Micheline McCarthy

CD4+ and CD8+ cells accumulate in the brains of acquired immunodeficiency syndrome patients with human immunodeficiency virus encephalitis.

To test the hypothesis that CD4+ T lymphocytes accumulate in brains of end-stage acquired immunodeficiency syndrome (AIDS) patients, we examined T-lymphocyte subsets in the CA1, CA3, and CA4 regions of the hippocampus of AIDS patients with (n = 10) and without (n = 11) human immunodeficiency virus encephalitis (HIVE) plus controls (n = 7). HIV p24 antigen was common in monocytic cells and rare in activated/memory CD45RO+ lymphocytes. Hippocampal activated/memory CD45RO+ T lymphocytes significantly increased (P < .001) in seven of the eight hippocampal subregions with hippocampal HIVE (1.14 ± 1.4 T cells/high-power field [hpf]), but AIDS hippocampus without HIVE were similar to controls (0.03 ± 0.07 T cells/hpf and 0.03 ± 0.09 T cells/hpf, respectively). CD45RO+ and CD3+ lymphocytes were similar in numbers and distribution, whereas CD4+ and CD8+ lymphocytes were weakly immunoreactive and less frequent. All four lymphocyte subtypes were present in perivascular spaces and microglial nodules of HIVE, and had direct contact with neurons. Monocytes, microglia, and multinucleated giant cells were immunoreactive for CD4 in AIDS cases with hippocampal HIVE but microglia in remaining AIDS cases and controls were CD4−. CD68+ macrophages significantly increased in hippocampus of HIVE patients (P < .05) and were predominately perivascular in the absence of local HIVE. These studies show that CD4+ T lymphocytes, as well as CD8+ T lymphocytes, participate in the local inflammatory response of HIVE in end-stage AIDS patients, and suggest that their recruitment requires local HIV infection. The perineuronal location of CD4+ cells provides the potential for lymphocyte-mediated neuronal injury or trans-receptor-mediated neuronal infection.

Brain CD8+ and cytotoxic T lymphocytes are associated with, and may be specific for, human immunodeficiency virus type 1 encephalitis in patients with acquired immunodeficiency syndrome

CD8+ T cells infiltrate brains with human immunodeficiency virus type-1 (HIV-1) encephalitis (HIVE) and related animal models; their perineuronal localization suggests cytotoxic T cell (CTL)-mediated neuronal killing. Because CTLs have not been identified in acquired immunodeficiency syndrome (AIDS) brains, the authors identified their cytotoxic granules in autopsy AIDS brains with HIVE and without HIVE (HIVnE) plus controls (7 to 13 cases/group) and determined gene expression profiles of CTL-associated genes in a separate series of cases. CD3+ and CD8+ T cells were significantly increased (P < .01) in perivascular spaces and inflammatory nodules in HIVE but were rare or absent in brain parenchyma in HIVnE and control brains. Eight HIVE brains contained granzyme B+ T cells and five contained perforin+ T cells. Their T-cell origin was confirmed by colocalization of CD8 and granzyme B in the same cell and the absence of CD56+ natural killer cells. The CTLs directly contacted with neurons, as the authors showed previously for CD3+ and CD8+ T cells. CTLs were rare or absent in HIV nonencephalitis (HIVnE) and controls. Granzyme B and H precursor gene expression was up-regulated and interleukin (IL)-12A precursor, a maturation factor for natural killer cells and CTLs, was down-regulated in HIVE versus HIVnE brain. This study demonstrates, for the first time, CTLs in HIVE and shows that parenchymal T cells and CTLs are sensitive biomarkers for HIVE. Consequently, CD8+ T cells and CTLs could mediate brain injury in HIVE and may represent an important biomarker for productive brain infection by HIV-1.

