-tagged and untagged TRIM5hu at MOI of 1.0 and 0.1. However, when passaged to fresh cells, further production of RT activity was not detected even after maintenance of log-phase cultures for 12 more weeks (data not shown). In contrast, transfer of these supernatants to parental CrFK cells did result in rescue of viral replication, which propagated to wild-type peak RT activity levels (data not shown). Some RT activity was also seen in supernatants from TRIM5hu- but not TRIM5rh-expressing CrFK cells infected at a high MOI, but these supernatants also yielded no persistent or amplified RT activity when passed to fresh TRIM5-expressing cells (data not shown). These data suggest that escape mutants were not generated, but that some wild-type virus was produced in TRIM5-expressing cell lines from integrated proviral progeny of virions that evaded the initial postentry restriction.
RNAi against endogenous TRIM5 in rhesus FRhK4 cells. Polymerase III-promoted shRNA constructs were derived and screened for RNAi activity against rhesus TRIM5, using Western blotting of transiently transfected HA-tagged protein. When effective transient knockdown was verified (data not shown), stable selection of rhesus FRhK4 cells was carried out. This had a strong positive effect on FIV and HIV-1 vector transduction (15.62- and 16.03-fold increases in titer, respectively) compared to the scrambled shRNA control cell line derived in parallel and to the parental cell line (Fig. 6). In contrast to these pronounced leftward shifts in lentiviral infectivity curves, NB-MLV was completely unaffected by the knockdown, which demonstrates specificity (Fig. 6). These results corroborate the specific restriction of FIV by TRIM5rh seen in our exogenous expression studies.
VLP cross-saturations of human and rhesus TRIM5. The restricting activity of both TRIM5 alleles was then tested for titration by coinfecting VLPs, a signature property that is consistent with saturability of the cellular restriction factors (two hit kinetic model) (6). In FRhK4 cells, both FIV and HIV-1 VLPs titrated endogenous rhesus TRIM5 (Fig. 7A). With HIV-1 VLPs, maximum increases were 23.0-fold and 16.5-fold for FIV and HIV-1, respectively; with FIV VLPs, the maximum increases were 12.7- and 6.43-fold. NB-MLV VLPs did not titrate the restrictions.
A comprehensive series of cross-saturation experiments were done next, with FIV, HIV-1, EIAV, and NB-MLV VLPs added at the time of transduction with FIV, HIV-1, EAIV, and NB-MLV GFP vectors of each TRIM5-expressing CrFK cell line (Fig. 7B). The results for rhesus TRIM5rh are summarized in Fig. 7B (top panel). FIV VLPs consistently increased transduction for each restricted (FIV, HIV-1, and EIAV) vector but had no effect on unrestricted NB-MLV GFP transduction. HIV-1 VLPs had an apparently greater effect on FIV transduction (8.5-fold increase) than HIV-1 and EIAV transduction (3.3- and 2.6-fold increases, respectively). EIAV VLPs also augmented transduction by all restricted vectors (3.9-, 4.7-, and 7.4-fold increases for HIV-1, FIV, and EIAV vector, respectively).
VLPs had a lesser effect in TRIM5hu-expressing CrFK cells (Fig. 7B, bottom panel). FIV VLPs had the strongest effect for restriction of FIV (3.7-fold increase), but they also increased HIV-1 and EIAV transduction (2.4- and 2.2-fold increases, respectively). EIAV VLPs were similar to FIV VLPs (3.6-fold increase) in relieving restriction of FIV. EIAV VLPs relieved restriction of EIAV more effectively than restriction of HIV-1 (3.5- versus 1.7-fold increase). NB-MLV GFP vector was not significantly affected by any species of VLP in any cell lines. Likewise, NB-MLV VLPs had no significant effect on the transduction efficiency of any of the vectors.
Effects of arsenic trioxide. The mitochondrial membrane destabilizing agent As2O3 has been reported to stimulate HIV-1 replication as well as increase HIV-1 and EIAV vector transduction of human cells (1, 2, 54). The mechanism is unknown, but As2O3 alters the subcellular localization and degradation of PML, another TRIM family member (54). We examined the effects of As2O3 on FIV, HIV-1, and NB-MLV GFP vector transduction in CrFK, CrFK-TRIM5hu, CrFK-TRIM5rh, HT1080, Vero, and FRhK4 cells (Fig. 8).
The effects were complex. Although As2O3 did increase FIV vector transduction of CrFK-TRIM5hu and CrFK-TRIM5rh cells (P < 0.04), transduction efficiency also increased commensurately in parental CrFK cells, indicating that in this cell line the effect is not TRIM5 specific (Fig. 8, top panels). It was, however, lentivirus-specific (NB-MLV was not affected). The magnitude of As2O3-induced increases were similar for FIV and HIV-1 vectors at low (1 to 2% GFP-positive cells at baseline) and higher (5 to 10% GFP-positive cells at baseline) vector inputs in CrFK-TRIM5hu cells. In addition, similar stimulation of transduction by both FIV and HIV vectors were seen in multiple human cell lines (HeLa, 293T, RD, and MT-4; data not shown). NB-MLV vectors were unaffected by As2O3 in any cell line. In contrast to human TRIM5, the effects of As2O3 on lentiviral vectors were input dependent as well as vector type dependent for rhesus TRIM5. This was seen in both FRhK4 cells and CrFK-TRIM5rh cells. With low vector input (1 to 2% GFP-positive cells at baseline), As2O3 treatment increased FIV but not HIV-1 vector transduction. Results were similar in FRhK4 cells. However, at higher vector input (5 to 10% GFP-positive cells) in either cell line, As2O3 treatment had no significant effect for either lentiviral vector in CrFK-TRIM5rh cells (data not shown). Again, NB-MLV transduction efficiency was unaffected by As2O3 treatment at both vector inputs. In Vero and HT1080 cells, both HIV-1 and FIV transductions were stimulated; note however that in both simian cell lines (Vero and FRhK4) significant stimulation required in excess of 2 μM, concentrations which were quite toxic to nonsimian cell lines (data not shown). We observed no significant effect for any of the vectors in 3T3 cells treated with 2.0 μM As2O3 (data not shown).
DISCUSSION
We have carried out a detailed first analysis of postentry restriction of FIV, extending such analyses to the third main group of lentiviruses. Although FIV-based lentiviral vectors have been shown to transduce primate cells and tissues with good practical efficiency (9, 21, 25, 27, 28, 36, 55), the species-specific parameters governing efficiency relative to HIV-1 have not been defined quantitatively, and the activities of known postentry restriction factors have not been characterized. Our data show that both Lv1 and Ref1 restrict FIV and that the major patterns can be accounted for by TRIM5. FIV is substantially restricted by both human and rhesus proteins, although considerably more so by the rhesus protein. In this discussion, we highlight novel virological aspects and then consider ramifications for human gene therapy with FIV and other nonprimate lentiviral vectors.
The extent of inhibition of HIV-1 and EIAV in our CrFK cells expressing TRIM5 variants agrees well with results obtained in CrFK lines generated independently by others (15), which supports the validity of the FIV findings. We were surprised to find in further experiments that canine D17 cells did not support significant restriction by exogenously introduced human TRIM5. Determining the reason for this difference will require further investigation. We verified that epitope-tagged TRIM5 proteins localize to similar, discrete structures consistent with cytoplasmic bodies in both canine and feline cells and that intact, full-length proteins were expressed at substantial levels in both (Fig. 3 and 4A). As previously observed by Javanbakht et al. (18) in a different canine cell line (Cf2Th thymocytes), we found that D17 cells did support some rhesus allele activity. The human TRIM5 allele was not tested in Cf2Th cells (18). It could be that some canine cells lack an additional factor or factors required for optimal TRIM5 antiviral function. Alternatively, since some TRIM variants (e.g., TRIM5) have transdominant activity (49), a canine factor might interfere with TRIM5 function.
Compared to HIV-1 vectors in RT activity-normalized infections, human cell susceptibility to FIV vectors varied from equivalent to HIV-1 (RD cells) to significantly less than HIV-1 (human T-cell lines). An interesting question that will require further study is whether TRIM5hu fully explains the relative restriction of FIV in human T-cell lines. The cell type-dependent pattern of restriction could be due to different TRIM5 expression levels or subcellular localization patterns or to other factors. No data have been published concerning relative endogenous TRIM5 expression levels. In rhesus FRhK4 cells, however, we found that stable RNAi against rhesus TRIM5 resulted in marked increases in FIV and HIV-1 infectivities but not NB-MLV infectivity, suggesting that rhesus TRIM5 is largely responsible for postentry restriction in these cells. Unlike FRhK4 cells, we could not significantly reverse restriction of FIV with stable shRNA expression in HeLa cells or human T-cell lines (data not shown). We do not know if this represented inadequate knockdown, because we have been unable to quantify endogenous TRIM5 protein levels with commercially available antibodies, which we have found to be inadequately sensitive and specific in immunoblotting (data not shown). Since T lymphocytes are prime targets for many retroviruses, evolution of stronger innate T-cell antiviral mechanisms in T cells would be advantageous to the primate host.
In contrast to primate cell lines, RT activity-normalized HIV-1 and FIV vectors had equivalent infectivities in nonprimate cell lines, with the concordant transduction efficiencies nevertheless varying greatly by cell type. For example, 18-fold variability in susceptibility was seen between canine cell lines (A72 compared to D17 and DK). Bovine MDBK cells are very resistant to both HIV-1 and FIV vectors.
In primate epithelial cell lines (HT1080, HeLa, 293T, and Vero), the absolute number of transducing units per RT unit were slightly lower (two to threefold) for FIV. However, comparing absolute infectivities for two different lentiviruses is problematic since the RT assay could score enzymatic activity differently for each reverse transcriptase, or be biased towards one or the other RT according to particular assay components (e.g., magnesium, pH, or ionic strength), or vary with the percentage of defective particles in a given preparation. Thus, it is not known whether the RT assay faithfully normalizes particle number in absolute terms.
However, it is important to appreciate that key aspects of these studies rule out any incidental effects of particle normalization or of the particular vector construct versions used. First, the variance in the ratios of FIV/HIV-1 vector infectivities, from a ratio of 1 in cells lacking primate restriction factors (CrFK, D17a, DK, 3T3, and MDBK) to considerably more than 1, e.g., 5 to 26, in human T-cell lines, securely identifies host cell factors as operative. The second aspect is the virus-specific restrictions conferred by selective expression of allelic TRIM5 proteins in the CrFK cell background. Expression of human TRIM5 was sufficient to markedly restrict FIV but not HIV-1 in engineered CrFK cell lines. Expression of rhesus TRIM5 restricted both lentiviruses to the same extent, and neither protein restricted NB-MLV. HIV-1 and FIV vectors are equivalent on an RT-normalized basis in restriction factor-lacking CrFK cells (and in human RD cells). In contrast, FIV vectors were significantly less infectious than HIV-1 vectors in CrFK-TRIM5hu cells and in human T-cell lines. This relative restriction of FIV (and EIAV) versus HIV-1 vectors in some but not all human cells is thus shown to be intrinsic to the viral biology. Insofar as it may be an issue for human gene therapy, it is not addressed by extant vector systems. Practical implications of Ref1 restriction for gene therapy are discussed below.
