CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

clinical indications. Mean retinal nerve fiber layer (RNFL) and ganglion cell inner plexiform layer (GC-IPL) thicknesses were compared between groups using ANCOVA, and means were correlated with pre-selected MRI brain volumes, NP domain scores, and CSF cytokines and neurofilament light chain. Results: There were no differences in age, race or visual acuity between the two groups; there were more women in the control group (p=0.006). In the HIV+ group, the median time since diagnosis was 19 years and all had an HIV RNA level <100 copies/ml for at least one year prior to the SD-OCT. Multiple regression analyses indicated that the HIV+ group had thinner adjusted-mean RNFL (78.17µm, 95% CI 76.3, 80.0; control = 84.0µm, 95% CI 81.3, 86.5; p < 0.005) and GC-IPL (90.0µm, 95% CI 87.0, 92.6; control = 96.6µm, 95% CI 92.2, 101.0; p = 0.01). In the HIV+ group, retinal thicknesses were negatively associated with the fraction of CSF volume (i.e. brain atrophy) on MRI (p=0.01 for RNFL and 0.006 for GC-IPL). There were few associations with NP domains and CSF measurements. Conclusion: PLWH on ART had thinning of the RNFL and the GC layer of the retina. This retinal thinning was asymptomatic but was strongly associated with measures of brain atrophy. This suggests that there is widespread neurodegeneration including the retina despite adequate ART.

and CD8+), B cells and myeloid cells in the brains of GF BLT (n=10), HuM-BLT (n=14), and MuM-BLT (n=10) mice. Results: Our results revealed higher levels of human hematopoietic cells in the brains of MuM-BLT (p=0.0076) and HuM-BLT (p=0.0534 [7.4x higher]) mice compared to GF BLT mice. Total human T cell, CD4+ T cell and CD8+ T cell numbers were significantly higher in the brains of HuM-BLT (p=0.0034, p=0.0034 and p=0.0106 respectively) and MuM-BLT (p=0.0041, p=0.0030, p=0.0076 respectively) mice compared to GF BLT mice. Human B cell and myeloid cell levels were not significantly different. We confirmed that these results were not due to the humanization procedure by performing a similar analysis of murine immune cell levels in the brains of GF and colonized wild-type mice. Conclusion: Collectively, our results demonstrate that gut microbiota regulate immune cell homeostasis in the CNS and provide the first evidence that gut microbiota may have a direct role in HIV pathogenesis and the establishment and maintenance of the CNS HIV reservoir. 420 CSF1R INHIBITION TARGETS CNS MACROPHAGES IN AN SIV/MACAQUE MODEL OF HIV CNS DISEASE Audrey C. Knight 1 , Samuel A. Brill 1 , Megan E. McCarron 1 , Suzanne E. Queen 1 , Lisa Mangus 1 , Kelly A. Metcalf Pate 1 , Sarah E. Beck 1 , Joseph Mankowski 1 1 Johns Hopkins University School of Medicine, Baltimore, MD, USA Background: CNS macrophages, including microglia, serve as a latent HIV cellular reservoir. Colony stimulating factor 1 receptor (CSF1R) is expressed exclusively on cells of myeloid lineage and is essential for microglial survival. CSF1R protein levels increase with SIV and HIV infection and remain elevated in the CNS despite suppressive ART. Consequently, CSF1R is an ideal drug target to reduce the CNS latent reservoir. Methods: Primary brain and spinal cord microglia were isolated from uninfected macaques. Cells were cultured for one week before treatment with 10uM PLX3397, a small molecule inhibitor of CSF1R, or vehicle. Calcein and ethidium staining was used to identify live and dead cells. Live cells were also quantified using interferon-beta qPCR. In vivo studies were conducted using daily oral treatment of 165mg/kg PLX3397 in ART-suppressed SIV-infected pigtailed macaques (N = 2). Two weeks after the start of PLX3397 treatment, ART was stopped while PLX3397 treatment continued. Plasma and CSF were collected every four days after release to measure viral loads. Animals were euthanized 16 days post-release from ART. Brain and spinal cord microglia were isolated after euthanasia to measure viral RNA, DNA, and replication competence. Results: PLX3397 treatment in vitro significantly reduced the number of primary microglia over 72 hours (P < 0.0001, Two-Way ANOVA). In vivo, PLX3397 treatment was well tolerated; animals did not show side-effects or develop monocytopenia. PLX3397 did not significantly affect plasma viral rebound kinetics. However, treatment did prevent CSF rebound in one animal. In addition, neither SIV RNA or DNA was detected in cultured primary microglia from this animal. In the second animal, viral RNA was isolated from CNS macrophages cultured from both brain and spinal cord. Conclusion: PLX3397 reduced CNS macrophage viability in vitro, demonstrating that targeting CSF1R may reduce CNS macrophages, including those harboring HIV. PLX3397 treatment was associated with a lack of SIV rebound from the CNS and a decrease in IBA1+ CNS macrophages in one of two animals after stopping ART. These studies demonstrate the potential of targeting CSF1R to reduce the HIV latent reservoir in the CNS. 421 METAGENOMIC NEXT-GENERATION SEQUENCING FOR DIAGNOSIS OF CNS INFECTION IN PLWH Jun Chen 1 , Renfang Zhang 1 , Li Liu 1 , Tangkai Qi 1 , Zhenyan Wang 1 , Wei Song 1 , Yang Tang 1 , Jianjun Sun 1 , Hongzhou Lu 1 1 Shanghai Public Health Clinical Center, Shanghai, China Background: Central nervous system (CNS) infection is still the leading cause of death in people living with HIV. Metagenomic next-generation sequencing (mNGS), which could identify a comprehensive spectrum of pathogens by a single assay, has recently shown its efficacy in the diagnosis of infectious diseases. However, its clinical utility in HIV-infected patients is still not well established. Methods: HIV-infected patients with clinically suspected CNS infection admitted to Shanghai Public Health Clinical Center, China underwent lumbar puncture. Cerebrospinal fluid (CSF) samples were sent to conventional testing,

