Search

It is well known that infection by the human immunodeficiency virus (HIV) dysregulates cell physiology, but little information is available on the consequences of HIV infection in primary macrophages and microglia. The authors examined the relationship between cell proliferation and HIV infection in primary cultures of microglia and in human central nervous system (CNS). In cultures infected with HIV (ADA and BaL), granulocyte-macrophage colony-stimulating factor (GM-CSF)-mediated cell proliferation was reduced in productively infected (p24+) cells as compared to p24- cells. The reduction was observed with both Ki67 and BrdU labeling, suggesting a G1/S block. The reduction was insignificant when microglia were infected with a Vpr- mutant virus. In human CNS, proliferating (Ki67+) cells were rare but were increased in the HIV+ and HIV encephalitis (HIVE) groups compared to the HIV- group. A positive correlation between GM-CSF immunoreactivity and Ki67 counts, implicating GM-CSF as a growth factor in human CNS was found. The relationship between total macrophage (CD68+) proliferation and infected macrophage (p24+) proliferation was assessed in HIVE by double labeling. Whereas 1.2% of total CD68+ cells were Ki67+, only 0.5% of HIV p24+ cells were Ki67+ (P < .05). Furthermore, staining for CD45RB (as opposed to CD68) facilitated the identification of Ki67+ microglia, indicating that CD68 could underestimate proliferating microglia. The authors conclude that although there is increased expression of GM-CSF and increased cell proliferation in the CNS of HIV-seropositive individuals, cell proliferation in the productively infected population is actually suppressed. These data suggest that there might be a viral gain in the suppressed host cell proliferation.

Macrophages and microglia are productively infected by HIV-1 and play a pivotal role in the pathogenesis of AIDS dementia. Although macrophages and microglia express CD45, a transmembrane protein tyrosine phosphatase, whether modulation of its activity affects human immunodeficiency virus type 1 (HIV-1) replication is unknown. Here, we report that of the five human CD45 isoforms, microglia express CD45RB and CD45RO (RB > RO) and treatment of microglia with a CD45 agonist antibody alphaCD45RO (UCHL-1) inhibits HIV-1 replication. alphaCD45RO prevented HIV-1 negative factor (Nef)-induced autophosphorylation of hematopoietic cell kinase (Hck), a myeloid lineage-specific Src kinase. Recombinant CD45 protein also inhibited HIV-1-induced Hck phosphorylation in microglia. Antennapedia-mediated delivery of Hck Src homology domain 3 (SH3), a domain that binds to the Nef PxxP motif with high affinity, reduced HIV-1-induced Hck phosphorylation and HIV-1 production in microglia. HIV-1-induced LTR transactivation was observed in U38 cells stably overexpressing wild-type Hck but not kinase-inactive Hck. In microglia, alphaCD45RO reduced activation of transcription factors (NF-kappaB and CCAAT enhancer binding protein) necessary for LTR transactivation in macrophages. These results establish that in myeloid lineage cells, Nef interacts with the Hck SH3 domain, resulting in autophosphorylation of Hck and an increase in HIV-1 transcription. alphaCD45RO-mediated inhibition of HIV-1 replication in microglia identifies the CD45 protein tyrosine phosphatase as a potential therapeutic target for HIV-1 infection/AIDS dementia.

Human immunodeficiency virus type 1 (HIV-1) infection of the brain produces a characteristic disease called acquired immunodeficiency syndrome (AIDS) dementia in which productive infection and inflammatory activation of microglia and macrophages play a central role. In this report, the authors demonstrate that minocycline (MC), a second-generation tetracycline with proven safety and penetration to the central nervous system, potently inhibited viral production from microglia. Inhibition of viral release was sustained through the entire course of infection and even when the drug exposure was limited to the first day of infection. Minocycline was effective even at low viral doses, and against R5- and X4R5-HIV, as well as in single-cycle reporter virus assays. Electrophoretic mobility shift analysis showed that minocycline inhibited nuclear factor (NF)-kappaB activation in microglia. HIV-1 long terminal repeat (LTR)-promoter activity in U38 cells was also inhibited. These results, combined with recently demonstrated in vivo anti-inflammatory effects of MC on microglia, suggest a potential utility for MC as an effective adjunct therapy for AIDS dementia.

