Approximately one quarter of the AIDS patients have severe HIV encepha
litis with diffuse neuronal damage that may be mediated by immune fact
ors secreted by CNS macrophages. Based on an in vitro brain microspher
e model, we developed an in vivo system in which human embryonic brain
tissue survives for several months in the interscapular fat pad of SC
ID mice. Coculture of human brain tissue with macrophages prior to tra
nsplantation resulted in infiltration of the microspheres by activated
macrophages. When the macrophages were infected in vitro with a neuro
tropic HIV strain, viral particles were detected in vivo up to 3 month
s after transplantation. HIV-infected transplants contained multinucle
ated giant cells similar to those seen in HIV encephalitis. However, t
he neuroglial component degenerated in the fat pad of SCID mice. The a
bsence of synaptogenesis in the human transplants suggests that the mu
rine fat pad lacks adequate stimuli or support for human neuronal diff
erentiation. To study neurologic damage associated with HIV infection,
sites of implantation that stimulate synaptogenesis (e.g. murine CNS)
will need to be explored. Based on these findings we conclude that tr
ansplantation of brain microspheres with HIV-infected macrophages into
SCID mice may be an achievable model of HIV encephalitis.