We found that arsenic trioxide had effects which are complex and not clearly TRIM5 dependent, preventing straightforward interpretation of the effects of this drug. While in our CrFK cell lines that express human or rhesus TRIM5, As2O3 did have generally stimulatory effects on both FIV and HIV-1 transduction, and NB-MLV was unaffected, these effects were cell type, vector type, and input amount dependent. Most importantly, arsenic enhancement is TRIM5 independent in the CrFK background because equivalent stimulation occurred in the parental CrFK cells. Our data on the effects of As2O3 on HIV-1 are, however, in good agreement with a number of observations of Berthoux et al. (1; compare Fig. 8 with Fig. 7 therein), i.e., we saw no effect at all in rhesus FRhK4 cells and a significant stimulation in AGM (Vero) cells, with a peak at 8 μM. At 12 μM, we began to observe significant toxicity in Vero and FRhK4 cells. As also observed previously by Berthoux et al. (1), these simian cells were the only cell type in which we could raise the As2O3 concentration above 2 μM without prohibitive toxicity (data not shown). Whether stimulation of infection in Vero cells can be securely attributed to an effect on AGM TRIM5 remains unclear; for example, restriction of N-MLV by AGM TRIM5 expressed in murine cells was not reversed by arsenic trioxide, whereas restriction by human TRIM5 was reversed (22). The individual lentiviruses differed as well in response to As2O3: FIV but not HIV-1 vector transduction was augmented in FRhK4 cells by 2 μM As2O3, and FIV was more stimulated than HIV-1 in CrFK-TRIM5rh cells.
Viral restriction determinants have been mapped to the exposed capsid surface for HIV-1 and N-MLV (11, 23, 33, 47, 51, 53). This region is quite different in FIV CA, which lacks the proline-rich cyclophilin A binding loop present in HIV-1 (10, 53). Interaction between this loop and cyclophilin A has been reported to protect HIV-1 from restriction by Ref1 (14, 53). Species-specific retroviral CA recognition appears to be mediated by the TRIM5 SPRY(B30.2) domain (41, 46, 50, 58), which may mediate degradation of the incoming viral core or reverse transcription complex or may alter kinetics of core disassembly. Direct interaction between the SPRY domain and CA was suggested by the recent discovery that a New World (owl) monkey TRIM5 contains a fusion to the complete cyclophilin A gene in place of the SPRY domain (30, 42) together with the finding that human cyclophilin A acts to interfere with Ref1 restriction by binding to an exposed proline-rich loop in HIV-1 CA (53). Recently, binding between a human TRIM5-GST fusion protein and N-MLV CA presented by detergent-stripped virion cores was demonstrated; SPRY domain deletion or use of B-MLV rather than N-MLV abrogated this binding (43). In contrast, free CA protein did not interact, suggesting that the SPRY domain recognizes a complex structure present at the face of multimerized CA (43). It may therefore be of interest to map specific determinants of restriction in FIV CA.
Transmission of FIV from cats to nonfelines has not been observed, despite presumably prevalent natural exposure (interfeline transmission is by biting). To some extent, this restricted tropism can be accounted for by quite limited FIV long terminal repeat activity in human cells (35, 36) and by different primary receptor usage (44), although some FIV isolates can also use human CXCR4 for entry (35, 56). High MOI infection with FIV can lead to infection and even lytic FIV envelope glycoprotein expression but not viral replication in CXCR4-positive human cells (35, 56). Consistent with this, one group has reported expression of FIV RNA and proteins after infection of human peripheral blood mononuclear cells as well, although passage of virus or continuous propagation were not demonstrated (19). Two rhesus macaques subsequently injected with FIV-infected autologous peripheral blood mononuclear cells were reported to show transient PCR positivity for RNA and DNA, with both animals becoming negative after 4 and 9 weeks, suggesting a self-limited viremia due to FIV (20). This latter result is surprising in view of the degree of restriction by rhesus TRIM5 shown here to an FIV clone that is also derived from the Petaluma strain that was used to infect the macaques and the inability to achieve significant HIV-1 replication in these primates in numerous other studies. Antibodies to FIV were not produced in one of these animals, however, and passage to other animals was not done.
The present results have implications for human gene therapy with FIV and EIAV vectors. Each have been shown to achieve useful levels of gene transfer in primate cells and tissues (reviewed in reference 37). Nevertheless, further development of strategies that employ lentiviral vectors will benefit from informed appraisal of the consequences of Ref1 restriction, which may be advantageous or disadvantageous depending on the application. In some situations, Ref1/human TRIM5 restriction will reduce the absolute efficiency of gene transfer by FIV or EIAV vectors. For example, intravenous administration of nonprimate lentivirus vectors may lead to dilution of particles to levels that are below Ref1/TRIM5 saturation thresholds. Such restriction may be surmountable by dosage adjustment, or simply by adding VLPs. In many gene therapy situations in vivo or ex vivo, however, localized inputs without VLPs result in particle concentrations well above those needed to saturate Ref1/TRIM5 activity. In this common situation, or where a desired target tissue expresses little TRIM5, Ref1 restriction would provide a net advantage: a highly evolved innate barrier to systemic propagation of replication-competent retroviruses, which are among the most concerning potential complications of retroviral vector-based therapy. In addition, relative levels of TRIM5 expression in different human tissues that are gene therapy targets are not well understood. Development of good antibodies to TRIM5 will be useful in determining which human tissues express significant amounts of this protein and will further clarify the implications of Ref1 for gene therapy. Similarly, the intracellular distributions we observed with the HA-tagged proteins cannot be taken as definitive until endogenous TRIM5 is characterized properly. Such studies may clarify whether overexpression (likely to be extreme with transient transfection, but probably less so with retroviral vectors as used here) and/or the use of GFP fusions accounts for the larger structures previously described (39).
Finally, it is not yet known exactly how TRIM5 proteins interact with capsid proteins, cyclophilins, and perhaps other factors to result in the observed restrictions. We can speculate that progress in understanding the actual molecular mechanisms, together with virus-specific determinant mapping and mutagenesis, could enable specialized engineering of nonprimate lentiviral vector capsid proteins for greater per-particle efficacy in TRIM5hu-expressing cells. Such modifications would, of course, negate the advantage of the innate block to systemic replication-competent retrovirus replication.
ACKNOWLEDGMENTS
We thank J. Sodroski and M. Stremlau for providing pLPCX vector plasmids encoding human and rhesus TRIM5, A. Fielding for D17a cells, Y. Ikeda for A72, DK, and MDBK cells, T. Whitwam for help with vector preparation and titration on multiple cell lines, and M. Peretz for technical assistance.
REFERENCES
Berthoux, L., S. Sebastian, E. Sokolskaja, and J. Luban. 2004. Lv1 inhibition of human immunodeficiency virus type 1 is counteracted by factors that stimulate synthesis or nuclear translocation of viral cDNA. J. Virol. 78:11739-11750.
Berthoux, L., G. J. Towers, C. Gurer, P. Salomoni, P. P. Pandolfi, and J. Luban. 2003. As2O3 enhances retroviral reverse transcription and counteracts Ref1 antiviral activity. J. Virol. 77:3167-3180.
Besnier, C., Y. Takeuchi, and G. Towers. 2002. Restriction of lentivirus in monkeys. Proc. Natl. Acad. Sci. USA 99:11920-11925.
Besnier, C., L. Ylinen, B. Strange, A. Lister, Y. Takeuchi, S. P. Goff, and G. J. Towers. 2003. Characterization of murine leukemia virus restriction in mammals. J. Virol. 77:13403-13406.
Best, S., P. Le Tissier, G. Towers, and J. P. Stoye. 1996. Positional cloning of the mouse retrovirus restriction gene Fv1. Nature 382:826-829.
Bieniasz, P. D. 2003. Restriction factors: a defense against retroviral infection. Trends Microbiol. 11:286-291.
Bock, M., K. N. Bishop, G. Towers, and J. P. Stoye. 2000. Use of a transient assay for studying the genetic determinants of Fv1 restriction. J. Virol. 74:7422-7430.
Cowan, S., T. Hatziioannou, T. Cunningham, M. A. Muesing, H. G. Gottlinger, and P. D. Bieniasz. 2002. Cellular inhibitors with Fv1-like activity restrict human and simian immunodeficiency virus tropism. Proc. Natl. Acad. Sci. USA 99:11914-11919.
Curran, M. A., S. M. Kaiser, P. L. Achacoso, and G. P. Nolan. 2000. Efficient transduction of nondividing cells by optimized feline immunodeficiency virus vectors. Mol. Ther. 1:31-38.
Gamble, T. R., F. F. Vajdos, S. Yoo, D. K. Worthylake, M. Houseweart, W. I. Sundquist, and C. P. Hill. 1996. Crystal structure of human cyclophilin A bound to the amino-terminal domain of HIV-1 capsid. Cell 87:1285-1294.
Hatziioannou, T., S. Cowan, and P. D. Bieniasz. 2004. Capsid-dependent and -independent postentry restriction of primate lentivirus tropism in rodent cells. J. Virol. 78:1006-1011.
Hatziioannou, T., S. Cowan, S. P. Goff, P. D. Bieniasz, and G. J. Towers. 2003. Restriction of multiple divergent retroviruses by Lv1 and Ref1. EMBO J. 22:385-394.
Hatziioannou, T., S. Cowan, U. K. Von Schwedler, W. I. Sundquist, and P. D. Bieniasz. 2004. Species-specific tropism determinants in the human immunodeficiency virus type 1 capsid. J. Virol. 78:6005-6012.
Hatziioannou, T., D. Perez-Caballero, S. Cowan, and P. D. Bieniasz. 2005. Cyclophilin interactions with incoming human immunodeficiency virus type 1 capsids with opposing effects on infectivity in human cells. J. Virol. 79:176-183.
Hatziioannou, T., D. Perez-Caballero, A. Yang, S. Cowan, and P. D. Bieniasz. 2004. Retrovirus resistance factors Ref1 and Lv1 are species-specific variants of TRIM5alpha. Proc. Natl. Acad. Sci. USA 101:10774-10779.
Hofmann, W., D. Schubert, J. LaBonte, L. Munson, S. Gibson, J. Scammell, P. Ferrigno, and J. Sodroski. 1999. Species-specific, postentry barriers to primate immunodeficiency virus infection. J. Virol. 73:10020-10028.
Ikeda, Y., M. K. Collins, P. A. Radcliffe, K. A. Mitrophanous, and Y. Takeuchi. 2002. Gene transduction efficiency in cells of different species by HIV and EIAV vectors. Gene Ther. 9:932-938.
Javanbakht, H., F. Diaz-Griffero, M. Stremlau, Z. Si, and J. Sodroski. 2005. The contribution of RING and B-box 2 domains to retroviral restriction mediated by monkey TRIM5alpha. J. Biol. Chem. 280:26933-26940.
Johnston, J., and C. Power. 1999. Productive infection of human peripheral blood mononuclear cells by feline immunodeficiency virus: implications for vector development. J. Virol. 73:2491-2498.
Johnston, J. B., M. E. Olson, E. W. Rud, and C. Power. 2001. Xenoinfection of nonhuman primates by feline immunodeficiency virus. Curr. Biol. 11:1109-1113.
