Construction of a series of site-specific HIV-1 vpu mutants from pNL4-3 (Adachi et al., 1986) was carried out by the QuickChange site-directed mutagenesis kit (Stratagene, La Jolla, USA). Amino acid and nucleotide substitutions are as follows: V9D and V13D (gta to gac), I19D and I24D (ata to gac), RK30AA (agg aaa to gcg gca), Q35A (caa to gca), D43A (gat to gct), E47A (gaa to gca), EDS50AAA (gaa gac agt to gca gca gct), S52A (agt to gct), S56G (agt to ggt), E59K (gaa to aaa), M66V and M70V (atg to gtg), and P75S (cct to tct). A vpu-deletion mutant, pNL-uE65 (Iida et al., 1999), was generated by replacing the initiation codon (ATG) to AGG. For flow cytometry analysis, vectors co-expressing Vpu by cytomegalovirus promoter (CMV P) and GFP via an internal ribosomal entry site (IRES) were constructed as follows. Wild-type (WT) and mutated vpu sequences were amplified by PCR using forward primer (Vpu-5-SacII: TCCCCGCGGATGCAACCTATAATAGTAGC), reverse primer (U down BamHI: CGCGGATCCCTACAGATCATCAATATCCC), and pNL4-3 derived sequences as template. Amplified products were then digested with SacII and BamHI, cloned into pIRES-hrGFP-1a vector (Stratagene), and resultant clones were designated pIRES-hrGFP-Vpu vectors. To analyze intracellular localization of Vpu, vectors expressing C-terminally FLAG-tagged Vpu were generated. Vpu genes were amplified by PCR using forward primer (EcoRI-5-Vpu: CGGAATTCAAGGATGCAACCTATAATAGTAGC), reverse primer (Vpu-3-XhoI: CCGCTCGAGCAGATCATCAATATCCCAAG), and proviral vpu mutant clones as template, and cloned into pSG-cFLAG as described previously (Wang et al., 2005). The resultant clones were designated pSG-VpucFLAG vectors.

293T (Lebkowski et al., 1985) and HEp2 cells (ATCC CCL-23) were maintained in Eagle's minimal essential medium (MEM) supplemented with 10% heat-inactivated fetal bovine serum (FBS). MAGI cells (Kimpton and Emerman, 1992) were cultured in MEM containing 10% FBS, 200 μg/mL G418 (SIGMA-ALDRICH, St. Louis, USA), and 100 μg/mL hygromycin B (SIGMA-ALDRICH). H9 cells (Mann et al., 1989) were maintained in RPMI1640 medium supplemented with 10% FBS.

Input viruses were prepared from 293T cells transfected with proviral clones by the calcium-phosphate co-precipitation method (CPC; Adachi et al., 1986). H9 cells (1 × 10⁶) were infected with equivalent amounts (5 × 10⁶ cpm) of various viruses as determined by reverse transcriptase (RT) assays, and monitored for virus replication by RT assays as previously described (Willey et al., 1988).

HEp2 cells were transfected with proviral clones by CPC followed by glycerol treatment (20% glycerol in MEM without FBS). On day 2 post-transfection, culture supernatants were collected and virions released from cells were determined by RT assays. Virion release assays in 293T cells were similarly performed; no glycerol treatment following CPC transfection. For analysis of Vpu expression vectors, subconfluent HEp2 cells in 24 well-dishes were transfected with 300 ng of pNL-uE65, 200 ng of pSG5, and 200 ng of pIRES-hrGFP-Vpu vectors or pSG-VpucFLAG vectors by Lipofectamine 2000 (Invitrogen, Carlsbad, USA). At 6 h post-transfection, culture medium was replaced with fresh 10% FBS-MEM and cells were incubated for 2 days. Viral production in the culture supernatants was determined by RT assays.