Regulation of cytokine and chemokine expression in microglia may have implications for CNS inflammatory disorders. In this study we examined the role of the cyclopentenone PG 15-deoxy-Delta(12,14)-PGJ(2) (15d-PGJ(2)) in microglial inflammatory activation in primary cultures of human fetal microglia. 15d-PGJ(2) potently inhibited the expression of microglial cytokines (IL-1, TNF-alpha, and IL-6). We found that 15d-PGJ(2) had differential effects on the expression of two alpha-chemokines; whereas the Glu-Lys-Arg (ELR)(-) chemokine IFN-inducible protein-10/CXCL10 was inhibited, the ELR(+) chemokine IL-8/CXCL8 was not inhibited. These findings were shown in primary human microglia and the human monocytic cells line THP-1 cells, using diverse cell stimuli such as bacterial endotoxin, proinflammatory cytokines (IL-1 and TNF-alpha), IFN-beta, and HIV-1. Furthermore, IL-8/CXCL8 expression was induced by 15d-PGJ(2) alone or in combination with TNF-alpha or HIV-1. Combined results from EMSA, Western blot analysis, and immunocytochemistry showed that 15d-PGJ(2) inhibited NF-kappaB, Stat1, and p38 MAPK activation in microglia. Adenoviral transduction of super-repressor IkappaBalpha, dominant negative MKK6, and dominant negative Ras demonstrated that NF-kappaB and p38 MAPK were involved in LPS-induced IFN-inducible protein 10/CXCL10 production. Interestingly, although LPS-induced IL-8/CXCL8 was dependent on NF-kappaB, the baseline or 15d-PGJ(2)-mediated IL-8/CXCL8 production was NF-kappaB independent. Our results demonstrate that 15d-PGJ(2) has opposing effects on the expression of two alpha-chemokines. These data may have implications for CNS inflammatory diseases.

IL-16 is a natural ligand for the CD4 molecule and is known for its chemotactic and anti-HIV-1 activities. We determined IL-16 expression in human brain tissue with HIV-1 encephalitis by specific immunocytochemistry and showed that infiltrating lymphocytes and activated microglia express IL-16. IL-16 immunoreactivity was particularly pronounced in microglial nodules. In vitro, human foetal microglia and not astrocytes produce IL-16, and HIV-1 infection up-regulates microglial IL-16 release in a Nef-dependent manner. These results support the notion that, in the brain, IL-16 is a macrophage-lineage specific modulator of the inflammatory response and HIV-1 expression. Recruitment of IL-16+ T cells and microglia/macrophages may represent an innate response to HIV-1 infection in the central nervous system that counterbalances viral stimulatory factors., (Copyright 2003 Blackwell Publishing Ltd)

Microglia are endogenous brain macrophages that show distinct phenotypes such as expression of myeloid antigens, ramified morphology, and presence within the neural parenchyma. They play significant roles in a number of human CNS diseases including AIDS dementia. Together with monocyte-derived (perivascular) macrophages, microglia represent a major target of HIV-1 infection. However, a recent report challenged this notion based on findings in SIV encephalitis. This study concluded that perivascular macrophages can be distinguished from parenchymal microglial cells by their expression of CD14 and CD45, and that macrophages, but not microglia, are productively infected in SIV and HIV encephalitis. To address whether parenchymal microglia are productively infected in HIV encephalitis, we analyzed expression of CD14, CD45 and HIV-1 p24 in human brain. Microglia were identified based on their characteristic ramified morphology and location in the neural parenchyma. We found that parenchymal microglia are CD14+ (activated), CD45+ (resting and activated), and constitute approximately two thirds of the p24+ cells in HIV encephalitis cases. These results demonstrate that microglia are major targets of infection by HIV-1, and delineate possible differences between HIVE and SIVE. Because productively infected tissue macrophages serve as the major viral reservoir, these findings have important implications for AIDS.

