Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2010 Jan 15.
Published in final edited form as: J Neurochem. 1998 Jun;70(6):2357–2368. doi: 10.1046/j.1471-4159.1998.70062357.x

Prostaglandin E2 and 4-Aminopyridine Prevent the Lipopolysaccharide-Induced Outwardly Rectifying Potassium Current and Interleukin-1β Production in Cultured Rat Microglia

Anthony O Caggiano *, Richard P Kraig *,†
PMCID: PMC2807138  NIHMSID: NIHMS166403  PMID: 9603200

Abstract

Brain inflammation includes microglial activation and enhanced production of diffusible chemical mediators, including prostaglandin E2. Prostaglandin E2 is generally considered a proinflammatory molecule, but it also promotes neuronal survival and down-regulates some aspects of microglial activation. It remains unknown, however, if and how prostaglandin E2 prevents microglial activation. In primary culture, microglial activation is predicted by a characteristic pattern of whole-cell potassium currents and interleukin-1β production. We investigated if prostaglandin E2 could alter these currents and, if so, whether these currents are necessary for microglial activation. Microglia were isolated from mixed cell cultures prepared from neonatal rat brains and exposed to 0–10 μM prostaglandin E2 and lipopolysaccharide for 24 h. Currents were elicited by using standard patch-clamp technique, and interleukin-1β production was measured by ELISA. Peak outward current densities in microglia treated with lipopolysaccharide plus prostaglandin E2 (10 nM) were reduced significantly from those of cells treated with lipopolysaccharide alone. Prostaglandin E2 and 4-aminopyridine (a blocker of outward potassium currents) also significantly reduced interleukin-1β production. Thus, although prostaglandin E2 is classified generally as a proinflammatory chemical, it has complex roles in brain inflammation that include preventing microglial activation, perhaps by reducing the outward potassium current.

Keywords: Prostaglandins, Gliosis, Patch-clamp, Inflammation, Eicosanoids

Prostaglandin E2 (PGE2) is a member of the eicosanoid family, chemicals produced from arachidonic acid by cyclooxygenase and lipoxygenase enzymes (for reviews, see Needleman et al., 1986; Funk, 1993; Piomelli, 1993). Eicosanoids mediate inflammation, causing vascular permeability and cellular recruitment (Goetzl et al., 1995), and contribute to normal brain processes, such as long-term potentiation (Luo and Vallano, 1995). Brain injury and inflammation cause increased expression of the enzymes that produce eicosanoids (Yamagata et al., 1993; Caggiano et al., 1996). Furthermore, ischemia (Aktan et al., 1993), concussive head injury (Ellis et al., 1981, 1989; Shohami et al., 1987; Dewitt et al., 1988), and bacterial endotoxin exposure (Van Dam et al., 1993) each change the quantity and relative proportion of PGE2 and other eicosanoids produced in the brain. Blocking the enzymes that produce prostaglandins has been known for decades to reduce injury from ischemia (Furlow and Hallenbeck, 1978) and slow the progression of Alzheimer’s disease (McGeer and McGeer, 1995); however, in vitro evidence exists that PGE2 may protect neurons from cytotoxic injury (Cazevieille et al., 1993; Akaike et al., 1994). Other evidence shows that nitric oxide (NO)-mediated cell injury by microglia may be reduced with PGE2 (Théry et al., 1994). Thus, although PGE2 is generally considered a proinflammatory molecule (Douglas, 1975), it has certain characteristics that are anti-inflammatory as well (Bonney and Davies, 1984; Kunkel et al., 1988).

Microglia are active participants in brain inflammation and injury. For example, with inflammation microglia undergo characteristic changes in morphology, expression of surface molecules (e.g., complement receptors and major immunohistocompatability complexes), and phagocytic activity (Graeber et al., 1988a,b; Morioka and Streit, 1991; Gehrmann et al., 1993; Caggiano and Kraig, 1996). In addition, microglia express mRNAs and enzymes that indicate they can produce cytokines, including tumor necrosis factor (Medana et al., 1997), interleukins (ILs) (Schluter et al., 1997), NO (Paakkari and Lindsberg, 1995), and transforming growth factor-β (Kiefer et al., 1993). The secretion of these substances has been shown directly in vitro following exposure to stimuli, such as γ-interferon (Wong et al., 1984) and lipopolysaccharide (LPS) (Hetier et al., 1988). When microglia show these characteristic changes to stimuli, they are referred to as “reactive” or “activated” microglia.

The functional state of cultured microglia (i.e., resting or reactive) is predicted by their whole-cell K + currents (Fischer et al., 1995). Microglia express an inwardly rectifying K+ current (K+ IR) under most nonstimulated (nonreactive) conditions (Kettenmann et al., 1990; Banati et al., 1991). This current is elicited with hyperpolarizing voltage steps, is K +-selective, displays time-dependent inactivation, is blocked by Ba2+ and Cs +, and is sensitive to tetraethylammonium but not 4-aminopyridine (4-AP) (Kettenmann et al., 1990; Nörenberg et al., 1992, 1994). Following activation by stimuli such as LPS, microglia also express a large outwardly rectifying K + current (K + OR) (Nörenberg et al., 1992). The K+ OR is K +-selective, shows time- and voltage-dependent activation and inactivation, is blocked by Cs+, and is inhibited by both 4-AP and tetraethylammonium (Nörenberg et al., 1994). Others have shown that a change in current profile is necessary for activation or proliferation of astrocytes (Pappas et al., 1994; MacFarlane and Sontheimer, 1997) and other cells (Chiu and Wilson, 1989; Puro et al., 1989; Gallo et al., 1996). The expression of the K+ OR is also thought to be necessary to counteract the depolarizing effects of certain metabolites present during inflammation, such as ATP and IgG (Young et al., 1983; Nörenberg et al., 1992; Illes et al., 1996).

Modulation of eicosanoid production can alter the activation of microglia in vivo (Caggiano and Kraig, 1996). Furthermore, PGE2 can reduce neurotoxic effects of microglia (Théry et al., 1994) and prevent increased expression and activity of inducible NO synthase in cultured microglia (Minghetti et al., 1997). We sought to determine if exposing cultured microglia to PGE2 before and during exposure to LPS might alter the expression of the K + OR that results from exposure to LPS alone and if the K + OR is necessary for activation by blocking it with 4-AP and measuring IL-1β production. In addition, we tested the ability of PGE2 to reduce the microglial IL-1β production and complement type 3 receptor (CR-3) expression following LPS exposure. Together, these parameters would demonstrate that PGE2 can prevent microglial activation and further previous in vivo (Caggiano and Kraig, 1996) and in vitro (Théry et al., 1994; Minghetti et al., 1997) observations. Furthermore, it would begin to demonstrate that reduction of the K + OR may be part of the mechanism by which PGE2 reduces the activation of microglia.

Here we demonstrate that PGE2 reduced the magnitude of the K + OR, the production of IL-1β, and the expression of CR-3 in microglia following exposure to LPS. 4-AP, an effective blocker of the K + OR, reduced the IL-1β production following LPS exposure. These results suggest that the K + OR may be necessary for microglial activation and that PGE2 may be able to prevent or reduce the activation of microglia by blocking the development of this current.

