Tetherin inhibits prototypic foamy virus release

PFV full-length infectious clone pcPFV was kindly provided by Maxine L. Linial [27]. Flag-HuTHN and Flag-AgmTHN was subcloned from pcDNA3.1-HuTHN and pcDNA3.1-AgmTHN into pCMV-Tag2B (Stratagene) vector. The BovTHN cDNA and CanTHN cDNA were amplified by reverse transcription (RT)-PCR from RNA samples that were extracted from fetal bovine lung (FBL) cells and a fetal canine thymus cell line (Cf2Th), followed by insertion into pCMV-Tag2B vector. The Flag-HuTHN delCT (20-180 aa), Flag-HuTHN delTM (47-180 aa), Flag-HuTHN delGPI (1-161 aa), Flag-AgmTHN delCT (26-182 aa), Flag-AgmTHN delTM (50-182 aa) and Flag-AgmTHN delGPI (1-159 aa) were constructed by inserting individual PCR fragment into pCMV-Tag2B vector. The human tetherin mutant C53/63/91A was provided by Klaus Strebel. HuTHN C53/63/91A was subcloned into pCMV-Tag2B vector. HuTHN mutants with cysteine or asparagine to alanine substitutions, C53A, C63A, C91A, C53/63A, C63/91A, C53/91A, N65/92A and C3A/N2A, were generated from Flag-HuTHN and Flag-HuTHN C53/63/91A, using a QuikChange site-directed mutagenesis kit (Stratagene) (Table 1). All the new constructs were verified by sequencing.

293T, BHK-21, and PFV indicator cell line (PFVL) cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum, 50 IU/ml penicillin, and 50 μg/ml streptomycin at 37°C in humidified air with 5% CO2. The indicator cell line PFVL was created by transfecting BHK-21 with reporter plasmid containing the firefly luciferase gene driven by a PFV long terminal repeat promoter. At 48 h post-transfection, the medium was changed with the addition of 600 μg/ml neomycin (G418, Promega) for additional three weeks. The cells were replaced with fresh medium every three days. To obtain positive stable monoclones, G418-resistant cells were isolated further by plating them at a limiting dilution on to 96-well plates. 293T cells were transfected with pcPFV proviruses along with either wild type tetherin or mutants. Cell lysates and viral supernatants were collected at between 44 and 48 h posttransfection. Transient transfection was performed using 1 mg/ml polyethyleneimine (PEI) (Sigma) as previously described [28].

Cells were washed twice with PBS, suspended in PBS, and mixed with an equal volume of sample buffer (4% sodium dodecyl sulfate, 125 mM Tris-HCl, pH 6.8, 10% 2-mercaptoethanol, 10% glycerol, and 0.002% bromophenol blue). For analysis of cysteine mutants under non-reducing conditions, cells were suspended in PBS and mixed with an equal volume of sample buffer that did not contain 2-mercaptoethanol. Proteins were solubilized by boiling for 20 min at 95°C. Western blot analyses were performed using anti-Gag polyclonal rabbit serum (provided by Maxine L. Linial) at a 1:5,000 dilution. The commercially available anti-Flag monoclonal antibody (Stratagene) was used at a 1:5,000 dilution. Anti-β-actin monoclonal mouse antibody (Sigma-Aldrich) was used at a 1:2,000 dilution as a loading control. After hybridizing the membrane with secondary goat anti-rabbit antibody and goat anti-mouse antibody at a 1:10,000 dilution, enhanced chemiluminescence reagents (Millipore) were used for signal detection with X-ray film.

For the preparation of viral proteins from supernatants of transfected cells, viral supernatants were collected, centrifuged at 2,000 rpm for 10 min to remove cell debris, and then pelleted through a 20% sucrose cushion for 2 h at 24,000 rpm and 4°C (Beckman). Viral pellets were resuspended in PBS and mixed with SDS sample buffer prior to loading onto a 10% protein gel. After proteins were separated by electrophoresis, they were transferred to nitrocellulose membranes (GE). The membranes were then blocked using Odyssey blocking buffer (LI-COR Biosciences) and incubated with anti-Gag polyclonal rabbit serum. Odyssey secondary antibodies were added according to the manufacturer's instructions (goat anti-rabbit IRDye 680). Blots were imaged using an Odyssey Infrared Imaging System (LI-COR Biosciences). Scan resolution of the instrument ranges from 21 to 339 μm, and in this study blots were imaged at 169 μm. Quantification was performed on single channels with the analysis software provided (LI-COR Biosciences).

