Quantitative Expression and Virus Transmission Analysis of DC-SIGN on Monocyte-Derived Dendritic Cells

Cells and cell lines.

To obtain MDDCs, we purified monocytes from peripheral blood mononuclear cells by discontinuous Percoll gradient centrifugation. The low-density fraction (monocyte enriched) was further depleted of lymphocytes by a 2-h plastic adhesion step at 37°C followed by extensive washing in prewarmed culture medium. This resulted in highly purified monocytes as determined by flow cytometry using anti-CD14 (90 to 95% positive) or anti-CD11c (95 to 98% positive) MAbs. To generate immature MDDC, purified monocytes were cultured in RPMI 1640 supplemented with granulocyte-macrophage colony-stimulating factor (50 ng/ml) and IL-4 (100 ng/ml) for 7 days. Each batch of immature MDDCs was analyzed by flow cytometry and had the following phenotype: HLA-DR^high CD11c^high CD14⁻ CD80⁺ CD83⁻ (data not shown). Previously described stably DC-SIGN- or DC-SIGNR-transfected 293 cells were used for flow cytometry, ICAM-3 and virus binding, and for virus transmission experiments (29, 30). THP cells expressing DC-SIGN were kindly provided by Dan Littman (New York University) and have been described previously (17).

Antibodies.

A phycoerythrin (PE)-conjugated goat anti-mouse Fab fragment (Caltag, Burlingame, Calif.) was used as a secondary antibody for standard fluorescence-activated cell sorter (FACS) assays. PE-conjugated DC11 was used for all quantitative FACS measurements (4). We used a previously characterized panel of MAbs specific for the lectin domain of DC-SIGN and/or DC-SIGNR (21). These MAbs were obtained by immunizing mice with murine 3T3 cells expressing human DC-SIGN or DC-SIGNR, and MAbs were produced as previously described (21). The name and isotypes of the antibodies are as follows: 506 (clone 120506, immunoglobulin G2a [IgG2a]), 507 (clone 120507, IgG2b), 516 (clone 120516, IgG2a), 518 (clone 120518, IgG2a), 526 (clone 120526 IgG2a), 531 (clone 120531, IgG1), 604 (clone 120604, IgG2b), and 612 (clone 120612, IgG2a). Alophycocyanin-conjugated HLA-DR, fluorescein isothiocyanate-conjugated CD14, fluorescein isothiocyanate-conjugated CD80, and tri-color-conjugated CD83 were all obtained from Caltag. Cy-chrome-conjugated CD11c was purchased from Pharmingen (San Diego, Calif.).

Generation of virus stocks.

Stocks of replication-competent luciferase reporter viruses were generated as described previously (29, 30). In brief, 293T cells were calcium phosphate-transfected with the proviral genomes, and the culture medium was replaced by fresh Dulbecco's modified Eagle medium (DMEM) 18 h after transfection. The culture supernatant was harvested 48 h after transfection, sterile filtered using 0.45-μm-pore-size filters, divided into aliquots, and frozen at −80°C. The amount of p24/p27 viral antigen in the supernatants was quantified using commercially available antigen capture enzyme-linked immunosorbent assay (ELISA) kits. HIV pseudotypes bearing the Ebola Zaire strain glycoprotein (EboZ-GP) were generated in a similar manner. However, 293T cells were cotransfected with an Env-defective proviral HIV genome harboring the luciferase gene in place of nef and an expression vector encoding the glycoprotein of the EboZ-GP.

Cell culture.

293T cells were maintained in DMEM supplemented with 10% fetal calf serum. T-REX cells (Invitrogen) expressing DC-SIGN or DC-SIGNR upon induction with doxycycline (Sigma, St. Louis, Mo.) were maintained in DMEM supplemented with 10% fetal calf serum, 100 μg of Zeocin (Invitrogen)/ml, and 5 μg of blasticidin (Invitrogen)/ml. Parental T-REX cells were cultivated in the same medium, but without Zeocin. CEMx174, C8166, and THP DC-SIGN as well as parental THP cells were cultured in RPMI 1640 medium containing 10% fetal calf serum and penicillin-streptomycin.

