HIV-associated damage to the central nervous system results in cognitive and motor deficits. Anti-retroviral therapies reduce the severity of symptoms, yet the proportion of patients affected has remained the same or increased. Although approximately half of HIV-infected patients worldwide are women, the question of whether biological sex influences outcomes of HIV infection has received little attention. We explored this question for both behavioral and cellular/morphologic endpoints, using a transgenic mouse that inducibly expresses HIV-1 Tat in the brain. After 3 months of HIV-1 Tat exposure, both sexes showed similar reduced open field ambulation. Male Tat+ mice also showed reduced forelimb grip strength and enhanced anxiety in a light–dark box assay. Tat+ males did not improve over 12 weeks of repeated rotarod testing, indicating a motor memory deficit. Male mice also had more cellular deficits in the striatum. Neither sex showed a change in volume or total neuron numbers. Both had equally reduced oligodendroglial populations and equivalent microglial increases. However, astrogliosis and microglial nitrosative stress were higher in males. Dendrites on medium spiny neurons in male Tat+ mice had fewer spines, and levels of excitatory and inhibitory pre- and post-synaptic proteins were disrupted. Our results predict sex as a determinant of HIV effects in brain. Increased behavioral deficits in males correlated with glial activation and synaptic damage, both of which are implicated in cognitive/motor impairments in patients. Tat produced by residually infected cells despite antiretroviral therapy may be an important determinant of the synaptodendritic instability and behavioral deficits accompanying chronic infection.Electronic supplementary materialThe online version of this article (doi:10.1007/s00429-013-0676-6) contains supplementary material, which is available to authorized users.

Opiate abuse and HIV-1 have been described as interrelated epidemics, and even in the advent of combined anti-retroviral therapy, the additional abuse of opiates appears to result in greater neurologic and cognitive deficits. The central nervous system (CNS) is particularly vulnerable to interactive opiate-HIV-1 effects, in part because of the unique responses of microglia and astroglia. Although neurons are principally responsible for behavior and cognition, HIV-1 infection and replication in the brain is largely limited to microglia, while astroglia and perhaps glial progenitors can be latently infected. Thus, neuronal dysfunction and injury result from cellular and viral toxins originating from HIV-1 infected/exposed glia. Importantly, subsets of glial cells including oligodendrocytes, as well as neurons, express µ-opioid receptors and therefore can be direct targets for heroin and morphine (the major metabolite of heroin in the CNS), which preferentially activate µ-opioid receptors. This review highlights findings that neuroAIDS is a glially driven disease, and that opiate abuse may act at multiple glial-cell types to further compromise neuron function and survival. The ongoing, reactive cross-talk between opiate drug and HIV-1 co-exposed microglia and astroglia appears to exacerbate critical proinflammatory and excitotoxic events leading to neuron dysfunction, injury, and potentially death. Opiates enhance synaptodendritic damage and a loss of synaptic connectivity, which is viewed as the substrate of cognitive deficits. We especially emphasize that opioid signaling and interactions with HIV-1 are contextual, differing among cell types, and even within subsets of the same cell type. For example, astroglia even within a single brain region are heterogeneous in their expression of µ-, δ-, and κ-opioid receptors, as well as CXCR4 and CCR5, and Toll-like receptors. Thus, defining the distinct targets engaged by opiates in each cell type, and among brain regions, is critical to an understanding of how opiate abuse exacerbates neuroAIDS.

