Syndecan-3 is a dendritic cell-specific attachment receptor for HIV-1

Abstract

Dendritic cells (DCs) efficiently capture HIV-1 and mediate transmission to T cells, but the underlying molecular mechanism is still being debated. The C-type lectin DC-SIGN is important in HIV-1 transmission by DCs. However, various studies strongly suggest that another HIV-1 receptor on DCs is involved in the capture of HIV-1. Here we have identified syndecan-3 as a major HIV-1 attachment receptor on DCs. Syndecan-3 is a DC-specific heparan sulfate (HS) proteoglycan that captures HIV-1 through interaction with the HIV-1 envelope glycoprotein gp120. Syndecan-3 stabilizes the captured virus, enhances DC infection in cis, and promotes transmission to T cells. Removal of the HSs from the cell surface by heparinase III or by silencing syndecan-3 by siRNA partially inhibited HIV-1 transmission by immature DCs, whereas neutralizing both syndecan-3 and DC-SIGN completely abrogated HIV-1 capture and subsequent transmission. Thus, HIV-1 exploits both syndecan-3 and DC-SIGN to mediate HIV-1 transmission, and an effective microbicide should target both syndecan-3 and DC-SIGN on DCs to prevent transmission.

Fig. 1.

HIV-1 exploits trypsin-sensitive molecules other than DC-SIGN to bind DCs. The 0.25 × 10⁶ cells per ml DCs were preincubated with 1 mg/ml mannan or trypsin, exposed to 1 ng of p24 HIV-1 [strains NL4.3 Δ-envelope, NL4.3 (X4), NL4.3BaL (R5), JR-CSF (R5), or NL-92UG021.6 (X4)] for 2 h at 37°C, washed, and lysed, and the p24 content of lysate was measured by ELISA. Results are expressed in percentage of p24 captured relative to the inoculum. Error bars represent the standard deviation of duplicates (n = 3 donors).

Fig. 2.

HSPGs mediate attachment of HIV-1 to DCs. (A) DCs were analyzed for HS expression by FACS. Open histograms represent isotype control; filled histograms represent specific antibody staining. (B) DCs were treated with trypsin and stained for DC-SIGN (DCN46) and HS (10E4). Expression is depicted as mean fluorescence intensity (MFI). (C) DCs were pretreated or not with heparinase I, exposed to 1 ng of p24 HIV-1 for 2 h at 37°C, washed, and lysed, and the p24 content was measured by ELISA. Results are expressed as percentage of p24 captured relative to the inoculum. Error bars represent the standard deviation of duplicates. (D) DCs as well as parental-, syndecan-2-, and syndecan-3-transfected Namalwa cells were treated with heparinase I, II, and III and stained for HS. (E) DCs were pretreated with heparinase III, mannan, or a combination of both; exposed to 1 ng of p24 HIV-1 for 2 h at 37°C; washed; and lysed; and the p24 content was measured by ELISA. Results are expressed as percentage of p24 captured relative to the inoculum. Error bars represent the standard deviation of duplicates (n = 3 donors).

Fig. 3.

Syndecan-3 represents the main HSPG on DCs. (A) Monocytes, macrophages, and immature DCs were stained with antibodies against syndecan-1 (1D4), syndecan-2 (10H4), syndecan-3 (1C7), syndecan-4 (8G3), and DC-SIGN (AZN-D1) and analyzed by FACS (n = 10 donors). The expression is depicted as MFI. Open histograms represent isotype control; filled histogram specific antibody staining. (B) Cryosections of human cervix were stained for syndecan-3 (1C7, red color) and DC-SIGN (DCN-46, green color) and analyzed by immunofluorescence microscopy.

Fig. 4.

HIV-1 binds to DCs by the HS chains of syndecan-3. (A) DCs were transfected with control or syndecan-3 siRNA; stained with antibodies against CD83 (HB-15E), HLA-DR (G46–6), DC-SIGN (DCN46), syndecan-3 (1C7), and HS (10E4); and analyzed by FACS. (B) DCs were transfected with control or syndecan-3 siRNA, treated with heparinase III or mannan, exposed to 1 ng of p24 NL4.3 (X4) for 2 h at 37°C, washed, and lysed, and the p24 content was measured by ELISA. Results are expressed in percentage of p24 captured relative to the initial inoculum. The results were normalized to the no treatment control, and the mean of three different donors is depicted as relative HIV-1 binding. Error bars represent standard deviation of three donors, and the means were analyzed by ANOVA. **, P < 0.01, versus the no treatment condition.

Fig. 5.

Binding of HIV-1 to syndecan-3 on DCs retains infectivity after long-term culture. DCs as well as 0.25 × 10⁶ cells per milliter parental-, syndecan-3-, and DC-SIGN-transfected Namalwa cells were exposed to 1 ng p24 NL4.3 (X4) for 2 h at 37°C, washed, and cultured at 37°C. Every day, aliquots of HIV-1-pulsed Namalwa cells were added to TZM cells. As a control, an amount of virus, identical to the virus initially attached to the cells, was incubated at 37°C in medium without cells. TZM infection was scored after 2 days. (B) DCs were treated with heparinase III, mannan, or the combination of both before virus exposure. (C) DCs were transfected with control or syndecan-3 siRNA or pretreated with heparinase III before virus exposure. The previous experiments are representative of three independent experiments by using three different DC donors.

Fig. 6.

Syndecan-3 facilitates HIV-1 infection of DCs. The 0.1 × 10⁶ cells per milliter DCs were transfected with control or syndecan-3 siRNA; treated with heparinase III, mannan, or a combination of both; and were exposed to 1 ng of p24 NL4.3 (X4). After 1 day, cells were washed and replated in a 96-well plate. Viral replication was monitored by p24 ELISA. Error bars represent the standard deviation of duplicates. These experiments are representative of three independent experiments by using three different DC donors.

Fig. 7.

HSPGs and C-type lectins mediate HIV-1 transmission by DCs. (A) The 0.25 × 10⁶ cells per milliter DCs were transfected with control or syndecan-3 siRNA; treated with heparinase III, mannan, or the combination of both; and exposed to 1 ng of p24 HIV-1. After 2 h, cells were washed and titrated on TZM cells to determine transmission. To compare the data of the different virus, results are depicted relative to the transmission observed with untreated DCs. The means of three different donors are depicted. Error bars represent standard deviation of three donors, and the means were analyzed by ANOVA. **, P < 0.01, versus no treatment control. (B and C) The 0.1 × 10⁶ cells per millilter DCs were transfected with control or syndecan-3 siRNA; treated with heparinase III, mannan, or the combination of both; and exposed to 1 ng of p24 NL4.3 (X4). Without washing, T cells were added, and cocultures were plated in 96-well plates. Supernatants were collected after different days, and viral replication was monitored by p24 ELISA. As controls, DCs were cultured without T cells, and T cells without DCs were infected with the initial inoculum as for DCs. Error bars represent the standard deviation of duplicates.