HIV-1 Tat-Induced Microgliosis and Synaptic Damage via Interactions between Peripheral and Central Myeloid Cells

PLOS ONE, September 2011

21912650

Shao-Ming Lu, Marie-Ève Tremblay, Irah L. King, Jin Qi, Holly M. Reynolds, Daniel F. Marker, John J. P. Varrone, Ania K. Majewska, Stephen Dewhurst, Harris A. Gelbard

Despite the ability of combination antiretroviral treatment (cART) to reduce viral burden to nearly undetectable levels in cerebrospinal fluid and serum, HIV-1 associated neurocognitive disorders (HAND) continue to persist in as many as half the patients living with this disease. There is growing consensus that the actual substrate for HAND is destruction of normal synaptic architecture but the sequence of cellular events that leads to this outcome has never been resolved. To address whether central vs. peripheral myeloid lineage cells contribute to synaptic damage during acute neuroinflammation we injected a single dose of the HIV-1 transactivator of transcription protein (Tat) or control vehicle into hippocampus of wild-type or chimeric C57Bl/6 mice genetically marked to distinguish infiltrating and resident immune cells. Between 8-24 hr after injection of Tat, invading CD11b(+) and/or myeloperoxidase-positive leukocytes with granulocyte characteristics were found to engulf both microglia and synaptic structures, and microglia reciprocally engulfed invading leukocytes. By 24 hr, microglial processes were also seen ensheathing dendrites, followed by inclusion of synaptic elements in microglia 7 d after Tat injection, with a durable microgliosis lasting at least 28 d. Thus, central nervous system (CNS) exposure to Tat induces early activation of peripheral myeloid lineage cells with phagocytosis of synaptic elements and reciprocal microglial engulfment of peripheral leukocytes, and enduring microgliosis. Our data suggest that a single exposure to a foreign antigen such as HIV-1 Tat can lead to long-lasting disruption of normal neuroimmune homeostasis with deleterious consequences for synaptic architecture, and further suggest a possible mechanism for enduring neuroinflammation in the absence of productive viral replication in the CNS.