CXCR4 in feline immunodeficiency virus infection and an experimental therapy

Abstract

Most strains of HIV-1 require CD4 and a chemokine receptor for attachment and entry into cells. The genetic adaptation of HIV-1 toward use of the CXCR4 chemokine receptor has been shown to play a critical role in the progression from latent infection to the development of AIDS. Whether CXCR4 receptor expression in different tissues in vivo correlates with increased susceptibility of those tissues to infection remains unclear, and is difficult to study in humans. Here we have employed the feline model for HIV to study the role of the CXCR4 chemokine receptor in lentiviral infection. In addition to mapping the distribution and infection rate of CXCR4 positive cells throughout lymphoid tissues, bone marrow and blood, we explored the effectiveness of chemokine receptor antagonists in blocking or ameliorating infection in vitro and in vivo. The natural CXCR4 ligand SD F-1α as well as the CXCR4-specific antagonist AMD3100 (1-1*-[1 ,4-phenylcncbis (methylene)]-bis(1,4,8,11-tetraazacyclotetradecane), were effective in inhibiting FIV infection of peripheral blood mononuclear cells (PBMC) in vitro. AMD3100 given to cats prior to intravenous virus challenge and continued twice daily for 30 days resulted in significantly decreased proviral loads in blood for the first 12 weeks post inoculation (p.i.). Blood proviral loads rebounded to levels similar to the controls after the drug was discontinued and tissue proviral levels were equal to controls at 6 months post-inoculation. Administration of AMD3100 to cats both prior to and for 30 days following oronasal challenge was not effective in reducing blood or tissue proviral loads. We hypothesize that failure of AMD3100 to inhibit mucosal infection resulted from relatively low drug concentrations outside of the plasma compartment that resulted in inferior blockade of CXCR4 in tissues. Furthermore, the drug failed to reduce viral loads when administered for seven days during chronic infection (22 weeks p.i.). The lack of antiviral activity in chronic FIV infection may be a reflection of low viral replication or sequestration of virus in macrophages during the latent phase rendering drugs targeting viral entry less effective during this time. It is also possible that, similar to HIV-1 infections, there may be a virus phenotype change in later stages of infection allowing the use of a broader spectrum of receptors. To study CXCR4 distribution and relationship to FIV infection, we sorted CXCR4 positive and negative fractions from lymph node, thymus, blood and bone marrow and measured proviral loads in lymphoid cell subsets by real time PCR. CXCR4 expression was significantly higher in lymph node than the other tissues examined both in proportion of cells staining and staining intensity per cell. Proviral burdens were significantly greater within the thymus than in lymph node, bone marrow or blood and higher concentrations of provirus were detected in T cells than B cells in blood and lymph node. CXCR4 was highly expressed in B and T cell subsets however proviral concentrations were similar between CXCR4 positive and negative fractions. Tissues from these studies were additionally used to compare distribution of provirus by real time PCR over time and with route of infection. The major viral reservoir in acute infection was the thymus whereas bone marrow had the highest proviral burden in chronic infection. Route of inoculation (intravenous versus oronasal) had no effect on proviral distribution when measured at 6 months post inoculation.