CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

Results: In contrast with MDM infection by 89.6, infection by macrophage-tropic clade D T/F viruses did not significantly decrease HO-1 expression or increase glutamate release, despite similar replication levels as 89.6. Unexpectedly, infection by clade D T/F viruses significantly increased MDM culture supernatant neurotoxicity despite no induction of glutamate release (n=4 donors, p<0.001 for equivalent replication to 89.6). This suggests that macrophage-tropic clade D T/F viruses can activate pathways of neurotoxin production independent of HO-1 and associated effects on glutamate metabolism. Conclusions: Our results suggest that macrophage-tropic clade D T/F viruses differ from clade B viruses in several potentially critical determinants of neurovirulence; differential ability to i) reduce expression of MDM HO-1 expression and to ii) induce glutamate release fromMDM. Nonetheless, macrophage-tropic clade D T/F viruses can induce release of neurotoxins other than glutamate (as yet unidentified) fromMDMwithout altering HO-1 expression, and this suggests potential clade- or T/F virus-specific mechanisms of HIV neurotoxin production. Such differences could contribute to variance in severity of neurological dysfunction among populations infected with different HIV clades. 395 Neurotoxicity Screening of Antiretroviral Drugs With Human iPSC-Derived Neurons Sandy Hinckley 1 ; Sean Sherman 1 ; Brookie M. Best 2 ; Jeremiah Momper 2 ; Qing Ma 3 ; Scott R. Letendre 2 ; Ron Ellis 2 ; Anne Bang 1 1 Sanford Burnham Prebys Med Discovery Inst, La Jolla, CA, USA; 2 Univ of California San Diego, San Diego, CA, USA; 3 Univ at Buffalo, State Univ of New York, Buffalo, NY, USA Background: While antiretroviral therapy (ART) has become increasingly effective and well-tolerated, there remains a patient subset that experience central nervous system (CNS) side effects, and neurocognitive performance may actually improve following change in specific ART agents. Understanding the risk associated with these drugs will allow informed selection of optimal therapy. Using human induced pluripotent stem cell-derived neurons (hiPSC), we screened 10 ART drugs and generated a neurotoxicity profile based on mitochondrial membrane potential, reactive oxygen species, cell health and neurite growth. Methods: hiPSC cortical neurons (Cellular Dynamics International ) were treated in dose response with ABC,ATV,COBI,DRV,DTG,EFV,EVG,RPV,RTV,TDF. Neurons were assayed for mitochondrial function and neurite growth using image based high content, high throughput screening. Statistical significance was determined as Z-score greater than 2 compared with vehicle control mean across replicate wells. Results: The majority of tested drugs demonstrated neurotoxicity: 7 caused mitochondrial toxicity and 3 affected neurite growth. Non-nucleoside reverse transcriptase inhibitor, protease inhibitor (PI) and pharmacoenhancer drugs exhibited the highest degree of mitochondrial toxicity. Only EFV was overtly cytotoxic suggesting mitochondrial dysfunction was a primary target, not a side effect of cell death. After 3 days exposure to ART drugs, morphology and cell health was assessed. DTG and EFV resulted in minor but significant neurite growth. RPV was cytotoxic at high dose but caused neurite inhibition at lower doses. Of note, neurite outgrowth or inhibition could contribute to CNS pathology. At high doses, COBI, EVG, RPV and EFV caused neurite inhibition coincident with cytoxicity suggesting the morphological effect was secondary to cell death. While PIs influenced mitochondrial function, no effect was found on morphology. In contrast, COBI, EVG, RPV and EFV presented a distinct profile with mitochondrial dysfunction followed by changes in morphology and/or cytoxicity. Conclusions: We characterized toxicity of 10 ART drugs and linked the majority with impacts either on mitochondrial function, neurite growth, or both. Although immature, hiPSC neurons are human and scalable for drug screening. Further studies are needed to determine whether our in vitro assays reflect neurotoxicity in vivo, but our results suggest that increased drug concentrations in the CNS could have adverse clinical consequences. 