BST-2 Expression Modulates Small CD4-Mimetic Sensitization of HIV-1-Infected Cells to Antibody-Dependent Cellular Cytotoxicity

ABSTRACT Antibodies recognizing conserved CD4-induced (CD4i) epitopes on human immunodeficiency virus type 1 (HIV-1) Env and able to mediate antibody-dependent cellular cytotoxicity (ADCC) have been shown to be present in sera from most HIV-1-infected individuals. These antibodies preferentially recognize Env in its CD4-bound conformation. CD4 downregulation by Nef and Vpu dramatically reduces exposure of CD4i HIV-1 Env epitopes and therefore reduce the susceptibility of HIV-1-infected cells to ADCC mediated by HIV-positive (HIV+) sera. Importantly, this mechanism of immune evasion can be circumvented with small-molecule CD4 mimetics (CD4mc) that are able to transition Env into the CD4-bound conformation and sensitize HIV-1-infected cells to ADCC mediated by HIV+ sera. However, HIV-1 developed additional mechanisms to avoid ADCC, including Vpu-mediated BST-2 antagonism, which decreases the overall amount of Env present at the cell surface. Accordingly, BST-2 upregulation in response to alpha interferon (IFN-α) was shown to increase the susceptibility of HIV-1-infected cells to ADCC despite the activity of Vpu. Here we show that BST-2 upregulation by IFN-β and interleukin-27 (IL-27) also increases the surface expression of Env and thus boosts the ability of CD4mc to sensitize HIV-1-infected cells to ADCC by sera from HIV-1-infected individuals. IMPORTANCE HIV-1 evolved sophisticated strategies to conceal Env epitopes from ADCC-mediating antibodies present in HIV+ sera. Vpu-mediated BST-2 downregulation was shown to decrease ADCC responses by limiting the amount of Env present at the cell surface. This effect of Vpu was shown to be attenuated by IFN-α treatment. Here we show that in addition to IFN-α, IFN-β and IL-27 also affect Vpu-mediated BST-2 downregulation and greatly enhance ADCC responses against HIV-1-infected cells in the presence of CD4mc. These findings may inform strategies aimed at HIV prevention and eradication.

A better understanding of the importance that the CD4-bound conformation of HIV-1 envelope glycoproteins has on ADCC responses prompted us to "force" this Env conformation on the surface of infected cells using small-molecule CD4 mimetics (CD4mc). CD4mc induction of the CD4-bound conformation results in enhanced recognition of HIV-1-infected cells by serum, breast milk, and cervicovaginal fluid samples from HIV-1-infected subjects. Most importantly, CD4mc sensitizes HIV-1-infected cells to ADCC responses mediated by these biological fluids (4,5,13).
The effect of CD4mc on ADCC responses may be influenced by the amount of Env available at the cell surface. Only limited amounts of Env are presented at the cell surface due to efficient Env internalization (12) and Vpu-mediated BST-2 downregulation (7)(8)(9); this places an upper limit on the amount of Env that can be rendered susceptible to ADCC by CD4mc. Interestingly, two BST-2 isoforms possessing distinct biological properties have been described (14,15). While the long isoform of BST-2 (L-BST-2) contains a cytoplasmic tyrosine motif mediating endocytic recycling, sensitivity to HIV-1 M Vpu and innate immune sensing, the short isoform of BST-2 (S-BST-2) lacks this motif due to the utilization of an alternative start codon (14,15). How these two isoforms modulate Env recognition on the surface of HIV-1-infected cells by HIV-positive (HIVϩ) sera and how this affects the activity of CD4mc remain unknown.
Type I interferons (IFNs) are an important part of the early host immune response observed during acute HIV-1 infection (16). The antiviral effect exerted by IFN is highlighted by the observation that transmitted/founder HIV-1 strains that initiate host infection have been shown to be more resistant to type I IFN responses than HIV-1 strains found during the chronic phase of infection (17)(18)(19). Furthermore, Vpu enhances viral replication particularly during early stages of infection, probably by counteracting the IFN-inducible restriction factor BST-2 (18,20). The induction of BST-2 expression by type I IFN treatment was also shown to sensitize infected cells to ADCC (8). In addition, interleukin-27 (IL-27) also enhances BST-2 levels on the surface of human monocytes and CD4 T cells (21). IL-27 is a member of the IL-12 family of cytokines and drives the differentiation of Th1 CD4 T cells (22,23). Interestingly, IL-27 induces an antiviral gene expression profile similar to that induced by alpha interferon (IFN-␣), including the apobec3g gene (24). Furthermore, IL-27 inhibited the replication of HIV-1 in cultures of primary CD4 ϩ T cells and monocytes/macrophages through the induction of APOBEC (apolipoprotein B mRNA-editing, enzyme-catalytic, polypeptide-like) proteins (24,25). Notably, IL-27-mediated BST-2 upregulation was shown to be independent from type I IFN responses (21). However, the effect of IL-27 on ADCC responses during viral infection has not been determined.