Chapter 26 Trophic signaling pathways activated by purinergic receptors in rat and human astroglia

Publisher Summary Mitogen-activated protein kinase (MAPK) cascades are intriguing candidates because emerging evidence points to a crucial role for these signaling pathways in cellular proliferation and differentiation (reviewed in Seger and Krebs; Neary) as well as in tissue injury and recovery (e.g., Hu and Wieloch). These cascades consist of at least three cytoplasmic protein kinases that are activated sequentially: MAPK kinase kinase→MAPK kinase→MAPK. At least three parallel MAPK pathways have been identified; these are frequently referred to as extracellular signal regulated protein kinase (ERK), stress-activated protein kinase (SAPK)—also known as JNK for c-Jun N-terminal kinase—and p38/MAPK. The latter two cascades have been implicated in growth arrest and apoptosis (Kyriakis and Avruch). MAPK cascades are stimulated by growth or stress signals, and the activated MAPKs can target proteins in the cytoplasm, membrane, or cytoskeleton, or they can translocate to the nucleus where they activate or induce transcription factors, thereby leading to the expression of genes important in cell growth or death. Our findings indicate that MAPK cascades mediate mitogenic signaling by purinergic receptors in both rat and human astrocytes. In rat astrocytes, P2 receptors are linked to the ERK cascade and mitogenesis, whereas in human astrocytes, both P1 and P2 receptors are coupled to ERK. The signaling pathway from rat and human purinergic receptors to the ERK cascade involves protein kinase C (PKC). Studies with rat astrocytes reveal that P2Y receptors are coupled to ERK by a pathway that is independent of the PI-PLC/inositol phosphate/calcium pathway.

HIV-1 strain-associated variability in infection of primary neuroglia.

Qualitative differences among strains of Human Immunodeficiency Virus type 1 (HIV-1) may influence viral infectivity for cells of the central nervous system (CNS) and determine or at least significantly influence the neuropathogenesis of brain infection. In this study, we compared infectivity for these cells in vitro among several different laboratory-adapted HIV-1 strains differing in cellular tropism. These strains included three lymphotropic strains (SF2, NL4-3, and SG3.1), two macrophage-tropic strains (SF128A, SF162), and one brain-derived strain (YU2). In microglia, macrophage-tropic strain SF128A established productive infection while the lymphotropic strain SF2 did not. In infected astrocytes, all HIV-1 strains transiently produced variable and much lower levels of p24 antigen. Viral DNA env or tat gene sequences were amplified from infected astrocytes; the amplified signals varied among HIV-1 strains, but the strongest viral DNA signals were obtained from cells infected by the lymphotropic strains SF2 and SG3.1. Transfection of astrocytes with infectious HIV-1 proviral DNA clones confirmed the observation that HIV-1 strains differ in their ability to replicate in astrocytes. Transfection revealed post-entry blocks to replication by macrophage-tropic proviruses pSF128A and pSF162. However, cytomegalovirus (CMV) superinfection of transfected astrocytes enhanced p24 production by lymphotropic HIV-1 proviruses twofold and stimulated p24 production by the otherwise inactive macrophage-tropic proviruses. This study demonstrates the spectrum of HIV-1 strain-associated variation in infectivity for neuroglia, and suggests, in addition, that herpesviral factors or viral-induced cellular factors may stimulate HIV-1 infection in astrocytes and expand the neural cell tropism of certain HIV-1 strains.

Media components influence viral gene expression assays in human fetal astrocyte cultures.

In vitro neurovirological studies of viral infectivity or viral gene expression may be confounded by the multiple neural cell types and/or fibroblast contamination present in early passage cultures prepared from dissociated human central nervous system (CNS) tissue. We have developed highly enriched astrocyte cultures for neurovirological study by culturing in a serum-free defined medium, B16, supplemented with basic fibroblast growth factor (FGF-2). Subculture in this medium selects against fibroblast proliferation and favors sustained proliferation of a highly enriched glial fibrillary acidic protein (GFAP)-positive cell population. These astrocytes support productive replication of cytomegalovirus (CMV) and transient expression of transfected CMV and human immunodeficiency virus type 1 (HIV-1) viral promoters. By comparison, CNS cultures developed in standard serum-containing medium initially contain predominantly astrocytes, but show increasing contamination with fibroblasts with sequential passage. These cultures support CMV viral synthesis in both fibroblasts and astrocytes, cell types distinguishable only by immunostaining for cell specific antigen. CMV or HIV-1 promoter activities, quantitated by transient gene expression assays, are distinctly lower in CNS cultures maintained in serum-containing medium.

