Mechanisms of prostaglandin E2-induced interleukin-6 release in astrocytes: possible involvement of EP4-like receptors, p38 mitogen-activated protein kinase and protein kinase C.

The expression of cyclooxygenase-2 (COX-2) and the synthesis of prostaglandin E2 (PGE2) as well as of cytokines such as interleukin-6 (IL-6) have all been suggested to propagate neuropathology in different brain disorders such as HIV-dementia, prion diseases, stroke and Alzheimer's disease. In this report, we show that PGE2-stimulated IL-6 release in U373 MG human astroglioma cells and primary rat astrocytes. PGE2-induced intracellular cAMP formation was mediated via prostaglandin E receptor 2 (EP2), but inhibition of cAMP formation and protein kinase A or blockade of EP1/EP2 receptors did not affect PGE2-induced IL-6 synthesis. This indicates that the cAMP pathway is not part of PGE2-induced signal transduction cascade leading to IL-6 release. The EP3/EP1-receptor agonist sulprostone failed to induce IL-6 release, suggesting an involvement of EP4-like receptors. PGE2-activated p38 mitogen-activated kinase (p38 MAPK) and protein kinase C (PKC). PGE2-induced IL-6 synthesis was inhibited by specific inhibitors of p38 MAPK (SB202190) and PKC (GF203190X). Although, up to now, EP receptors have only rarely been linked to p38 MAPK or PKC activation, these results suggest that PGE2 induces IL-6 via an EP4-like receptor by the activation of PKC and p38 MAPK via an EP4-like receptor independently of cAMP.

Lipopolysaccharide-induced switch between retinoid receptor (RXR) alpha and glucocorticoid attenuated response gene (GARG)-16 messenger RNAs in cultured rat microglia.

Glucocorticoid-attenuated response genes (GARG) belong to a recently described family of genes responsive to the action of dexamethasone. Full-length cDNA of one member of this family, GARG16, has been cloned from rat microglia and regulation of its mRNA expression has been studied. Moreover, regulation of retinoid/retinoic acid activated transcription factor (RXR/RAR) mRNAs in mixed astrocyte and in purified microglia cultures has been investigated. RARbeta mRNA was undetectable in microglia by RT-PCR, whereas clearly present in the mixed cultures. RXRalpha, RARgamma, and GARG16 mRNAs were found in both culture systems. RXRalpha mRNA was strongly expressed in control microglia but rapidly declined upon treatment with LPS. Conversely, GARG16 mRNA was almost untraceable in control microglia but rapidly increased by LPS. Time-course studies revealed an oscillating behavior of expression of both mRNAs during the first 6 hr, which receded to control levels (RXRalpha high, GARG16 low) at 72 hr of LPS-treatment. Additionally, p38 MAPK and SEK phosphorylations peaked at 1 hr followed by steady declines, whereas MEK and c-Jun showed double peaks at 1+4 hr and 1+6 hr, respectively, before subsiding to control levels. This behavior was not observed in comparative studies with TNF-alpha, interleukin-10 (IL-10), or interferon-gamma inducible protein 10 (IP-10). Finally, inhibitors of p38 MAPK, p42/p44 ERK, and PKCalpha as well as the use of dexamethasone revealed major influences of the p38 MAPK-c-Jun-AP-1 signaling pathway on RXRalpha and GARG16 mRNA expressions. The counter regulatory control of GARG16 and RXRalpha mRNA expression is believed to be an example of a fine-tuned cellular mechanism to react to inflammatory stimuli.

Differential expression, activity and regulation of the sodium/myo-inositol cotransporter in astrocyte cultures from different regions of the rat brain.

The high-affinity sodium/myo-inositol cotransporter (SMIT) is involved in osmoregulation in several cells and tissues. In the CNS the activity of SMIT also determines the individual susceptibility of neural cells to the inositol depleting effect of lithium, which is considered to be important in lithium's therapeutic effects in manic-depressive illness. Among neural cells SMIT is particularly active in astrocytes. In the present work we have cloned the cDNA of SMIT of the rat and assessed its activity, expression and regulation in primary astroglia cultures derived from five different rat brain regions: cerebellum, cortex, diencephalon, hippocampus and tegmentum. After an incubation period of 24 h in medium containing 3[H]labeled myo-inositol different steady-state concentrations were detected which were dependent on the brain region from which the astrocytes were cultured. In addition, myo-inositol uptake in astrocytes from different areas was characterized by two different Km values (27 microM for cerebellum and diencephalon, 50 microM for cortex, hippocampus and tegmentum) and by three different v(max) values (approx. 200 pmol/mg protein/min for astrocytes from cerebellum and tegmentum, 298 for hippocampus and 465 for cortex), indicating that the active myo-inositol uptake into astroglial cells is distinct in the various brain regions. The efficacy of uptake as determined by v(max) values of 3[H]myo-inositol uptake correlated with the level of mRNA of SMIT in the astrocyte cultures from the various brain regions as determined by semiquantitative reverse transcription-polymerase chain reaction (RT-PCR). Both 3[H]myo-inositol uptake and SMIT mRNA content was upregulated by incubation of astrocytes in medium of increased osmolarity. In astrocytes from cerebellum, cortex, hippocampus and tegmentum 3[H]myo-inositol uptake was downregulated by chronic incubation with 400 microM inositol. This effect was not observed in astrocytes from diencephalon. Furthermore, in astrocytes from cortex and hippocampus but not from cerebellum, diencephalon and tegmentum incubation with corticosterone for three days upregulated 3[H]myo-inositol uptake. It is concluded that SMIT is differentially expressed and regulated in astrocytes from distinct brain regions. These regional differences suggest particular consideration of localized effects in investigations of the role of myo-inositol in the mechanism of action of antibipolar drugs.