The indicator cells were plated in 96-well plates (10⁴/well) and cultured at 37°C for 18 h. Then cells were infected with viral supernatants. After replacing the virus stock with the maintenance medium, cells were incubated at 37°C for 48 h. The luciferase activity was measured by using a 96-channel chemiluminescense measurement machine Glomax (Promega, USA). Readout was counts per second. The relative light unit was determined automatically.

We first examined whether human tetherin inhibits PFV release. To this end, we transfected 293T cells with the full-length PFV DNA construct named pcPFV and a human tetherin cDNA plasmid HuTHN. At 48 h after transfection, the amount of virus in supernatants was analyzed by Western blotting of Gag protein, viral DNA copy number assay and by infecting an indicator cell line named PFVL. PFV production and infectivity was reduced by human tetherin in a dose dependent manner, without affecting the expression of cell-associated Gag protein (Figure 1A, B).

HIV-1 Vpu counteracts tetherin by causing downregulation of cell surface tetherin [10, 16, 29, 30]. To determine whether the antiviral activity of tetherin against PFV is antagonized by Vpu, HIV-1 Vpu was co-expressed with pcPFV and human tetherin. As expected, Vpu enhanced the release of PFV in the presence of tetherin (Figure 1C, D). Together, these data suggest that human tetherin inhibits the release of PFV and this inhibition is overcome by HIV-1 Vpu.

Tetherin was originally identified as a specific marker of late-stage B-cell maturation and a potential target for the immunotherapy of multiple myeloma [24, 31]. It has an N-terminal transmembrane domain (TM) and a C-terminal GPI anchor. It associates with lipid rafts at the cell surface and on internal membranes [22]. In order to test whether these two domains are required for inhibition of PFV, we generated the cytoplasmic tail deletion mutant (delCT), transmembrane domain deletion mutant (delTM) and GPI deletion mutant (delGPI) (Figure 2A). The delCT mutant suppressed PFV production as efficiently as the wild type tetherin; in contrast, the delTM and the delGPI mutants only moderately inhibited PFV production (Figure 2B, C). These data suggest that the transmembrane domain and the GPI anchor of human tetherin are important for the inhibition of PFV.

Three cysteine residues C53, C63 and C91 contribute to tetherin dimerziation by forming disulfide bonds. To test their role in the antiviral activity of tetherin, we created three groups of mutants. The first group includes C53A, C63A and C91A that change each cysteine into an alanine. The second group includes C53/63A, C63/91A and C53/91A that change two cysteines into alanines. The third group is a C53/63/91A mutant that changes all three cysteines into alanines. We also mutated the two glycosylation sites at N65 and N92, and created the N65/92A mutant. Lastly, we generated the C3A/N2A mutant that changed the three cysteines and the two asparagines to alanines (Figure 3A). These mutants were expressed in 293T cells and the cell lysates were prepared in a buffer with or without β-ME. As shown in Figure 3B, all mutants showed a dominant 27-35 kDa monomer pattern under the reducing condition (+β-ME), while N65/92A and C3A/N2A reduced the apparent molecular weight to 17-20 kDa due to the lack of glycosylation. However, under the non-reducing condition, all mutants, except for C53/63/91A and C3A/N2A, formed the 60-70 kDa dimer. This result suggests that mutation of any individual cysteine or any two of the three cysteines in combination does not abolish tetherin dimerization. We next tested whether these mutants were able to inhibit PFV production. Since the mutants were expressed at different levels when the same amounts of plasmid DNA were transfected into 293T cells (Figure 3B), we thus adjusted the amount of plasmid DNA used for transfection such as to achieve similar expression levels of different mutants in order to compare the effect of these mutants on PFV release (data not shown). Interestingly, all mutants suppressed PFV production efficiently (Figure 3C). Furthermore, mutants C53/63/91A, N65/92A and C3A/N2A inhibited PFV release in a dose dependent manner (Figure 3D-F). These results indicate that dimerization and glycosylation are not essential for human tetherin to inhibit PFV release.

PFV productively infects various cell lines that are derived from human, chimpanzees, monkeys, cows and dogs. We thus wished to test whether tetherins from these latter species could also inhibit PFV production. We cloned the cDNA of tetherins from bovine and canine from fetal bovine lung (FBL) cells and a fetal canine thymus cell line (Cf2Th). Tetherins of human, simian, bovine or canine origin displayed 59.5% identity at the amino acid level. To measure their ability to suppress PFV release, pcPFV was cotransfected with cDNA clones of human tetherin, African green monkey tetherin, bovine tetherin, or canine tetherin into 293T cells. The results show that all tetherins tested significantly inhibit the release of PFV (Figure 4) and that the inhibition effect of tetherin is dose-dependent (Figure 5A-C). Furthermore, both the TM domain and the GPI anchor of African green monkey tetherin are important for the inhibition of PFV (Figure 5D).