Poster Abstracts

419 GUT MICROBIOTA REGULATE IMMUNE CELL HOMEOSTASIS IN THE CNS Angela Wahl 1 , Baolin Liao 1 , Cara Richardson 1 , Wenbo Yao 1 , Allison Rogala 1 , R. Balfour Sartor 1 , J. V. Garcia 1 1 University of North Carolina at Chapel Hill, Chapel Hill, NC, USA Background: The CNS is an important site of HIV infection, pathogenesis, and persistence. We recently demonstrated that CD4+ T cells are present in the brain and that they are susceptible to HIV infection. We also showed that HIV infection results in CD4+ T cell depletion in the CNS and that CD4+ T cell homeostasis could be restored by ART. Recently, it was shown that gut microbiota influences T cell trafficking to the brain in certain disease states via the gut-brain axis. Based on these observations, we hypothesized that gut microbiota regulates immune cell homeostasis in the CNS which could modulate HIV infection, pathogenesis, and persistence in the CNS. Methods: Direct experimentation in humans to establish gut microbiota’s role in CNS immune homeostasis is not possible. We established an in vivo platform to investigate gut microbiota’s role in human hematopoietic cell homeostasis in the brain. We generated germ-free (GF) bone marrow/liver/thymus (BLT) humanized mice and BLT mice colonized with human (HuM-BLT mice) or murine (MuM-BLT mice) gut microbiota. First, we rederived GF immunodeficient NSG mice. GF NSG mice were implanted with human thymus/liver tissue and transplanted with autologous stem cells in a GF surgical isolator. HuM-BLT and MuM-BLT mice were constructed by colonizing GF mice with human or mouse fecal microbiota. Using flow cytometry, we quantitated human T cells (CD4+

CROI 2020 146