Johnston, J. C., M. Gasmi, L. E. Lim, J. H. Elder, J. K. Yee, D. J. Jolly, K. P. Campbell, B. L. Davidson, and S. L. Sauter. 1999. Minimum requirements for efficient transduction of dividing and nondividing cells by feline immunodeficiency virus vectors. J. Virol. 73:4991-5000.
Keckesova, Z., L. M. Ylinen, and G. J. Towers. 2004. The human and African green monkey TRIM5alpha genes encode Ref1 and Lv1 retroviral restriction factor activities. Proc. Natl. Acad. Sci. USA 101:10780-10785.
Kozak, C. A., and A. Chakraborti. 1996. Single amino acid changes in the murine leukemia virus capsid protein gene define the target of Fv1 resistance. Virology 225:300-305.
Llano, M., M. Vanegas, O. Fregoso, D. Saenz, S. Chung, M. Peretz, and E. M. Poeschla. 2004. LEDGF/p75 determines cellular trafficking of diverse lentiviral but not murine oncoretroviral integrase proteins and is a component of functional lentiviral pre-integration complexes. J. Virol. 78:9524-9537.
Loewen, N., C. Bahler, W. Teo, T. Whitwam, M. Peretz, R. Xu, M. Fautsch, D. H. Johnson, and E. M. Poeschla. 2002. Preservation of aqueous outflow facility after second-generation FIV vector-mediated expression of marker genes in anterior segments of human eyes. Investig. Ophthalmol. Vis. Sci. 43:3686-3690.
Loewen, N., R. Barraza, T. Whitwam, D. Saenz, I. Kemler, and E. Poeschla. 2003. FIV vectors. Methods Mol. Biol. 229:251-271.
Loewen, N., M. Fautsch, M. Peretz, C. Bahler, J. D. Cameron, D. H. Johnson, and E. M. Poeschla. 2001. Genetic modification of human trabecular meshwork with lentiviral vectors. Hum. Gene Ther. 12:2109-2119.
Lotery, A. J., T. A. Derksen, S. R. Russell, R. F. Mullins, S. Sauter, L. M. Affatigato, E. M. Stone, and B. L. Davidson. 2002. Gene transfer to the nonhuman primate retina with recombinant feline immunodeficiency virus vectors. Hum. Gene Ther. 13:689-696.
Munk, C., S. M. Brandt, G. Lucero, and N. R. Landau. 2002. A dominant block to HIV-1 replication at reverse transcription in simian cells. Proc. Natl. Acad. Sci. USA 99:13843-13848.
Nisole, S., C. Lynch, J. P. Stoye, and M. W. Yap. 2004. A Trim5-cyclophilin A fusion protein found in owl monkey kidney cells can restrict HIV-1. Proc. Natl. Acad. Sci. USA 101:13324-13328.
Olsen, J. C. 1998. Gene transfer vectors derived from equine infectious anemia virus. Gene Ther. 5:1481-1487.
O'Rourke, J. P., H. Hiraragi, K. Urban, M. Patel, J. C. Olsen, and B. A. Bunnell. 2003. Analysis of gene transfer and expression in skeletal muscle using enhanced EIAV lentivirus vectors. Mol. Ther. 7:632-639.
Owens, C. M., P. C. Yang, H. Gottlinger, and J. Sodroski. 2003. Human and simian immunodeficiency virus capsid proteins are major viral determinants of early, postentry replication blocks in simian cells. J. Virol. 77:726-731.
Perron, M. J., M. Stremlau, B. Song, W. Ulm, R. C. Mulligan, and J. Sodroski. 2004. TRIM5alpha mediates the postentry block to N-tropic murine leukemia viruses in human cells. Proc. Natl. Acad. Sci. USA 101:11827-11832.
Poeschla, E. M., and D. J. Looney. 1998. CXCR4 is required by a nonprimate lentivirus: heterologous expression of feline immunodeficiency virus in human, rodent, and feline cells. J. Virol. 72:6858-6866.
Poeschla, E., F. Wong-Staal, and D. Looney. 1998. Efficient transduction of nondividing cells by feline immunodeficiency virus lentiviral vectors. Nat. Med. 4:354-357.
Poeschla, E. M. 2003. Non-primate lentiviral vectors. Curr. Opin. Mol. Ther. 5:529-540.
Remington, K. M., Y. Q. Zhu, T. R. Phillips, and T. W. North. 1994. Rapid phenotypic reversion of zidovudine-resistant feline immunodeficiency virus without loss of drug-resistant reverse transcriptase. J. Virol. 68:632-637.
Reymond, A., G. Meroni, A. Fantozzi, G. Merla, S. Cairo, L. Luzi, D. Riganelli, E. Zanaria, S. Messali, S. Cainarca, A. Guffanti, S. Minucci, P. G. Pelicci, and A. Ballabio. 2001. The tripartite motif family identifies cell compartments. EMBO J. 20:2140-2151.
Saenz, D. T., and E. M. Poeschla. 2004. FIV: from lentivirus to lentivector. J. Gene Med. 6(Suppl. 1):S95-S104.
Sawyer, S. L., L. I. Wu, M. Emerman, and H. S. Malik. 2005. Positive selection of primate TRIM5alpha identifies a critical species-specific retroviral restriction domain. Proc. Natl. Acad. Sci. USA 102:2832-2837.
Sayah, D. M., E. Sokolskaja, L. Berthoux, and J. Luban. 2004. Cyclophilin A retrotransposition into TRIM5 explains owl monkey resistance to HIV-1. Nature 430:569-573.
Sebastian, S., and J. Luban. 2005. TRIM5alpha selectively binds a restriction-sensitive retroviral capsid. Retrovirology 2:40.
Shimojima, M., T. Miyazawa, Y. Ikeda, E. L. McMonagle, H. Haining, H. Akashi, Y. Takeuchi, M. J. Hosie, and B. J. Willett. 2004. Use of CD134 as a primary receptor by the feline immunodeficiency virus. Science 303:1192-1195.
Song, B., B. Gold, C. O'hUigin, H. Javanbakht, X. Li, M. Stremlau, C. Winkler, M. Dean, and J. Sodroski. 2005. The B30.2(SPRY) domain of the retroviral restriction factor TRIM5 exhibits lineage-specific length and sequence variation in primates. J. Virol. 79:6111-6121.
Song, B., H. Javanbakht, M. Perron, D. H. Park, M. Stremlau, and J. Sodroski. 2005. Retrovirus restriction by TRIM5 variants from Old World and New World primates. J. Virol. 79:3930-3937.
Stevens, A., M. Bock, S. Ellis, P. LeTissier, K. N. Bishop, M. W. Yap, W. Taylor, and J. P. Stoye. 2004. Retroviral capsid determinants of Fv1 NB and NR tropism. J. Virol. 78:9592-9598.
Stoye, J. P. 2002. An intracellular block to primate lentivirus replication. Proc. Natl. Acad. Sci. USA 99:11549-11551.
Stremlau, M., C. M. Owens, M. J. Perron, M. Kiessling, P. Autissier, and J. Sodroski. 2004. The cytoplasmic body component TRIM5alpha restricts HIV-1 infection in Old World monkeys. Nature 427:848-853.
Stremlau, M., M. Perron, S. Welikala, and J. Sodroski. 2005. Species-specific variation in the B30.2(SPRY) domain of TRIM5 determines the potency of human immunodeficiency virus restriction. J. Virol. 79:3139-3145.
Towers, G., M. Bock, S. Martin, Y. Takeuchi, J. P. Stoye, and O. Danos. 2000. A conserved mechanism of retrovirus restriction in mammals. Proc. Natl. Acad. Sci. USA 97:12295-12299.
Towers, G., M. Collins, and Y. Takeuchi. 2002. Abrogation of Ref1 retrovirus restriction in human cells. J. Virol. 76:2548-2550.
Towers, G. J., T. Hatziioannou, S. Cowan, S. P. Goff, J. Luban, and P. D. Bieniasz. 2003. Cyclophilin A modulates the sensitivity of HIV-1 to host restriction factors. Nat. Med. 9:1138-1143.
Turelli, P., V. Doucas, E. Craig, B. Mangeat, N. Klages, R. Evans, G. Kalpana, and D. Trono. 2001. Cytoplasmic recruitment of INI1 and PML on incoming HIV preintegration complexes: interference with early steps of viral replication. Mol. Cell 7:1245-1254.
Wang, G., V. Slepushkin, J. Zabner, S. Keshavjee, J. C. Johnston, S. L. Sauter, D. J. Jolly, T. W. Dubensky, Jr., B. L. Davidson, and P. B. McCray, Jr. 1999. Feline immunodeficiency virus vectors persistently transduce nondividing airway epithelia and correct the cystic fibrosis defect. J. Clin. Investig. 104:R55-R62.
Willett, B. J., L. Picard, M. J. Hosie, J. D. Turner, K. Adema, and P. R. Clapham. 1997. Shared usage of the chemokine receptor CXCR4 by the feline and human immunodeficiency viruses. J. Virol. 71:6407-6415.
Yap, M. W., S. Nisole, C. Lynch, and J. P. Stoye. 2004. Trim5alpha protein restricts both HIV-1 and murine leukemia virus. Proc. Natl. Acad. Sci. USA 101:10786-10791.
Yap, M. W., S. Nisole, and J. P. Stoye. 2005. A single amino acid change in the SPRY domain of human Trim5-alpha leads to HIV-1 restriction. Curr. Biol. 15:73-78.
Zennou, V., C. Petit, D. Guetard, U. Nerhbass, L. Montagnier, and P. Charneau. 2000. HIV-1 genome nuclear import is mediated by a central DNA flap. Cell 101:173-185.
Zufferey, R., D. Nagy, R. J. Mandel, L. Naldini, and D. Trono. 1997. Multiply attenuated lentiviral vector achieves efficient gene delivery in vivo. Nat. Biotechnol. 15:871-875.(Dyana T. Saenz, Wulin Teo)
Cystic Fibrosis Research and Treatment Center, University of North Carolina, Chapel Hill, North Carolina
ABSTRACT
The Ref1 and Lv1 postentry restrictions in human and monkey cells have been analyzed for lentiviruses in the primate and ungulate groups, but no data exist for the third (feline) group. We compared feline immunodeficiency virus (FIV) to other restricted (human immunodeficiency virus type 1 [HIV-1], equine infectious anemia virus [EIAV]) and unrestricted (NB-tropic murine leukemia virus [NB-MLV]) retroviruses across wide ranges of viral inputs in cells from multiple primate and nonprimate species. We also characterized restrictions conferred to permissive feline and canine cells engineered to express rhesus and human TRIM5 proteins and performed RNA interference (RNAi) against endogenous TRIM5. We find that expression of rhesus or human TRIM5 proteins in feline cells restricts FIV, impairing pseudotyped vector transduction and viral replication, but rhesus TRIM5 is more restricting than human TRIM5. Notably, however, canine cells did not support restriction by human TRIM5 and supported minimal restriction by rhesus TRIM5, suggesting that these proteins may not function autonomously or that a canine factor interferes. Stable RNAi knockdown of endogenous rhesus TRIM5 resulted in marked increases in FIV and HIV-1 infectivities while having no effect on NB-MLV. A panel of nonprimate cell lines varied widely in susceptibility to lentiviral vector transduction, but normalized FIV and HIV-1 vectors varied concordantly. In contrast, in human and monkey cells, relative restriction of FIV compared to HIV-1 varied from none to substantial, with the greatest relative infectivity deficit for FIV vectors observed in human T-cell lines. Endogenous and introduced TRIM5 restrictions of FIV could be titrated by coinfections with FIV, HIV-1, or EIAV virus-like particles. Arsenic trioxide had complex and TRIM5-independent enhancing effects on lentiviral but not NB-MLV infection. Implications for human gene therapy are discussed.
