Inhibition of HIV-1 transmission in trans from dendritic cells to CD4+ T lymphocytes by natural antibodies to the CRD domain of DC-SIGN purified from breast milk and intravenous immunoglobulins

Abstract

The present study demonstrates that human breast milk and normal human polyclonal immunoglobulins purified from plasma [intravenous immunoglobulins (IVIg)] contain functional natural immunoglobulin A (IgA) and IgG antibodies directed against the carbohydrate recognition domain (CRD) domain of the dendritic cell-specific intercellular adhesion molecule-3-grabbing non-integrin (DC-SIGN) molecule, which is involved in the binding of human immunodeficiency virus (HIV)-1 to dendritic cells (DCs). Antibodies to DC-SIGN CRD were affinity-purified on a matrix to which a synthetic peptide corresponding to the N-terminal CRD domain (amino-acid 342-amino-acid 371) had been coupled. The affinity-purified antibodies bound to the DC-SIGN peptide and to the native DC-SIGN molecule expressed by HeLa DC-SIGN+ cells and immature monocyte-derived dendritic cells (iMDDCs), in a specific and dose-dependent manner. At an optimal dose of 200 microg/ml, natural antibodies to DC-SIGN CRD peptide purified from breast milk and IVIg stained 25 and 20% of HeLa DC-SIGN+ cells and 32 and 12% of iMDDCs, respectively. Anti-DC-SIGN CRD peptide antibodies inhibited the attachment of virus to HeLa DC-SIGN by up to 78% and the attachment to iMDDCs by only 20%. Both breast milk- and IVIg-derived natural antibodies to the CRD peptide inhibited 60% of the transmission in trans of HIV-1(JRCSF), an R5-tropic strain, from iMDDCs to CD4+ T lymphocytes. Taken together, these observations suggest that the attachment of HIV to DCs and transmission in trans to autologous CD4+ T lymphocytes occur through two independent mechanisms. Our data support a role of natural antibodies to DC-SIGN in the modulation of postnatal HIV transmission through breast-feeding and in the natural host defence against HIV-1 in infected individuals.

Figure 1

Specificity of the interaction between the dendritic cell-specific intercellular adhesion molecule-3-grabbing non-integrin (DC-SIGN) CRD peptide and human immunodeficiency virus (HIV). (a) An enzyme-linked immunosorbent assay (ELISA) plastic plate, coated with DC-SIGN peptide, was incubated with gp160 in a buffer containing 0·02 m Tris, 0·15 m NaCl and 10 mm CaCl₂ supplemented or not with ethylenediaminetetraacetic acid (EDTA). Biotin-conjugated immunoglobulin G (IgG) to gp160 was then added before addition of steptavidin and measurement of the optical density (OD) at 490–650 nm. (b) HIV preincubated or not with DC-SIGN peptide (10 or 50 µg/ml) was added to HeLa DC-SIGN⁺ cells. After 1 hr of incubation at 37°, cells were washed and lysed, and the p24 concentration was determined by ELISA. Results are expressed as mean ± standard deviation calculated from four independent experiments. Statistical significance as determined by Mann–Whitney test is indicated. *P ≤ 0·05.

Figure 2

Anti-dendritic cell-specific intercellular adhesion molecule-3-grabbing non-integrin (DC-SIGN) CRD peptide reactivity and specific activity in breast-milk samples from HIV-seronegative and HIV-seropositive women. The anti-DC-SIGN CRD peptide reactivities of 11 samples from HIV-negative women (a) and 13 samples from HIV-seropositive women (b) were investigated by enzyme-linked immunosorbent assay (ELISA) as described in the Materials and methods. Results are expressed in μg/ml. (c) Specific activity of the same samples to DC-SIGN peptide was assayed by ELISA as described in the Materials and methods. Specific activity (SA) was calculated as SA = (anti-DC-SIGN antibodies)/(total immunoglobulins) × 100. Results are expressed in arbitrary units (AU). The statistical significance as determined by the Mann–Whitney U-test is indicated.

