Previously, it was shown that the cell-membrane-expressed glycosphingolipid, globotriaosylceramide (Gb(3)/P(k)/CD77), protects against HIV-1 infection and may be a newly described natural resistance factor against HIV infection. We have now investigated the potential of a novel, water soluble, non-toxic and completely synthetic analogue of Gb(3)/P(k) (FSL-Gb(3)) to inhibit HIV-1 infection in vitro. A uniquely designed analogue, FSL-Gb(3), of the natural Gb(3)/P(k) molecule was synthesized. HIV-1(IIIB) (X4 virus) and HIV-1(Ba-L) (R5 virus) infection of PHA/interleukin-2-activated, peripheral blood mononuclear cells (PBMCs) and Jurkat T cells in vitro was assessed, as well as infection of U87.CD4.CCR5 by various clinical R5 tropic viruses after treatment with FSL-Gb(3). We monitored Gb(3), CD4 and CXCR4 expression by fluorescent antibody cell sorting and viral replication by p24(gag) ELISA. Total cellular Gb(3) was examined by glycosphingolipid extraction and thin layer chromatography. In vivo toxicity was monitored in mice by histological assessment of vital organs and lymphoid tissue. FSL-Gb(3) blocked X4 and R5 of both lab and clinical viral strains in activated PBMCs or the U87.CD4.CCR5 cell line with a 50% inhibitory concentration (IC(50)) of approximately 200-250 microM. FACS and TLC overlay showed that FSL-Gb(3) can insert itself into cellular plasma membranes and that cellular membrane-absorbed FSL-Gb(3) is able to inhibit subsequent HIV-1 infection. There was no effect of FSL-Gb(3) on cell surface levels of CD4 or CXCR4. Thus, FSL-Gb(3) can inhibit HIV-1 by two mechanisms: direct inhibition of virus and inhibition of viral entry. Infusion of FSL-Gb(3) into laboratory mice at doses well in excess of theoretical therapeutic doses was tolerated with no untoward reactions. Our results demonstrate the potential utility of using a completely synthetic, water soluble globotriaosylceramide analogue, FSL-Gb(3), having low toxicity, for possible future use as a novel therapeutic approach for the systemic treatment of HIV/AIDS.

Studies of potential HIV mucosal microbicides are difficult to undertake due to the requirement for a suitable animal model and the use of biosafety level 3 containment, which are not always available to researchers. Here we show the use of a mouse model of vaginal and rectal transmission of an HIV chimeric virus that does not require level 3 biosafety containment, to test the ex vivo efficacy of soluble Gb3 analogs for the prevention of mucosal HIV infection. The model uses a pseudoenvelope-typed vesicular stomatitis virus (VSV)/HIV recombinant virus that can infect all murine cell types. We demonstrate that the envelope glycoproteins VSV-G of VSV and gp-120 of HIV both bind Gb3. We show that soluble Gb3 analogs inhibit in vitro infection of cervical and vaginal-derived cell lines by both intact HIV and the VSV/HIV recombinant virus. Soluble Gb3 analogs incorporated into gel or used alone and applied directly to the vaginal and rectal mucosal tissue of mice were able to resist viral infection as monitored by PCR and quantitative real-time PCR copy number of HIV cDNA extracted from mouse tissue. Only a trend towards significant efficacy for prevention of mucosal transmission through lower copy number in the treatment groups was evident from these studies; however, this finding warrants further evaluation. In addition, we illustrate a methodology to evaluate inflammatory responses in either vagina or rectum after administration of soluble microbicidal compounds. These studies provide a potential new ex vivo methodology suitable for animal facilities in general, to screen microbicide drug candidates, including drug candidates that target viral proteins, for efficacy and safety, in order to accelerate development and discovery of prophylactic and therapeutic agents for HIV/AIDS.