Flow cytometry and quantification.

Stably DC-SIGN- or DC-SIGNR-transfected 293 cells, treated overnight with doxycycline, or MDDCs were stained in FACS buffer (phosphate-buffered saline supplemented with 3% fetal calf serum and 0.02% sodium azide) for 30 min on ice with MAbs at a final concentration of 10 μg/ml. The samples were washed and incubated with PE-conjugated goat anti-mouse Fab fragments (Caltag) (1/100) for 30 min on ice and then washed and resuspended in FACS buffer containing 2% paraformaldehyde. The samples were analyzed with a FACScan (Becton Dickinson, San Jose, Calif.) cell analyzer using the CellQuest software for data evaluation. Dead cells were excluded on the basis of their forward- and side-scatter characteristics. For quantification we used the Quantum Simply Cellular microbead kit from Sigma according to the manufacturer's instructions and as previously described (24, 30). Briefly, quantification was performed by converting the geometrical mean channel fluorescence (GMCF) into antibody-binding sites (ABS). The kit contains a mixture of five microbead populations of uniform size, coated with goat anti-mouse antibodies, that have differing abilities to bind mouse antibodies (i.e., ranging from 0 to ∼250,000 molecules). PE-conjugated CD11 was added at saturating amounts to cells and separately to 100,000 beads. All samples including the beads were processed and analyzed identically. The binding capacities of the stained microbeads were then regressed against the corresponding GMCF of each bead population, and the GMCF of the antigen analyzed on the sample cells was converted to ABS per cell by comparison with the regression curve generated. The GMCF of the isotype control for each experiment was converted to ABS and subtracted from the ABS value obtained with the experimental sample.

ICAM-3 binding assays.

We used the assay we have previously described (2). Briefly, soluble Fc-ICAM-3 protein (R&D Systems, Minneapolis, Minn.) was iodinated by using Iodogen (Pierce). Specific activities of 500 to 2,000 Ci/mmol were obtained by using 5 μg of protein with 500 μCi of Na ¹²⁵I for 20 min in 5-ml glass tubes precoated with 10 μg of Iodogen by chloroform evaporation. Radiolabeled proteins were purified from free Na ¹²⁵I by separation through a 0.3-ml Dowex column prepared in a 1-ml syringe and preequilibrated in a mixture containing 50 mM HEPES (pH 7.4), 5 mM MgCl₂, 1 mM CaCl₂, 1% bovine serum albumin (BSA), and 150 mM NaCl. Protein fractions were eluted in the void volume of the column, and the fractions containing peaks of labeled protein were combined. Stably DC-SIGN- or DC-SIGNR-transfected cells were induced overnight with doxycycline, washed once with phosphate-buffered saline, and resuspended in binding buffer (50 mM HEPES [pH 7.4], 2 mM magnesium chloride, 2 mM calcium chloride,; 0.5% BSA). Cells (10⁶) were incubated with 50,000 cpm of Fc-ICAM-3 for 60 min at room temperature. Cells were collected onto Brandel grade GF/B filters with wash buffer (same as binding buffer plus 150 mM sodium chloride and no BSA) using a cell harvester. Filters were counted using a Wallac Wizard 1470 automatic gamma counter. Percent binding was determined by dividing the counts from the filters by the input radioactivity after deduction of the background counts obtained on parental 293 T-REX cells.

Inhibition of DC-SIGN/DC-SIGNR engagement by HIV and SIV particles and EboZ-GP-pseudotyped HIV particles.