MATERIALS AND METHODS

Mixed-cell cultures

Mixed-cell cultures were prepared with a technique modified from Giulian and Baker (1986). In brief, Wistar rat pups (Charles River Laboratories, Inc., Wilmington, MA, U.S.A.) on postnatal days 0–2 were deeply anesthetized with halothane, washed with ethanol, and decapitated. Brains were removed and placed in medium (Sato’s medium; Bottenstein and Sato, 1979). When all brains had been removed, they were transferred to Hanks’ balanced salt solution. Meninges were carefully removed, cortices bluntly dissected away from the midbrain and brainstem, and any remaining meninges removed. Cortices were stored in medium until all dissection was complete. Tissue was dissociated by agitation in Hanks’ balanced salt solution with trypsin (Sigma, St. Louis, MO, U.S.A.) for 15 min at 37°C in a 50-ml Falcon tube. Filtered fetal bovine serum (FBS) and DNase (Sigma) were added, and the solution was vortexed and spun at 1,100 rpm for 3 min. The supernate was removed, and 5 ml of Sato’s medium with 10% FBS (Sato’s/FBS) was added with 0.5 ml of DNase. The solution was triturated until tissue was dissociated. Sato’s/FBS (25 ml) was added, and the tube was vortexed and spun at 1,100 rpm for 10 min. The supernate was removed. Tissue was resuspended in 25 ml of Sato’s/FBS, vortexed, and spun at 1,100 rpm for 10 min. Again, supernate was removed and tissue resuspended and spun at 1,100 rpm for 5 min. Supernate was removed and Sato’s/FBS was added. Cells were counted on a hemacytometer and diluted to ~2 × 107 cells per 12 ml. Cells were plated in 75-cm2 flasks (12 ml per flask). Medium (Sato’s/FBS) was changed every 3–4 days.

Isolation of microglia

At 10–16 days, plates contained a monolayer of astrocytes with microglia loosely attached and floating in the medium (Giulian and Baker, 1986; Levison and McCarthy, 1991). At this point, flasks were shaken at 150 rpm for 10 min to free adherent microglia. The medium containing microglia was placed on 35-mm tissue culture dishes for 20 min. Dishes were washed several times to remove nonadherent cells. Microglia were maintained in Sato’s/FBS until used.

Identity of microglia

Culture purity was assessed by labeling microglia with 1,1′ - dioctadecyl - 3,3,3′,3′ - tetramethylindocarbocyanine perchlorate–low-density lipoprotein (DiI-Ac-LDL; Biomedical Technologies Inc., Stoughton, MA, or Molecular Probes, Eugene, OR, U.S.A.) and labeling all cells with calcein-AM (Molecular Probes). Fluorescent images (DiI-Ac-LDL: excitation at 550 nm, emission at 570 nm; calcein-AM: excitation at 500 nm, emission at 520 nm) were superimposed, and the number of microglia out of the total number of cells was assessed. Cultures were ~95–99% microglia. Other cells were usually astrocytes as assessed by glial fibrillary acidic protein staining and morphology.

Microglia stained positively with the OX-42 (MCA OX-42; Serotec, Oxford, U.K.) and ED-1 (MCA 341; Serotec) antibodies. Microglia showed increased staining intensity for both OX-42 and ED-1 following LPS stimulation. Normal microglial cultures produced low to nonmeasurable quantities of IL-1, as assessed by ELISA (Biosource International, Camarillo, CA, U.S.A.); however, following LPS stimulation, microglia produced large amounts of IL-1. These procedures confirmed that the culturing technique produced relatively pure microglia that could be activated by methods consistent with those in the current literature.

Experimental manipulations

At 3 days following isolation and plating, medium was changed to contain varying concentrations (1 nM to 10 μM) of PGE2 or prostaglandin B2 (PGB2) for 1 h. Medium was then changed to contain the prostaglandin with 100 ng/ml LPS. All experimental medium was Sato’s with 1% FBS. Microglial cultures were also treated with medium containing no prostaglandins or LPS and with medium containing LPS alone.

Electrophysiological recordings

At 24 h after experimental manipulations, current traces were elicited by using standard whole-cell patch-clamp technique following methods similar to that of Lascola and Kraig (1996). Patch pipettes were pulled from thin-walled borosilicate glass (1.5 mm o.d., 0.86 mm i.d.; A-M Systems, Inc., Everett, WA, U.S.A.) on a micropipette puller (model P-87; Sutter Instrument Co., Novato, CA, U.S.A.). Pipettes were heat-polished to tip diameters of 1–2 μm.

Recording solutions were derived from Nörenberg et al. (1994). The bath solution contained the following (in mM): NaCl, 160; KCl, 4.5; HEPES, 5; MgCl2, 1; CaCl2, 2; dextrose, 11; adjusted to pH 7.35 with NaOH. The pipette solution contained the following (in mM): KCl, 150; CaCl2, 1 (free Ca2+, 0.01 μM); MgCl2, 2; EGTA, 11; HEPES, 10; adjusted to pH 7.35 with KOH. Both bath and pipette solutions were between 295 and 310 mOsm as measured by a vapor-pressure osmometer (model 550, Wescor, Logan, UT, U.S.A.). Bath solutions also contained 200 μM 4,4′-diisothi-ocyanatostilbene-2,2′-disulfonic acid (DIDS), to block Cl− conductances, as Cl − currents have been reported in microglia under some conditions (Visentin et al., 1995; Schlichter et al., 1996).

Tissue culture dishes (35 mm) were mounted in a movable open perfusion system (Medical Systems, Greenvale, NY, U.S.A.) and visualized by using an inverted Leitz microscope (Fluovert FU; Ernst Leitz, Germany) with phase-contrast optics. Bath solution was perfused at a rate of ~1 ml/min. After whole-cell access was achieved and gigaohm seals confirmed, currents were elicited using an Axopatch 200A integrating patch-clamp amplifier (Axon Instruments, Foster City, CA, U.S.A.). Membrane potential was stepped from −160 mV to +40 mV (20-mV increments). Each step was 75 ms, and cells were held at −80 mV before and between steps. Voltage steps were separated by 5 s to avoid voltage-dependent inactivation characteristic of the K + OR. The output was filtered at 2 kHz and then digitally sampled at 5.2 kHz by a Digidata 1200 A/D converter (Axon Instruments) connected to an IBM-compatible computer (AST Premmia 4/66d; AST Research, Irvine, CA, U.S.A.). pClamp software (version 6.3; Axon Instruments) was used to control voltage-clamp protocol, data acquisition, storage, and analysis.

To eliminate potential variability associated with different cell shapes, all cells recorded in each experimental group were relatively round, flat, and phase-dark and lacked easily visible organelles or inclusion bodies. In this manner, selection bias between experimental groups was minimized.

IL-1β measurements

Microglia were prepared in the same manner as described above, but were plated on 24-well tissue culture dishes. Three days after plating, cells were exposed to 100 ng/ml LPS, 1 mM 4-AP, LPS plus 1 μM PGE2, LPS plus 4-AP, or LPS plus PGE2 plus 4-AP, in medium with 1% FBS. After 24 h, medium was removed and processed for IL-1β by using an ELISA kit (Biosource International). All values (four per experimental group) were averaged and compared by Student’s t test to the group treated with LPS alone.