In this study we determine whether morphine alone or in combination with HIV-1 Tat or gp120 affects the expression of Toll-like receptors (TLRs) by astrocytes and to assess whether TLRs expressed by astrocytes function in the release of inflammatory mediators in vitro. TLR profiling by immunofluorescence microscopy, flow cytometry, in-cell westerns, and RT-PCR showed that subpopulations of astrocytes possessed TLR 2, TLR3, TLR4, and TLR9 antigenicity. Exposure to HIV-1 Tat, gp120, and/or morphine significantly altered the proportion of TLR-immunopositive and/or TLR expression by astroglia in a TLR-specific manner. Subsets of astroglia displayed significant increases in TLR2 with reciprocal decreases in TLR9 expression in response to Tat or gp120 ± morphine treatment. TLR9 expression was also significantly decreased by morphine alone. Exposing astrocytes to the TLR agonists LTA (TLR2), poly I:C (TLR3), LPS (TLR4) and unmethylated CpG ODN (TLR9) resulted in increased secretion of MCP-1/CCL2 and elevations in reactive oxygen species. TLR3 and TLR4 stimulation increased the secretion of TNF-α, IL-6, and RANTES/CCL5, while activation of TLR2 caused a significant increase in nitric oxide levels. The results suggest that HIV-1 proteins and/or opioid abuse disrupt the innate immune response of the central nervous system (CNS) which may lead to increased pathogenicity.

A rigorously controlled, cell culture paradigm was used to assess the role of HIV-1 gp120 ± morphine in mediating opioid-HIV interactive toxicity in striatal neurons. Computerized time-lapse microscopy tracked the fate of individual neurons co-cultured with mixed-glia from mouse striata during opioid and gp120 exposure. Subpopulations of neurons and astroglia displayed μ-opioid receptor, CXCR4, and CCR5 immunoreactivity. While gp120 alone was or tended to be neurotoxic irrespective of whether X4-tropic gp120IIIB, R5-tropic gp120ADA, or dual-tropic gp120MN was administered, interactive toxicity with morphine differed depending on HIV-1 strain. For example, morphine only transiently exacerbated gp120IIIB-induced neuronal death; however, in combination with gp120MN, morphine caused sustained increases in the rate of neuronal death compared to gp120MN alone that were prevented by naloxone. Alternatively, gp120ADA significantly increased the rate of neuron death, which was unaffected by morphine. The transient neurotoxic interactions between morphine and gp120IIIB were abrogated in the absence of glia suggesting that glia contribute significantly to the interactive pathology with chronic opiate abuse and neuroAIDS. To assess how mixed-glia might contribute to the neurotoxicity, the effects of morphine and/or gp120 on the production of reactive oxygen species (ROS) and on glutamate buffering were examined. All gp120 variants, and to a lesser extent morphine, increased ROS and/or decreased glutamate buffering, but together failed to show any interaction with morphine. Our findings indicate that HIV-1 strain-specific differences in gp120 are critical determinants in shaping both the timing and pattern of neurotoxic interactions with opioid drugs.

Objective We explored whether the opiate, morphine, affects the actions of maraviroc, as well as a recently synthesized bivalent derivative of maraviroc linked to an opioid antagonist, naltrexone, on HIV-1 entry in primary human glia. Methods HIV-1 entry was monitored in glia transiently transfected with an LTR construct containing a luciferase reporter gene under control of a promoter for the HIV-1 transactivator protein Tat. The effect of maraviroc and the bivalent ligand ± morphine on CCR5 surface expression and cytokine release was also explored. Results Maraviroc inhibits HIV-1 entry into glial cells, while morphine negates the effects of maraviroc leading to a significant increase in viral entry. We also demonstrate that the maraviroc-containing bivalent ligand better inhibits R5-tropic viral entry in astrocytes than microglia compared to maraviroc when coadministered with morphine. Importantly, the inhibitory effects of the bivalent compound in astrocytes were not compromised by morphine. Exposure to maraviroc decreased the release of pro-inflammatory cytokines and restricted HIV-1-dependent increases in CCR5 expression in both astrocytes and microglia, while exposure to the bivalent had similar effect in astrocytes but not in microglia. CCR5-MOR stoichiometric ratio varied among the two cell types with CCR5 expressed at much higher levels than MOR in microglia, which could explain the effectiveness of the bivalent ligand in astrocytes compared to microglia. Conclusion A novel bivalent compound reveals fundamental differences in CCR5-MOR interactions and HIV-1 infectivity among glia, and has unique therapeutic potential in opiate abuse-HIV interactive comorbidity.