Quantitative expression and virus transmission analysis of DC-SIGN on monocyte-derived dendritic cells

Abstract

The C-type lectins DC-SIGN and DC-SIGNR efficiently bind human immunodeficiency virus (HIV) and simian immunodeficiency virus (SIV) strains and can transmit bound virus to adjacent CD4-positive cells. DC-SIGN also binds efficiently to the Ebola virus glycoprotein, enhancing Ebola virus infection. DC-SIGN is thought to be responsible for the ability of dendritic cells (DCs) to capture HIV and transmit it to T cells, thus promoting HIV dissemination in vitro and perhaps in vivo as well. To investigate DC-SIGN function and expression levels on DCs, we characterized a panel of monoclonal antibodies (MAbs) directed against the carbohydrate recognition domain of DC-SIGN. Using quantitative fluorescence-activated cell sorter technology, we found that DC-SIGN is highly expressed on immature monocyte-derived DCs, with at least 100,000 copies and often in excess of 250,000 copies per DC. There was modest variation (three- to fourfold) in DC-SIGN expression levels between individuals and between DCs isolated from the same individual at different times. Several MAbs efficiently blocked virus binding to cell lines expressing human or rhesus DC-SIGN, preventing HIV and SIV transmission. Interactions with Ebola virus pseudotypes were also blocked efficiently. Despite their ability to block virus-DC-SIGN interactions on cell lines, these antibodies only inhibited transmission of virus from DCs by approximately 50% or less. These results indicate that factors other than DC-SIGN may play important roles in the ability of DCs to capture and transmit HIV.

FIG. 1.

Quantitative FACS measurements of DC-SIGN on MDDCs. Monocytes were purified from peripheral blood mononuclear cells by discontinuous Percoll gradient centrifugation, and MDDCs were derived by GMCSF and IL-4 treatment for 7 days (see Materials and Methods). MDDCs were obtained from seven donors three to five times each over a period of 3 months and phenotyped, i.e., MDDCs were HLA-DR^high CD11c^high CD14⁻ CD80⁺ CD83⁻ (data not shown). For each batch of MDDCs we quantified the surface level of DC-SIGN using PE-coupled DC11 (4) in conjunction with the Quantum Simply Cellular microbead kit (Sigma). Shown are the numbers of ABS per cell for each individual (A to G). Each measuring is represented by a symbol, and a horizontal bar shows the average number of ABS for each individual. Because of the technical characteristics of the quantification (see the text), numbers above 250,000 ABS/cell should be considered with caution.

FIG. 2.

Inefficient inhibition of ICAM-3 binding by the lectin domain-specific MAbs. The abilities of the lectin domain-specific MAbs to block binding of iodinated soluble, monomeric ICAM-3 to 293 cells expressing DC-SIGN is shown. Each measure represents the average from three independent experiments, each done in triplicate, ± the standard error of the mean. The number of each MAb is shown as well as its specificity. −, no MAb; mIg, mouse immunoglobulin; S, DC-SIGN specific; R, DC-SIGNR specific; B, DC-SIGN and DC-SIGNR specific. Within a representative experiment, DC-SIGN-expressing cells specifically bound 5.1% of the input material whereas preincubation with antibody 531 reduced specific binding to 1.7%. ICAM-3 was added in excess so that it would not be limiting for binding.

FIG. 3.