Microglia are pivotal in the pathogenesis of AIDS dementia, as they serve as the major target of HIV infection in the CNS. In addition, activation of microglia correlates best with clinical dementia. Although the beta-chemokine RANTES/CCL5 is important in modulating HIV infection as well as cellular activation, no information is available regarding how its expression is regulated in microglia by HIV-1. Here we report that RANTES/CCL5 expression is induced in microglia by HIV-1, but that this requires infection by HIV-1. This conclusion was supported by (1) the delayed kinetics coinciding with viral replication; (2) the lack of effect of X4 viruses; (3) inhibition by the reverse transcriptase inhibitor AZT, and (4) the lack of effect of cytokine antagonists or antibodies. Interestingly, RANTES/CCL5 production was dependent on the viral accessory protein Vpr, in addition to Nef, demonstrating a novel role for Vpr in chemokine induction in primary macrophage-type cells. Furthermore, the specific p38 MAP kinase inhibitor SB203580 augmented chemokine expression in microglia, indicating a negative role played by p38. These data suggest unique features of RANTES/CCL5 regulation by HIV-1 in human microglial cells.

Significant numbers of patients with acquired immunodeficiency syndrome (AIDS) develop CNS infection primarily in macrophages and microglial cells. Therefore, the regulation of human immunodeficiency virus type 1 (HIV-1) infection and activation of the brain mononuclear phagocytes subsequent to infection are important areas of investigation. In the current report, we studied the role of granulocyte-macrophage colony-stimulating factor (GM-CSF) and macrophage-CSF (M-CSF) in the expression of antiviral beta-chemokines and HIV-1 p24 in cultures of primary human fetal microglia. We found that stimulation with GM-CSF or M-CSF induced macrophage inflammatory proteins (MIP-1alpha and MIP-1beta) and augmented RANTES expression, after HIV-1 infection of microglia. This was not due to the effect of GM-CSF on viral expression because GM-CSF was neither necessary nor stimulatory for viral infection, nor did GM-CSF enhance the expression of env-pseudotyped reporter viruses. Blocking GM-CSF-induced microglial proliferation by nocodazole had no effect on beta-chemokine or p24 expression. The functional significance of the GM-CSF-induced beta-chemokines was suggested by the finding that, in the presence of GM-CSF, exogenous beta-chemokines lost their anti-HIV-1 effects. We further show that although HIV-1-infected microglia produced M-CSF, they failed to produce GM-CSF. In vivo, GM-CSF expression was localized to activated astrocytes and some inflammatory cells in HIV-1 encephalitis, suggesting paracrine activation of microglia through GM-CSF. Our results demonstrate a complex interplay between CSFs, chemokines, and virus in microglial cells and may have bearing on the interpretation of data derived in vivo and in vitro., (Copyright 2002 Wiley-Liss, Inc.)

