The dual role of tetraspanin CD63 in HIV-1 replication

Background Previously, we showed that the tetraspanin membrane protein CD63 mediates both early and post-integration stages of the HIV-1 replication cycle. The temporal roles of CD63 were discerned using monoclonal antibodies and small interfering RNAs (siRNAs) to block CD63 function, and determining which of the sequential steps in HIV-1 replication were disrupted. Inhibition was shown to occur during early infection, suggestive of involvement in virus entry or reverse transcription. In addition, we have shown that treatment with CD63 siRNA post-infection, significantly inhibited virus production in supernatant, suggesting an important role for CD63 in macrophages during HIV-1 replication events occurring after proviral integration, and possibly during egress. Results In this study we used CD63 siRNA to investigate the infectivity of pseudotyped viruses (carrying an NL4-3 Env-negative luciferase backbone) in primary human macrophages. We demonstrated that lab adapted R5- and R5X4-tropic HIV-1 strains are significantly inhibited by CD63 silencing. However, the infectivity of MLV or VSV-pseudotyped strains, which enter though receptor-mediated endocytosis, is unaffected by silencing CD63. These results indicate that CD63 may support Env-mediated entry or fusion events facilitated though CD4 and CCR5. Also, antibody and siRNA-based CD63 inhibition studies indicate a potential role for CD63 following proviral integration. Further, we show that CD63 expression is key for efficient replication in primary CD4+ T cells, complementing our prior studies with primary human macrophages and immortalized cell lines. Conclusions Collectively, these findings indicate that CD63 may support Env-mediated fusion as well as a late (post-integration) step in the HIV-1 replication cycle.

CD63 is a type II cellular membrane protein belonging to the tetraspanin superfamily [12,13]. CD63 associates with tetraspanin-enriched microdomains (TEMs), which have been shown to act as gateways for HIV-1 budding at the plasma membrane [14]. Numerous studies have shown that CD63 readily incorporates into HIV-1 virions [15][16][17] and partially colocalizes with HIV-1 Env and Gag proteins in HIV-1 producing cells [16,[18][19][20][21]. Colocalization and virion incorporation data indirectly suggest a positive role for CD63 in the HIV-1 replication cycle, although this has not always been borne out experimentally, and conflicting results have been observed. For example, a recent study showed that CD63 suppresses trafficking of the chemokine receptor CXCR4 to the cell surface, which affects HIV-1 entry into T lymphocytes [22][23][24]. However, we have demonstrated that anti-CD63 monoclonal antibody or siRNA treatment does not affect the expression of the other major coreceptor CCR5, or the primary receptor CD4 [25][26][27]. Another recent paper showed CD63 is not required for HIV-1 infection of human MØ [28], in contrast with our prior studies.
Our laboratory and others have demonstrated that CD63 is important for HIV-1 replication in primary MØ and CD4 + cell lines [25][26][27]29]. Our earlier studies showed that anti-CD63 monoclonal antibody treatment 30 min prior to and during infection markedly reduced HIV-1 replication in MØ [25]. Inhibition was shown to occur during early infection, suggestive of involvement in virus entry or reverse transcription. Subsequently, we confirmed the requirement of CD63 in HIV-1 replication in primary human MØ and an immortalized CD4 + cell line following CD63 down regulation by siRNA [26,27], and presently show its requirement by HIV-1 in primary human CD4 + T lymphocytes. Here, we show that CD63 expression supports Env-mediated fusion during MØ transduction with pseudotyped viruses. Down modulating CD63 expression in latently infected U1 cells blocked viral egress, but not intracellular p24 production, indicating CD63 also regulates a late replication step (s). Implications of our findings with regard to mechanisms of HIV-1 replication are discussed below.

