The availability of highly active antiretroviral therapies (HAART) has not eliminated HIV-1 infection of the central nervous system (CNS) or the occurrence of HIV-associated neurological problems. Thus, the neurobiology of HIV-1 is still an important issue. Here, we review key features of HIV-1-cell interactions in the CNS and their contributions to persistence and pathogenicity of HIV-1 in the CNS. HIV-1 invades the brain very soon after systemic infection. Various mechanisms have been proposed for HIV-1 entry into the CNS. The most favored hypothesis is the migration of infected cells across the blood-brain barrier ("Trojan horse" hypothesis). Virus production in the CNS is not apparent before the onset of AIDS, indicating that HIV-1 replication in the CNS is successfully controlled in pre-AIDS. Brain macrophages and microglia cells are the chief producers of HIV-1 in brains of individuals with AIDS. HIV-1 enters these cells by the CD4 receptor and mainly the CCR5 coreceptor. Various in vivo and cell culture studies indicate that cells of neuroectodermal origin, particularly astrocytes, may also be infected by HIV-1. These cells restrict virus production and serve as reservoirs for HIV-1. A limited number of studies suggest restricted infection of oligodendrocytes and neurons, although infection of these cells is still controversial. Entry of HIV-1 into neuroectodermal cells is independent of the CD4 receptor, and a number of different cell-surface molecules have been implicated as alternate receptors of HIV-1. HIV-1-associated injury of the CNS is believed to be caused by numerous soluble factors released by glial cells as a consequence of HIV-1 infection. These include both viral and cellular factors. Some of these factors can directly induce neuronal injury and death by interacting with receptors on neuronal membranes (neurotoxic factors). Others can activate uninfected cells to produce inflammatory and neurotoxic factors and/or promote infiltration of monocytes and T-lymphocytes, thus amplifying the deleterious effects of HIV-1 infection. CNS responses to HIV-1 infection also include mechanisms that enhance neuronal survival and strengthen crucial neuronal support functions. Future challenges will be to develop strategies to prevent HIV-1 spread in the brain, bolster intrinsic defense mechanisms of the brain and to elucidate the impact of long-term persistence of HIV-1 on CNS functions in individuals without AIDS.