infected cells was higher in Jurkat L-BST-2 cells that express the BST-2 isoform susceptible to Vpu action (Fig. 2C). Accumulation of Env (as measured by 2G12) correlated significantly with recognition of infected cells by HIVϩ sera (Fig. 2D). Recognition of infected cells by HIVϩ sera also correlated with BST-2 expression (Fig. 2E).
BST-2 levels regulate Env accumulation and its recognition by HIV؉ sera on the surface of HIV-1-infected primary CD4 ؉ T cells. Figures 1 and 2 showed that BST-2 levels, and its sensitivity to Vpu downregulation, dictated Env accumulation on the surfaces of HIV-1-infected cell lines. Moreover, Env accumulation on the surfaces of infected cells increased the amount of Env available to engage CD4mc and henceforth sample the CD4-bound conformation, which is preferentially recognized by HIVϩ sera (1,27). IFN-␣ treatment has been shown to enhance BST-2 levels, resulting in an accumulation of Env on the surfaces of HIV-1-infected cells and thus increasing the sensitivity of HIV-1-infected cells to ADCC (8). Similar observations were recently  reported (29). Therefore, we decided to take advantage of the type 1 interferon responsiveness of BST-2 (10, 11). Primary CD4 ϩ T cells from healthy HIV-1 uninfected individuals were mock infected or infected with HIV-1 (CH58 TF), and BST-2 levels were modulated by stimulation with type 1 IFNs (IFN-␣ and IFN-␤) or with IL-27 (21). As expected, BST-2 levels were significantly higher in uninfected than HIV-1-infected cells.
Interestingly, IFN-␣, IFN-␤, and IL-27 treatment enhanced BST-2 detection in both uninfected and HIV-1-infected cells (Fig. 3A). BST-2 upregulation resulted in Env accumulation on the surfaces of infected cells, as measured with the 2G12 antibody (Fig. 3B). We then evaluated whether IFN-␣, IFN-␤, and IL-27 treatment enhanced recognition of HIV-1-infected cells by HIVϩ sera. Despite a significant increase in Env accumulation on the surfaces of infected cells (Fig. 3B), treatment with IFN-␤ and IL-27 failed to enhance recognition of infected cells by HIVϩ sera, and the effect of IFN-␣ treatment was relatively minor (Fig. 4). However, addition of the CD4mc BNM-III-170 significantly increased recognition of HIV-1-infected cells by HIVϩ sera; these results are in agreement with previous reports demonstrating the ability of HIVϩ sera to recognize CD4i epitopes on primary HIV-1 Env that are not spontaneously exposed (1, 3) and the capacity of CD4mc to promote the CD4-bound conformation of Env on the surfaces of HIV-1-infected cells (4,5,13,30). Remarkably, the combination of IFN-␣, IFN-␤, or IL-27 with BNM-III-170 further increased recognition of HIV-1-infected cells by all sera tested compared to any one of these treatments (Fig. 4).
BST CD4 ϩ T cells with HIV-1 CH58 TF and evaluated their susceptibility to ADCC mediated by autologous peripheral blood mononuclear cells (PBMCs) using a previously described fluorescence-activated cell sorting (FACS)-based assay (5, 31). As reported (5,13), CD4mc BNM-III-170 significantly increased ADCC mediated by all HIVϩ sera tested (Fig. 5). In agreement with the recognition of infected cells by HIVϩ sera (Fig. 4), IFN-␣ treatment alone had a minor but significant effect on ADCC responses (Fig. 5), but IFN-␤ and IL-27 treatment failed to do so (Fig. 5). Remarkably, addition of BNM-III-170 further enhanced the susceptibility of infected cells to ADCC for cells treated with IFN-␣, IFN-␤, or IL-27 (Fig. 5). As expected, enhanced recognition of HIV-1-infected cells by HIVϩ sera positively correlated with enhanced ADCC responses (Fig. 6). These results highlight the potential of combining type I IFNs and IL-27 with CD4mc to sensitize HIV-1-infected cells to ADCC.

