BackgroundNeuroinflammation is central to the aetiology of HIV-associated neurocognitive disorders (HAND) that are prevalent in late stage AIDS. Anti-retroviral (ARV) treatments are rolled out relatively late in the context of neuroinflammatory changes, so that their usefulness in directly preventing HAND is probably limited. It is common practice for HIV+ individuals in developing countries to make use of traditional medicines. One such medicine is Sutherlandia frutescens - commonly consumed as a water infusion. Here its efficacy as an anti-inflammatory modality in this context was investigated in an in vitro co-culture model of the blood–brain barrier (BBB).MethodsSingle cultures of human astrocytes (HA), HUVECs and primary human monocytes, as well as co-cultures (BBB), were stimulated with HIV-1 subtype B & C Tat protein and/or HL2/3 cell secretory proteins after pre-treatment with S.frutescens extract. Effects of this pre-treatment on pro-inflammatory cytokine secretion and monocyte migration across the BBB were assessed.ResultsIn accordance with others, B Tat was more pro-inflammatory than C Tat, validating our model. S.frutescens decreased IL-1β secretion significantly (P < 0.0001), but exacerbated both monocyte chemoattractant protein-1 (P < 0001) – a major role player in HIV-associated neuroinflammation – and CD14+ monocyte infiltration across the BBB (P < 0.01).ConclusionsCurrent data illustrates that the combined use of HL2/3 cells and the simulated BBB presents an accurate, physiologically relevant in vitro model with which to study neuroinflammation in the context of HIV/AIDS. In addition, our results caution against the use of S.frutescens as anti-inflammatory modality at any stage post-HIV infection.