DC-SIGN/DC-SIGNR-expressing cells were seeded in 96-well plates and preincubated with 20 μg of the DC-SIGN/DC-SIGNR-specific MAbs or mannan/ml for 30 min. Thereafter the cells were typically pulsed with 15 ng of p24-normalized or 5 ng of p27-normalized luciferase reporter virus stocks/well for 3 h at 37°C. Unbound virus was removed by washing the cells three times with DMEM culture medium, and virus binding was quantified by lysing the cells in 1% Triton X-100 and assessing the amount of bound viral antigen by antigen capture ELISA. To determine transmission, the virus-pulsed cells were cocultivated with CEMx174 target cells and the luciferase activity in the cultures was measured 3 days after the start of the cocultivation. Blocking of EboZ-GP-mediated binding to DC-SIGN/DC-SIGNR was determined in a cis infection assay. T-REX cells were seeded in 96-well plates, induced with doxycycline to express the respective lectins, and preincubated with the MAbs as described above. Subsequently the cells were challenged with 5 ng of EboZ-GP-bearing HIV luciferase pseudotypes/well, and luciferase activity in the cultures was quantified 3 days after infection.

Affinities of lectin domain-specific MAbs for DC-SIGN.

We previously described the generation of eight MAbs directed against the lectin domain of DC-SIGN (21). One is specific for DC-SIGNR, four are specific for DC-SIGN, and three recognize both DC-SIGN and DC-SIGNR (Table 1). The DC-SIGN-specific MAbs also recognize rhesus macaque DC-SIGN by FACS and by immunofluorescent staining of frozen tissue sections (21). To further characterize this panel of MAbs, we determined their respective affinities for DC-SIGN or, in the case of the DC-SIGNR-specific MAb 604, DC-SIGNR. Stably transfected 293 cells were stimulated overnight with doxycycline to induce DC-SIGN or DC-SIGNR expression (29, 30), stained with different concentrations of each MAb, and processed for FACS analysis. We found that all MAbs exhibited similar affinities for DC-SIGN and/or DC-SIGNR (Table 1), with approximately 1.5 μg of each MAb/ml being needed to obtain 50% of the maximum staining intensity, and that there were no significant differences in maximum staining intensities between the MAbs (2, 23). MAbs that recognized both DC-SIGN and DC-SIGNR bound to both proteins with similar affinities (data not shown).

DC-SIGN quantification on monocyte-derived DCs reveals very high copy numbers of DC-SIGN.

The ability of DC-SIGN to bind and transmit HIV is related in part to DC-SIGN expression levels (29). To determine if there are significant differences in DC-SIGN expression on MDDCs that could impact virus transmission efficiency, we isolated monocytes from seven different donors three to five times each. MDDCs were derived by GM-CSF and IL-4 treatment for 7 days, and the MDDCs were shown by FACS to have the expected phenotype of immature DCs, i.e., HLA-DR^high CD11c^high CD14⁻ CD80⁺ CD83⁻ (data not shown). Maturation of MDDCs as judged by CD83 upregulation (data not shown) was achieved by exposure to 100ng of lipopolysaccharides (LPS)/ml. DC-SIGN expression levels were measured using quantitative FACS, which provides a reproducible means for converting mean channel fluorescence values into ABS through the use of synthetic beads that bind fixed numbers of antibody molecules (24). To eliminate errors associated with the use of secondary antibodies, we used MAb DC11 directly conjugated to PE at saturating levels. However, similar results were also obtained with a MAb directed to the carbohydrate recognition domain of DC-SIGN (data not shown). While there was variation in DC-SIGN expression on MDDCs from the same donor over time as well as between donors (Fig. 1), DC-SIGN was always expressed in excess of 100,000 copies per cell, and often at much higher levels. On cell lines, we have found that approximately 60,000 copies of DC-SIGN are needed for efficient virus transmission activity (29). It is important to note that the quantitative FACS assay can accurately measure ABS values to approximately 250,000. Values in excess of this are derived from projecting the standard curve beyond the last standard and so should not be deemed as definitive. These experiments show that for the individuals examined, DC-SIGN is expressed at high levels on MDDCs, with some variability over time and between individuals.