Cultured rat microglia express functional beta-chemokine receptors.

We have investigated the functional expression of the beta-chemokine receptors CCR1 to 5 in cultured rat microglia. RT-PCR analysis revealed constitutive expression of CCR1, CCR2 and CCR5 mRNA. The beta-chemokines MCP-1 (1-30 nM) as well as RANTES and MIP-1alpha (100-1000 nM) evoked calcium transients in control and LPS-treated microglia. Whereas, the response to MCP-1 was dependent on extracellular calcium the response to RANTES was not. The effect of MCP-1 but not that of RANTES was inhibited by the calcium-induced calcium release inhibitor ryanodine. Calcium responses to MCP-1- and RANTES were observed in distinct populations of microglia.

Abnormal G protein alpha(s) - and alpha(i2)-subunit mRNA expression in bipolar affective disorder.

Disturbances of events associated with intracellular signaling pathways have been suspected of involvement in the development or progression of affective disorders. Often, heterotrimeric G proteins are located at the beginning of these pathways as modulators of extracellular messages. For this reason, messenger RNA expression of three G protein alpha-subunits and of phosphatidylinositol-3 kinase (PI-3 K) regulatory subunit p85 was examined in granulocytes from patients with bipolar or unipolar affective disorder and compared to healthy controls. Messenger RNA expression of the G protein subunit alpha(q) and of p85 was identical in unipolar and bipolar patients and in controls. Furthermore, mRNAs of G protein subunits alpha(s) and alpha(i2) were not different in unipolar patients as compared to healthy controls. Alpha(s) mRNA, however, was markedly increased in bipolar patients. This increase was observed in lithium-treated (more than 12 months) and in unmedicated patients. Elevated levels of alpha(i2) mRNA in unmedicated bipolar patients did not reach statistical significance, whereas mRNA in bipolar patients receiving lithium was significantly above controls. Finally, long-term medication of unipolar patients with lithium had no influence on alpha(i2) mRNA levels. The data reveal elevated mRNA levels of G alpha(s) as a robust feature of bipolar affective disorder. Moreover, despite responsiveness of alpha(i2) gene expression to cAMP-related events, no substantial upregulation of alpha(i2) mRNA was observed in bipolar patients. The lack of alpha(i2) mRNA upregulation, hence, could be an additional abnormality in these patients. Even though lithium was able to reinstate this upregulation, there was no feedback downregulation of alpha(s). This strongly supports the notion of major disturbances of the cAMP signaling system in bipolar illness.

Cloning of rat HIV-1-chemokine coreceptor CKR5 from microglia and upregulation of its mRNA in ischemic and endotoxinemic rat brain.

Chemokine receptors play a crucial role in the recruitment of immune cells to sites of inflammation. Although chronic diseases of the brain are often accompanied by inflammatory events, there is presently no information about the occurrence and regulation of these receptors in the central nervous system (CNS). Moreover, one CC-chemokine receptor, CKR5, has recently been identified as coreceptor for HIV-1 entry into macrophages. HIV-1 target cells in brain are macrophage-related microglia, which suggests that they are infected by the same mechanism (He et al.,: Nature 385:645-649, 1997). Although rats are not susceptible to HIV-1 infection, they can be used to study chemokine receptor regulation in a variety of brain pathologies. After cloning CC-CKR5 and establishing reverse transcriptase polymerase chain reaction (RT-PCR) for its ligands macrophage inflammatory protein (MIP)-1alpha, MIP-1beta, and regulated on activation, normal T cell-expressed and secreted (RANTES), we studied expression of these four mRNAs in purified microglia and compared it with their expression in rat brain. Lipopolysaccharide (LPS)-treated microglia showed transiently increased mRNA levels of both CKR5 and its ligands. Similar data were obtained from brains of LPS-injected rats. In middle cerebral artery occluded (MCAO)-animals, RANTES mRNA was unaffected, whereas CKR5 mRNA showed a sustained rise until 96 hr after surgery. MIPs exogenously added to microglial cultures markedly reduced CKR5 mRNA expression, whereas RANTES did not. MIP mRNAs, in contrast to RANTES and CKR5 mRNAs, were undetectable in normal brain. RANTES appears to play a role distinct from MIPs in brain. In summary, upregulation of CC-chemokines and CKR5 in the CNS upon bacterial infection or in ischemia may impact on microglial activation stage and result in increased risk of HIV-1 infection.

Molecular biology of microglia cytokine and chemokine receptors and microglial activation.

Activation of brain microglial cells can be subdivided into a number of stages. Early stages likely are proliferation and migration to sites of cell damage. These two stages have been studied exemplarily on the IL-3 receptor beta-subunit and on the CC-chemokine receptor 5 using molecular biological methods. First, IL-3 receptor beta-subunit cDNA has been cloned in full length from rat microglia. Since cultured microglia are already activated to some extent, mRNA of this subunit has been detected in the isolated cells, but was absent in normal rat brain. Lipopolysaccharide (LPS) increased this mRNA in the cultured cells and LPS injected into the circulation of rats induced the mRNA specifically in brain microglia as revealed by in situ hybridizations. Next, we obtained partial cDNAs of receptor-coupled protein tyrosine kinases JAK 1 and JAK 2. These mRNAs were present both in cultured microglia and in rat brain, but were not influenced by LPS. Finally, a full-length cDNA of the rat chemokine receptor 5 has been obtained by PCR methodology. Its mRNA was increased by administration of LPS both in cultured microglia and in vivo. It is expected, that further investigations on these receptors could help to develop improved strategies to combat chronic inflammatory events in the brain.