CD63 facilitates virus Env-mediated entry in human MØ
We studied the effect of silencing CD63 expression on the transduction of various pseudotyped viruses carrying an NL4-3 envelope-negative luciferase backbone. Primary human MØ were transfected with siRNAs (CD63, CD4, or ERBB2IP). The efficiency of CD63-specific siRNA down regulation has been shown in our previous studies [26,27]. Forty-eight hours post-transfection, cells were transduced with various Env-specific reporter pseudoviruses (MLV, VSV, HIV-1 ADA , or HIV-1 89.6 ) for single-cycle infection. Luciferase activities were analyzed 3 days post-infection to monitor pseudovirus integration. MLV and VSV (carrying the same NL4-3 envelope-negative luciferase backbone) were used as controls because they enter the cell via endocytosis, independently of CD4 and chemokine receptor usage. ERBB2IP siRNA was used as a cellular target negative control, as our previous studies showed that silencing ERBB2IP does not inhibit HIV-1 [30,31]. Figure 1 shows that when viral entry is mediated though CD4 and coreceptor CCR5 using R5 (HIV-1 ADA ) or dual-tropic (HIV-1 89.6 ) pseudotyped envelopes, infectivity is significantly reduced in CD63-silenced MØ; consistent with the CD4 siRNA control. However, when viral entry is redirected though the endocytic pathway by pseudotyping virions with MLV or VSV, infectivity is not affected in CD63-or CD4-silenced MØ. Like human MØ, CD4 + T lymphocytes are also target reservoirs for HIV-1 infection and replication. In our previous studies we showed that CD63 silencing affects HIV-1 production in monocyte-derived MØ and a CD4 + cell line [25][26][27], and we now extend our investigations to study the role of CD63 in primary human peripheral blood lymphocytes (PBLs).
Efficient silencing of CD63 mRNA expression in PBLs with siRNA was confirmed by quantitative reverse transcriptase PCR (qRT-PCR). CD63 mRNA expression was reduced by >92% following CD63 siRNA transfections of PBLs ( Figure 2A). CD63 mRNA was not significantly reduced by control siRNA transfections (CD4, ERBB2IP) indicating specificity for CD63 siRNA down regulation in PBLs. Western blot analysis revealed that CD63 protein expression (53 kDa) was significantly reduced in CD63-siRNA transfectants compared to untransfected cells ( Figure 2B). Reducing CD63 expression in PBLs did not affect viability, as assayed by coupling GAPDH to 3-phosphoglyceric phosphokinase and measuring ATP production (data not shown).

CD63 down regulation affects HIV-1 production in human PBLs
We sought to elucidate the effect of CD63 down regulation on viral production in PBLs. PBLs were transfected with CD63 siRNA or control siRNAs (ERBB2IP or CD4) for 48 h, followed by infection with HIV-1 89.6 . HIV-1 production was assessed in culture supernatants by p24 ELISA at 7 days post-infection. As shown in Figure 2C, virus production in PBLs was significantly reduced following CD63 or CD4 silencing compared to ERBB2IP siRNA transfected cells (P < 0.05). Thus, our data strongly suggest that CD63 expression supports HIV-1 replication in PBLs. Cell lysates were collected on day 3 post-transduction, and luciferase activities were analyzed to monitor pseudovirus integration. MLV and VSV were used as controls for the entry step because they enter the cells via receptor-mediated endocytosis. In contrast, pseudoyped viruses coated with R5 (ADA) and R5/X4 dual-tropic (89.6) HIV-1 Env proteins gain entrance into MØs by sequential Env interactions with CD4 and CCR5, resulting in viral-cell membrane fusion. All experiments were performed in quadruplicate. *P < 0.05, compared with ERBB2IP group.
CD63 down regulation also affects later events of the HIV-1 replication cycle To further assess the role of CD63 in late stage HIV-1 replication, CD63 siRNA was transfected into U1/HIV-1 cells, which are chronically infected monocytoid cells harboring 2 integrated copies of provirus per cell [32,33]. Early steps of HIV-1 replication (such as entry and reverse transcription) are not required for replication in these cells, and virus production can be induced with 3 phorbol 12-myristate 13-acetate (PMA). Silencing CD63 did not significantly inhibit intracellular p24 production ( Figure 3A), suggesting that CD63 does not regulate viral p24 protein production. However, extracellular HIV-1 p24 obtained from supernatants was significantly reduced in CD63-and TSG101-depleted U1 cells (p < 0.05) compared to the ERBB2IP siRNA control ( Figure 3B). TSG101 is involved in post-integration trafficking and release of HIV-1 Gag proteins [34]. Similar to TSG101, these findings indicate that CD63 may also serve a role in the release of HIV-1 from the cell, which may relate to why HIV-1 readily incorporates CD63 into nascent virions. These outcomes are also consistent with our previous studies performed in primary MØ. Treatment with CD63-specific siRNA 3 days post-infection, significantly inhibited virus production in culture supernatants [26], suggesting an important role for CD63 in MØ during HIV-1 replication events occurring after proviral integration, and possibly during egress. Previously, we have shown that CD63 silencing significantly inhibits HIV-1 replication in MØ [25,27], U1/ HIV-1 producing cells [26], U373 CD4 + cell line [26], and presently in T lymphocytes ( Figure 2C), revealing that CD63 expression is broadly required for replication in relevant HIV-1 target cells.
We further demonstrate dual roles for CD63, showing that it contributes to both early and late events of the HIV-1 replication cycle. CD63 mediated efficient entry of HIV-1 Env pseudotyped viruses into CCR5 + MØ (Figure 1). When viral entry was redirected though the endocytic pathway by pseudotyping virions with MLV or VSV, infectivity in MØ was not affected by CD63 silencing. However, CD63 silencing did reduce infectivity with R5-and R5X4-tropic viral pseudotypes that depend on CD4 and CCR5 for entry. This indicates (i) that CD63 supports Env-mediated entry or fusion events facilitated though CD4 and CCR5, and (ii) that the entry events are specific for the HIV-1 Env protein. In support, other investigators have demonstrated that the extracellular domain 2 of CD63 partially colocalizes with CD4 at the cell surface of macrophages [35]. CD63 also exerts a role in later stages of the HIV-1 replication cycle. As CD63 is a major component of late endosomes, it is possible that CD63 is also involved in HIV-1 maturation and budding in MØ. In this study, we show that silencing CD63 expression decreases extracellular, but not intracellular, HIV-1 p24 production ( Figure 3A, B), indicating that CD63 may influence viral egress. In MØ, nascent HIV-1 virions accumulate and assemble in multi-vesicular bodies (MVBs) or endosomes containing tetraspanins such as CD63 and CD81, and may be released subsequently as exosomes. However, in T-cells HIV-1 Gag assembly is localized to the plasma membrane [14] where CD63 is also found. As CD63 is located in endosomes and the plasma membrane where HIV assembly takes place, it is possibly involved in virus assembly at these sites. This complex localization pattern of CD63 suggests that its intracellular trafficking and distribution must be tightly regulated.
Several studies have shown that CD63 supports HIV-1 replication [25][26][27], is located at assembly initiation sites [36][37][38], and is incorporated into nascent virions [15,16], yet viral binding partners with CD63 have remained elusive. Exactly how CD63 is targeted to viral assembly sites and incorporated into budding HIV-1 particles also remains unclear. A plausible explanation is that virus budding may occur in TEMs and that CD63 must be present for the completion of the recycling loop with return of both CD63 and viral proteins to a common site for virus release. CD63-containing TEMs may be required for budding of virus from these cells.