396 Background: Brain volumetric changes occur following HIV-infection and often persist in individuals despite cART. The underlying mechanism(s) for these structural changes are not understood, but it has been suggested that chronic low-level expression of the HIV-1 trans -activating protein Tat, may contribute to neural damage over time, but this has not been tested. Here we evaluated the effects of chronic low-level tat gene expression on neuropathology. Methods: Tetracycline-inducible GFAP-driven HIV-1 Tat transgenic mice (rtTA-Tat) and mice expressing only the rtTa (rtTA) promoter (3-12 month of age) were used in these studies. Volumetric measures of macrostructural brain regions and cortical thickness were obtained by T2-weighted in vivo MRI. Neuronal and synaptic integrity were determined by immunoblot analysis of βIII-Tubulin, synaptophysin, and PSD95. Inflammation was assessed by qRT-PCR of proinflammatory cytokines. Brain sphingolipid metabolismwas determined by mass spectrometry. Results: We took advantage of a leaky tetracycline-inducible gene system to produce chronic low-level tat expression in the absence of gene induction. Tat mRNA was undetectable in rtTA mice, detectable in non-induced rtTA-Tat mice, and increased three-fold following a 30-day doxycycline-induced gene induction. All brain volume measures were similar in 3-month old rtTA, non-induced rtTA-Tat and induced-rtTA-Tat mice. In 11-12 month old mice non-induced rtTA-Tat mice exhibited increased total ventricular and dentate gyrus volumes and decreased motor cortical thickness compared to control rtTA mice. Induction of tat gene expression with doxycycline did not further alter regional brain volumes. Cortical thinning observed in non-induced rtTA-Tat mice corresponded with decreased βIII-Tubulin, synaptophysin, and PSD95 levels. Expression of inflammatory cytokines were similar rtTA and non-induced rtTA-Tat mice, but were increased following Induction of tat gene expression. Hierarchal clustering analysis of cortical lipid content revealed that brain lipid content of non-induced rtTA-Tat mice closely resembled those of doxycycline-induced rtTA-Tat mice. Conclusions: These findings demonstrate that chronic low-level tat expression is associated with changes in brain volume, alterations in bioactive lipid content, and synaptic simplification that were independent of measurable changes in inflammatory gene expression. These data suggest that chronic low level tat expression is sufficient to produce structural changes in the brain. 397 Glutaminase Regulates Extracellular Vesicles Release in HIV-1–Infected Macrophages BeiqingWu ;Yunlong Huang; Runze Zhao;Yuju Li; Jialin C. Zheng Univ of Nebraska Med Cntr, Omaha, NE, USA Background: Extracellular vesicles (EVs) are important in the intercellular communication in the central nervous system and their release is increased upon neuroinflammation and neurological disorders. Our previous data demonstrated an increased release of EVs from HIV-1-infected macrophages that have neurotoxic effects. However, the mechanism of elevation of EV release in those HIV-1-infected cells remains unknown. In the current studies, we investigated glutaminase (GLS), which is a mitochondria enzyme critical for glutamine metabolism. GLS is upregulated in HIV-1-infected macrophages and microglia. We propose that HIV-1 infection increases GLS, leading to a metabolic status that favors the EVs generation and release. The new understanding of the metabolic control of EV release in HIV-1-infected cells will shed light on HIV-1 pathobiology and neurological complications. Methods: Human primary microglia and monocyte-derived macrophages culture system and macrophage-tropic HIV-1 ADA were used to study the regulation of EVs during HIV-1 infection. EVs were isolated through differential centrifugations. A gene overexpression system, delivered via adenovirus vector, was utilized to overexpress GLS in the cell culture to mimic the upregulation of GLS during HIV-1 infection. A brain-specific GLS transgenic mouse line was created to model GLS elevation in vivo . BPTES was used to specifically inhibit GLS activity. Transmission electron microscopy and Western blot were used to quantify the EVs released from cells and brain tissues. Glutamate and glutamine levels were determined by reverse phase high performance liquid chromatography. Results: An elevated number of EVs was found in the supernatants of HIV-1-infected macrophages and microglia when compared with controls. Overexpression of GLS in macrophages and microglia cultures leads to increased release of EVs. Conversely, blocking the GLS activity by BPTES significantly reduced EV release and glutamate generation in HIV-1-infected macrophages and microglia, suggesting a critical role of GLS in EV release. Interestingly, we detected an elevated release of EVs in the brain tissues of GLS transgenic mice, suggesting that GLS is also important for EV release in vivo . Chronic Low-Level HIV-1 Tat Expression Promotes a Neurodegenerative Phenotype Alfred Chin ; Alex M. Dickens; Amanda L.Trout; Jackie Lovett; Joelle Dorskind; Norman Haughey Johns Hopkins Univ Sch of Med, Baltimore, MD, USA

Poster Abstracts

150

CROI 2016