Cell death determination

Following removal of medium for IL-1β processing, cells were loaded with the Sytox Green Fluorescent Dead Cell Stain at 250 pM (S-7020; Molecular Probes), in a HEPES buffer (10 mM HEPES, 25 mM glucose, 137 mM NaCl, 5.3 mM KCl, 3 mM CaCl2, 1 mM MgCl2, pH 7.4). After 20 min, phase-contrast and fluorescent (excitation at 488 nm, dichroic cutoff at 510 nm) images (three per well) were acquired by using a TILL Photonics Polychrome II Monochrometer (Applied Scientific Instruments, Inc., Eugene, OR, U.S.A.), a Leica DMIRB/E (Leica Mikroskopie, Wetzlar, Germany), a PXL CCD camera (Photometries, Inc., Tucson, AZ, U.S.A.), and MetaFluor Imaging System software (Universal Imaging Corp., West Chester, PA, U.S.A.) with an IBM-compatible computer (AST Premmia 4/66d). Sytox-positive cells were counted and divided by the total number of cells present (phase image) to give the percentage of cell death.

OX-42 immunocytochemistry and densitometry

Some cells were treated with LPS and 1 or 10 μM PGE2 (or PGE2 alone) for 24 h, washed with phosphate-buffered saline, and incubated for 1 h with the OX-42 antibody. These plates were washed and incubated with a peroxidase-labeled goat anti-mouse antibody (Biosource International). The peroxidase antibody was visualized by incubation in phosphate-buffered saline containing 0.05% diaminobenzidine and 0.01% H2O2. Plates were cover-slipped with glycerol gelatin (Sigma). Images were acquired using a 510/40 nm bandpass filter with a CCD camera (CH250; Photometrics) using PMIS software (version 3.0; Photometrics). Images were background-corrected with a blank dish and coverslip. Areas of interest were drawn around each cell in the field by using Image Pro Plus (version 1.2; Media Cybernetics, Silver Spring, MD, U.S.A.), and the optical density of each cell was measured. For details on optical density analysis, see Caggiano and Kraig (1996).

Statistical procedures

Peak currents were determined by using Clampfit software (pClamp; Axon Instruments). Given the variable cell size of cultured microglia, currents were normalized for cell size by dividing currents by cell capacitance (pA/pF). Data reported are means ± SE. Group means were compared by using an unpaired Student’s t test.

DiI-Ac-LDL and calcein-AM images were acquired by fluorescent excitation at 550 nm and 500 nm with a CCD camera (CH250; Photometrics) using PMIS software (version 3.0; Photometrics Inc.). Images were background-corrected and colored with Image Pro Plus (version 1.2; Media Cybernetics). For the construction of current density versus voltage plots, data arrays from current traces were imported into Origin software (version 4.0; Microcal Software, Inc., Northampton, MA, U.S.A.). Data points representing currents at 20 ms into each voltage step were copied into Excel (version 2.0; Microsoft Corp.), corrected for cell size (capacitance), and copied into SigmaPlot software (versions 2.0 and 4.0; SPSS Inc., Chicago, IL, U.S.A.), where histograms and current density versus voltage plots were constructed. Representative current traces (see Figs. 3 and 5) were exported from Clampfit and imported into Corel Draw (version 5.0; Corel Corp., Salinas, CA, U.S.A.), where pipette capacitance transients were trimmed and figures were labeled and assembled.

FIG. 3.

FIG. 3

Open in a new tab

Representative current traces elicited during whole-cell patch-clamping from a normal (non-LPS-stimulated) microglial cell and a microglial cell exposed to 100 ng/ml LPS for 24 h. At the top of the figure is a schematic of the voltage steps applied to each cell. Cells were held at −80 mV and stepped in 20-mV increments between −160 mV and +40 mV at 5-s intervals. Current traces labeled “Normal” are from a representative cell not exposed to LPS and are dominated by an inwardly rectifying current. Current traces labeled “24 hrs LPS” are from a representative cell exposed to LPS (100 ng/ml) for 24 h. Notice the emergence of a large outwardly rectifying current. Amplitude and time scale are as shown.

FIG. 5.

FIG. 5

Open in a new tab

Representative current traces elicited during whole-cell patch-clamping from two cells treated with LPS plus 100 nM PGE2. At the top of the figure is a schematic of the voltage steps applied to each cell. Cells were held at −80 mV and stepped in 20-mV increments between −160 mV and +40 mV at 5-s intervals. Current traces are from two representative cells exposed to LPS (100 ng/ml) and 100 nM PGE2. Notice that the outward currents are reduced in comparison with those of cells treated with LPS alone (Fig. 3). Some cells express a lower magnitude outward current (top traces), whereas in other cells the outward current is completely prevented (bottom traces). Amplitude and time scale are as shown.

Materials

PGE2 and PGB2 were purchased from Cayman Chemical Co. (Ann Arbor, MI, U.S.A.). PGE2 and PGB2 were dissolved in pure ethanol. LPS from Salmonella enteritidis was purchased from Sigma.

RESULTS

Microglia cultures

The purity of microglial cultures shaken from mixed cell cultures between 10 and 16 days was assessed by labeling of the cells with a cytoplasmic marker calcein-AM, which fills all cell types, and DiI-Ac-LDL, a marker of microglia. Figure 1 is a composite digital photomicrograph showing DiI-Ac-LDL (red) and calcein-AM (green) fluorescence. Notice that all of the cells, except for one (indicated with the arrow), stained positive for both markers (yellow). Cultures that had large numbers (>5% by visual inspection) of other cell types were discarded. Similarly, microglia that displayed prominent organelles or intracellular inclusions before experimental manipulation were also discarded, because this is generally considered a sign of activation.

FIG. 1.

FIG. 1

Open in a new tab

Pseudocolored digital photomicrograph of microglial primary culture processed for DiI-Ac-LDL (red) and calcein-AM (green). DiI-Ac-LDL marks microglia, and calcein-AM fills all viable cells. Notice that all but one cell (arrow) are positive for both markers (yellow). The non-DiI-Ac-LDL cell appears to be an astrocyte based upon its morphology. Scale bar = 30 μm.

Microglia in all cultures displayed morphologic variability. To reduce variability in cell shape and to minimize any potential selection bias between experimental and control groups associated with morphologic differences, all cells chosen for recording were relatively round, flat, and phase-dark. Cell capacitance ranged from 20 to 99 pF (n = 175; mean ± SE = 54.94 ± 1.73 pF). As cell size varied widely, measured currents were normalized to current density (pA/pF) via division by cell capacitance. Membrane potentials measured upon achieving whole-cell access were distributed around two points [−80 mV (59%) and −35 mV (41%); Fig. 2A]. The initial membrane potential was not correlated (r2 = 1.6 × 10−7) with the presence or size of the K + OR (Fig. 2B). To confirm that these cells were one population, we used whole-cell patches to monitor membrane potential continuously. This showed that many cells spontaneously switched between these two potential ranges (Fig. 2C).

FIG. 2.

FIG. 2

Open in a new tab

Demonstration of the bimodal distribution of resting membrane potential of cultured microglia. A: Frequency histogram showing distribution of initial membrane potentials around −80 mV and −35 mV. Membrane potentials were measured after whole-cell access to each cell was achieved. B: Regression analysis of initial membrane potential versus the size of the outward K+ current measured from each microglia. Best fit line (r2 = 1.6 × 10−7) demonstrates that there is no significant relationship between initial membrane potential and peak K+ current. C: Voltage trace of a single microglia demonstrating the switching of the membrane potential between the hyperpolarized and depolarized stable states.