For analysis of Vpu expression by proviral clones, transfected HEp2 cells were lysed with CHAPS buffer (10 mM CHAPS, 150 mM NaCl, 1 mM NaF, 1 mM Na₃VO₄, and 10 mM Na-phosphate pH 7.4). For analysis of Vpu expression by subgenomic vectors, transfected 293T cells were lysed in TNE buffer (10 mM Tris–HCl pH 8.0, 1% Nonidet P-40, 150 mM NaCl, 1 mM EDTA, and 1% protease inhibitor cocktail (SIGMA-ALDRICH)]. Vpu in cell lysates were monitored as previously described (Yamashita et al., 2008) using anti-FLAG (SIGMA-ALDRICH), anti-β actin (SIGMA-ALDRICH), anti-p24 (NIH AIDS Research and References Reagent Program), or anti-Vpu (NIH AIDS Research and References Reagent Program) antibody.

For analysis of cell surface CD4, MAGI cells in 60 mm dishes were transfected with 5 μg of pIRES-hrGFP-Vpu vectors by CPC. On day 2 post-transfection, cells were trypsinized, washed with PBS, and resuspended in FACS washing buffer (10% FBS in PBS). Cells were then stained with a PE conjugate mouse anti-human CD4 antibody (BD Biosciences, San Jose, USA). For analysis of cell surface tetherin, HEp2 cells in 60 mm dishes were transfected with pIRES-hrGFP-Vpu vectors and harvested as described above. Cells were then stained with an anti-HM1.24 monoclonal antibody (a generous gift from Chugai Pharmaceutical Co., Tokyo, Japan) and a secondary PE conjugate anti-mouse Ig antibody (BD Biosciences). Stained cells were analyzed by FACSCalibur and CELLQuest software (BD Biosciences).

HEp2 cells were seeded onto 4-well Lab-Tek chambered coverglasses (Nalge Nunc International, Rochester, USA), incubated overnight, and transfected with a pNL-uE65 (300 ng), pSG-VpucFLAG vectors (200 ng), and pSG5 (200 ng) by Lipofectamine 2000 (Invitrogen). On the next day, cells were fixed with 4% paraformaldehyde, permeabilized with 0.2% Triton X-100, and incubated with a first antibody [anti-TGN46 (Novus Biologicals, Inc., Littleton, USA), anti-FLAG, or anti-HM1.24 antibody]. Cells were then incubated with an appropriate secondary antibody [Alexa Fluor 488 or 555 SFX kit (Invitrogen)] according to manufacturer's instructions, and cell nuclei were counterstained with DAPI. Cell images were obtained by DIGITAL ECLIPSE C1si-Ready (Nikon Co., Tokyo, Japan) and analyzed by EZ-C1 viewer (Nikon Co.).

HIV-1 Vpu has been shown to positively modulate viral infectivity by mediating CD4 degradation and also to promote virion release by down-modulating tetherin (Levesque et al., 2003; Neil et al., 2008; Van Damme et al., 2008). The correlation between the viral replication and the two Vpu functions has been examined for a limited number and kind of vpu mutants (Terwilliger et al., 1989; Schubert et al., 1996; Van Damme et al., 2008; Dubé et al., 2010). To study functional role of Vpu in viral replication more clearly and extensively, we generated a series of proviral vpu mutants (Figure 1A). Substitutions were introduced into residues that are well-conserved among viruses of HIV-1 group M. For mutations within TMD, Val9, Val13, Ile19, and Ile24 were replaced with Asp, respectively. By these substitutions, TMD was changed to contain a negatively charged amino acid in place of non-charged amino acids. Some CTD mutants (ArgLys30, Gln35, Asp43, Glu47, GluAspSer50, and Ser52) were generated by Ala-substitution, and the other mutants (Ser56, Glu59, Met66, Met70, and Pro75) contained amino acids with physicochemical property distinct from authentic ones (Gly, Lys, Val, Val, and Ser, respectively). Since Ser52 and Ser56 are key residues to interact with β-TrCP (Schubert and Strebel, 1994; Margottin et al., 1998), it is expected that S52A and S56G mutations disrupt the ability to recruit E3 ubiquitin ligase complex and the following degradation of target proteins.

