Soluble complexes of regulated upon activation, normal T cells expressed and secreted (RANTES) and glycosaminoglycans suppress HIV-1 infection but do not induce Ca(2+) signaling

Abstract

Chemokines comprise a family of low-molecular-weight proteins that elicit a variety of biological responses including chemotaxis, intracellular Ca(2+) mobilization, and activation of tyrosine kinase signaling cascades. A subset of chemokines, including regulated upon activation, normal T cell expressed and secreted (RANTES), macrophage inflammatory protein-1alpha (MIP-1alpha), and MIP-1beta, also suppress infection by HIV-1. All of these activities are contingent on interactions between chemokines and cognate seven-transmembrane spanning, G protein-coupled receptors. However, these activities are strongly inhibited by glycanase treatment of receptor-expressing cells, indicating an additional dependence on surface glycosaminoglycans (GAG). To further investigate this dependence, we examined whether soluble GAG could reconstitute the biological activities of RANTES on glycanase-treated cells. Complexes formed between RANTES and a number of soluble GAG failed to induce intracellular Ca(2+) mobilization on either glycanase-treated or untreated peripheral blood mononuclear cells and were unable to stimulate chemotaxis. In contrast, the same complexes demonstrated suppressive activity against macrophage tropic HIV-1. Complexes composed of (125)I-labeled RANTES demonstrated saturable binding to glycanase-treated peripheral blood mononuclear cells, and such binding could be reversed partially by an anti-CCR5 antibody. These results suggest that soluble chemokine-GAG complexes represent seven-transmembrane ligands that do not activate receptors yet suppress HIV infection. Such complexes may be considered as therapeutic formulations for the treatment of HIV-1 infection.

Figure 1

RANTES–heparin complexes do not induce cytoplasmic Ca²⁺ mobilization in glycanase-treated or untreated PBMC. Normal PBMC were activated, as described in Materials and Methods, and harvested after 11 days in culture to measure the intracellular Ca²⁺ responses. RANTES-GAG complexes (○), then were analyzed with glycanase-treated (A) or untreated (B) PBMC. In control assays, glycanase-treated PBMC (A) were stimulated with heparin alone (▵) or with PBS (thin line). Untreated PBMC (B) were stimulated with RANTES–heparin complexes (○), heparin alone (▵), PBS (thin line), or RANTES (thick line). Fluorescence changes over time were measured by flow cytometry in the FL-1 window. One time unit is equal to 0.2 sec.

Figure 2

RANTES–heparin complexes inhibit chemotaxis in differentiated HL-60 cells. HL-60 clone 15 cells were differentiated (see Materials and Methods) by the addition of butyric acid and IL-5. On day 7 of treatment the cells were harvested and assayed for chemotaxis in response to RANTES (A), RANTES–heparin complexes (B), or heparin alone (C). For these experiments, RANTES-GAG complexes were formed as described in the text by mixing RANTES (3 μM) with a 10-fold excess (30 μM) of clinical grade, endotoxin-free heparin (Fujisawa Pharmaceutical, Osaka). Serial concentrations of the complexes were tested in parallel with matching concentrations of either RANTES or heparin alone. Each assay condition was performed in triplicate. Each data point represents the mean value of triplicate assays. SD values are shown with bars.

Figure 3

Binding of ¹²⁵I-RANTES–heparin complexes to glycanase-treated PBMC. Normal human PBMC were activated, harvested on day 6 of culture, and treated with glycanase mixture as described in Materials and Methods. (A) Binding of serial concentrations of ¹²⁵I-RANTES complexes (0.03–10 nM of the chemokine component) to treated cells. (B) Competition binding of 0.3 nM ¹²⁵I-RANTES complexes vs. serial concentrations of unlabeled RANTES complexes (0.01–100 nM chemokine component). Each data point represents the mean value of triplicate assays. SD values are shown with bars.

Figure 4

Inhibition of ¹²⁵I-RANTES complex binding to glycanase-treated PBMC by anti-CCR5 mAb 2D7. Normal human PBMC were activated as described previously and harvested on day 10 of culture to assay binding. Cells were treated with glycanase mixture for 1 hr. ¹²⁵I-RANTES was incubated with heparin (1 mg/ml) for 1 hr at 4°C. ¹²⁵I-RANTES complexes (0.3 nM RANTES component) were added to cells that had been treated with anti-CCR5 mAb 2D7 (10 μg/ml) (solid), normal mouse IgG (10 μg/ml) (striped), or PBS only (open). The values shown reflect the average of triplicate assays. SD values are shown with bars.

Figure 5

Inhibition of HIV-1 replication by RANTES-GAG complexes. PBMC were infected with HIV-1 NSI.03 and treated with RANTES, GAG, or RANTES-GAG complexes as described in Materials and Methods. (A) RANTES was tested after incubation with heparin (▴), heparan sulfate (●), chondroitin sulfate (□), or dermatan sulfate (○). Untreated RANTES was tested as control (■). The plots reflect the amount of the RANTES component present in each assay. (B) Results obtained with sham formulations containing only heparin (solid), heparan sulfate (open), chondroitin sulfate (striped), or dermatan sulfate (shaded) at a concentration matching the highest amount tested in the RANTES-GAG preparations (4 μM) are shown for comparison. Levels of infection determined 6 days postinfection by HIV-1 p24 antigen-capture ELISA are shown. Percent inhibition was determined for each assay relative to the control assays carried out in medium alone. The results obtained with RANTES, heparin, and RANTES plus heparin represent the mean of duplicate assays, whereas all other results represent the mean of triplicate assays.