Whole-cell currents

The whole-cell currents of normal (non-LPS-stimulated) microglia (n = 30) elicited with voltage steps from −160 mV to +40 mV were dominated by a voltage and time-dependent inactivating inwardly rectifying current and a lack of an outward current, as previously described. A typical example of currents from a normal microglial cell is shown in Fig. 3 (“Normal”). The capacitance transients in this figure were trimmed. Occasionally, a small outward current was observed in nonstimulated cells, and on rare occasions (1/30) a large outward current was observed. The reversal of these currents occurred between −80 and −60 mV (Fig. 4). Figure 4 shows the current density/voltage relationship of normal, LPS-stimulated, and each LPS plus PGE2 group. Values were taken from each voltage step at 20 ms into each current trace. Note that the normal trace (solid circles) is inwardly rectifying, reverses between −80 and −60 mV, and displays only a small outward current even with voltage steps to +40 mV. These data are consistent with previous observations (Kettenmann et al., 1990; Nörenberg et al., 1992, 1994).

FIG. 4.

FIG. 4

Open in a new tab

Current density versus voltage relationships of currents elicited from normal, LPS-stimulated, and LPS-plus-PGE2-treated microglia. Currents in this figure have been normalized to current density size via division by cell capacitance (pA/pF). Symbols for each group are as indicated in the figure. Notice that the normal group (black circles) shows a relatively large inward current and little outward current. The group treated with LPS alone (light gray squares) has this inward current along with a large outward current that is elicited with voltage steps above −40 mV. Notice that groups treated with LPS plus PGE2 show an incremental reduction in the magnitude of this outward current. Values are means ± SE.

Following 24 h of LPS stimulation, microglia displayed a similar, but slightly larger, inward current compared with normal microglia (Fig. 3, bottom). In addition to the inward current, the LPS-stimulated cells also displayed a large outwardly rectifying activating and inactivating current that was not seen in normal microglia (Fig. 3, “24 hrs LPS”). The currents in the LPS-stimulated cells (Fig. 4, gray squares) showed inward rectification, reversed between −80 and −60 mV, but at voltage steps of −20 mV or higher, showed a large outward current as well (Figs. 3 and 4). These results also agree with previous observations (Nörenberg et al., 1992).

PGE2 + LPS stimulation

Cells treated for 1 h before LPS stimulation with PGE2 showed whole-cell currents similar in character to those of normal and LPS-stimulated microglia (Fig. 5). The activating and inactivating characteristics, as well as the reversal potentials, of the currents in the PGE2-treated cells were also similar to those of normal and LPS-stimulated cells (Figs. 4 and 5). As can be seen in Figs. 4 and 5, however, the magnitude (pA/pF) of the outward current in cells treated with LPS and PGE2 was reduced in comparison with that in cells treated with LPS alone. Microglia treated with LPS plus 1 or 10 μM PGE2 (Fig. 4, octagons and gray circles) were nearly identical to normal microglia (solid circles), showing inward rectification and little outward current. Cells treated with LPS plus 1 nM PGE2 were nearly identical to cells treated with LPS alone, whereas cells treated with 10 and 100 nM PGE2 (Fig. 4, triangles and diamonds) showed intermediate values.

To quantify and statistically compare these observations, peak inward and outward current densities (pA/pF) were calculated for each experimental group and compared with those values from cells treated with LPS alone. Figure 6 shows the means ± SE of peak outward (top) and inward (bottom) current densities from normal and stimulated microglia. The peak outward current density of cells treated with LPS and 10 nM PGE2 (n = 24) was reduced significantly (p < 0.005) from that of cells treated with LPS alone (n = 22). The peak outward current densities of cells treated with 0.1, 1.0, or 10 μM PGE2 (n = 18, 15, and 17, respectively) were reduced further and also were significantly different from that of cells treated with LPS alone (p < 0.0005). The average peak outward current density in cells treated with LPS and 1 or 10 μM PGE2 were indistinguishable from that of normal cells (Fig. 6). Cells treated with LPS plus 1 nM PGE2 (n = 14) were not significantly different (p = 0.95) from cells treated with LPS alone.

FIG. 6.

FIG. 6

Open in a new tab

Histograms of the average peak K+ current densities from normal and LPS-stimulated microglia treated with 0–10 μM PGE2. Top histogram shows the peak outward current densities. The large outward K+ current of LPS-stimulated cells is reduced significantly with 10 nM PGE2 and reduced to normal levels with 1 and 10 μM PGE2. Bottom histogram shows the peak inward current densities. The inward K+ current densities in cells treated with LPS and ≥ 10 nM PGE2 are reduced to levels observed in normal cells. Values are means ± SE. *p < 0.005; **p < 0.0005.

The percentage of cells in each experimental group expressing any outwardly rectifying current were as follows: LPS alone, 95%; LPS plus 1 nM PGE2, 100%; LPS plus 10 nM PGE2, 88%; LPS plus 100 nM PGE2, 67%; LPS plus 1 μM PGE2, 54%; LPS plus 10 μM PGE2, 38%; and normal, 26%. Comparing the current densities of only the cells expressing the K + OR from the groups producing intermediate values (10 and 100 nM PGE2), current densities were still reduced significantly from that of the group treated with LPS alone (p < 0.005 and 0.001, respectively).

The peak inward current density of cells treated with LPS was slightly larger, but not significantly different from that of normal cells (n = 30, p = 0.23; Fig. 6, bottom). The peak inward current densities of cells treated with LPS plus 0.1 or 1 μM PGE2 were reduced significantly from that of cells treated with LPS alone (p < 0.01). PGE2 at 10 nM and 10 μM showed small, but not significant, reductions, but the averages of these groups were about that of the normal group.

As a control for the experimental manipulations and the solvent in which the PGE2 was carried (100% ethanol), experiments were also conducted using 0.1, 1.0, and 10 μM PGB2, a prostaglandin with similar structure to PGE2 but lacking known biological activity. Both prostaglandins were dissolved in 100% ethanol at similar concentrations so that the final concentration of ethanol in the PGE2 and PGB2 manipulations was the same.

Microglia treated with LPS and PGB2 had currents nearly identical to microglia treated with LPS alone. The current density/voltage relationship was also similar between these groups (Fig. 7). These currents all reversed between −80 and −60 mV and showed both large inwardly and outwardly rectifying currents. Currents were quantified as with PGE2, comparing the peak inward and outward current densities. Figure 8 shows the means ± SE of the group peak inward and outward current densities. Microglia treated with 0.1, 1.0, and 10 μM PGB2 (n = 12, 14, and 15, respectively) showed no significant differences (p > 0.05) from microglia treated with LPS alone (Fig. 8).

FIG. 7.

FIG. 7

Open in a new tab

Current density versus voltage relationships of currents elicited from normal, LPS-stimulated, and LPS-plus-PGB2-treated microglia. Cells were treated with PGB2 as a control for the solvent used with PGE2 (100% ethanol) and the other experimental procedures. Currents in this figure have been normalized to current density via division by cell capacitance (pA/pF). Symbols for each group are as indicated in the figure. Notice that the normal group (black circles) shows a relatively large inward current and little outward current. The group treated with LPS alone (light gray squares) has this inward current along with a large outward current that is elicited with voltage steps above −40 mV. Groups treated with LPS plus PGB2 show only a small nonsignificant reduction in the magnitude of this outward current when compared with changes observed with PGE2. Values are means ± SE.

FIG. 8.

