REDUCED PROVIRUS BURDEN AND ENHANCED HUMORAL IMMUNE FUNCTION IN AZT-TREATED SCID-FELINE MICE INOCULATED WITH FELINE IMMUNODEFICIENCY VIRUS

The lack of a safe, economical murine lentivirus model for human immun odeficiency virus type 1 (HIV-1) infection of humans has hampered the preclinical evaluation of potential antiviral compounds, vaccines, and biological response modifiers. A small animal model that does not emp loy HIV-1 is needed to minimize risk of accidental human exposure, enh ance efficient use of scarce experimental compounds, and reduce labora tory space necessary to conduct statistically significant in vivo tria ls. Feline immunodeficiency virus (FIV), an immunosuppressive lentivir us of domestic cats, has been used extensively as an animal model for the pathogenesis and therapy of human HIV-1 infection. Cats, however, are not amenable to large-scale efficacy trials because of their relat ively large size, high cost, and limited degree of physiologic charact erization, particularly with regard to drug metabolism, To adapt the f eline immune system to a small laboratory animal host, severe combined immunodeficient mice (SCID mice) were engrafted with feline lymphoid tissues (forming the SCID-fe mouse) and inoculated with FIV. Two quant itative parameters, the incidence of provirus detection in feline tiss ue grafts and the level of feline IgG in plasma, were used to demonstr ate the antiviral efficacy of 3'-azido-3'-deoxythymidine (AZT, azidoth ymidine, Retrovir(R), zidovudine) in the SCID-fe system, Of 17 SCID-fe mice inoculated with 7 X 10(6) peripheral blood mononuclear cells (PB MC) from an FIV-infected cat, eight had detectable FIV provirus in bot h the feline thymus and feline lymph node implants, as measured by pol ymerase chain reaction (PCR)/Southern blot analysis. Treatment of thes e mice with AZT at a dose of 125 mg kg(-1) day(-1) in drinking water b eginning 1 day prior to FIV inoculation and continuing throughout the study interval prevented the dual detection of provirus in feline lymp h node and thymus grafts of all mice tested. In a separate experiment, the level of spontaneous feline IgG production was quantified by ELIS A 2 weeks after FIV inoculation with and without AZT treatment. Mean p lasma feline IgG level of five SCID-fe mice inoculated with 10(3) TCID 50 cell-free FIV was 2.23 mg ml(-1). Mean feline IgG level of five mic e inoculated with the same quantity of FIV and treated with AZT beginn ing 1 day prior to virus inoculation and continuing for 2 weeks therea fter was 14.98 mg ml(-1). AZT significantly (P < 0.05) enhanced feline humoral immune function at a virus inoculum titer of 10(3) TCID50. Th ese data indicate that a 4 week preclinical study using the FIV-infect ed SCID-fe model can generate quantitative indicators of provirus burd en and immunologic function following treatment with an efficacious hu man antiretroviral compound.