Figure 3

Dose-dependent and specific binding of antibodies to dendritic cell-specific intercellular adhesion molecule-3-grabbing non-integrin (DC-SIGN) CRD peptide. (a) Variable amounts (from 25 to 1000 µg/ml) of anti-DC-SIGN CRD peptide antibodies (Abs) purified from intravenous immunoglobulins (IVIg) and breast milk of HIV-seronegative women were incubated with DC-SIGN CRD peptide for 4 hr at 37°. Wells were washed and incubated with goat anti-human F(ab′)₂ coupled to peroxidase for 1 hr at 37°. (b) The same experiment was performed with two irrelevant peptides (CCR5 and gp120). Results are expressed as mean ± standard deviation calculated from two independent experiments.

Figure 4

Characterization of purified anti-dendritic cell-specific intercellular adhesion molecule-3-grabbing non-integrin (DC-SIGN) CRD peptide antibodies. (a) The isotype composition of breast-milk and intravenous immunoglobulins (IVIg) purified anti-DC-SIGN antibodies was evaluated by enzyme-linked immunosorbent assay (ELISA) as described in the Materials and methods. (b, c) The specific activity of breast-milk and IVIg purified anti-DC-SIGN antibodies was assayed as described in the Materials and methods. Specific activity (SA) was calculated as: SA = (anti-DC-SIGN IgA or IgG)/(total IgA or IgG) × 100. Results are expressed in arbitrary units (AU). (d) Determination of secretory immunoglobulin A (SIgA) subclass composition of DC-SIGN and CCR5 antibodies purified from breast milk using an SIgA1 ELISA quantification test. Results are expressed as mean ± standard deviation calculated from two independent experiments. OD, optical density.

Figure 5

Cytofluorometric analysis of the binding of purified anti-dendritic cell-specific intercellular adhesion molecule-3-grabbing non-integrin (DC-SIGN) CRD peptide antibodies to HeLa DC-SIGN⁺ cells and immature monocyte-derived dendritic cells (iMDDCs). HeLa DC-SIGN⁺ cells (a) or iMDDCs (b) were incubated with increasing amounts of purified anti DC-SIGN antibodies (50, 200, 400 µg/ml), anti-DC-SIGN monoclonal antibody (mAb) (positive control; 10 µg/ml) and column effluent (400 µg/ml) for 1 hr on ice. Cells were incubated with fluorescein isothiocyanate (FITC)-conjugated rabbit anti-human Ab for 30 min on ice and analysed using FACSCalibur flow cytometer and the cellquest software. (c) Purified anti-DC-SIGN antibodies (200 µg/ml) were preincubated for 30 min at room temperature with DC-SIGN peptide before addition to iMDDCs and binding determination by cytofluorometry. A representative experiment from five independent experiments is shown. MFI, mean fluorescence intensity.

Figure 6

Inhibition of human immunodeficiency virus (HIV) attachment on HeLa dendritic cell-specific intercellular adhesion molecule-3-grabbing non-integrin (DC-SIGN)⁺ cells and of HIV transmission in trans from immature monocyte-derived dendritic cells (iMDDCs) to CD4+ T cells by natural anti-DC-SIGN peptide antibodies. (a) HeLa DC-SIGN⁺ cells were preincubated with different amounts of anti-DC-SIGN antibodies (50, 200, 400 µg/ml) purified from breast milk and intravenous immunoglobulins (IVIg), with mannan (20, 200 µg/ml), with anti-DC-SIGN monoclonal antibody (mAb) (50, 200, 400 µg/ml) or with column effluent anti-DC-SIGN peptide-depleted immunoglobulins (400 µg/ml) before addition of R5-tropic HIV_JRCSF for 1 hr at 37°. The cells were then washed and lysed, and p24 antigen was quantified by capture enzyme-linked immunosorbent assay (ELISA). (b) iMDDCs were preincubated with differents amounts of breast milk and IVIg purified anti-DC-SIGN antibodies (20 or 50 µg/ml), mannan (200 µg/ml), anti-DC-SIGN mAb (20 µg/ml) or column effluent anti-DC-SIGN peptide-depleted immunoglobulins (400 µg/ml), before addition of R5-tropic HIV_JRCSF for 3 hr at 37°. The cells were washed and autologous CD4+ T cells were added. After 3 days of coculture the production of p24 antigen in culture supernatants was measured. Results are expressed as mean ± standard deviation calculated from five independent experiments for different donors. Statistical significance as determined by the Mann–Whitney U-test is indicated. *P ≤ 0·05.