Microglia, the resident brain macrophages, are the principal cells involved in the regulation of inflammatory and antimicrobial responses in the CNS. Interferon-beta (IFNbeta) is an antiviral cytokine induced by viral infection or following non-specific inflammatory challenges of the CNS. Because of the well-known anti-inflammatory properties of IFNbeta, it is also used to treat multiple sclerosis, an inflammatory CNS disease. Despite the importance of IFNbeta signaling in CNS cells, little has been studied, particularly in microglia. In this report, we investigated the molecular mechanisms underlying IFNbeta-induced beta-chemokine expression in primary human fetal microglia. Multiple signaling cascades are activated in microglia by IFNbeta, including nuclear factor-kappaB (NF-kappaB), activator protein-1 (AP-1) and Jak/Stat. IFNbeta induced IkappaBalpha degradation and NF-kappaB (p65:p50) DNA binding. Inhibition of NF-kappaB by either adenoviral transduction of a super repressor IkappaBalpha, or an antioxidant inhibitor of NF-kappaB reduced expression of the beta-chemokines, regulated upon activation, normal T-cell expressed and secreted (RANTES) and macrophage inflammatory protein (MIP)-1beta. IFNbeta also induced phosphorylation of extracellular signal-regulated kinase (ERK) mitogen-activated protein kinase, and the MAP kinase kinase 1 (MEK1) inhibitor PD98059 dose-dependently inhibited beta-chemokine mRNA and protein expression. PD98059 did not inhibit NF-kappaB binding, demonstrating that ERK was not responsible for NF-kappaB activation. Two downstream targets of ERK were identified in microglia: AP-1 and Stat1. IFNbeta induced AP-1 nuclear binding activity in microglia and this was suppressed by PD98059. Additionally, IFNbeta induced Stat1 phosphorylation at both tyrosine 701 (Y701) and serine 727 (S727) residues. S727 phosphorylation of Stat1, which is known to be required for maximal transcriptional activation, was inhibited by PD98059. Our results demonstrating multiple signaling cascades initiated by IFNbeta in primary human microglia are novel and have implications for inflammatory and infectious diseases of the CNS.

Nitric oxide (NO) produced by microglia has been implicated in the pathogenesis of various central nervous system diseases; however, the intracellular signal pathways for the production of NO are not well known. Protein kinase C (PKC) plays a key role in a variety of signal transduction processes. To elucidate how PKC regulates microglial NO production, we examined the effects of PKC inhibitors on lipopolysaccharide (LPS)-stimulated NO production by primary cultured rat microglia. Staurosporine, a non-selective PKC inhibitor, increased LPS-induced production of NO at 0.1-10 nM range of concentration. Protein kinase A (PKA) inhibitor, H89, did not affect LPS-induced NO production, suggesting that staurosporine effect is not mediated by inhibition of PKA. However, other two PKC inhibitors, whose specificities for PKC isoforms were different, Gö6976 and Ro-32-0432, exhibited different effects on NO production from staurosporine; the former inhibited and the latter showed no effect. Interestingly, an activator of PKC, phorbol 12-myristate 13-acetate (PMA) also increased LPS-induced production of NO at 1-10 nM range of concentration, suggesting that prolonged incubation with PMA caused down-regulation of PKC. These results indicate that the inhibition or down-regulation of some PKC isoforms causes the enhancement of NO production. The different effects of PKC inhibitors on the NO production suggest that the different PKC isoforms play different roles in regulation of NO production in microglia.

Microglial activation has recently been recognized as a cause of damage in various neurodegenerative diseases. A possible mechanism underlying this damage is the activation of microglia by serum factors leaked through a disruption of the blood-brain barrier, which in turn trigger microglial cell proliferation and the release of various substances toxic to neurons, such as superoxide (O(2)(-)). We recently reported that serum albumin enhanced O(2)(-) production in cultured rat microglia stimulated by phorbol ester. In the present report, we identify the active site of this enhancement within the albumin molecule. We purified an active subfragment from trypsin-treated bovine serum albumin that was composed of 12-mer and 33-mer peptides connected by a disulfide bond. The chemically synthesized 12-mer peptide showed activity within a concentration range ( approximately 10(-7) M:) equivalent to that of albumin. The activities of a series of synthesized peptides conclusively indicated that the minimum active sequence was Leu-His-Thr-Leu. The present study may shed light on the mechanism of neuronal cell damage in various neurodegenerative diseases.