To determine replication kinetics, mutant virus stocks were prepared from transfected 293T cells and inoculated into a Vpu-dependent lymphocyte cell line H9 (Figures 1B–E). Out of fifteen mutants shown in Figure 1A, seven (V9D, V13D, I19D, EDS50AAA, S52A, S56G, and E59K) exhibited a replication-defect similar to that of a vpu-deletion mutant (ΔVpu in Figures 1B–E) in H9 cells. The other mutant viruses were found to have a replication potential similar to that of WT virus. Viral clone carrying I24D mutation had a small replication-defect but appeared not to be significant (Figure 1E). In total, the results described above demonstrate that narrow regions in Vpu TMD and CTD are required for efficient viral replication.

It has been shown previously that Vpu is required to promote virion release in tetherin-positive cells (e.g., HeLa, HEp2, T cells, and macrophages) but dispensable in tetherin-negative cells (e.g., 293T, HT1080, and COS-7; Sakai et al., 1995; Neil et al., 2008). To determine if mutations in vpu affect progeny production, proviral mutant clones were transfected into HEp2 and 293T cells, and virion level in the culture supernatants was determined by RT assays. In HEp2 cells, control ΔVpu and seven TMD (V9D, V13D, and I19D) and CTD (EDS50AAA, S52A, S56G, and E59K) mutants exhibited two- to three-fold reduction of virus production relative to WT (Figure 2A). These seven mutants were replication-defective in H9 cells (Figures 1B–E). The other eight mutants behaved similarly with WT with respect to virion production in HEp2 cells. As expected, all mutant clones were indistinguishable from WT in 293T cells (Figure 2B). Vpu expression level in HEp2 cells was different among proviral clones in this particular experiment (Figure 2C), but virion level was not correlated to the amount of Vpu (Figures 2A–C). The results described above indicate that, in the presence of tetherin, efficient viral replication coincidentally occurred with promotion of virion release by Vpu.

Down-modulation of CD4 appears to be important for HIV-1 replication, since numbers of viral proteins (Env, Nef, and Vpu) show this particular activity. For analysis of CD4 and tetherin down-regulation by Vpu, we generated constructs that co-express Vpu (target for study) and GFP (transfection marker). To confirm the activity of these constructs, Vpu expression and virion production level were examined. As shown in Figure 3, these vectors expressed the Vpu proteins and complemented virion production of ΔVpu to a similar extent with that observed in the provirus experiment (Figure 2A).

To determine whether Vpu mutant proteins down-modulate CD4, we evaluated its expression level at the surface of transfected MAGI cells (CD4-positive HeLa cells) (Figure 4). Some mutants (I24D, RK30AA, and M70V) were able to decrease the cell surface level of CD4 equally efficiently with WT Vpu. In contrast, mutants V9D, I19D, S56G, and E59K failed to down-modulate cell surface CD4. These four mutants were all replication-defective in H9 cells (Figure 1). Thus, viral replication potential and ability to down-modulate CD4 by Vpu are correlated.

Down-modulation of cell surface tetherin by Vpu has been shown to be important for enhancement of virion release from HeLa cells (Van Damme et al., 2008; Mitchell et al., 2009). It also has been reported that, during infection of T cell lines with viruses carrying S52 and S56 mutations, cell surface level of tetherin was partially reduced (Douglas et al., 2009) or unchanged (Miyagi et al., 2009). To assess the potential of mutant Vpu proteins, we analyzed cell surface tetherin on HEp2 cells expressing Vpu. HEp2 cells were transfected with vectors co-expressing Vpu and GFP as above, and monitored for expression level of cell surface tetherin by flow cytometry (Figure 5). Some mutants (I24D, RK30AA, and M70V), which promote virion release (Figure 3B), down-modulated cell surface tetherin as effective as WT Vpu. In addition, mutants S56G and E59K effectively reduced the tetherin. In contrast, some TMD mutants (V9D and I19D) did not decrease tetherin from cell surface. While these two TMD mutants did not enhance virion release as expected, CTD mutants S56G and E59K did so considerably (Figure 3B). In sum, TMD mutants V9D and I19D did not promote virion release and did not down-modulate cell surface tetherin. CTD mutants S56G and E59K did not enhance virion release but did reduce cell surface tetherin. These results show that, two activities of Vpu, i.e., down-modulation of cell surface tetherin and enhancement of virion release, are not always correlated.