FIG. 8

Open in a new tab

Histograms of the average peak K+ current densities from normal and LPS-stimulated microglia treated with 0–10 μM PGB2. Top histogram shows the peak outward current densities. Notice the large outward K+ current is not reduced significantly with any concentration of PGB2. Bottom histogram shows the peak inward current densities. Inward K+ current density values in the groups treated with LPS and PGB2 are not significantly different from values in the group treated with LPS alone. Values are means ± SE.

OX-42 immunocytochemistry

Average optical densities of OX-42 staining in cells treated with LPS and 1 μM (n = 85) or 10 μM (n = 62) PGE2 were reduced significantly from that of cells treated with LPS alone (n = 79; p < 0.0001). Figure 9 is a histogram demonstrating optical densities from each group. Average optical densities of groups treated with PGE2 but no LPS were similar to those of the normal (0 LPS and 0 PGE2) group.

FIG. 9.

FIG. 9

Open in a new tab

Histogram showing average optical density of OX-42 staining in microglia treated with 0–10 μM PGE2, with and without LPS. As confirmation of the electrophysiological data presented, complement receptor expression was measured by the optical density of the OX-42 antibody immunocytochemical staining. OX-42 recognizes CR-3. Cells treated with LPS and 1 or 10 μM PGE2 show significantly reduced optical densities when compared with cells treated with LPS alone. Cells treated with PGE2 but no LPS show optical densities very similar to those of cells treated with no PGE2 and no LPS. Values are means ± SE. **p < 0.0001.

IL-1β production

Cells treated with normal medium or medium with 1 mM 4-AP produced low to nondetectable levels (12.7 and 20.1 pg/ml, respectively) of IL-1β 24 h after initial exposure (Fig. 10A). Cells treated with 100 ng/ml LPS produced 136.2 pg/ml IL-1β. Treatment of cells with LPS plus 1 μM PGE2, LPS plus 1 mM 4-AP, or LPS plus PGE2 plus 4-AP significantly reduced IL-1β production (54.8, 55.0, and 43.7 pg/ml; p < 0.001, 0.05, and 0.05, respectively) after 24 h (Fig. 10A).

FIG. 10.

FIG. 10

Open in a new tab

The effects of PGE2 and 4-AP on the production of IL-1β and cell death in cultured microglia. A: Histogram showing average IL-1β production from each experimental group after 24 h of stimulation. Both PGE2 and 4-AP, a blocker of the K+ OR, significantly reduced the IL-1β production following stimulation with 100 ng/ml LPS. Notice that treating microglia with LPS plus PGE2 plus 4-AP resulted in the same level of reduction as LPS plus either agent alone. Values are means ± SE. Each experimental group was compared with the LPS group by Student’s t test. *p < 0.05; **p < 0.005. B: Histogram demonstrating average percentage of dead cells from each experimental group after 24 h of stimulation. The percentage of dead cells was examined to determine if PGE2 current prevents the apoptotic-like cell death that occurs ~24 h after macrophage activation. Furthermore, it was used to determine if 4-AP would result in cell death not found with PGE2 due to the inability of the cell to maintain polarity during activation. LPS treatment resulted in 15.56% of cells dying after 24 h. PGE2 significantly reduced this number to 5.10%. Treatment of cells with 4-AP resulted in some cell death in both the unstimulated and LPS-stimulated groups. Values are means ± SE. Each group was compared with the LPS group by Student’s t test. *p < 0.05; **p < 0.01.

Cell death

The percentages of microglia dying in each experimental group after 24 h are displayed in Fig. 10B. After treatment with 100 ng/ml LPS, 15.56% of microglia were Sytox-positive after 24 h, compared with 4.25% of unstimulated cells. Microglia treated with LPS plus 4-AP showed 16.96% cell death, whereas cells treated with LPS plus PGE2 or LPS plus PGE2 plus 4-AP showed 5.10 and 9.13% cell death, respectively.

DISCUSSION

Our results demonstrate that PGE2 acts directly on microglia to reduce the K+ OR, and to a lesser extent the K+ IR, expressed in these cells after activation from LPS. This reduction of K+ OR was dose-dependent, with significant reductions at 10 nM and complete reduction at 1 μM. PGE2 was also able to reduce the production of IL-1β and expression of the CR-3 following LPS exposure, demonstrating that the activation of the microglia had definitely been reduced. Furthermore, the ability of 4-AP to block IL-1β production suggests that the reduced K+ OR may be part of the mechanism by which PGE2 prevents microglial activation. These results support earlier in vivo work that suggests that cyclooxygenase products (e.g., PGE2) prevent the activation of microglia from inflammatory stimuli (Caggiano and Kraig, 1996). When these results are considered with other published studies of eicosanoids and cultured microglia (Théry et al., 1994; Minghetti et al., 1997), it is clear that eicosanoids, specifically PGE2, act directly upon microglia, perhaps by reducing K+ OR, to modulate not only components of the activated state, but also the activation process. Given both the pro- and anti-inflammatory properties of PGE2, understanding how PGE2 affects one important component of brain inflammation (i.e., microglial activation) will further our understanding of how eicosanoids and anti-inflammatory drugs function in normal and pathologic brain tissue processes.

Microglia are activated by diverse stimuli, including nerve transection (Graeber et al., 1988a,b; Morioka and Streit, 1991), various neurological diseases (for review, see McGeer et al., 1993), and even a noninjurious stimulus, such as spreading depression (Gehrmann et al., 1993; Caggiano and Kraig, 1996), which may be associated with migraine headache (Moskowitz et al., 1993). In vitro stimuli, such as γ-interferon (Wong et al., 1984) and LPS (Hetier et al., 1988), also activate microglia causing them to produce IL-1 (Hetier et al., 1988) and tumor necrosis factor (Sawada et al., 1989), alter expression of surface molecules (Imamura et al., 1991), and enhance their NO production (Zielasek et al., 1992). Previous work in this laboratory suggests that modulation of eicosanoid production alters the extent of in vivo reactive microgliosis (Caggiano and Kraig, 1996). Other in vitro reports demonstrate that the PGE2 modulates microglial-mediated neurotoxicity (Théry et al., 1994), an observation that may be explained by the recent finding that PGE2 reduces NO production from reactive microglia (Minghetti et al., 1997). It remains unknown, however, if and how PGE2 prevents the activation of microglia.

Following LPS stimulation, microglia are transformed into a reactive state that is characterized electrophysiologically by the expression of the K+ OR (Nörenberg et al., 1992), which is lacking in most nonstimulated microglia (Kettenmann et al., 1990; Banati et al., 1991). The presence of this K+ OR is a robust indicator of the functional state of the cell (Fischer et al., 1995). It is possible that the expression of this outward current is necessary for the activation of microglia, as has been observed in astrocytes (MacFarlane and Sontheimer, 1997). It may also promote the survival of microglia in an activated state. The binding of ATP, IgG, and other molecules causes cellular depolarization (Young et al., 1983; Woodroofe et al., 1989). Activated microglia produce ATP and stimulate themselves in an autocrine loop (Ferrari et al., 1997). Therefore, microglia’s own activation will lead to cellular depolarization, requiring a mechanism to maintain cell polarity to prevent detrimental effects (Illes et al, 1996).