INTRODUCTION
Innate cellular defenses against retroviruses include early postentry restrictions to infection that are dependent upon the identity of the viral capsid protein and are mediated by dominantly acting, species-specific host factors. Murine leukemia virus (MLV) variants provide the classic model of this kind of restriction. N-tropic MLV (N-MLV) is restricted by the Friend virus susceptibility factor 1 b (Fv1b) allele in BALB/c mice but not in NIH Swiss mice, whereas B-MLV is restricted by the Fv1n allele in NIH Swiss mice but not in BALB/c mice. NB-MLV is restricted by neither allele. The n and b Fv1 alleles encode retroviral Gag-like proteins similar to murine endogenous retroviral Gag (5). The critical viral determinant maps to capsid residue 110 (7, 23, 51).
Human and simian cells display similar activities that restrict N-MLV but not B-MLV or NB-MLV (3, 4, 8, 11-13, 16, 48). Although the viral determinant of N-MLV restriction in nonmurine cells also maps to capsid residue 110 (47, 51), the mechanisms appear to differ. Fv1 acts after reverse transcription, whereas the human activity Ref1 and simian activity Lv1 block infection prior to or concurrent with reverse transcription (6). In addition to restricting N-MLV, Lv1 has been shown to restrict lentiviruses: human immunodeficiency virus type 1 (HIV-1) and type 2 (HIV-2) and equine infectious anemia virus (EIAV) as well as certain simian lentiviruses (12, 15, 16, 29, 52). These data are consistent with the inability of HIV-1 to productively infect Old World Monkeys. In contrast, the human cell activity of Ref1 restricts EIAV and N-MLV but does not restrict HIV-1, HIV-2, or simian immunodeficiency viruses (SIVs) (8, 12, 16).
An informative characteristic of Fv1, Lv1, and Ref1 is their saturability when cells are infected with large amounts of virus or vector, or coinfected with other restricted retroviruses or virus-like particles (VLPs). Titration of the restricting activity in this way requires only that the coinfecting particles be entry competent and contain mature capsids of any retrovirus that is also restricted; they need not contain viral genomes or enter via the same receptor pathway (1, 3, 4, 12, 52). Together, these compelling infectivity and cross-titration phenotypes have supported a model in which cytoplasmic factors can block infection at low viral inputs by interacting directly or indirectly with the viral capsid protein (6).
Recently, the tripartite motif (39) protein TRIM5 was identified as the major factor restricting HIV-1 infection in rhesus cells (49). Its properties appear to account for the major phenotypic features of Lv1 restriction (15, 49). In addition, orthologs of TRIM5 have subsequently been shown to mediate Ref1 (human TRIM5 [TRIM5hu]) as well as the broad restriction to HIV-1, N-MLV, EIAV, and SIVmac in African green monkey (AGM) cells (TRIM5agm) (15, 22, 34, 46, 57). Rhesus TRIM5 (TRIM5rh) expression in normally HIV-1-permissive HeLa cells confers strong restriction of HIV-1 but not SIVmac (49). Introduction of TRIM5rh into Crandell feline kidney cells (CrFK), which lack restriction factors (7, 11, 15, 51), confers potent restriction of HIV-1, SIVagm, EIAV, and N-MLV (15, 22, 41, 47). Human TRIM5hu expression results in a weak effect on HIV-1 and strong restriction of EIAV and N-MLV (22, 49). B-MLV and NB-MLV are not restricted by any TRIM5 alleles so far described, which has led to their use as negative controls experimentally (12).
Three groups of lentiviruses infect primates (e.g., HIV-1), ungulates (e.g., EIAV), and felines (feline immunodeficiency virus [FIV]) respectively. While Lv1 and Ref1 restriction patterns and TRIM5-mediated blocks to viruses in the first two lentiviral groups have been well described, no data have been reported for the third. FIV is the cause of a pandemic in multiple feline species. The virus is highly divergent at the nucleotide and protein levels from primate and ungulate lentiviruses, including the capsid (CA) region implicated in mediating Lv1/Ref1 restriction. Uniquely among the nonprimate lentiviruses, FIV causes CD4+ T-cell depletion and AIDS. Here we report a detailed analysis of FIV postentry restriction, which we carried out by characterizing Lv1 and Ref1 restrictions compared to other retroviruses that served as positive (HIV-1 and EIAV) and negative (NB-MLV) controls and by examining specific effects of TRIM5 cDNA expression and TRIM5-directed RNA interference (RNAi). We also studied the effects of arsenic trioxide, a mitochondrial membrane potential-disrupting agent that enhances HIV-1 vector transduction in some cells (54), apparently by attenuating postentry restrictions (1, 2, 22). Finally, we consider implications for use of FIV vectors (36, 40) in human gene therapy.
MATERIALS AND MATHODS
Cells. Cell lines were cultured in RPMI or Dulbecco's modified Eagle medium with 10% fetal calf serum. 293T human embryonic kidney cells were used as lentivirus and subgenomic lentivector producer cells. For NB-MLV production, the Moloney murine leukemia virus-based retrovirus producer cell line Phoenix A, a gift of G. Nolan, was used. 293T, HeLa (human epithelial adenocarcinoma), Crandell feline kidney (CrFK), HT1080 (human epithelial fibrosarcoma), RD (human rhabdomyosarcoma), Vero (African green monkey kidney), FRhK4 (fetal rhesus kidney), 3T3 (NIH Swiss murine embryo fibroblast), human leukemic T-cell (H9, Jurkat clone E6-1, MT-4, Molt-4 clone 8, and SupT1), MDBK (bovine kidney), and canine kidney (DK), skin fibroma (A72), and osteosarcoma (D17a) cell lines were used in transduction experiments.
Vector stocks. Transfections were done with the calcium phosphate transient transfection method in two-chamber cell factories (CF2) (26). Medium was changed 12 to 16 h later, and supernatants were collected 48 h thereafter, filtered through a 0.2-μm-pore-size filter, concentrated by ultracentrifugation over a sucrose cushion in a swinging bucket SW28 rotor at 25,000 rpm for 2 h, and aliquoted and frozen at –80°C. FIV-based lentiviral vectors were produced in 293T cells using a 5' U3-substituted system (26, 35, 36). Enhanced green fluorescent protein (eGFP) transfer vector GiNW was cotransfected with the pFP93 packaging construct (26) and vesicular stomatitis virus G protein (VSV-G) envelop construct pMD.G (60). The HIV-1 GFP vectors, which have the same internal promoter (cytomegalovirus [CMV]), were produced by transfection of 293T cells with HIV-1 transfer construct pHR · -hCMV-eGFP (59) containing gfp, HIV-1 packaging construct pCMVR8.91 (60), and pMD.G. For NB-MLV GFP vector, MLV transfer construct pLEGFP (Clontech, Palo Alto, CA) and pMD.G were cotransfected into Phoenix A cells. The EIAV GFP vector SIN6.1eGFPW and the EIAV packaging plasmid pEV53B have been described previously (31, 32). EIAV vectors were produced by transfection of pEV53B, SIN6.1eGFPW, and pCI-VSV-G (31) into 293T cells at a ratio of 1:1:0.6. The day following transfection, the medium was changed to include 10 mM sodium butyrate, and viral supernatants were collected 24 h later.
VLPs. To generate FIV VLPs, packaging construct pFP93 was cotransfected by calcium phosphate coprecipitation with pMD.G into 293T cells in CF2 chambers. For HIV-1 VLPs, HIV-1 packaging construct pCMVR8.91 was cotransfected with pMD.G. For NB-MLV VLPs, pMD.G was transfected into Phoenix A cells. Supernatants were collected 48 h after changing the medium, filtered through a 0.2-μm-pore-size filter, concentrated by ultracentrifugation over a sucrose cushion in a swinging bucket SW28 rotor at 25,000 rpm for 2 h, and then aliquoted and frozen at –80°C. EIAV VLPs were harvested from 293T cells in a manner similar to EIAV GFP vector following transfection of pEV53B and pCI-VSV-G.
Replicating virus stocks. Wild-type, replication-competent FIV was produced in 293T cells by transfection of pCT5, in which replacement of the 5' FIV U3 element by the human CMV immediate-early promoter enables human cell expression (35, 36). The medium was replaced 14 h after plasmid transfection, and supernatants were collected 32 h later. Supernatants were filtered (0.45 μM) and frozen in 500-μl aliquots. Cotransfection of pCT5 with pMD.G was used in the preparation of some stocks to boost the initial round infectivity for CrFK cell replication studies. Titers were determined for virus stocks on CrFK cells using a focal infectivity assay (38) as described previously (35).
Generation of stable TRIM5 allele-expressing cell lines. Stable cell lines expressing TRIM5 alleles were generated with MLV-based retroviral vectors kindly provided by J. Sodroski (49); 7.5 μg of each pLPCX vector encoding hemagglutinin (HA) epitope-tagged or untagged human or rhesus TRIM5 and puromycin resistance was transfected by calcium phosphate coprecipitation with 2.5 μg pMD.G into 1 x 106 Phoenix A cells in 6-cm-diameter dishes. Medium was changed 16 h later, and supernatants were collected 48 h thereafter and filtered. CrFK, D17, 293T, or HT1080 cells were seeded, 2.5 x 105 per well, into three wells of a six-well plate 24 h before transduction. The cells were transduced a second time 12 h later. Selection in 5 μg/ml (for CrFK), 0.3 μg/ml (for D17), 3 μg/ml (for 293T), or 0.5 μg/ml (for HT1080) puromycin (Sigma) was begun 48 h after the second transduction. Lines were robustly polyclonal, with outgrowth of more than 105 colonies from initial selection of each line.
Normalization of vector and virus stocks. Reverse transcriptase (RT) activities were used to normalize viral preparations. Undiluted or diluted supernatants (10 μl) were assayed by mixing with 50 μl of RT cocktail consisting of poly(A) template (5 μg/ml) and pd(t)12-18 oligonucleotide (1.57 μg/ml) in 50 mM Tris (pH 7.8), 7.5 mM KCl, 2 mM dithiothreitol, 5 mM MgCl2, 0.05% NP-40, and 1 μCi [32P]TTP. Samples were incubated in a Falcon flexible plastic 96-well plate for 1.5 h at 37°C. Five microliters of the reaction was blotted onto a Whatman DE-81 ion-exchange filter paper 96-well plate. After drying in a 37°C oven for 15 min, the plate was washed with 2x SSC (1x SSC is 0.15 M NaCl plus 0.015 M sodium citrate) and ethanol using a Filtermate Harvester. The plate was dried in a 37°C oven for 30 min. The bottom of the plate was sealed, and 35 μl of Microscint-O scintillation fluor was pipetted into each well. A TopCount NXT microplate luminescence-scintillation counter (Packard) was used to measure the radioactivity of each sample.