Recent studies have suggested that the ability of Vpu to localize to TGN and to colocalize with tetherin is important to elicit anti-tetherin activity, and that this property of Vpu correlates with virion release enhancement (Dubé et al., 2009, 2010). Therefore, we examined mutational effects of Vpu on its intracellular localization. For microscopic analysis, we generated vectors encoding C-terminally FLAG-tagged Vpu, and verified their protein expression and ability to promote virion production from cells. As shown in Figure 6, these vectors expressed the Vpu proteins and complemented virion production of ΔVpu to a similar extent with that observed in the provirus experiment (Figure 2A).

Using these FLAG-tagged Vpu constructs, we analyzed intracellular localization of Vpu in HEp2 cells. Experiments for microscopic analysis were performed similarly with those for virion release assays to ensure direct functional comparison between the virion release and intracellular localization. As shown in Figure 7, WT Vpu localized in intracellular compartments stained with the TGN marker TGN46, and colocalized with tetherin in good agreement with the results reported previously (Dubé et al., 2009). None of WT and mutant Vpu proteins localized with the ER marker calreticulin, and their colocalization with the late endosome marker CD63 was rarely seen (data not shown). Mutants I24D and RK30AA, which have the ability to promote virion release and to down-modulate cell surface tetherin, were detected in the TGN46-positive compartment and colocalized with tetherin like WT Vpu.

A Vpu TMD mutant I19D, defective for enhancing virion release and for inducing down-modulation of tetherin, showed altered localization and its expression was diffusely observed in the cytoplasm. A portion of the mutant was noted to exist with TGN46. Furthermore, this mutant colocalized less extensively with tetherin. Another Vpu TMD mutant V9D exhibited a similar phenotype with the I19D mutant (data not shown). Such a clear change in subcellular localization of Vpu TMD mutants has not been reported previously (Tiganos et al., 1998; Dubé et al., 2010). A third TMD mutant I24D exhibited a similar intracellular distribution with WT Vpu. Taken together, the evident alteration of intracellular localization by V9D and I19D mutations may have resulted from introducing a negatively charged amino acid into these residues, and show their critical importance for function of TMD.

In contrast to TMD mutants, a CTD mutant E59K was defective for virion release enhancement but efficiently down-modulated cell surface tetherin. As expected, E59K was co-stained with TGN46. Interestingly, this mutant colocalized with tetherin more extensively than WT Vpu. Another CTD mutant S56G also colocalized with tetherin extensively (data not shown). The results for localization of our CTD mutants (S56G and E59K) were consistent with a previous report (Dubé et al., 2010). Extensive colocalization of these CTD mutants could explain efficient down-modulation of cell surface tetherin (Figure 5). The result for E59K also suggests that amino acid residue(s) close to β-TrCP binding motif may play a role in recruiting β-TrCP and subsequent degradation of the target protein. In total, the activity of Vpu to localize optimally in TGN and to colocalize with tetherin is parallel with the capability to down-modulate cell surface tetherin but not to enhance virion release. Colocalization of Vpu and tetherin may be sufficient to induce down-modulation of tetherin from cell surface but subsequent degradation of the restriction factor may also be required for efficient virion release.

On the other hand, it has been reported that Vpu retains newly synthesized CD4 in ER and targets it to proteasomal degradation (Lenburg and Landau, 1993; Bour et al., 1995; Paul and Jabbar, 1997; Margottin et al., 1998; Magadán et al., 2010). Thus, we were interested in analyzing whether Vpu colocalizes with CD4 intracellularly. However, as mentioned earlier, Vpu did not localize the ER marker calreticulin in our assays. Furthermore, colocalization of Vpu and CD4 was not detected in MAGI cells transfected with a WT Vpu construct (data not shown). Evidently, it is necessary to examine more thoroughly the intracellular localization of Vpu and CD4.