Limitations of the human-PBL-SCID mouse model for vaginal transmission of HIV-1

Problem SCID mice reconstituted with human peripheral blood lymphocytes (PBL) are amenable to vaginal transmission of HIV-1. We investigated the effectiveness of this model to establish systemic HIV-1 infection. Method of study Eighty progesterone-primed C.B-17 SCID mice were reconstituted with human-PBLs and intravaginally inoculated with CCR5 HIV-1 (BaL or 92BR09) infected human-PBLs in the presence of human semen. After two weeks, viral RNA load in spleen, peritoneal lavage (PL), and serum was quantitated by the nucleic acid sequence-based amplification method. Results In five independent experiments, spleen from 8/60 (13.3%), PL from 7/60 (11.6%), and serum from 16/56 (28.5%) mice were positive for BaL HIV-1 infection. Similarly, spleen from 4/20 (20%), PL from 1/20 (5%) and serum from 5/20 (25%) mice vaginally inoculated with 92BR09-infected human-PBLs were positive for HIV-1. A one-sided power analysis using normal approximation revealed that at 5% significance level, the overall response rate need to increase form 0.29 to 0.9 and 80% of the control groups needs to achieve a response rate between 6/10 and 9/10 to make the assay feasible. Conclusion The incidence of vaginal transmission of CCR5 HIV-1 in the human-PBL-SCID mouse was low and variable, which constitutes a major disadvantage for preclinical evaluation of vaginal microbicides.