Inhibition of virus binding and transfer by DC-SIGN/DC-SIGNR-specific antibodies. (A) Blocking of HIV-1 NL4-3 interaction with DC-SIGN expressed on T-REX cells. T-REX cells were seeded in 96-well plates and induced to express the indicated lectins by an overnight incubation in medium containing 0.1 μg of doxycycline/ml. The cells were preincubated in 20 μg of the indicated MAbs or mannan/ml and then pulsed with NL4-3 luciferase reporter virus for 3 h. Unbound virus was washed away and to determine binding, the cells were lysed, and the amount of bound viral antigen was quantified by p24-ELISA (solid bars). Alternatively, the virus-bearing cells were cocultivated with CEMx174 target cells, and the efficiency of virus transmission was assessed by quantifying the luciferase activity in the cultures 3 days after the addition of the target cells (white bars). Virus binding and transmission are shown relative to the values obtained for the mock control. The data represent the average ± SEM of at least three independent experiments performed in quadruplicates. In separate, representative experiments, T-REX DC-SIGN cells recovered 7.4% of the input virus in the absence of inhibitors, and a luciferase activity of 1,232 cps was measured after transmission to target cells. In contrast, T-REX cells that did not express DC-SIGN bound 0.71% of the input virus, and transmission to target cells resulted in a luciferase activity of 71 cps. (B) Blocking of HIV-1 NL4-3 interaction with DC-SIGNR expressed on T-REX cells. The experiment was performed as for panel A except that DC-SIGNR-expressing cells were used. Binding is shown by solid bars; virus transmission is shown by white bars. In a representative experiment, mock-treated DC-SIGNR cells bound 4.2% of the input virus, and transmission to target cells resulted in a luciferase activity of 1,965 cps. Control cells bound 0.71% of the input virus, and after transmission to target cells a luciferase activity of 387 cps was measured. (C) inhibition of HIV-1 engagement of DC-SIGN ex-pressed on THP cells. THP-DC-SIGN cells and parental THP cells (no DC-SIGN) were used in this experiment. Antibody blocking of NL4-3 binding and transmission was assessed as described for panel A. A representative binding experiment resulted in 8.8% of virus recovery by mock-treated THP DC-SIGN cells compared to 0.65% recovered by control cells. After THP DC-SIGN-mediated virus transmission to target cells, a luciferase activity of 7,962 cps was measured whereas transmission from control cells resulted in 251 cps.

FIG. 4.

Inhibition of SIV transmission mediated by rhesus macaque DC-SIGN. The ability of the MAbs to inhibit transmission of a SIVmac239 MER Env reporter virus was assessed as described in the legend to Figure 3; however, T-REX cells expressing rhesus macaque DC-SIGN were used and were preincubated with 30 μg of the MAbs/ml. The infection efficiency is presented relative to mock-treated cells. The average ± standard error of the mean for three independent experiments, each performed in quadruplicate, is shown. In a representative experiment transmission from rhesus macaque DC-SIGN cells in the absence of inhibitors resulted in a luciferase activity of 1,920 cps whereas pretreatment with mannan reduced luciferase activity to 531 counts.

FIG. 5.

Transmission of virus from MDDCs is partially inhibited by anti-DC-SIGN MAbs. MDDCs were prepared from blood monocytes and matured by overnight incubation in medium containing LPS. The cells were pulsed with NL4-3 luciferase reporter virus. After a 1.5-h incubation the cells were washed and cocultivated with CEMssR5 cells, and luciferase activity in the cultures was determined 3 days after the start of the cocultures. CEMssR5 cells express both major HIV coreceptors, CXCR4 and CCR5. Transmission efficiency is presented relative to the mock control; the values represent the average ± standard error of the mean for three independent experiments performed in quadruplicate. Within a representative experiment a luciferase activity of 585 cps was detected after virus transmission by mock-treated cells. Pretreatment with mannan reduced luciferase activity in target cells to 245 cps.

FIG. 6.

Inhibition of EboZ-GP engagement of DC-SIGN and DC-SIGNR. The capacity of the MAbs to inhibit the EboZ-GP-mediated infection of T-REX cells expressing either DC-SIGN (black bars) or DC-SIGNR (white bars) was assessed. T-REX cell lines were induced to express human DC-SIGN/DC-SIGNR, preincubated with 20 μg of the indicated MAbs or mannan/ml, and infected with a HIV luciferase reporter virus pseudotyped with EboZ-GP. Infection efficiency was determined by assessing the luciferase activity 3 days after infection. Infection relative to control cells is presented. The average ± standard error of the mean for three independent experiments is shown. In a single representative experiment infection of untreated T-REX DC-SIGN and DC-SIGNR cells resulted in luciferase activities of 5,135 and 8,797 cps, respectively. Preincubation with mannan diminished luciferase production to 323 cps in DC-SIGN T-REX cells and 1,424 cps in DC-SIGNR T-REX cells.