The lectin domain-specific MAbs do not efficiently inhibit the binding of ICAM-3.

Having identified MAbs to the lectin domain of DC-SIGN and showing that all bind with similar affinities, we tested their abilities to block binding of soluble ICAM-3 to 293 cells expressing DC-SIGN or DC-SIGNR (Fig. 2). To average data between experiments, we normalized results to the positive control in each experiment. Thus, the amount of ICAM-3 or virus (Fig.2 through 6) bound to cells expressing DC-SIGN in the absence of inhibitors was set to 100%, making it possible to average results between experiments that used different ICAM-3 or virus stocks. Raw data values for representative experiments are included in the legends for Fig. 2 through 6. We found that only MAb 531 inhibited significantly the binding of ICAM-3 to DC-SIGN (by 63% ± 12%). MAbs 604 and 612 competed the binding of ICAM-3 to DC-SIGNR by 67% ± 8% and 52% ± 11%, respectively (data not shown). Thus, none of these MAbs potently inhibited binding of soluble ICAM-3 to DC-SIGN or DC-SIGNR.

Identification of MAbs which inhibit HIV binding and transmission by DC-SIGN/DC-SIGNR.

We investigated the ability of the MAbs to inhibit HIV binding and potentiation of infection in trans by DC-SIGN/DC-SIGNR-expressing cell lines in order to identify MAbs that could be used to investigate the role of DC-SIGN in virus transmission on MDDCs. T-REX cells were induced to express DC-SIGN or DC-SIGNR and were preincubated with the indicated MAbs or mannan, a carbohydrate that blocks virus interactions with both DC-SIGN and DC-SIGNR (17). The cells were subsequently pulsed with HIV-1 NL4-3 luciferase reporter virus, extensively washed, and either lysed with the amount of bound virus being quantified by p24 ELISA or cocultivated with target C8166 cells. The efficiency of virus transmission was then assessed by quantifying the luciferase activity in the cocultures 3 days later (Fig. 3A ).

We found that five out of the eight antibodies that were tested efficiently blocked virus binding and transmission (Fig. 3A) from DC-SIGN-expressing cells, inhibiting both functions by at least 70% compared to the mock control. Preincubation with mannan inhibited binding and transmission to a comparable degree. In contrast, a control mouse immunoglobulin as well as the DC-SIGNR-specific MAb 604 and the DC-SIGN-specific MAb 507 had no significant effect on HIV binding to DC-SIGN. The MAb 518 exhibited detectable blocking activity in this cellular context, inhibiting virus binding to about 50% and transmission to about 30% relative to the mock control. HIV interactions with DC-SIGNR were efficiently blocked by the DC-SIGNR-specific MAb 604 and the dual-specific MAb 612, both of which reduced binding and transmission by at least 85% compared to the mock and mouse IgG (mIgG) controls (Fig. 3B). MAb 526 also reduced virus binding and transmission via DC-SIGNR, though less efficiently than was observed with MAbs 604 and 612.

Finally we examined HIV binding and transmission from THP cells stably expressing DC-SIGN as well as from DC-SIGN-negative, parental THP cells. DC-SIGN-positive THP cells transmitted bound virus four- to eightfold more efficiently than 293 cells despite similar expression levels. Despite this, the MAbs inhibited HIV binding to and transmission from DC-SIGN-positive THP cells more efficiently than from the inducible 293 cell lines (Fig. 3C). Seven out of eight MAbs reduced binding and transmission to background levels. Even MAb 507, which did not significantly inhibit DC-SIGN engagement on T-REX cells, reduced binding and transmission from THP cells to about 10%, whereas the MAbs 516 and 526 diminished both functions to about 3% compared to the mock control.

Antibody inhibition of SIV transmission by rhesus macaque DC-SIGN.

