HIV-1 infects the cells of the immune system, primarily CD4+
T cells and cells of the monocyte-macrophage lineage, and progressive immunodeficiency marks disease progression [71
]. During acute HIV-1 infection virus replicates rapidly, leading to an abundance of virus in the peripheral blood. Despite the immune privileged status of the brain, HIV-1 has been linked to a number of neurological symptoms, even in the absence of confounding disease. HIV-1 penetrates the BBB, enters the CNS and causes varying degrees of neurocognitive impairment or HAND, characterized by cognitive, motor and behavioral disorders [71
]. In parallel, as few common antiretroviral drugs adequately penetrate the CNS, creating a reservoir for virus until replication is reactivated within the CNS, in such, antiretroviral therapy (ART) has lessened the incidence, but not the prevalence of HAD [72
]. Furthermore, information regarding extended, low level HIV-1 infection in the CNS is limited and the effects of ART on non-infected neural cells such as astrocytes are poorly studied [73
]. One recent study reports that certain antiretroviral medications increase β-amyloid (Aβ) expression by neurons; Aβ plaque accumulation is a common feature of HIV-1 CNS infection [74
]. Understanding the molecular and cellular mechanisms involved in response to low level, chronic viral expression in the brain along with the effects of ART on neural cells may further our understanding of HIV-1-associated neurodegeneration.
HAD, a clinical correlate of HIV-associated encephalitis (HIVE), often occurs during active, chronic HIV-1 infection of the brain. Approximately 15% of individuals infected with HIV-1 develop HAD, and approximately 30% to 60% of individuals develop a less severe form of minor cognitive motor disorders. During the pre-ART era, up to 25% of HIV-1-infected patients developed HAD as the first acquired immune deficiency syndrome (AIDS)-defining illness and early evidence of HIV-1 neuropathogenicity included HIV-1 in cerebrospinal fluid (CSF), abnormal neuroimaging, and the presence of HIV-1 in tissue obtained by brain biopsy [75
]. CNS manifestations observed during HIV-1 viremia associated with seroconversion include meningitis, encephalitis, facial palsy, and peripheral nerve disorders; proposing HIV-1 neuroinvasion occurs early during infection.
Following HIV-1 penetration through the BBB, the virus productively infects and resides primarily in microglia and macrophages, playing a central role in neuroinflammatory disorders (reviewed by [76
] and [77
]). HIV-infected or immune-stimulated macrophages/microglia produce neurotoxins, and inflammatory mediators [78
]. Macrophage- and microglia-derived factors including but not limited to: inflammatory chemokines and cytokines (such as, IL-1β and TNF-α), arachidonate and its metabolites (such as platelet-activating factor), free radicals, and viral proteins have been investigated in HIV-1-associated neurodegeneration (reviewed by [77
]). While TNF-α, IL-1β, and IL-6 regulatory mechanisms suppress acute HIV-1 infection in microglia, the observations seen in chronically infected monocytic cell lines exacerbate disease [81
While the BBB controls infiltration of substances/cells from the periphery, the HIV-1 disease process impairs the BBB, facilitating entry of HIV-1-infected monocytes as described by the classical Trojan horse hypothesis. HIV-1 proteins, trans
-acting protein (Tat
) and envelope glycoprotein (gp)120
, and host proinflammatory cytokines and chemokines compromise the BBB, allowing free HIV-1 and transmigration of HIV-1-infected cells into the CNS [71
]. Astrocytes interact with endothelial cells, as part of the gliovascular unit, and regulate the BBB physiology in attempt to protect its integrity [20
]. Recently, Eugenin et al.
describe a novel mechanism for bystander BBB toxicity, which is mediated by low numbers of HIV-1-infected astrocytes, amplified by gap junctions, and induces apoptosis of non-infected BBB cells [20
]. Furthermore, Ju et al
. demonstrate upregulation of both TNF-α and MMP-9 in response to HIV-1 Tat
], a protein expressed by infected astrocytes, which can in turn affect other astrocytes.
Although acute HIV-1 replication is attenuated in the brain, HIV-1 mRNA expression in the CSF has been observed within eight days post-HIV-1 transmission [83
]. The mechanisms underlying neurological and behavioral characteristics present in the first months following HIV-1 neuroinvasion are associated with HIV-1 Tat
]. HIV-1 Tat
is a key mediator of neurotoxicity promoting potentiation of glutamate and NMDA-triggered Ca2+
, excitotoxic neuron apoptosis [85
] and activation of GFAP expression in astrocytes [86
]. Thus, reactive astrogliosis and glutamate excitotoxicity are both implicated in HIV-1 CNS neurodegeneration.
The abundance of astrocytes in the brain would make them an ideal target for HIV-1 infection and Hao et al.
reported HIV-1 enters astrocytes through a receptor mediated endocytic pathway involving HIV-1 gp120
in a concentration dependent manner [87
]. Electron microscopy revealed HIV-1 virions in clathrin-coated pits of cytoplasmic vacuoles of astrocytes. While non-productive or latent infections of astrocytes were reported in the early decades of HIV/AIDS [88
], many now believe that only ~2% of the brain astrocytes are likely infected by HIV-1 since astrocytes lack the ability to actively replicate HIV-1 and express viral regulatory genes [89
In an artificial system, the WNT/β-catenin/transcription factor (TCF)-4 signaling pathway significantly inhibits HIV-1 replication in astrocytes; however, little is known about the direct mechanism. Knockdown of β-catenin and TCF-4 resulted in HIV-1 transcription in transiently and stably transfected cell lines with an integrated HIV long terminal repeat (LTR)-luciferase reporter construct. IFN-γ induced transcription of an antagonist of the β-catenin pathways, DKK1, in a signal transducers and activators of transcription (STAT)-3-dependent manner, which resulted in increased HIV-1 LTR activity through attenuated β-catenin signaling [90
]. Costimulation of primary human astrocytes with METH and HIV-1 Tat
exacerbated β-catenin downregulation by greater than 50%. While METH alone had no effect, HIV-1 induced TCF-4 downregulation in astrocytes [91
Despite limited infection of astrocytes by HIV-1, numerous studies, including ours demonstrate that HIV-1 virions do affect astrocyte inflammatory responses [92
]. HIV-1 proteins such as HIV-1 gp120
and HIV-1 Tat
have also been shown to affect astrocyte function in multiple ways [95
]. Furthermore, astrocytes are exquisitely sensitive to the proinflammatory cytokines present in the brain as a result of infection and many of the functional changes are amplified via
autocrine and paracrine loops [18