Infection of Human Neural Cell Aggregate Cultures with a Clinical Isolate of Cytomegalovirus

Human neural cell aggregate cultures were prepared from dissociated fetal brain tissue and maintained in rotation culture. After 35 days in culture, aggregates had the histologic appearance of dense, immature, neural cells in a tightly packed neuropil. Electron microscopy revealed ultrastructural features suggestive of immature neurons and neuroglia. In addition, neuron-specific enolase and glial fibrillary acidic protein associated with radial glial cells were detected within the aggregates by immunoperoxidase staining. When infected with a laboratory-adapted strain of cytomegalovirus (CMV), [AD169], cells containing large, bizarre, nuclei and CMV-induced intranuclear inclusion bodies were dispersed throughout the aggregates at 16 days postinfection. In situ hybridization using a CMV-specific DNA probe and electron microscopy confirmed the presence of virus sequences as well as virus particles at histologic sites of cytopathology. In sharp contrast, aggregate cultures infected with a CMV strain recovered from the retina of an acquired immune deficiency syndrome (AIDS) patient with CMV retinitis and encephalitis displayed distinct foci of cytopathology at 23 days postinfection, a pattern not observed in CMV [AD169]-infected aggregates. Our findings suggest that human neural cell aggregates represent a promising multicellular non-neoplastic culture system in which to study the replication of human neurotropic viruses within neural tissue.

Neurologic Consequences of the Immune Reconstitution Inflammatory Syndrome (IRIS)

Although major advancements have been made in the treatment of HIV infection, graft-versus-host reactions, and autoimmune diseases, an unexpected consequence of treatment has been the emergence of a devastating inflammatory syndrome, termed the immune reconstitution inflammatory syndrome (IRIS). The pathophysiology of the syndrome is poorly understood, and the syndrome poses unique challenges for diagnosis and treatment. We have reviewed the neurologic manifestations of IRIS in the context of HIV infection as well as in the setting of treatment of autoimmune diseases such as multiple sclerosis, in which compartmental immune suppression may occur without an obvious underlying immune suppression. The purpose of this review is to identify common themes that may assist in the diagnosis and management of these IRIS syndromes.

Maturing neurons are selectively sensitive to human immunodeficiency virus type 1 exposure in differentiating human neuroepithelial progenitor cell cultures.

Human immunodeficiency virus type 1 (HIV-1) infection of the brain is associated with neuronal injury manifested by dendritic pruning, aberrant neurofilament metabolism, and decreased synaptic density. The central nervous system (CNS) responds to neuronal injury by differentiating new neurons and astrocytes from resident populations of multipotent neuroepithelial progenitor cells (NEP) located in regions such as the subventricular zone or hippocampus. In vitro studies have demonstrated that the HIV-1 virion or envelope glycoprotein gp120 can injure differentiated human neurons and astrocytes, suggesting that HIV-1 proteins could similarly injure NEP or NEP-derived glial and neuronal lineage-committed precursor cells. To answer this question, human fetal brain-derived “neurospheres” containing NEP and NEP-derived precursor cells were cultured in low serum differentiation medium containing lymphotropic HIV-1(SF2), macrophage-tropic HIV-1(SF128A), or recombinant gp120SF2 from HIV-1(SF2). These experiments indicate that exposure to HIV-1 does not affect the ability of the NEP to differentiate into cells expressing either astrocyte-specific or neuron-specific cytoskeletal antigens. However prolonged exposure to HIV-1 does selectively decrease expression of neuronal antigens (microtubule β-III-tubulin and intermediate filament neurofilament-L) but not astrocyte antigens (intermediate filament glial fibrillary acidic protein). The effects of continuous exposure to HIV-1 or gp120 may result from injury to developing neurons and/or impairment of the neuronal developmental process itself. By depressing neuronal microtubule and neurofilament protein expression, HIV-1 and gp120 exposure compromise the potential for postmitotic neuronal dendrite and axon development.