The pathological activation of microglia has been implicated in ischemic neuronal damage and some neurodegenerative diseases; however, the mechanism of microglial activation is not well understood. Previously, we showed that a serum factor, albumin, increased O(2)(-) production by cultured microglia (Si et al., 1997, Glia 21: 413-418). In the present study, we found that serum also enhanced lipopolysaccharide (LPS)-induced production of nitric oxide and tumor necrosis factor-alpha, which are other important neurotoxins released by activated microglia. In the presence of 0.1% normal rat serum, the half-effective concentration for LPS decreased from 300 to 1 ng/ml. The factor seemed to be a relatively high-molecular-weight protein because the factor was retained after a molecular sieve (50 kDa) membrane separation. The factor was labile to trypsinization and heat treatment at 72 degrees C for 5 min but was stable at 56 degrees C for 60 min. Several purified serum proteins including albumin could not mimic the enhancing effect of serum. Acute-phase serum showed a potent enhancing effect at a 10 times lower concentration than the normal serum. By gel filtration chromatography, the enhancing effect observed was a single peak at about 60 kDa. These results suggest that some serum protein infiltrates into brain parenchyma after blood-brain barrier disruption and such protein may result in neuronal damage by activating microglia to release neurotoxins in some central nervous system diseases., (Copyright 2000 Academic Press.)

Microglial activation has been considered as a result of neuronal damage, however, recently it becomes to recognize as a possible cause of the damage in various neurodegenerative diseases. To elucidate the mechanism of the microglial activation, we examined the time course of lipopolysaccharide (LPS)-induced change in morphology and the release of cytokines and nitric oxide (NO) in cultured microglia from neonatal rat brains. With addition of 1 microg/ml LPS, the cell morphology was drastically changed within 3 h from amoeboid shape to bipolar rod shape. The peak time of such morphological change was at 6 h and then returned to small round shape gradually. This transient change in morphology was completely inhibited by 0.1 mM dibutyryl-cAMP. On the other hand, the release of cytokines and NO showed different time courses after stimulation by LPS; at first tumor necrosis factor (TNF)-alpha was released within 1 h lag time, secondly interleukin (IL)-1beta within 3 h, thirdly IL-6, and at last NO was released with about 6 h lag time. The addition of dibutyryl cAMP markedly inhibited the release of TNF-alpha and IL-1beta, but not IL-6 and NO at all. These results suggest that there are at least two different intracellular signaling pathways of LPS-induced microglial activation; one for early release of TNF-alpha and IL-1beta sensitive to dibutyryl-cAMP and the other for late release of IL-6 and NO insensitive to dibutyryl-cAMP. The transient morphological change seems to be associated with the early release based on the sensitivity to dibutyryl-cAMP.

A pathological microglial activation is believed to contribute to progressive neuronal damage in neurodegenerative diseases by the release of potentially toxic agents and by triggering reactive astrocytic changes. Using cultured microglia from neonatal rat brains, we investigated the mode of propentofylline action in strengthening cAMP-dependent intracellular signaling. We compared this action with the effects of dibutyryl-cAMP, a cell-permeable cAMP analog. Propentofylline inhibited lipopolysaccharide (LPS)-induced release of both tumor necrosis factor (TNF)-alpha and interleukin (IL)-1beta in a dose-dependent manner within the therapeutic low micromolar range. However, LPS-induced release of IL-6 and NO were not affected by propentofylline. All these differential effects of propentofylline on LPS-induced microglial release were mimicked by the addition of dibutyryl-cAMP. Microglial proliferation and phorbol myristate acetate (PMA)-induced O2- release were also dose-dependently inhibited by propentofylline as well as dibutyryl-cAMP. These results suggest that propentofylline, probably via reinforcement of cAMP intracellular signaling, alters the profile of the newly adopted immune properties in a way that it inhibits potentially neurotoxic functions while maintaining beneficial functions. This differential regulation of microglial activation may explain the neuroprotective mechanism exerted by propentofylline., (Copyright 1998 Elsevier Science B.V.)