PGE2 (10 nM; Fig. 6) significantly reduced the expression of the K+ OR, whereas higher concentrations nearly or completely prevented K+ OR. Furthermore, PGE2 was able to reduce IL-1β production and CR-3 expression, demonstrating that, in addition to preventing K+ OR, the overall activation status of the cells was reduced. The K + IR was also reduced by 0.1 and 1.0 μM PGE2 (Fig. 6, bottom). Similar treatments with PGB2 produced only small, nonsignificant changes in these currents (Figs. 7 and 8). These small effects may be due to specific, but weaker, activity of this prostaglandin, or may be simply the result of nonspecific experimental and sampling fluctuations. In either case, the dramatic differences between the effects of PGE2 and PGB2 show that the PGE2 effects are not the result of experimental manipulation, such as the ethanol solvent.

The intermediate K+ OR values produced by 10 and 100 nM PGE2 appear to be the result of both a reduced magnitude of the current and fewer cells expressing the K+ OR. Comparing the current densities of only the cells expressing the K+ OR in the groups producing intermediate values (10 and 100 nM PGE2; see Resuits), current densities were still reduced significantly from those of the group treated with LPS alone (p < 0.005 and 0.001, respectively). Thus, we can determine that PGE2 acts to reduce both the magnitude of the K+ OR present in activated microglia and the proportion of cells expressing the K+ OR following LPS stimulation.

The ability of PGE2 to reduce the immune activity of microglia, as assessed by CR-3 expression and IL-1β production, confirms that, along with preventing expression of the K+ OR, the activation of the cells has also been reduced. As these observations are offered only in confirmation of the electrophysiological data and other reports (Théry et al., 1994; Minghetti et al., 1997), the full spectrum of PGE2 concentrations was not examined. However, these data taken together introduce the question of whether the K+ OR is required for microglial activation and/or is serving some other cellular purpose. To determine if the K+ OR is necessary for activation, LPS-induced IL-1β production was measured following exposure to 1 mM 4-AP. This 4-AP concentration blocks ~90% of the K+ OR (Nörenberg et al., 1994). Figure 10A shows that if microglia are exposed to LPS while 1 mM 4-AP is present, the quantity of IL-1β produced is reduced significantly from that in microglia exposed to LPS alone. The fact that treatment of cells with LPS plus PGE2 and 4-AP together produced little additional reduction in IL-1β production (Fig. 10A) further supports the notion that PGE2 may be acting through a mechanism mimicked by 4-AP. These data suggest that the K+ OR may be required to complete activation of microglia.

The K+ OR may also be serving to maintain cell polarity during activation. Blocking this current in an activated cell would lead to depolarization from self-or non-self-stimulation and eventually to cell death. Even in the nonstimulated condition, one would expect to find a certain percentage of somewhat activated microglia, a point confirmed by the fact that 26% of the nonstimulated microglia displayed some K+ OR; therefore, blocking the K+ OR should lead to a small degree of enhanced cell death. Indeed, 4-AP induced a small amount of cell death in the nonstimulated group (Fig. 10B). Macrophages, B-cells, and other cells undergo apoptotic cell death following activation (Bingisser et al., 1996; Yokochi et al., 1996; Williams et al., 1997). Microglia likely also die by similar mechanisms ~24 h after activation (Fig. 10B). As PGE2 is reducing the general activation status of microglia (reduced K+ OR, IL-1β, and CR-3), it should also reduce the number of cells dying following exposure to LPS (Fig. 10B). We believe that even though 4-AP reduced microglial activation (IL-1β production), it prevented those cells that became activated from maintaining their cellular polarity and therefore resulted in cell death due to uncompensated depolarization. This was observed when microglia were treated with LPS plus 4-AP (Fig. 10B).

The membrane potential of a microglial cell will affect how the cell reacts to stimuli that produce membrane currents. The microglia used in this investigation had bimodally distributed membrane potentials, showing peaks around −80 mV and −35 mV (Fig. 2A). This has been observed before (Nörenberg et al., 1994) and is likely the reason that others have reported intermediate mean membrane potentials around −55 mV (Kettenmann et al., 1990; Stewart, 1994). We confirmed that the initial membrane potential was not correlated with the presence or size of the outward K+ current (Fig. 2B) and demonstrated that individual cells fluctuate between these two states (Fig. 2C). Thus, we can be reasonably sure that the membrane potentials recorded immediately upon achieving a whole-cell seal with the microglia do not predict whether these cells are resting or activated and do not confound the present findings. The function or physiologic consequences of this membrane potential fluctuation are not fully understood, but are discussed in some detail by Nörenberg et al. (1994).

The PGE2 concentrations at which positive responses were observed here are similar to those reported elsewhere (Cazevieille et al., 1993; Akaike et al., 1994; Minghetti et al., 1997). Furthermore, the dose-dependent manner in which the K+ OR current density was reduced suggests that microglia may not be activated in an “all or none” fashion, but rather in a graded fashion perhaps reflective of a balance of factors promoting and preventing their activation. The many factors that promote microglial activation have been explored in some depth (for review, see Zielasek et al., 1992); however, knowledge of the cellular, chemical, and electrical influences that reduce or prevent their activation is limited (Théry et al., 1994; Caggiano and Kraig, 1996; Minghetti et al., 1997). Given the present results, it appears that PGE2 can reduce the general activation of microglia perhaps by reducing the expression or activity of K+ OR.

In summary, we have shown that PGE2 reduces the proportion of cells expressing the K+ OR, the magnitude of the K+ OR current density, IL-1β production, and CR-3 expression following LPS stimulation. Blocking the K+ OR with 4-AP can prevent at least some components of activation (IL-1β) and also results in cell death with activation. These results demonstrate that PGE2 reduces the activation of microglia from inflammatory stimuli, such as LPS, by preventing K+ OR expression or activity. This has wide-ranging implications regarding how the generally proinflammatory eicosanoids may affect brain function and disease, serving as inflammatory molecules on target brain tissue while serving to regulate the activity of the immune effector cell of the brain, the microglia.

Acknowledgments

This work was supported by grants from the National Institute of Neurological Disorders and Stroke (NS-19108), a Zenith Award from the Alzheimer’s Association, and the Brain Research Foundation of the University of Chicago. A. O. Caggiano was supported in part by a Medical Scientist National Research Service Award (T32-GM-07281) from the National Institute of General Medical Health. We thank Dr. P. Kunkler for helpful discussions and readings of the manuscript and Dr. C. D. Lascola for technical assistance. Marcia P. Kraig assisted with animal care and cell cultures, and Raymond Hulse assisted with electronic images. We also thank Dr. D. Nelson for the use of equipment in her laboratory and Dr. Dana Giulian for advice on producing microglial cultures.