Stable TRIM5 RNAi. Two 63-nucleotide (nt) oligonucleotides encoding a short hairpin RNA (shRNA1) predicted to be processed by Dicer to the small interfering RNA (siRNA1) of Stremlau et al. (49) were synthesized such that annealing resulted in BamHI and HindIII adaptor termini (sense, GATCCGCTCAGGGAGGTCAAGTTGTTCAAGAGACAACTTGACCTCCCTGAGCT TTTTTGGAAA; antisense, AGCTTTTCCAAAAAAGCTCAGGGAGGTCAAGTTGTCTCTTGAACAACTTGACCTCCCTGAGCG. This adaptor was inserted into pSilencer (Ambion). Stable knockdown cell lines were generated as previously described for LEDGF/p75 (24) by stable transfection of plasmid linearized in prokaryotic sequences. A control cell line was developed using the same construct with a control 21-nt scrambled sequence shRNA in place of the 21-nt anti-TRIM5 sequence. Selection was in 0.125 mg hygromycin (Sigma) per ml, starting 48 h after transfection.
Vector transduction of different cell lines. A72, D17a, DK, CrFK, RD, 293T, HT1080, HeLa, H9, MT-4, Molt-4, Jurkat, Vero, FRhK4, 3T3, and MDBK cells, cell lines expressing HA-tagged human and rhesus TRIM5, rhesus TRIM5 shRNA knockdown FRhK4 cells, and scramble shRNA FRhK4 cell lines were seeded in 24-well plates at 45,000 cells/well. Cells were transduced with twofold or fourfold serial dilutions of FIV, HIV-1, EIAV, and NB-MLV GFP vectors. Cells were fixed in 1% formaldehyde and analyzed by fluorescence-activated cell sorting 48 h posttransduction. Mean titers per 1,000 RT units were calculated from the linear ranges of infectivity curves (25-27). In some cases, cells were treated with different doses of As2O3 (0.1 to 12 μM; Sigma) at the time of transduction. As2O3 was prepared as described by Berthoux et al. (2).
Assay for FIV replication. CrFK-derived cell lines that express human or rhesus TRIM5 were infected overnight at multiplicities of infection (MOI) of 0.01, 0.1, and 1.0. Cells were washed three times with phosphate-buffered saline and new medium was added. Supernatants were collected and frozen periodically over 3 weeks and RT activities were determined.
Immunoblotting. Cells were lysed in 150 mM NaCl, 1% NP-40, 0.5% deoxycholate, 0.1% sodium dodecyl sulfate, 50 mM Tris, pH 8.0, containing protease inhibitors (Complete Mini; Roche). After normalization for protein concentration, lysates were combined with ?-mercaptoethanol, boiled, and separated on a 10% sodium dodecyl sulfate-polyacrylamide gel. Proteins were transferred to a polyvinylidene difluoride membrane (Immobilon-P; Millipore) using a Bio-Rad Trans-Blot SD semi-dry electrophoretic transfer cell. The transferred blot was blocked in 10% milk before probing with anti-HA antibody (1:2,000; Novus Biologicals) in 5% milk-0.1% Tween. Peroxidase-conjugated goat anti-mouse immunoglobulin G (1:2,000; Calbiochem) was used as the secondary antibody. Blots were developed with a chemiluminescence kit (Amersham Pharmacia) and exposed to radiographic film.
Confocal immunofluorescence microscopy. CrFK or D17 cell lines (2 x 105) (parental, or lines stably expressing HA-tagged TRIM5hu or TRIM5rh) were seeded into Lab-Tek II chamber slides (Nalge Nunc International). Cells were fixed in 4% paraformaldehyde (Tousimis Research Corp.) then methanol before blocking in 10% fetal bovine serum (Gibco) and 20 mM ammonium chloride (Sigma). Cells were probed with rat anti-HA primary antibody (1:500; Roche) and goat anti-rabbit immunoglobulin G conjugated to Cy3 (1:500; Chemicon International). Slides were sealed with ProLong Gold mounting medium (Molecular Probes) and imaged on a Zeiss LSM 510 confocal laser scanning microscope mounted on an Axiovert 100 M (Zeiss, Thornwood, NY).
RESULTS
Comparative restriction of FIV vectors in cells from primate and nonprimate species. We initially carried out single-round VSV-G pseudotyped FIV vector (26, 27, 36) infections of human (293T, RD, HeLa, HT1080, Jurkat, Molt-4, MT-4, SupT1, and H9), simian (FRhK4 and Vero), feline (CrFK), canine (D17a, A72, and DK), bovine (MDBK), and murine (3T3) cell lines. Each cell line was transduced with eGFP vector over a wide range of inputs in two- or fourfold increments. We observed wide variations in susceptibility, both between and within individual species (Fig. 1A). Some of the variation could be grouped as cell type dependent. For example, human epithelial-derived cell lines (HT1080 and HeLa) were significantly more transducible by FIV vectors than transformed human T-cell lines (Jurkat, H9, MT-4, Molt-4, and SupT1). However, considerable variation was also seen between fibroeptihelial cell lines from the same species. For example, FIV vector transduction efficiencies differed 18-fold between canine DK and D17a cells. Some cell lines were notably permissive (293T, CrFK, RD, D17a, and A72). HeLa, HT1080, MT-4, and 3T3 cells exhibited intermediate resistance to FIV vectors. The simian cell lines (FRhK4 and Vero), the canine DK cell line, and human leukemic T-cell lines (H9, Jurkat, and Molt-4) were significantly more resistant to FIV vector transduction. The most resistant was the bovine MDBK cell line.
The same cell lines were tested in parallel for HIV-1 vector transduction efficiency (Fig. 1B). The HIV-1 vector has the same genetic structure, i.e., with an internal CMV promoter driving expression of an eGFP reporter placed upstream of the 3' long terminal repeat, thus removing promoter activity as a variable. Those cells found to be highly permissive for FIV (293T, CrFK, RD, and A72) were also highly permissive for HIV-1. Most of the intermediate-susceptibility cell lines (HeLa, HT1080, and 3T3) retained the same relative position. However, infectivity curves in certain cell lines shifted markedly to the left compared to FIV (notably, the human T-cell lines H9, Jurkat, MT-4, Molt-4, and SupT1). The simian cell lines (FRhK4 and Vero) were also comparatively resistant to HIV-1, consistent with their previously described Lv1 phenotype (1), as were canine DK cells. Again, the most resistant cell line was MDBK, with FIV and HIV-1 having equivalent infectivity on an RT-normalized basis.
We then compared FIV and HIV-1 vectors as a function of infectivity (transducing units per 103 RT activity units). They were equivalent (no significant difference, P > 0.05) in permissive feline CrFK and human RD cells as well as the more-resistant murine 3T3, canine DK, and bovine MDBK cells (Fig. 1C). Both vectors were comparatively less efficient in the monkey cell lines (FRhK4 and Vero cells), with FIV displaying twofold less infectivity than HIV-1. This approximate twofold difference was also seen in 293T, A72, HeLa, and HT1080 cells. Both vectors were markedly impaired in H9 cells, consistent with prior observations that H9 cells are relatively resistant to HIV-1 infection but allow chronic viral production once infected (59). Interestingly, we documented a strong bias for HIV-1 vectors over FIV in all of the human T-cell lines tested. HIV-1 vectors were 5-, 5-, 9.5-, 12.4-, and 26-fold more efficient than FIV in H9, MT-4, Jurkat, SupT1, and Molt-4 cells respectively.
Human and rhesus TRIM5 restrict FIV. After characterizing these differential restriction phenotypes for FIV versus HIV-1, we asked whether TRIM5 proteins could account for them. Since CrFK cells are widely reported to lack postentry restriction factors (7, 11, 15, 51), which is consistent with the data in Fig. 1, we stably expressed HA epitope-tagged and untagged TRIM5hu and TRIM5rh in this cell line, using MLV-based retroviral vectors prepared with the plasmids of Stremlau et al. (49). The four cell lines were selected in puromycin, and HA-tagged TRIM5 expression was verified by immunoblotting (Fig. 2A). Expression of the human and primate proteins was equivalent (Fig. 2A; corroborated in immunofluorescence analyses shown subsequently in Fig. 3). In addition, these cell lines were robustly polyclonal with respect to MLV vector insertion, representing outgrowth of more than 105 independent colonies from each transduction. FIV, HIV-1, EIAV, and NB-MLV GFP vector transductions were carried out, with twofold dilutions over a wide range. FIV infectivity shifted markedly to the right in CrFK cells expressing TRIM5hu or TRIM5rh (Fig. 2B). Rhesus TRIM5 caused a greater shift (64.6- ± 6.5-fold) than human TRIM5 (8.72- ± 2.4-fold). Similar to the findings of Hatziioannou et al. (15), analogously constructed vectors derived from the ungulate lentivirus EIAV (31) were shifted 25.9-fold and 11.6-fold in rhesus and human TRIM5-expressing CrFK cells, respectively, compared to parental CrFK cells. While HIV-1 was also strongly shifted (59.96- ± 16.6-fold to the right) by rhesus TRIM5, it was only weakly (approximately 2.4- ± 1.3-fold) blocked by human TRIM5, which is in agreement with prior results in HeLa (49) and CrFK (15) cells. In contrast, the negative control vector NB-MLV was unaffected by either rhesus or human TRIM5, establishing specificity (Fig. 2B). Furthermore, the same restricting activities were seen in CrFK cells expressing TRIM5 proteins lacking the HA tag, although lack of a good antibody to the native proteins prevented verification of equal expression (data not shown).
Subcellular distributions of human and rhesus TRIM5 proteins. Immunofluorescence analyses of the stable CrFK cell lines using anti-HA antibody showed similar, abundant expression and cytoplasmic distributions of human and rhesus TRIM5 (Fig. 3, top panels). The finely punctate foci we observed in stable cell lines are somewhat more diffusely distributed and smaller than the large and more-discrete structures, termed "cytoplasmic bodies," that were identified after transient transfection of a TRIM5-GFP fusion cDNA (39). In addition, we noted some variation on this point. For example, rhesus TRIM5 had a more-diffuse (although still heterogeneous) distribution in CrFK cells than did the human protein (Fig. 3); moreover, a minority of cells lacking discernible large foci in favor of more diffuse patterns could be found in each cell line and in the canine and human cell lines described below.
Canine D17 cells do not support restriction by introduced human TRIM5. Like feline CrFK cells, canine D17 cells are also reported to lack restriction factors and are highly permissive for HIV-1, FIV, and EIAV transduction (reference 17 and Fig. 1). Examinations of current drafts of the dog genome have not revealed any TRIM5 ortholog (45). We therefore derived stable D17 cell lines with the same TRIM5 retroviral vectors used for CrFK cells. Intracellular immunoreactivities and localizations of TRIM5hu and TRIM5rh were indistinguishable from CrFK cells (Fig. 3, middle panels). However, clearly different viral infectivity phenotypes were seen in these cell lines (Fig. 4A). Human TRIM5 did not restrict FIV or HIV-1 vectors (1.37- and 0.99-fold change in titer for FIV and HIV-1, respectively). Rhesus TRIM5 also had lesser effects in the canine cells than in the feline cells (2.21- and 2.10-fold decrease in titer for FIV and HIV-1, respectively). We additionally tested for restriction by TRIM5rh when retrovirally expressed in the relatively permissive human cell lines 293T and HT1080. TRIM5rh was similarly distributed in 293T cells (Fig. 3, bottom panels) and was able to restrict appreciably (Fig. 4B; 9.54- and 8.64-fold decrease in titer for FIV and HIV-1, respectively). TRIM5rh also restricted significantly in HT1080 cells (Fig. 4C; 35.01- and 35.64-fold decrease in titer for FIV and HIV-1, respectively). Again, NB-MLV titer was unaffected in any cell line.