DISCUSSION
Increasing evidence suggests that Fc␥ receptor-dependent functions of antibodies play a role in controlling human immunodeficiency virus type 1 (HIV-1) infection and replication (32)(33)(34)(35)(36)(37)(38)(39)(40). Analysis of the correlates of protection in the RV144 vaccine trial suggested that decreased HIV-1 acquisition was linked to increased ADCC activity in protected vaccinees (41). ADCC-mediating antibodies (Abs) targeting anti-cluster A epitopes were isolated from some RV144 vaccinees (42) and were shown to preferentially recognize the HIV-1 envelope glycoproteins sampling the CD4-bound conforma-   (7). CD4i antibodies represent a significant portion of the anti-Env Abs elicited during natural HIV-1 infection (1,27,43). This elicitation of CD4i Abs could result from transitional exposure of CD4i Env epitope during viral entry (44) or, most likely, after binding of shed gp120 with CD4 on uninfected bystander cells (30). However, not all CD4i antibodies are able to mediate ADCC against HIV-1-infected cells. While anticluster A antibodies have been shown to mediate potent ADCC responses against infected cells exposing Env in the CD4-bound conformation (3,4,7,45), CD4i antibodies targeting the coreceptor binding site appear to be unable to do so (3,4,45,46). While the reasons for these differences are not fully understood, the angle of approach of the antibody toward Env might differentially expose the Fc region which must be engaged by the Fc␥ receptor in order to activate effector cells. Nevertheless, to limit the exposure of anti-cluster A epitopes that are exposed in the CD4-bound conformation of Env on the surfaces of infected cells, HIV-1 evolved sophisticated mechanisms to efficiently internalize Env (12) to counteract the host restriction factor BST-2 with the viral Vpu protein (7-9) and to downregulate CD4 by Nef and Vpu (1, 7). The requirement to evade ADCC provides one plausible explanation of why the vast majority of circulating HIV-1 strains worldwide express functional Nef and Vpu proteins, which limit the exposure of CD4i Env epitopes on the surfaces of infected cells.
In agreement with the necessity for HIV-1 to avoid exposing the CD4-bound conformation of Env, we recently showed that forcing Env to adopt this conformation with CD4mc sensitizes HIV-1-infected cells to ADCC by sera from HIV-1-infected sub- jects (5). Here we show that increasing the amounts on Env at the cell surface, once this Env is induced by CD4mc to adopt the CD4-bound conformation, results in increased recognition of HIV-1-infected cells by HIVϩ sera. We found that enhanced recognition of infected cells by HIVϩ sera translates into enhanced susceptibility of infected cells to ADCC. This was achieved by exploiting the type 1 interferon responsiveness of the restriction factor BST-2, known to trap mature viral particles on the surfaces of infected cells. IFN-␣ and -␤ enhance BST-2 levels on the surfaces of infected cells, which translates into enhanced levels of Env potentially able to be targeted by ADCC after engaging the CD4mc. Interestingly, similar results were obtained using IL-27, a cytokine known to modulate BST-2 levels in an IFN-independent manner. Altogether, our results suggest a model (Fig. 7) where the conformation and availability of Env at the cell surface dictates the sensitivity of HIV-1-infected cells to ADCC. HIV-1 limits the amount of Env present at the cell surface and tightly controls its conformation. By preventing Env from assuming the CD4-bound conformation, HIV-1 avoids Env recognition by CD4i ADCC-mediating Abs present in the sera of the majority of HIV-1-infected individuals. Small CD4mc sensitize HIV-1-infected cells to ADCC by forcing Env to expose CD4i anti-cluster A-mediating epitopes. IFN-␣, IFN-␤, and IL-27 treatment, through upregulation of BST-2, increases the total amounts of Env available for CD4mc to induce ADCC-susceptible Env conformations on the surfaces of infected cells. While HIV-1infected cells are protected from ADCC responses, we recently demonstrated that uninfected bystander CD4 ϩ T cells bind gp120 shed from productively infected cells and are efficiently recognized by ADCC-mediating antibodies (30). Importantly, we also demonstrated that this phenomenon can be blocked by CD4mc that abrogates the binding of gp120 to uninfected cells and effectively redirects the immune system to infected cells. Therefore, the combination of CD4mc and type I IFN or IL-27 would represent an effective strategy to specifically target and eliminate HIV-1-infected cells by ADCC. Robust type I interferon responses are among the earliest host immune defenses observed during acute HIV-1 infection (16). Accordingly, transmitted/founder viruses, including those used in the present study, were found to be more resistant to IFN treatment than viruses from chronic HIV-1 infection (17)(18)(19). In that context, Vpu

