BACKGROUND: HIV-infected macrophages persist in people with HIV in tissue such as the brain, gut and lung despite suppressive antiretroviral therapy and can contribute to viral rebound when therapy is stopped. What governs the establishment and reactivation of latent HIV infection in different types of tissue macrophages is poorly understood, but essential to inform HIV elimination strategies which adequately target this reservoir.
METHODS: An in vitro model of HIV latency in human macrophages was utilised, where primary human serum-derived monocyte-derived macrophages (MDM), GM-CSF-derived alveolar like-macrophages (AlvMDM) or multiple cytokine-differentiated monocyte-derived microglia were infected with a GFP-HIV reporter virus. Non-productively infected cells were FACS sorted and latency reactivation quantified longitudinally using fluorescence microscopy. The impact of stimulation with M1 (TNF + IFN?) and M2 (IL-4) polarising cytokines was assessed. HIV-infected MDM from 3 donors were further analysed using single cell transcriptomics.
RESULTS: MDM and AlvMDM harboured comparable levels of latent HIV and exhibited a similar rate of spontaneous reactivation (+0.27 vs 0.15% cells/day respectively, p=0.3). Modulation of macrophage phenotype using M2 polarising conditions enhanced HIV reactivation in both MDM and AlvMDM (p=0.02 and 0.01 respectively), whilst M1 polarising conditions inhibited reactivation in MDM (p=0.03). HIV reactivation from non-productively infected microglial cells trended lower than for MDM (p=0.06). Single cell transcriptomic analysis of non-productively infected (GFP-) MDM revealed substantial numbers of cells (>10%) expressing HIV transcripts, indicating latent HIV infection. Productively HIV-infected MDM were distinguished with 20 differentially expressed host genes. Unique cellular markers identifying latently infected macrophages were not identified.
CONCLUSIONS: Tissue-specific macrophage sub-populations may differ in their ability to harbour reactivatable latent HIV reservoirs. The ability of polarising cytokines to modulate HIV reactivation suggests local tissue microenvironments may influence latency in macrophages, but also indicates the potential for macrophage reservoirs to be reactivated by exogenous stimulation as part of shock and kill cure approaches. Unique data from single cell analysis of latently infected macrophages indicates a substantial proportion of macrophages can harbour latent HIV infection but remain phenotypically ‘silent’ and evade immune recognition. Productively-infected macrophages displayed a distinct gene signature, which may identify targets for novel therapeutics targeting the productive HIV macrophage reservoir.

Download the e-Poster (PDF)