Human and rhesus TRIM5 block FIV replication. We infected CrFK cells, which are permissive for FIV replication, and the CrFK-derived lines expressing TRIM5 proteins with replication-competent FIV at three MOI. Consistent with the single-round vector analyses, spreading infection was severely impaired in CrFK cells expressing either tagged or untagged TRIM5hu or TRIM5rh at MOI 0.01, 0.1, and 1.0 compared to parental CrFK cells (Fig. 5). Some RT activity accumulated in supernatants collected at 3 weeks from CrFK cells expressing untagged TRIM5rh at MOI 0.01 and from CrFK cells expressing HA-tagged and untagged TRIM5hu at MOI of 1.0 and 0.1. However, when passaged to fresh cells, further production of RT activity was not detected even after maintenance of log-phase cultures for 12 more weeks (data not shown). In contrast, transfer of these supernatants to parental CrFK cells did result in rescue of viral replication, which propagated to wild-type peak RT activity levels (data not shown). Some RT activity was also seen in supernatants from TRIM5hu- but not TRIM5rh-expressing CrFK cells infected at a high MOI, but these supernatants also yielded no persistent or amplified RT activity when passed to fresh TRIM5-expressing cells (data not shown). These data suggest that escape mutants were not generated, but that some wild-type virus was produced in TRIM5-expressing cell lines from integrated proviral progeny of virions that evaded the initial postentry restriction.
RNAi against endogenous TRIM5 in rhesus FRhK4 cells. Polymerase III-promoted shRNA constructs were derived and screened for RNAi activity against rhesus TRIM5, using Western blotting of transiently transfected HA-tagged protein. When effective transient knockdown was verified (data not shown), stable selection of rhesus FRhK4 cells was carried out. This had a strong positive effect on FIV and HIV-1 vector transduction (15.62- and 16.03-fold increases in titer, respectively) compared to the scrambled shRNA control cell line derived in parallel and to the parental cell line (Fig. 6). In contrast to these pronounced leftward shifts in lentiviral infectivity curves, NB-MLV was completely unaffected by the knockdown, which demonstrates specificity (Fig. 6). These results corroborate the specific restriction of FIV by TRIM5rh seen in our exogenous expression studies.
VLP cross-saturations of human and rhesus TRIM5. The restricting activity of both TRIM5 alleles was then tested for titration by coinfecting VLPs, a signature property that is consistent with saturability of the cellular restriction factors (two hit kinetic model) (6). In FRhK4 cells, both FIV and HIV-1 VLPs titrated endogenous rhesus TRIM5 (Fig. 7A). With HIV-1 VLPs, maximum increases were 23.0-fold and 16.5-fold for FIV and HIV-1, respectively; with FIV VLPs, the maximum increases were 12.7- and 6.43-fold. NB-MLV VLPs did not titrate the restrictions.
A comprehensive series of cross-saturation experiments were done next, with FIV, HIV-1, EIAV, and NB-MLV VLPs added at the time of transduction with FIV, HIV-1, EAIV, and NB-MLV GFP vectors of each TRIM5-expressing CrFK cell line (Fig. 7B). The results for rhesus TRIM5rh are summarized in Fig. 7B (top panel). FIV VLPs consistently increased transduction for each restricted (FIV, HIV-1, and EIAV) vector but had no effect on unrestricted NB-MLV GFP transduction. HIV-1 VLPs had an apparently greater effect on FIV transduction (8.5-fold increase) than HIV-1 and EIAV transduction (3.3- and 2.6-fold increases, respectively). EIAV VLPs also augmented transduction by all restricted vectors (3.9-, 4.7-, and 7.4-fold increases for HIV-1, FIV, and EIAV vector, respectively).
VLPs had a lesser effect in TRIM5hu-expressing CrFK cells (Fig. 7B, bottom panel). FIV VLPs had the strongest effect for restriction of FIV (3.7-fold increase), but they also increased HIV-1 and EIAV transduction (2.4- and 2.2-fold increases, respectively). EIAV VLPs were similar to FIV VLPs (3.6-fold increase) in relieving restriction of FIV. EIAV VLPs relieved restriction of EIAV more effectively than restriction of HIV-1 (3.5- versus 1.7-fold increase). NB-MLV GFP vector was not significantly affected by any species of VLP in any cell lines. Likewise, NB-MLV VLPs had no significant effect on the transduction efficiency of any of the vectors.
Effects of arsenic trioxide. The mitochondrial membrane destabilizing agent As2O3 has been reported to stimulate HIV-1 replication as well as increase HIV-1 and EIAV vector transduction of human cells (1, 2, 54). The mechanism is unknown, but As2O3 alters the subcellular localization and degradation of PML, another TRIM family member (54). We examined the effects of As2O3 on FIV, HIV-1, and NB-MLV GFP vector transduction in CrFK, CrFK-TRIM5hu, CrFK-TRIM5rh, HT1080, Vero, and FRhK4 cells (Fig. 8).
The effects were complex. Although As2O3 did increase FIV vector transduction of CrFK-TRIM5hu and CrFK-TRIM5rh cells (P < 0.04), transduction efficiency also increased commensurately in parental CrFK cells, indicating that in this cell line the effect is not TRIM5 specific (Fig. 8, top panels). It was, however, lentivirus-specific (NB-MLV was not affected). The magnitude of As2O3-induced increases were similar for FIV and HIV-1 vectors at low (1 to 2% GFP-positive cells at baseline) and higher (5 to 10% GFP-positive cells at baseline) vector inputs in CrFK-TRIM5hu cells. In addition, similar stimulation of transduction by both FIV and HIV vectors were seen in multiple human cell lines (HeLa, 293T, RD, and MT-4; data not shown). NB-MLV vectors were unaffected by As2O3 in any cell line. In contrast to human TRIM5, the effects of As2O3 on lentiviral vectors were input dependent as well as vector type dependent for rhesus TRIM5. This was seen in both FRhK4 cells and CrFK-TRIM5rh cells. With low vector input (1 to 2% GFP-positive cells at baseline), As2O3 treatment increased FIV but not HIV-1 vector transduction. Results were similar in FRhK4 cells. However, at higher vector input (5 to 10% GFP-positive cells) in either cell line, As2O3 treatment had no significant effect for either lentiviral vector in CrFK-TRIM5rh cells (data not shown). Again, NB-MLV transduction efficiency was unaffected by As2O3 treatment at both vector inputs. In Vero and HT1080 cells, both HIV-1 and FIV transductions were stimulated; note however that in both simian cell lines (Vero and FRhK4) significant stimulation required in excess of 2 μM, concentrations which were quite toxic to nonsimian cell lines (data not shown). We observed no significant effect for any of the vectors in 3T3 cells treated with 2.0 μM As2O3 (data not shown).
DISCUSSION
We have carried out a detailed first analysis of postentry restriction of FIV, extending such analyses to the third main group of lentiviruses. Although FIV-based lentiviral vectors have been shown to transduce primate cells and tissues with good practical efficiency (9, 21, 25, 27, 28, 36, 55), the species-specific parameters governing efficiency relative to HIV-1 have not been defined quantitatively, and the activities of known postentry restriction factors have not been characterized. Our data show that both Lv1 and Ref1 restrict FIV and that the major patterns can be accounted for by TRIM5. FIV is substantially restricted by both human and rhesus proteins, although considerably more so by the rhesus protein. In this discussion, we highlight novel virological aspects and then consider ramifications for human gene therapy with FIV and other nonprimate lentiviral vectors.
The extent of inhibition of HIV-1 and EIAV in our CrFK cells expressing TRIM5 variants agrees well with results obtained in CrFK lines generated independently by others (15), which supports the validity of the FIV findings. We were surprised to find in further experiments that canine D17 cells did not support significant restriction by exogenously introduced human TRIM5. Determining the reason for this difference will require further investigation. We verified that epitope-tagged TRIM5 proteins localize to similar, discrete structures consistent with cytoplasmic bodies in both canine and feline cells and that intact, full-length proteins were expressed at substantial levels in both (Fig. 3 and 4A). As previously observed by Javanbakht et al. (18) in a different canine cell line (Cf2Th thymocytes), we found that D17 cells did support some rhesus allele activity. The human TRIM5 allele was not tested in Cf2Th cells (18). It could be that some canine cells lack an additional factor or factors required for optimal TRIM5 antiviral function. Alternatively, since some TRIM variants (e.g., TRIM5) have transdominant activity (49), a canine factor might interfere with TRIM5 function.
Compared to HIV-1 vectors in RT activity-normalized infections, human cell susceptibility to FIV vectors varied from equivalent to HIV-1 (RD cells) to significantly less than HIV-1 (human T-cell lines). An interesting question that will require further study is whether TRIM5hu fully explains the relative restriction of FIV in human T-cell lines. The cell type-dependent pattern of restriction could be due to different TRIM5 expression levels or subcellular localization patterns or to other factors. No data have been published concerning relative endogenous TRIM5 expression levels. In rhesus FRhK4 cells, however, we found that stable RNAi against rhesus TRIM5 resulted in marked increases in FIV and HIV-1 infectivities but not NB-MLV infectivity, suggesting that rhesus TRIM5 is largely responsible for postentry restriction in these cells. Unlike FRhK4 cells, we could not significantly reverse restriction of FIV with stable shRNA expression in HeLa cells or human T-cell lines (data not shown). We do not know if this represented inadequate knockdown, because we have been unable to quantify endogenous TRIM5 protein levels with commercially available antibodies, which we have found to be inadequately sensitive and specific in immunoblotting (data not shown). Since T lymphocytes are prime targets for many retroviruses, evolution of stronger innate T-cell antiviral mechanisms in T cells would be advantageous to the primate host.
In contrast to primate cell lines, RT activity-normalized HIV-1 and FIV vectors had equivalent infectivities in nonprimate cell lines, with the concordant transduction efficiencies nevertheless varying greatly by cell type. For example, 18-fold variability in susceptibility was seen between canine cell lines (A72 compared to D17 and DK). Bovine MDBK cells are very resistant to both HIV-1 and FIV vectors.
In primate epithelial cell lines (HT1080, HeLa, 293T, and Vero), the absolute number of transducing units per RT unit were slightly lower (two to threefold) for FIV. However, comparing absolute infectivities for two different lentiviruses is problematic since the RT assay could score enzymatic activity differently for each reverse transcriptase, or be biased towards one or the other RT according to particular assay components (e.g., magnesium, pH, or ionic strength), or vary with the percentage of defective particles in a given preparation. Thus, it is not known whether the RT assay faithfully normalizes particle number in absolute terms.