Protection from ADCC ADCC response
Nef Vpu

FIG 7
Env conformation and its accumulation at the cell surface dictates sensitivity of HIV-1-infected cells to ADCC. ADCC-mediating Abs present in sera from HIV-1-infected individuals preferentially recognize Env in its CD4-bound conformation (1). To limit the exposure of this conformation, HIV-1 has evolved sophisticated mechanisms to counteract the host restriction factor BST-2 with the viral Vpu protein (7)(8)(9) and to downregulate CD4 by Nef and Vpu (7). Nef and Vpu decrease the accumulation of Env and its interaction with CD4 at the cell surface, two factors that determine the susceptibility of HIV-1-infected cells to ADCC. Small CD4 mimetics sensitize HIV-1-infected cells to ADCC mediated by HIVϩ sera by forcing Env to sample its CD4-bound conformation (5). Type I IFN or IL-27 treatment, through upregulation of BST-2 despite Vpu activity, boosts the ability of CD4mc by increasing the amounts of CD4mc-sensitized Env available on the cell surface.
Effect of Type 1 IFN and IL-27 on ADCC Journal of Virology counteraction of BST-2 was recently identified as a major determinant of this IFN resistance (18,20) and was found to play a crucial role in enhancing virus replication and release in human CD4 ϩ T cells, particularly in the presence of IFN (18). Here we found that IFN-␣, IFN-␤, or IL-27 treatment enhanced BST-2 levels and, in combination with CD4mc, similarly sensitized HIV-1-infected cells to ADCC. However, there are many other IFN-␣ subtypes, and some of them inhibit HIV-1 replication more efficiently in vitro and in animal models than IFN-␣2 (47,48). Thus, it will be important to evaluate to what extent the different IFN-␣ subtypes sensitize HIV-1-infected cells to ADCC in the presence of CD4mc. CD4mc were recently shown to enhance the viral neutralization and ADCC activities of antibodies elicited in nonhuman primates (NHP) by several different Env immunogens (49), suggesting that combining a vaccine with a small-molecule CD4mc, administered orally or in a microbicide formulation, might be useful as a prophylactic strategy against HIV-1 transmission. Interestingly, mucosal application of IFN-␤ protected macaques from intrarectal and intravaginal simian-human immunodeficiency virus (SHIV) challenges (50). Similarly, IFN-␣2 treatment of rhesus macaques prevented systemic infection by simian immunodeficiency virus (SIV) (51). Whereas a combination of IFNs or IL-27 with CD4mc might further limit HIV-1 transmission or help decrease the size of the viral reservoir in HIV-1-infected individuals remains to be evaluated, our results support performing future experiments aimed at evaluating whether sensitization of HIV-infected cells to ADCC could affect viral transmission and/or replication in animal models.

MATERIALS AND METHODS
Cell lines and isolation of primary cells. HEK293T human embryonic kidney (obtained from ATCC) and primary cells were grown as previously described (7,52). Peripheral blood mononuclear cells (PBMCs) were obtained by leukapheresis. All participants provided informed written consent prior to enrollment in accordance with Institutional Review Board approval. CD4 T lymphocytes were purified from resting PBMCs by negative selection and activated as previously described (5). Jurkat Tag cells stably expressing the long isoform of BST-2 (L-BST-2) or the short isoform of BST-2 (S-BST-2) and the Jurkat Tag empty vector (EV) cell line expressing no BST-2 were previously described (15). Viral production, infections, and detection of infected cells. In order to achieve the same level of infection between wild-type (wt) and Vpu Ϫ viruses, vesicular stomatitis virus G (VSVG)-pseudotyped HIV-1 replicating competent viruses were produced. Briefly, proviral vectors and a VSVG-encoding plasmid were cotransfected in 293T cells by standard calcium phosphate transfection. Two days after transfection, cell supernatants were harvested, clarified by low-speed centrifugation (5 min at 1,500 rpm), and concentrated by ultracentrifugation for 1 h at 4°C at 100,605 ϫ g over a 20% sucrose cushion. Pellets were resuspended in fresh RPMI 1640 medium, and aliquots were stored at Ϫ80°C until use (1). Viruses were then used to infect Jurkat Tag cell lines or primary CD4 T cells from healthy donors by spin infection at 800 ϫ g for 1 h in 96-well plates at 25°C.
