Sulfated polysaccharides inhibit lymphocyte-to-epithelial transmission of human immunodeficiency virus-1.

We have previously suggested that sulfated polysaccharides could be used in a vaginal formulation to inhibit infection by human immunodeficiency virus (HIV-1). This supposition was based on studies in which we developed and employed an in vitro model to simulate the mechanism of HIV-1 transmission during coitus. We found that adhesion of mononuclear cells to epithelia was the initial step in infection and speculated that blocking adhesion would prevent HIV-1 transmission. We observed that certain sulfated polysaccharides prevented adhesion of lymphoma cell lines to epithelial cell lines, which were derived from the genital tract, in concentrations of a few milligrams per milliliter; and we theorized that sulfated polysaccharides could thus be used as active ingredients in a topical "microbicide." In the present in vitro study, evidence is presented that a number of sulfated polysaccharides, including carrageenan, dextran sulfate, heparin, fucoidan, and pentosan polysulfate, are capable of blocking infection by mechanisms other than adhesion at concentrations of a thousand times lower than the dosages that are needed to block cell adhesion. One of these compounds, iota carrageenan, is capable not only of blocking infection of epithelia at concentrations of 1-2 micrograms, but of blocking adhesion to a far greater extent than the other sulfated polysaccharides tested. For this reason, as well as for considerations of safety, stability, and gelling properties, we suggest that iota carrageenan may be the best choice of the sulfated polysaccharides tested for use as a vaginal microbicide. The same in vitro model was employed to decipher the cell surface molecules involved in lymphocyte-to-epithelial adhesion. To accomplish this, we screened for the presence of cell adhesion molecules (CAMs), carbohydrates, proteoglycans, and carbohydrate-binding sites. HIV-1-infected lymphocytic cells expressed a CAM profile typical of activated, infected cells (e.g., HLA-DR+, CD4-, LFA-1+, ICAM-1+, LFA-3+, CD2+) whereas epithelia expressed few CAMs (LFA-3, ICAM-1, VLA-5, CD44, CD26, sLEX). Both cell types expressed heparan sulfate and chondroitin sulfate proteoglycans. A variety of sugars (mannose, fucose, galactose, Nac-galactosamine, Nac-glucosamine) were also present, but these cells expressed few carbohydrate-binding sites; lymphocytes bound beta-galactose. We were unable to block the adhesion with anti-CAM antibodies or with exogenous sugars. When enzymes were used against sulfated cell surface molecules, chondroitinase was found to block the adhesion. Our evidence suggests that this CAM-independent adhesion may be a lectin-glycosaminoglycan interaction.