Abbreviations used

4-AP

4-aminopyridine

CR-3

complement type 3 receptor

DiI-Ac-LDL

1,1′-dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine perchlorate–low-density lipoprotein

FBS

fetal bovine serum

IL

interleukin

K+ IR

inwardly rectifying K+ current

K+ OR

outwardly rectifying K+ current

LPS

lipopolysaccharide

NO

nitric oxide

PGB2

prostaglandin B2

PGE2

prostaglandin E2

References

  1. Akaike A, Kaneko S, Tamura Y, Nakata N, Shiomi H, Ushikubi F, Narumiya S. Prostaglandin E2 protects cultured cortical neurons against N-methyl-D-aspartate receptor-mediated glutamate cytotoxicity. Brain Res. 1994;663:237–243. doi: 10.1016/0006-8993(94)91268-8. [DOI] [PubMed] [Google Scholar]
  2. Aktan S, Aykut C, Yegen BC, Okar I, Ozkutlu U, Ercan S. The effect of nordihydroguaiaretic acid on leukotriene C4 and prostaglandin E2 production following different reperfusion periods in rat brain after forebrain ischemia correlated with morphological changes. Prostaglandins Leukot Essent Fatty Acids. 1993;49:633–641. doi: 10.1016/0952-3278(93)90171-r. [DOI] [PubMed] [Google Scholar]
  3. Banati RB, Hoppe D, Gottmann K, Kreutzberg GW, Kettenmann H. A subpopulation of bone marrow-derived macrophage-like cells shares a unique ion channel pattern with microglia. J Neurosci Res. 1991;30:593–600. doi: 10.1002/jnr.490300402. [DOI] [PubMed] [Google Scholar]
  4. Bingisser R, Stey C, Weller M, Groscurth P, Russi E, Frei K. Apoptosis in human alveolar macrophages is induced by endotoxin and is modulated by cytokines. Am J Respir Cell Mol Biol. 1996;15:64–70. doi: 10.1165/ajrcmb.15.1.8679223. [DOI] [PubMed] [Google Scholar]
  5. Bonney RJ, Davies P. Possible autoregulatory functions of the secretory products of mononuclear phagocytes. Contemp Top Immunobiol. 1984;13:199–223. doi: 10.1007/978-1-4757-1445-6_10. [DOI] [PubMed] [Google Scholar]
  6. Bottenstein JE, Sato GH. Growth of a rat neuroblastoma cell line in serum-free supplemented medium. Proc Natl Acad Sci USA. 1979;76:514–517. doi: 10.1073/pnas.76.1.514. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Caggiano AO, Kraig RP. Eicosanoids and nitric oxide influence induction of reactive gliosis from spreading depression in microglia but not astrocytes. J Comp Neurol. 1996;369:93–108. doi: 10.1002/(SICI)1096-9861(19960520)369:1<93::AID-CNE7>3.0.CO;2-F. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Caggiano AO, Breder CD, Kraig RP. Long-term elevation of cyclooxygenase-2, but not lipoxygenase, in regions synaptically distant from spreading depression. J Comp Neurol. 1996;376:447–462. doi: 10.1002/(SICI)1096-9861(19961216)376:3<447::AID-CNE7>3.0.CO;2-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Cazevieille C, Muller A, Bonne C. Prostacyclin (PGI2) protects rat cortical neurons in culture against hypoxia/reoxygenation and glutamate-induced injury. Neurosci Lett. 1993;160:106–108. doi: 10.1016/0304-3940(93)90924-a. [DOI] [PubMed] [Google Scholar]
  10. Chiu SY, Wilson GF. The role of potassium channels in Schwann cell proliferation in Wallerian degeneration of explant rabbit sciatic nerves. J Physiol (Lond) 1989;408:199–222. doi: 10.1113/jphysiol.1989.sp017455. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Dewitt DS, Kong DL, Lyeth BG, Jenkins LW, Hayes RL, Wooten ED, Prough DS. Experimental traumatic brain injury elevates brain prostaglandin E2 and thromboxane B2 levels in rats. J Neurotrauma. 1988;5:303–313. doi: 10.1089/neu.1988.5.303. [DOI] [PubMed] [Google Scholar]
  12. Douglas WW. Pharmacologic Basis of Therapeutics. Macmillan; New York: 1975. p. 664. [Google Scholar]
  13. Ellis EF, Wright KF, Wei EP, Kontos HA. Cyclooxygenase products of arachidonic acid metabolism in cat cerebral cortex after experimental concussive brain injury. J Neurochem. 1981;37:892–896. doi: 10.1111/j.1471-4159.1981.tb04476.x. [DOI] [PubMed] [Google Scholar]
  14. Ellis EF, Police RJ, Rice LY, Grabeel M, Holt S. Increased plasma PGE2, 6-keto-PGF1 alpha, and 12-HETE levels following experimental concussive brain injury. J Neurotrauma. 1989;6:31–37. doi: 10.1089/neu.1989.6.31. [DOI] [PubMed] [Google Scholar]
  15. Ferrari D, Chiozzi P, Falzoni S, Hanau S, Di Virgilio F. Purinergic modulation of interleukin-1 beta release from microglial cells stimulated with bacterial endotoxin. J Exp Med. 1997;185:579–582. doi: 10.1084/jem.185.3.579. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Fischer HG, Eder C, Hadding U, Heinemann U. Cytokine-dependent K+ channel profile of microglia at immunologically defined functional states. Neuroscience. 1995;64:183–191. doi: 10.1016/0306-4522(94)00398-o. [DOI] [PubMed] [Google Scholar]
  17. Funk CD. Molecular biology in the eicosanoid field. Prog Nucleic Acid Res. 1993;45:67–98. doi: 10.1016/s0079-6603(08)60867-3. [DOI] [PubMed] [Google Scholar]
  18. Furlow TW, Jr, Hallenbeck JM. Indomethacin prevents impaired perfusion of the dog’s brain after global ischemia. Stroke. 1978;9:591–594. doi: 10.1161/01.str.9.6.591. [DOI] [PubMed] [Google Scholar]
  19. Gallo V, Zhou JM, McBain CJ, Wright P, Knutson PL, Armstrong RC. Oligodendrocyte progenitor cell proliferation and lineage progression are regulated by glutamate receptor-mediated K+ channel block. J Neurosci. 1996;16:2659–2670. doi: 10.1523/JNEUROSCI.16-08-02659.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Gehrmann J, Mies G, Bonnekoh P, Banati R, Iijima T, Kreutzberg GW, Hossmann KA. Microglial reaction in the rat cerebral cortex induced by cortical spreading depression. Brain Pathol. 1993;3:11–17. doi: 10.1111/j.1750-3639.1993.tb00720.x. [DOI] [PubMed] [Google Scholar]
  21. Giulian D, Baker TJ. Characterization of ameboid microglia isolated from developing mammalian brain. J Neurosci. 1986;6:2163–2178. doi: 10.1523/JNEUROSCI.06-08-02163.1986. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Goetzl EJ, An S, Smith WL. Specificity of expression and effects of eicosanoid mediators in normal physiology and human diseases. FASEB J. 1995;9:1051–1058. doi: 10.1096/fasebj.9.11.7649404. [DOI] [PubMed] [Google Scholar]
  23. Graeber MB, Streit WJ, Kreutzberg GW. Axotomy of the rat facial nerve leads to increased CR3 complement receptor expression by activated microglial cells. J Neurosci Res. 1988a;21:18–24. doi: 10.1002/jnr.490210104. [DOI] [PubMed] [Google Scholar]
  24. Graeber MB, Tetzlaff W, Streit WJ, Kreutzberg GW. Microglial cells but not astrocytes undergo mitosis following rat facial nerve axotomy. Neurosci Lett. 1988b;85:317–321. doi: 10.1016/0304-3940(88)90585-x. [DOI] [PubMed] [Google Scholar]