However, it is important to appreciate that key aspects of these studies rule out any incidental effects of particle normalization or of the particular vector construct versions used. First, the variance in the ratios of FIV/HIV-1 vector infectivities, from a ratio of 1 in cells lacking primate restriction factors (CrFK, D17a, DK, 3T3, and MDBK) to considerably more than 1, e.g., 5 to 26, in human T-cell lines, securely identifies host cell factors as operative. The second aspect is the virus-specific restrictions conferred by selective expression of allelic TRIM5 proteins in the CrFK cell background. Expression of human TRIM5 was sufficient to markedly restrict FIV but not HIV-1 in engineered CrFK cell lines. Expression of rhesus TRIM5 restricted both lentiviruses to the same extent, and neither protein restricted NB-MLV. HIV-1 and FIV vectors are equivalent on an RT-normalized basis in restriction factor-lacking CrFK cells (and in human RD cells). In contrast, FIV vectors were significantly less infectious than HIV-1 vectors in CrFK-TRIM5hu cells and in human T-cell lines. This relative restriction of FIV (and EIAV) versus HIV-1 vectors in some but not all human cells is thus shown to be intrinsic to the viral biology. Insofar as it may be an issue for human gene therapy, it is not addressed by extant vector systems. Practical implications of Ref1 restriction for gene therapy are discussed below.
We found that arsenic trioxide had effects which are complex and not clearly TRIM5 dependent, preventing straightforward interpretation of the effects of this drug. While in our CrFK cell lines that express human or rhesus TRIM5, As2O3 did have generally stimulatory effects on both FIV and HIV-1 transduction, and NB-MLV was unaffected, these effects were cell type, vector type, and input amount dependent. Most importantly, arsenic enhancement is TRIM5 independent in the CrFK background because equivalent stimulation occurred in the parental CrFK cells. Our data on the effects of As2O3 on HIV-1 are, however, in good agreement with a number of observations of Berthoux et al. (1; compare Fig. 8 with Fig. 7 therein), i.e., we saw no effect at all in rhesus FRhK4 cells and a significant stimulation in AGM (Vero) cells, with a peak at 8 μM. At 12 μM, we began to observe significant toxicity in Vero and FRhK4 cells. As also observed previously by Berthoux et al. (1), these simian cells were the only cell type in which we could raise the As2O3 concentration above 2 μM without prohibitive toxicity (data not shown). Whether stimulation of infection in Vero cells can be securely attributed to an effect on AGM TRIM5 remains unclear; for example, restriction of N-MLV by AGM TRIM5 expressed in murine cells was not reversed by arsenic trioxide, whereas restriction by human TRIM5 was reversed (22). The individual lentiviruses differed as well in response to As2O3: FIV but not HIV-1 vector transduction was augmented in FRhK4 cells by 2 μM As2O3, and FIV was more stimulated than HIV-1 in CrFK-TRIM5rh cells.
Viral restriction determinants have been mapped to the exposed capsid surface for HIV-1 and N-MLV (11, 23, 33, 47, 51, 53). This region is quite different in FIV CA, which lacks the proline-rich cyclophilin A binding loop present in HIV-1 (10, 53). Interaction between this loop and cyclophilin A has been reported to protect HIV-1 from restriction by Ref1 (14, 53). Species-specific retroviral CA recognition appears to be mediated by the TRIM5 SPRY(B30.2) domain (41, 46, 50, 58), which may mediate degradation of the incoming viral core or reverse transcription complex or may alter kinetics of core disassembly. Direct interaction between the SPRY domain and CA was suggested by the recent discovery that a New World (owl) monkey TRIM5 contains a fusion to the complete cyclophilin A gene in place of the SPRY domain (30, 42) together with the finding that human cyclophilin A acts to interfere with Ref1 restriction by binding to an exposed proline-rich loop in HIV-1 CA (53). Recently, binding between a human TRIM5-GST fusion protein and N-MLV CA presented by detergent-stripped virion cores was demonstrated; SPRY domain deletion or use of B-MLV rather than N-MLV abrogated this binding (43). In contrast, free CA protein did not interact, suggesting that the SPRY domain recognizes a complex structure present at the face of multimerized CA (43). It may therefore be of interest to map specific determinants of restriction in FIV CA.
Transmission of FIV from cats to nonfelines has not been observed, despite presumably prevalent natural exposure (interfeline transmission is by biting). To some extent, this restricted tropism can be accounted for by quite limited FIV long terminal repeat activity in human cells (35, 36) and by different primary receptor usage (44), although some FIV isolates can also use human CXCR4 for entry (35, 56). High MOI infection with FIV can lead to infection and even lytic FIV envelope glycoprotein expression but not viral replication in CXCR4-positive human cells (35, 56). Consistent with this, one group has reported expression of FIV RNA and proteins after infection of human peripheral blood mononuclear cells as well, although passage of virus or continuous propagation were not demonstrated (19). Two rhesus macaques subsequently injected with FIV-infected autologous peripheral blood mononuclear cells were reported to show transient PCR positivity for RNA and DNA, with both animals becoming negative after 4 and 9 weeks, suggesting a self-limited viremia due to FIV (20). This latter result is surprising in view of the degree of restriction by rhesus TRIM5 shown here to an FIV clone that is also derived from the Petaluma strain that was used to infect the macaques and the inability to achieve significant HIV-1 replication in these primates in numerous other studies. Antibodies to FIV were not produced in one of these animals, however, and passage to other animals was not done.
The present results have implications for human gene therapy with FIV and EIAV vectors. Each have been shown to achieve useful levels of gene transfer in primate cells and tissues (reviewed in reference 37). Nevertheless, further development of strategies that employ lentiviral vectors will benefit from informed appraisal of the consequences of Ref1 restriction, which may be advantageous or disadvantageous depending on the application. In some situations, Ref1/human TRIM5 restriction will reduce the absolute efficiency of gene transfer by FIV or EIAV vectors. For example, intravenous administration of nonprimate lentivirus vectors may lead to dilution of particles to levels that are below Ref1/TRIM5 saturation thresholds. Such restriction may be surmountable by dosage adjustment, or simply by adding VLPs. In many gene therapy situations in vivo or ex vivo, however, localized inputs without VLPs result in particle concentrations well above those needed to saturate Ref1/TRIM5 activity. In this common situation, or where a desired target tissue expresses little TRIM5, Ref1 restriction would provide a net advantage: a highly evolved innate barrier to systemic propagation of replication-competent retroviruses, which are among the most concerning potential complications of retroviral vector-based therapy. In addition, relative levels of TRIM5 expression in different human tissues that are gene therapy targets are not well understood. Development of good antibodies to TRIM5 will be useful in determining which human tissues express significant amounts of this protein and will further clarify the implications of Ref1 for gene therapy. Similarly, the intracellular distributions we observed with the HA-tagged proteins cannot be taken as definitive until endogenous TRIM5 is characterized properly. Such studies may clarify whether overexpression (likely to be extreme with transient transfection, but probably less so with retroviral vectors as used here) and/or the use of GFP fusions accounts for the larger structures previously described (39).
Finally, it is not yet known exactly how TRIM5 proteins interact with capsid proteins, cyclophilins, and perhaps other factors to result in the observed restrictions. We can speculate that progress in understanding the actual molecular mechanisms, together with virus-specific determinant mapping and mutagenesis, could enable specialized engineering of nonprimate lentiviral vector capsid proteins for greater per-particle efficacy in TRIM5hu-expressing cells. Such modifications would, of course, negate the advantage of the innate block to systemic replication-competent retrovirus replication.
ACKNOWLEDGMENTS
We thank J. Sodroski and M. Stremlau for providing pLPCX vector plasmids encoding human and rhesus TRIM5, A. Fielding for D17a cells, Y. Ikeda for A72, DK, and MDBK cells, T. Whitwam for help with vector preparation and titration on multiple cell lines, and M. Peretz for technical assistance.
REFERENCES
Berthoux, L., S. Sebastian, E. Sokolskaja, and J. Luban. 2004. Lv1 inhibition of human immunodeficiency virus type 1 is counteracted by factors that stimulate synthesis or nuclear translocation of viral cDNA. J. Virol. 78:11739-11750.
Berthoux, L., G. J. Towers, C. Gurer, P. Salomoni, P. P. Pandolfi, and J. Luban. 2003. As2O3 enhances retroviral reverse transcription and counteracts Ref1 antiviral activity. J. Virol. 77:3167-3180.
Besnier, C., Y. Takeuchi, and G. Towers. 2002. Restriction of lentivirus in monkeys. Proc. Natl. Acad. Sci. USA 99:11920-11925.
Besnier, C., L. Ylinen, B. Strange, A. Lister, Y. Takeuchi, S. P. Goff, and G. J. Towers. 2003. Characterization of murine leukemia virus restriction in mammals. J. Virol. 77:13403-13406.
Best, S., P. Le Tissier, G. Towers, and J. P. Stoye. 1996. Positional cloning of the mouse retrovirus restriction gene Fv1. Nature 382:826-829.
Bieniasz, P. D. 2003. Restriction factors: a defense against retroviral infection. Trends Microbiol. 11:286-291.
Bock, M., K. N. Bishop, G. Towers, and J. P. Stoye. 2000. Use of a transient assay for studying the genetic determinants of Fv1 restriction. J. Virol. 74:7422-7430.
Cowan, S., T. Hatziioannou, T. Cunningham, M. A. Muesing, H. G. Gottlinger, and P. D. Bieniasz. 2002. Cellular inhibitors with Fv1-like activity restrict human and simian immunodeficiency virus tropism. Proc. Natl. Acad. Sci. USA 99:11914-11919.
Curran, M. A., S. M. Kaiser, P. L. Achacoso, and G. P. Nolan. 2000. Efficient transduction of nondividing cells by optimized feline immunodeficiency virus vectors. Mol. Ther. 1:31-38.
Gamble, T. R., F. F. Vajdos, S. Yoo, D. K. Worthylake, M. Houseweart, W. I. Sundquist, and C. P. Hill. 1996. Crystal structure of human cyclophilin A bound to the amino-terminal domain of HIV-1 capsid. Cell 87:1285-1294.
Hatziioannou, T., S. Cowan, and P. D. Bieniasz. 2004. Capsid-dependent and -independent postentry restriction of primate lentivirus tropism in rodent cells. J. Virol. 78:1006-1011.
Hatziioannou, T., S. Cowan, S. P. Goff, P. D. Bieniasz, and G. J. Towers. 2003. Restriction of multiple divergent retroviruses by Lv1 and Ref1. EMBO J. 22:385-394.
Hatziioannou, T., S. Cowan, U. K. Von Schwedler, W. I. Sundquist, and P. D. Bieniasz. 2004. Species-specific tropism determinants in the human immunodeficiency virus type 1 capsid. J. Virol. 78:6005-6012.
Hatziioannou, T., D. Perez-Caballero, S. Cowan, and P. D. Bieniasz. 2005. Cyclophilin interactions with incoming human immunodeficiency virus type 1 capsids with opposing effects on infectivity in human cells. J. Virol. 79:176-183.