CD4 mimetic, type I IFN, or IL-27 treatments. The CD4mc BNM-III-170 was synthesized as described previously (28). BNM-III-170 was dissolved in dimethyl sulfoxide (DMSO) at a stock concentration of 10 mM, aliquoted, and stored at Ϫ20°C. BNM-III-170 was then diluted to 50 M in phosphate-buffered saline (PBS) for cell surface staining or in complete RPMI 1640 medium for ADCC assays. IFN-␣ (PBL Assay Science) was reconstituted in complete RPMI 1640 medium at 1 ϫ 10 7 U/ml, aliquoted, and stored at Ϫ80°C. IFN-␣ was then added to the cells at 1,000 U/ml. IFN-␤ (Rebif; EMD Serono Inc.) (19) was added to the cells at 1 ng/ml. IL-27 (R&D Systems) was reconstituted at 100 g/ml in sterile PBS containing 0.1% bovine serum albumin and stored at Ϫ80°C. IL-27 was then added to the cells at 100 ng/ml. Type I IFN or IL-27 was added to the cells 24 h postinfection, 24 h before cell surface staining or ADCC assays.
Antibodies and sera. The following antibodies (Abs) were used as the primary Abs for cell surface staining: 5 g/ml of human anti-HIV-1 Env monoclonal antibody (MAb) 2G12 (National Institutes of Health [NIH] AIDS and Research and Reference Reagent Program), 2 g/ml rabbit anti-BST-2 Ab (sc-99191; Santa Cruz), or sera from HIV-1-infected individuals (1:1,000 dilution), whereas 1 g/ml of either Alexa Fluor 647-labeled goat anti-human MAbs (Invitrogen, San Diego, CA, USA) or Brilliant Violet 421-labeled donkey anti-rabbit MAbs (Biolegend, San Diego, CA, USA) were used as secondary Abs, and AquaVivid (Invitrogen, San Diego, CA, USA) was used as a viability dye. All sera were heat inactivated for 30 min at 56°C and stored at 4°C until ready to use in subsequent experiments. Written informed consent was obtained from all study participants (the Montreal Primary HIV Infection Cohort [53,54] and the Canadian Cohort of HIV Infected Slow Progressors [55][56][57]), and research adhered to the ethical guidelines of Centre de Recherche du CHUM (CRCHUM) and was reviewed and approved by the CRCHUM institutional review board (ethics committee). A random-number generator (QuickCalcs; GraphPad) was used to randomly select a number of sera for experiments.
Flow cytometry analysis of cell surface staining and ADCC responses. Cell surface staining was performed as previously described (1,5). Binding of HIV-1-infected cells by HIVϩ sera, anti-Env MAbs (2G12) or anti-BST-2 MAbs was performed 48 h after infection, 24 h after treatment with type I IFN or IL-27, in the presence or absence of BNM-III-170 (50 M) or an equivalent volume of vehicle (DMSO). Detection of p24ϩ infected cells was performed as described previously (5). The percentage of infected cells (p24ϩ cells) was determined by gating the living cell population based on the viability dye staining (Aqua Vivid; Invitrogen). Samples were analyzed on a LSRII cytometer (BD Biosciences, Mississauga, ON, Canada), and data analysis was performed using FlowJo vX.0.7 (Tree Star, Ashland, OR, USA).
Measurement of ADCC-mediated killing was performed with a previously described assay (5). Briefly, primary CD4 ϩ T cells infected for 48 h and treated for 24 h with type I IFN or IL-27 or not treated with type I IFN or IL-27 were incubated with autologous PBMCs (effector/target cell ratio of 10:1) in the presence or absence of HIVϩ sera (1:1,000), in the presence of CD4mc BNM-III-170 (50 M), or with an equivalent volume of vehicle (DMSO). The percentage of cytotoxicity was calculated as described previously (5).
Statistical analyses. Statistics were analyzed using GraphPad Prism version 6.01 (GraphPad, San Diego, CA, USA). Every data set was tested for statistical normality, and this information was used to apply the appropriate (parametric or nonparametric) statistical test. P values of Ͻ0.05 were considered significant; significance values are indicated as follows: *, P Ͻ 0.05; **, P Ͻ 0.01, ***, P Ͻ 0.001; ****, P Ͻ 0.0001.