Conclusions
In conclusion, we have shown that CD63 plays a role both in the early entry stages of the HIV-1 replication cycle in concert with CD4 and/or CCR5, and in a late were detected using a p24 Capture ELISA kit (ImmunoDiagnostics, Woburn, MA). TSG101-specific siRNA was used as a control because it is involved in post-integration trafficking, assembly, and release of HIV-1 Gag proteins [34]. Experiments were performed in triplicate. *P < 0.05, compared to ERBB2IP group.
replicative step, possibly at the level of egress. Further research is needed to gain insight into the molecular mechanisms of CD63 that facilitate early and late events in the HIV-1 replication cycle.
Our experimental research has been approved by the Institutional Biosafety Committee (IBC) at the University of Texas Medical Branch, Galveston, Texas. The IBC ref- Dual-tropic (R5/X4) HIV-1 89.6 is an HIV-1 laboratory adapted strain originally isolated from an infected individual. The original preparation was prepared from molecularly cloned virus, and grown in CEMx174 cells [40] and donated to the NIH AIDS Research and Reference Reagent Program. We purchased HIV-1 89.6 from the Virology Core Facility, Center for AIDS Research at Baylor College of Medicine, Houston, TX, which was prepared and propagated in human PBMCs. HIV-1 89.6 stock containing 49.977 ng/ml of HIV p24 with 261,300 TCID50/ ml was used to infect PBLs at an m.o.i of 0.02. In order to measure down regulation of target genes in PBLs, cell lysates were prepared with cell lysis buffer (Promega, WI) at 2 days post-transfection, and were collected for Western blot analysis, or RNA was extracted for real time quantitative PCR analysis. PBLs were transfected with 50 nM siRNAs in 6-well plates (5 × 10 5 cells/ well) using Nucleofector®, and standard applications of manufacture's protocol (Amaxa) were followed. For Western blot analysis, anti-human CD63 monoclonal antibody (Santa Cruz) was used.

Real time quantitative PCR
Total mRNA was isolated from siRNA-transfected cells and MØ using RNeasy Mini Kits (Qiagen). CD63 specific primers and probe were purchased from Applied Biosytems (Carlsbad, CA). All reactions were performed using Applied Biosystems TaqMan Universal Master Mix and run using an Applied Biosystems 7500 Fast Real Time PCR system and 7500 Fast System Software. Silencing of target genes was determined by normalizing target gene expression to GAPDH expression (n = 3).

Statistical analysis
The results are expressed as mean ± SD of at least four wells. Two-tailed, paired Student's t-test was used to determine statistical significance. P values of <0.05*, and <0.01** were considered significant.