The experimental inoculation of macaques with SIV is one of the most commonly used animal models for the HIV infection in humans. DCs in the rectal and vaginal mucosa of rhesus macaques express DC-SIGN (21). The importance of DC-SIGN for virus transmission in vivo could thus be evaluated by investigating whether interference with DC-SIGN function inhibits viral spread in macaques after a vaginal or rectal challenge. We have previously shown that MAb 507 inhibits binding and transmission of SIV to and from rhesus macaque DC-SIGN (21). Here, we tested the entire panel of MAbs for the ability to inhibit SIV transmission by rhesus macaque DC-SIGN (Fig. 4). T-REX cells expressing rhesus DC-SIGN were preincubated with the indicated MAbs, challenged with SIV luciferase reporter virus bearing the M-tropic 239 MER Env protein, vigorously washed, and cocultivated with target cells. Virus transmission was assessed by quantifying the luciferase activity in the cocultures 3 days later. We found that only MAbs 507 and 526 efficiently inhibited virus transmission by rhesus macaque DC-SIGN (Fig. 4).

Virus transfer by MDDCs does not solely depend on DC-SIGN.

Dendritic cells express DC-SIGN, CD4, and perhaps other molecules that could serve as attachment factors for HIV. To examine the role of DC-SIGN in this process, MDDCs were matured by exposure to LPS and preincubated with the indicated MAbs or mannan. The cells were pulsed with the virus strain NL4-3. After binding for 2 h, the cells were washed extensively and cocultured with CEMssR5 target cells. Luciferase activity in the cultures was quantified 3 days later. Mannan reduced transmission of NL4-3 by only 56% compared to results with the mIgG controls (Fig. 5). MAb 506 blocked transmission with the greatest efficiency, reducing infection in trans by 69% for HIV-1 NL4-3 relative to the mIgG controls. Direct addition of virus to DCs did not result in efficient infection (data not shown), indicating that the luciferase activity in the cocultures was due to infection of the target cells. When immature MDDCs were used, inhibition of virus transmission by the MAbs and by mannan was somewhat less efficient and more variable (data not shown). Thus, even though mannan treatment as well as some DC-SIGN-specific MAbs efficiently blocked DC-SIGN-mediated virus binding and transmission to and from THP and 293 cells, inhibition of virus transmission from either mature or immature MDDCs was inefficient. These results indicate that other molecules play important roles in virus transmission from MDDCs or that the MAbs and mannan fail to inhibit virus interactions with DC-SIGN in the context of dendritic cells.

Antibody blocking of EboZ-GP engagement of DC-SIGN and DC-SIGNR.

We recently found that DC-SIGN/DC-SIGNR bind to the glycoprotein (GP) of Ebola virus and strongly enhance infection of target cells in cis (Simmons et al., submitted). The expression of these lectins on critical targets of Ebola infection suggests that DC-SIGN/DC-SIGNR might play an important role in Ebola virus dissemination in and between hosts. We investigated whether the MAbs inhibited EboZ-GP mediated engagement of these lectins. T-REX cells expressing the indicated lectins were preincubated with the panel of MAbs or mannan and challenged with Env-defective HIV luciferase reporter viruses bearing the EboZ-GP, and luciferase activity was determined 3 days after infection (Fig. 6). Expression of DC-SIGN/DC-SIGNR rendered the cells 50- to 200-fold more permissive than parental T-REX control cells. Addition of mannan reduced infection to background levels, whereas mIgG did not significantly inhibit viral entry. Similar to the situation observed for HIV transmission, MAb 604 blocked binding to DC-SIGNR, while MAbs 612 and 526 inhibited EboZ-GP usage of DC-SIGN and DC-SIGNR. MAb 507 weakly inhibited entry into DC-SIGN-expressing T-REX cells and had no effect on infection of DC-SIGNR-positive cells. These results indicate that Ebola GP and HIV Env can efficiently interact with DC-SIGN/DC-SIGNR, leading to enhanced infection of target cells, and viral engagement of these lectins seems to require similar determinants in the respective lectin domains.