A Human iPSC-derived Neurosphere Model for Studying HIV-1 CNS Infection and Persistence

HIV‑1 remains incurable, and HIV‑associated neurocognitive disorders (HAND) continue to affect nearly half of patients despite combination antiretroviral therapy (cART), a consequence of persistent neuroinflammation and the release of neurotoxic viral proteins. Conventional 2-D cultures fail to capture the cellular complexity of the human CNS, limiting their utility for modeling viral pathogenesis and therapeutic response. This ongoing burden underscores the need for advanced, human‑relevant 3-D models that more accurately mimic CNS biology and HIV‑1 infection dynamics. To address this gap, we developed an induced pluripotent stem cell (iPSC)‑derived 3-D neurosphere platform for modeling HIV‑1 infection. iPSC-derived neural progenitor cells (NPCs) were aggregated into ultra‑low‑ attachment plates to form single, uniform spheres and were matured for at least two weeks. Differentiated neurospheres exhibited consistent morphology, and cellular diversity was confirmed by immunostaining for SOX2 (neural progenitors), TH (dopaminergic neurons), GFAP (astrocytes), and IBA‑1 (microglia‑like cells). Neurospheres were challenged with three different HIV‑1 strains (89.6, JR‑CSF, CHO40; MOI 10), and infection was assessed by western blot detection of Pr55, p24, and Nef proteins and RT‑qPCR quantification of TAR and env transcripts. Robust viral replication was observed across strains, with high‑level RNA (-10⁹ TAR; -10⁷ env) and accumulation of viral proteins compared to uninfected controls. Treatment with a clinically relevant cART cocktail suppressed viral protein expression and significantly reduced TAR/env transcripts, demonstrating pharmacologic responsiveness. Importantly, Alu‑gag PCR confirmed integration of HIV‑1 proviral DNA into the host genome, and immunoprecipitation localized TAR DNA primarily to astrocytic and microglia‑like fractions. These findings underscore the model’s relevance for studying CNS reservoirs and persistence and its potential for dissecting reservoir dynamics within cell types implicated in HAND. Collectively, this platform offers a scalable, human-relevant system for studying HIV-1 neuropathogenesis and evaluating antiviral strategies. Furthermore, it has the potential to be adapted to study other neurotropic pathogens, neuroinflammatory conditions, and CNS disorders, advancing disease modeling beyond conventional 2-D cultures and animal models.