Microglial activation and disruption of blood-brain barrier (BBB) are known to occur and contribute to neuronal damages in cerebral ischemic conditions and in some neurodegenerative diseases. To investigate whether a serum factor leaked out from circulation enhances microglial activation, we examined the effect of normal rat serum on superoxide (O2-) production by cultured microglia. Microglia cultured from neonatal rat brains were studied on their O2- production induced by the addition of phorbol myristate acetate by a method of acetyl-cytochrome c reduction. The O2- production was significantly increased by the addition of 0.01% rat serum, and the maximal enhancement was observed at about 0.1% rat serum. After the serum was fractionated using a molecular sieve membrane, we observed the enhancing effect only in a greater molecular weight fraction (>50 kDa). Furthermore, three kinds of bovine serum albumin (BSA) with different purity, and human serum and plasma albumins, also enhanced O2- production to a similar extent to that by rat serum. However, other proteins tested showed no significant effect. The enhancement of O2- production by BSA was observed dose-dependently, and the effect of 50 microg/ml of purified BSA was equivalent to that of 0.1% rat serum, suggesting that albumin itself enhances O2- production by microglia. These results imply that albumin leaked out through impaired BBB may activate microglia and that the potentiation of O2- production by albumin results in the pathogenesis of neuronal damage in cerebral ischemia and some neurodegenerative diseases.

In order to elucidate the mechanism(s) of neuronal protection by hypothermia against ischemic damage, we examined the effect of lowering temperature on the microglial activation that is thought to cause the development of ischemia-induced neuronal damages. Cultured microglia from neonatal rats were measured for microglial activation by the following indices: production of superoxide and nitric oxide by the methods of acetyl-cytochrome c reduction and nitrite accumulation in the culture medium, respectively, and cell proliferation evaluated by [3H]thymidine uptake. At 30 degrees C, superoxide production induced by phorbol ester was approximately as low as 30% of the control at 37 degrees C, and nitric oxide production after addition of lipopolysaccharide was decreased to approximately 25% of the control. The time course of nitric oxide production indicates that the induction of nitric oxide synthase seemed to be significantly suppressed by lowering temperature. In addition, the proliferation of microglia was remarkably inhibited at 30 degrees C. The level of proliferation in the hypothermic condition is much lower in microglia (14% of the control) than those in astrocytes cultured from brain cortices (96%) and fibroblasts cultured from brain meninges (53%), suggesting that the microglial activation is highly susceptible to lowering temperature. The present study indicates that hypothermia potently inhibits proliferation, superoxide and nitric oxide production of cultured microglia and that the hypothermic protection against postischemic neuronal damage might be, at least in part, due to the suppression of microglial activation.

Propentofylline is a xanthine derivative that has been known to protec t neurons against ischemia-induced damage. To assess its neuroprotective mechanisms, we examined the effect of propentofylline on microglial proliferation that is thought to play an important role in neuronal damage. We determined the proliferation of microglia cultured from neonatal rat brains by measuring [3H]thymidine update. Propentofylline inhibited microglial proliferation in a dose dependent manner; EC50 was about 3 mu M. Similar results were observed with 2-chloroadenosine (agonist for A1 and A2 adenosine receptors) and 2-chloro-N6-cyclopentyladenosine (A1 receptor agonist) but not with 2-p-(2-carboxyethyl)phenethylamino-5'-N-ethyl-carboxamidoadenosine hydrochloride (A2 receptor agonist). However, 8-cyclopentyl-1,3-dipropylxathine (A1 receptor antagonist) could not reverse the inhibitory effect of propentofylline. Our results suggest that the neuroprotection by propentofylline is, as least in part, due to the direct effect of the drug on microglia and that the drug inhibits the proliferation via a certain mechanism not directly mediated by adenosine receptors.