Diagnostic value for culture of cerebrospinal fluid from HIV-1-infected individuals for opportunistic viruses : a prospective study

Objective.To investigate the diagnostic value of cerebrospinal fluid (CSF) culture for opportunistic viruses from HIV-1-infected individuals. Methods.A 4-year prospective study was conducted using a participant population consisting of 186 HIV-1-infected individuals without neurologic disease, 73 HIV-1-infected individuals with encephalopathy, myelopathy, and/or peripheral neuropathy, and 10 controls. CSF samples recovered at 1-year intervals were subjected to virus culture using techniques commonly used in the clinical laboratory setting. Results.CSF samples obtained from only 15 of the 269 (5.6%) participants yielded an opportunistic virus upon culture. Cytomegalovirus, herpes simplex virus types 1 and 2, adenovirus, and presumptive enteroviruses were identified. No consistent correlation was observed between the detection of an opportunistic virus within a CSF sample and the presence or future development of neurologic disease. However, a significant correlation was observed between culture of virus from CSF and the future development of abnormal CD4+ (X2, P= 0.0286) and CD8+ (X2, P= 0.0018) lymphocyte counts in HIV-1-infected participants without neurologic disease. Conclusion.These results show that culture of CSF to screen for opportunistic viruses is neither diagnostic nor predictive of neurologic disease in HIV-1-infected individuals. Nevertheless, the presence of virus within CSF may be an indicator of HIV-1-mediated immune dysfunction and a predictor for future development of abnormal CD4+ and/or CD8+ lymphocyte counts.

Cytomegalovirus and human herpesvirus-6 trans- activate the HIV-1 long terminal repeat via multiple response regions in human fetal astrocytes

Cytomegalovirus (CMV) and human herpesvirus-6 (HHV-6) infection stimulated HIV-1 replication and trans-activated the HIV-1 promoter (the long terminal repeat or LTR) to a similar extent in transfected, nonimmortalized, human fetal astrocytes. CMV infection increased basal LTR expression by approximately sevenfold, while HHV-6 infection increased basal LTR expression by fourfold. This enhancing effect required cell-cell contact between CMV-infected or HHV-6-infected and LTR-containing cells. To determine the target regions on the HIV promoter that respond to CMV and HHV-6 trans-activation, several modified LTR-reporter gene constructs were tested. Loss of functional NFkappaB, Sp1, or upstream modulatory sites on the LTR caused significant reduction ofbasal LTR expression in astrocytes. These elements also mediated the trans-activation events during HHV-6 or CMV infection in astrocytes, though to varying degrees. Electrophoretic mobility shift assays (EMSA) indicated that core, enhancer, and upstream modulatory regions of the LTR interacted specifically with nuclear proteins from both uninfected and CMV- or HHV-6-infected human fetal astrocytes. CMV or HHV-6 infection did not appear to induce unique, LTR-specific nuclear binding proteins, but rather enhanced the relative proportion of some of the existing protein complexes, in particular, the complexes formed with the AP-1 binding sites on the HIV-1 LTR (nt - 354 to - 316). Our data suggest that CMV or HHV-6 trans-activation of HIV LTR activity in human fetal astrocytes proceeds via intracellular molecular interactions involving herpesviral gene products, cellular proteins, and multiple sites on the LTR upstream of the TATA box. The pattern of LTR activity in astrocytes suggests that host cell factors modulating HIV expression may differ from those dominant in T-cells or immortalized astroglia, and this could contribute to differences in the astrocytes ability to support HIV replication.