Hatziioannou, T., D. Perez-Caballero, A. Yang, S. Cowan, and P. D. Bieniasz. 2004. Retrovirus resistance factors Ref1 and Lv1 are species-specific variants of TRIM5alpha. Proc. Natl. Acad. Sci. USA 101:10774-10779.
Hofmann, W., D. Schubert, J. LaBonte, L. Munson, S. Gibson, J. Scammell, P. Ferrigno, and J. Sodroski. 1999. Species-specific, postentry barriers to primate immunodeficiency virus infection. J. Virol. 73:10020-10028.
Ikeda, Y., M. K. Collins, P. A. Radcliffe, K. A. Mitrophanous, and Y. Takeuchi. 2002. Gene transduction efficiency in cells of different species by HIV and EIAV vectors. Gene Ther. 9:932-938.
Javanbakht, H., F. Diaz-Griffero, M. Stremlau, Z. Si, and J. Sodroski. 2005. The contribution of RING and B-box 2 domains to retroviral restriction mediated by monkey TRIM5alpha. J. Biol. Chem. 280:26933-26940.
Johnston, J., and C. Power. 1999. Productive infection of human peripheral blood mononuclear cells by feline immunodeficiency virus: implications for vector development. J. Virol. 73:2491-2498.
Johnston, J. B., M. E. Olson, E. W. Rud, and C. Power. 2001. Xenoinfection of nonhuman primates by feline immunodeficiency virus. Curr. Biol. 11:1109-1113.
Johnston, J. C., M. Gasmi, L. E. Lim, J. H. Elder, J. K. Yee, D. J. Jolly, K. P. Campbell, B. L. Davidson, and S. L. Sauter. 1999. Minimum requirements for efficient transduction of dividing and nondividing cells by feline immunodeficiency virus vectors. J. Virol. 73:4991-5000.
Keckesova, Z., L. M. Ylinen, and G. J. Towers. 2004. The human and African green monkey TRIM5alpha genes encode Ref1 and Lv1 retroviral restriction factor activities. Proc. Natl. Acad. Sci. USA 101:10780-10785.
Kozak, C. A., and A. Chakraborti. 1996. Single amino acid changes in the murine leukemia virus capsid protein gene define the target of Fv1 resistance. Virology 225:300-305.
Llano, M., M. Vanegas, O. Fregoso, D. Saenz, S. Chung, M. Peretz, and E. M. Poeschla. 2004. LEDGF/p75 determines cellular trafficking of diverse lentiviral but not murine oncoretroviral integrase proteins and is a component of functional lentiviral pre-integration complexes. J. Virol. 78:9524-9537.
Loewen, N., C. Bahler, W. Teo, T. Whitwam, M. Peretz, R. Xu, M. Fautsch, D. H. Johnson, and E. M. Poeschla. 2002. Preservation of aqueous outflow facility after second-generation FIV vector-mediated expression of marker genes in anterior segments of human eyes. Investig. Ophthalmol. Vis. Sci. 43:3686-3690.
Loewen, N., R. Barraza, T. Whitwam, D. Saenz, I. Kemler, and E. Poeschla. 2003. FIV vectors. Methods Mol. Biol. 229:251-271.
Loewen, N., M. Fautsch, M. Peretz, C. Bahler, J. D. Cameron, D. H. Johnson, and E. M. Poeschla. 2001. Genetic modification of human trabecular meshwork with lentiviral vectors. Hum. Gene Ther. 12:2109-2119.
Lotery, A. J., T. A. Derksen, S. R. Russell, R. F. Mullins, S. Sauter, L. M. Affatigato, E. M. Stone, and B. L. Davidson. 2002. Gene transfer to the nonhuman primate retina with recombinant feline immunodeficiency virus vectors. Hum. Gene Ther. 13:689-696.
Munk, C., S. M. Brandt, G. Lucero, and N. R. Landau. 2002. A dominant block to HIV-1 replication at reverse transcription in simian cells. Proc. Natl. Acad. Sci. USA 99:13843-13848.
Nisole, S., C. Lynch, J. P. Stoye, and M. W. Yap. 2004. A Trim5-cyclophilin A fusion protein found in owl monkey kidney cells can restrict HIV-1. Proc. Natl. Acad. Sci. USA 101:13324-13328.
Olsen, J. C. 1998. Gene transfer vectors derived from equine infectious anemia virus. Gene Ther. 5:1481-1487.
O'Rourke, J. P., H. Hiraragi, K. Urban, M. Patel, J. C. Olsen, and B. A. Bunnell. 2003. Analysis of gene transfer and expression in skeletal muscle using enhanced EIAV lentivirus vectors. Mol. Ther. 7:632-639.
Owens, C. M., P. C. Yang, H. Gottlinger, and J. Sodroski. 2003. Human and simian immunodeficiency virus capsid proteins are major viral determinants of early, postentry replication blocks in simian cells. J. Virol. 77:726-731.
Perron, M. J., M. Stremlau, B. Song, W. Ulm, R. C. Mulligan, and J. Sodroski. 2004. TRIM5alpha mediates the postentry block to N-tropic murine leukemia viruses in human cells. Proc. Natl. Acad. Sci. USA 101:11827-11832.
Poeschla, E. M., and D. J. Looney. 1998. CXCR4 is required by a nonprimate lentivirus: heterologous expression of feline immunodeficiency virus in human, rodent, and feline cells. J. Virol. 72:6858-6866.
Poeschla, E., F. Wong-Staal, and D. Looney. 1998. Efficient transduction of nondividing cells by feline immunodeficiency virus lentiviral vectors. Nat. Med. 4:354-357.
Poeschla, E. M. 2003. Non-primate lentiviral vectors. Curr. Opin. Mol. Ther. 5:529-540.
Remington, K. M., Y. Q. Zhu, T. R. Phillips, and T. W. North. 1994. Rapid phenotypic reversion of zidovudine-resistant feline immunodeficiency virus without loss of drug-resistant reverse transcriptase. J. Virol. 68:632-637.
Reymond, A., G. Meroni, A. Fantozzi, G. Merla, S. Cairo, L. Luzi, D. Riganelli, E. Zanaria, S. Messali, S. Cainarca, A. Guffanti, S. Minucci, P. G. Pelicci, and A. Ballabio. 2001. The tripartite motif family identifies cell compartments. EMBO J. 20:2140-2151.
Saenz, D. T., and E. M. Poeschla. 2004. FIV: from lentivirus to lentivector. J. Gene Med. 6(Suppl. 1):S95-S104.
Sawyer, S. L., L. I. Wu, M. Emerman, and H. S. Malik. 2005. Positive selection of primate TRIM5alpha identifies a critical species-specific retroviral restriction domain. Proc. Natl. Acad. Sci. USA 102:2832-2837.
Sayah, D. M., E. Sokolskaja, L. Berthoux, and J. Luban. 2004. Cyclophilin A retrotransposition into TRIM5 explains owl monkey resistance to HIV-1. Nature 430:569-573.
Sebastian, S., and J. Luban. 2005. TRIM5alpha selectively binds a restriction-sensitive retroviral capsid. Retrovirology 2:40.
Shimojima, M., T. Miyazawa, Y. Ikeda, E. L. McMonagle, H. Haining, H. Akashi, Y. Takeuchi, M. J. Hosie, and B. J. Willett. 2004. Use of CD134 as a primary receptor by the feline immunodeficiency virus. Science 303:1192-1195.
Song, B., B. Gold, C. O'hUigin, H. Javanbakht, X. Li, M. Stremlau, C. Winkler, M. Dean, and J. Sodroski. 2005. The B30.2(SPRY) domain of the retroviral restriction factor TRIM5 exhibits lineage-specific length and sequence variation in primates. J. Virol. 79:6111-6121.
Song, B., H. Javanbakht, M. Perron, D. H. Park, M. Stremlau, and J. Sodroski. 2005. Retrovirus restriction by TRIM5 variants from Old World and New World primates. J. Virol. 79:3930-3937.
Stevens, A., M. Bock, S. Ellis, P. LeTissier, K. N. Bishop, M. W. Yap, W. Taylor, and J. P. Stoye. 2004. Retroviral capsid determinants of Fv1 NB and NR tropism. J. Virol. 78:9592-9598.
Stoye, J. P. 2002. An intracellular block to primate lentivirus replication. Proc. Natl. Acad. Sci. USA 99:11549-11551.
Stremlau, M., C. M. Owens, M. J. Perron, M. Kiessling, P. Autissier, and J. Sodroski. 2004. The cytoplasmic body component TRIM5alpha restricts HIV-1 infection in Old World monkeys. Nature 427:848-853.
Stremlau, M., M. Perron, S. Welikala, and J. Sodroski. 2005. Species-specific variation in the B30.2(SPRY) domain of TRIM5 determines the potency of human immunodeficiency virus restriction. J. Virol. 79:3139-3145.
Towers, G., M. Bock, S. Martin, Y. Takeuchi, J. P. Stoye, and O. Danos. 2000. A conserved mechanism of retrovirus restriction in mammals. Proc. Natl. Acad. Sci. USA 97:12295-12299.
Towers, G., M. Collins, and Y. Takeuchi. 2002. Abrogation of Ref1 retrovirus restriction in human cells. J. Virol. 76:2548-2550.
Towers, G. J., T. Hatziioannou, S. Cowan, S. P. Goff, J. Luban, and P. D. Bieniasz. 2003. Cyclophilin A modulates the sensitivity of HIV-1 to host restriction factors. Nat. Med. 9:1138-1143.
Turelli, P., V. Doucas, E. Craig, B. Mangeat, N. Klages, R. Evans, G. Kalpana, and D. Trono. 2001. Cytoplasmic recruitment of INI1 and PML on incoming HIV preintegration complexes: interference with early steps of viral replication. Mol. Cell 7:1245-1254.
Wang, G., V. Slepushkin, J. Zabner, S. Keshavjee, J. C. Johnston, S. L. Sauter, D. J. Jolly, T. W. Dubensky, Jr., B. L. Davidson, and P. B. McCray, Jr. 1999. Feline immunodeficiency virus vectors persistently transduce nondividing airway epithelia and correct the cystic fibrosis defect. J. Clin. Investig. 104:R55-R62.
Willett, B. J., L. Picard, M. J. Hosie, J. D. Turner, K. Adema, and P. R. Clapham. 1997. Shared usage of the chemokine receptor CXCR4 by the feline and human immunodeficiency viruses. J. Virol. 71:6407-6415.
Yap, M. W., S. Nisole, C. Lynch, and J. P. Stoye. 2004. Trim5alpha protein restricts both HIV-1 and murine leukemia virus. Proc. Natl. Acad. Sci. USA 101:10786-10791.
Yap, M. W., S. Nisole, and J. P. Stoye. 2005. A single amino acid change in the SPRY domain of human Trim5-alpha leads to HIV-1 restriction. Curr. Biol. 15:73-78.
Zennou, V., C. Petit, D. Guetard, U. Nerhbass, L. Montagnier, and P. Charneau. 2000. HIV-1 genome nuclear import is mediated by a central DNA flap. Cell 101:173-185.
Zufferey, R., D. Nagy, R. J. Mandel, L. Naldini, and D. Trono. 1997. Multiply attenuated lentiviral vector achieves efficient gene delivery in vivo. Nat. Biotechnol. 15:871-875.(Dyana T. Saenz, Wulin Teo)