Background: Activation of the kynurenine pathway (KP), especially the production of the neurotoxin quinolinic acid (QUIN), has been pathogenetically linked to HIV-related brain damage most particularly AIDS dementia complex (ADC). Recently, both the products of the KP, particularly QUIN, and KP-related tryptophan depletion have been found to cause T cell apoptosis and hyporesponsiveness. We hypothesized that strains of HIV differ in their ability to induce the KP leading to neurotoxicity and immune paresis, which further contributes to viral persistence through impaired T cell-related clearance.

Methods: Macrophages isolated from peripheral blood mononuclear cells (PBMC) and adult microglia from adult brain biopsies were infected overnight by various HIV isolates. The neurotropic strains JRFL, C158, and C124 and the non-neurotropic strain BaL were all added at the same viral TCID₅₀ concentration. QUIN levels were analyzed by gas chromatography-mass spectrometry by collecting culture supernatants at days 3, 4, 5, and 6 following infection. T cell apoptosis was assessed by adding culture supernatants from infections at day 3 to phytohemagglutinin (PHA) -activated CD4⁺ cells. CD4⁺ cells were stained with carboxy-fluorescein diacetate, succinimidyl ester (CFSE) and incubated for 5 days and the T cell repertoire was assessed by flow cytometry using CD3, CD4, CD45, CD25, CD127, and caspase 3 antibodies.

Results:  Our results for macrophages consistently indicate significantly low levels of QUIN production in Bal compared to JRFL (p <0.007), C158 (p <0.005), and C124 (p <0.003). Similarly, for microglia QUIN was significantly low for BaL as compared to JRFL (p <0.001), C158 (p <0.01), and C124 (p <0.01). Preliminary results from T cell apoptosis experiments using supernatants from macrophages indicate higher levels of caspase 3-positive levels for CD4RO⁺ memory cells for JRFL (30%) than those from BaL (19%). Caspase-positive levels were also greater for T regulatory cells from JRFL (9%) than BaL (4%).

Conclusions: There are significant differences in the ability of strains of HIV to induce the KP, with ADC-related strains having the greatest ability. This has implications for understanding the basis of neurovirulence. Furthermore, such KP induction probably plays a significant role in viral persistence in the brain through induction of tolerance and facilitation of neurotropism. Finally, the poor induction of the KP by BaL has significant implications for studies of neuropathogenesis that have used BaL.