Microglia, the resident brain macrophages, are the principal cells involved in the regulation of inflammatory and antimicrobial responses in the CNS. Interferon-β (IFNβ) is an antiviral cytokine induced by viral infection or following non-specific inflammatory challenges of the CNS. Because of the well-known anti-inflammatory properties of IFNβ, it is also used to treat multiple sclerosis, an inflammatory CNS disease. Despite the importance of IFNβ signaling in CNS cells, little has been studied, particularly in microglia. In this report, we investigated the molecular mechanisms underlying IFNβ induced βchemokine expression in primary human fetal microglia. Multiple signaling cascades are activated in microglia by IFNβ, including nuclear factor-κB (NF-κB), activator protein-1 (AP-1) and Jak/Stat. IFNβ induced IκBα degradation and NF-κB (p65:p50) DNA binding. Inhibition of NF-κB by either adenoviral transduction of a super repressor IκBα, or an antioxidant inhibitor of NF-κB reduced expression of the βchemokines, regulated upon activation, normal T-cell expressed and secreted (RANTES) and macrophage inflammatory protein (MIP)-Iβ. IFNβ also induced phosphorylation of extracellular signal-regulated kinase (ERK) mitogen-activated protein kinase, and the MAP kinase kinase 1 (MEK1) inhibitor PD98059 dose-dependently inhibited β-chemokine mRNA and protein expression. PD98059 did not inhibit NF-κB binding, demonstrating that ERK was not responsible for NF-κB activation. Two downstream targets of ERK were identified in microglia: AP-1 and Stat1. IFNβ induced AP-1 nuclear binding activity in microglia and this was suppressed by PD98059. Additionally, IFNβ induced Stat1 phosphorylation at both tyrosine 701 (Y701) and serine 727 (S727) residues. S727 phosphorylation of Stat1, which is known to be required for maximal transcriptional... [ABSTRACT FROM AUTHOR]

The article presents an abstract of the research paper "Serum Enhancement of Microglial Production of Superoxide, Nitric Oxide, and Tumor Necrosis Factor-alpha in Culture." It will be presented in the joint meeting of the International Society for Neurochemistry and the European Society of Neurochemistry that will be held in Berlin, Germany from August 8-14, 1999. By the release of neurotoxic mediators, activated microglia can have an impact on neurons.

Viruses often alter cell death pathways in order to ensure their propagation, and HIV-1 is no exception. Whereas, the role of HIV-1 in T cell apoptosis have been thoroughly examined, there is little known of its effects on apoptosis of cells in the monocyte/macrophage lineage. Here we examine the mechanisms by which HIV1 modifies apoptotic pathways in human microglia in vivo and in vitro. Microglia are the resident macrophages of the central nervous system (CNS). In HIV encephalitis (HIVE), both microglia and perivascular macrophages become infected by HIV-1, and the soluble products generated following cell-virus interactions could induce diffuse glial activation and apoptosis of neural cells. By influencing host mechanisms of cell death, HIV-1 could significantly alter the homeostasis of the CNS and guarantee its own survival in a drug-inaccessible viral reservoir. Exposure to HIV-1 induced productive infection of cultured microglia as evident by syncytia formation and p24 expression. TUNEL staining determined that p24[sup +] microglia and multinucleate giant cells were TUNEL-negative. To confirm that this was relevant in vivo, double-labeling studies on HIV encephalitic brain demonstrated that p24[sup +] cells were negative for TUNEL. Double immunohistochemistry for TUNEL and cell specific markers such as GFAP, Von Willebrand Factor, beta-tubulin isotype III, and CD68 show that astrocytes, endothelial cells, neurons, and macrophages, respectively, are TUNEL-positive. Preliminary in vitro studies showed that the HIV accessory protein vpr is not responsible, but that the protein nef may play a role in the ability of HIV-1 to affect the survival of microglia. These results demonstrate that the productive infection of microglia by HIV-1 protects them from undergoing apoptosis. Therefore, HIV-1 may be able to perpetuate its success by creating an apoptosis-resistant viral reservoir in the CNS. [ABSTRACT FROM AUTHOR]