  25. Hetier E, Ayala J, Denefle P, Bousseau A, Rouget P, Mallat M, Prochiantz A. Brain macrophages synthesize interleukin-1 and interleukin-1 mRNAs in vitro. J Neurosci Res. 1988;21:391–397. doi: 10.1002/jnr.490210230. [DOI] [PubMed] [Google Scholar]
  26. Illes P, Nörenberg W, Gebicke-Haerter PJ. Molecular mechanisms of microglial activation. B Voltage- and purino-ceptor-operated channels in microglia. Neurochem Int. 1996;29:13–24. doi: 10.1016/0197-0186(95)00133-6. [DOI] [PubMed] [Google Scholar]
  27. Imamura K, Suzumura A, Sawada M, Marunouchi T, Takahashi A. Functional and immunohistochemical studies of cultured rat microglia. Rinsho Shinkeigaku. 1991;31:127–134. [PubMed] [Google Scholar]
  28. Kettenmann H, Hoppe D, Gottmann K, Banati R, Kreutzberg G. Cultured microglial cells have a distinct pattern of membrane channels different from peritoneal macrophages. J Neurosci Res. 1990;26:278–287. doi: 10.1002/jnr.490260303. [DOI] [PubMed] [Google Scholar]
  29. Kiefer R, Lindholm D, Kreutzberg GW. Interleukin-6 and transforming growth factor-beta 1 mRNAs are induced in rat facial nucleus following motoneuron axotomy. Eur J Neurosci. 1993;5:775–781. doi: 10.1111/j.1460-9568.1993.tb00929.x. [DOI] [PubMed] [Google Scholar]
  30. Kunkel SL, Spengler M, May MA, Spengler R, Larrick J, Remick D. Prostaglandin E2 regulates macrophage-derived tumor necrosis factor gene expression. J Biol Chem. 1988;263:5380–5384. [PubMed] [Google Scholar]
  31. Lascola CD, Kraig RP. Whole-cell chloride currents in rat astrocytes accompany changes in cell morphology. J Neurosci. 1996;16:2532–2545. doi: 10.1523/JNEUROSCI.16-08-02532.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Levison SW, McCarthy KD. Characterization and partial purification of AIM: a plasma protein that induces rat cerebral type 2 astroglia from bipotential glial progenitors. J Neurochem. 1991;57:782–794. doi: 10.1111/j.1471-4159.1991.tb08220.x. [DOI] [PubMed] [Google Scholar]
  33. Luo Y, Vallano ML. Arachidonic acid, but not sodium nitroprusside, stimulates presynaptic protein kinase C and phosphorylation of GAP-43 in rat hippocampal slices and synaptosomes. J Neurochem. 1995;64:1808–1818. doi: 10.1046/j.1471-4159.1995.64041808.x. [DOI] [PubMed] [Google Scholar]
  34. MacFarlane SN, Sontheimer H. Electrophysiological changes that accompany reactive gliosis in vitro. J Neurosci. 1997;17:7316–7329. doi: 10.1523/JNEUROSCI.17-19-07316.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. McGeer PL, McGeer EG. The inflammatory response system of brain: implications for therapy of Alzheimer and other neurodegenerative diseases. Brain Res Brain Res Rev. 1995;21:195–218. doi: 10.1016/0165-0173(95)00011-9. [DOI] [PubMed] [Google Scholar]
  36. McGeer PL, Kawamata T, Walker DG, Akiyama H, Tooyama I, McGeer EG. Microglia in degenerative neurological disease. Glia. 1993;7:84–92. doi: 10.1002/glia.440070114. [DOI] [PubMed] [Google Scholar]
  37. Medana IM, Hunt NH, Chaudhri G. Tumor necrosis factor-alpha expression in the brain during fatal murine cerebral malaria: evidence for production by microglia and astrocytes. Am J Pathol. 1997;150:1473–1486. [PMC free article] [PubMed] [Google Scholar]
  38. Minghetti L, Nicolini A, Polazzi E, Creminon C, Maclouf J, Levi G. Inducible nitric oxide synthase expression in activated rat microglial cultures is downregulated by exogenous prostaglandin E2 and by cyclooxygenase inhibitors. Glia. 1997;19:152–160. [PubMed] [Google Scholar]
  39. Morioka T, Streit WJ. Expression of immunomolecules on microglial cells following neonatal sciatic nerve axotomy. J Neuroimmunol. 1991;35:21–30. doi: 10.1016/0165-5728(91)90158-4. [DOI] [PubMed] [Google Scholar]
  40. Moskowitz MA, Nozaki K, Kraig RP. Neocortical spreading depression provokes the expression of c-Fos protein-like immunoreactivity within trigeminal nucleus caudalis via trigeminovascular mechanisms. J Neurosci. 1993;13:1167–1177. doi: 10.1523/JNEUROSCI.13-03-01167.1993. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Needleman P, Turk J, Jakschik BA, Morrison AR, Lefkowith JB. Arachidonic acid metabolism. Annu Rev Biochem. 1986;55:69–102. doi: 10.1146/annurev.bi.55.070186.000441. [DOI] [PubMed] [Google Scholar]
  42. Nörenberg W, Gebicke-Haerter PJ, Illes P. Inflammatory stimuli induce a new K+ outward current in cultured rat microglia. Neurosci Lett. 1992;147:171–174. doi: 10.1016/0304-3940(92)90587-w. [DOI] [PubMed] [Google Scholar]
  43. Nörenberg W, Gebicke-Haerter PJ, Illes P. Voltage-dependent potassium channels in activated rat microglia. J Physiol (Lond) 1994;475:15–32. doi: 10.1113/jphysiol.1994.sp020046. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Paakkari I, Lindsberg P. Nitric oxide in the central nervous system. Ann Med. 1995;27:369–377. doi: 10.3109/07853899509002590. [DOI] [PubMed] [Google Scholar]
  45. Pappas CA, Ullrich N, Sontheimer H. Reduction of glial proliferation by K+ channel blockers is mediated by changes in pHi. Neuroreport. 1994;6:193–196. doi: 10.1097/00001756-199412300-00049. [DOI] [PubMed] [Google Scholar]
  46. Piomelli D. Arachidonic acid in cell signaling. Curr Opin Cell Biol. 1993;5:274–280. doi: 10.1016/0955-0674(93)90116-8. [DOI] [PubMed] [Google Scholar]
  47. Puro DG, Roberge F, Chan CC. Retinal glial cell proliferation and ion channels: a possible link. Invest Ophthalmol Vis Sci. 1989;30:521–529. [PubMed] [Google Scholar]
  48. Sawada M, Kondo N, Suzumura A, Marunouchi T. Production of tumor necrosis factor-alpha by microglia and astrocytes in culture. Brain Res. 1989;491:394–397. doi: 10.1016/0006-8993(89)90078-4. [DOI] [PubMed] [Google Scholar]
  49. Schlichter LC, Sakellaropoulos G, Ballyk B, Pennefather PS, Phipps DJ. Properties of K+ and Cl− channels and their involvement in proliferation of rat microglial cells. Glia. 1996;17:225–236. doi: 10.1002/(SICI)1098-1136(199607)17:3<225::AID-GLIA5>3.0.CO;2-#. [DOI] [PubMed] [Google Scholar]
  50. Schluter D, Kaefer N, Hof H, Wiestler OD, Deckert-Schluter M. Expression pattern and cellular origin of cytokines in the normal and Toxoplasma gondii-infected murine brain. Am J Pathol. 1997;150:1021–1035. [PMC free article] [PubMed] [Google Scholar]
  51. Shohami E, Shapira Y, Sidi A, Cotev S. Head injury induces increased prostaglandin synthesis in rat brain. J Cereb Blood Flow Metab. 1987;7:58–63. doi: 10.1038/jcbfm.1987.8. [DOI] [PubMed] [Google Scholar]
  52. Stewart RR. Membrane properties of microglial cells isolated from the leech central nervous system. Proc R Soc Lond [Biol.] 1994;255:201–208. doi: 10.1098/rspb.1994.0029. [DOI] [PubMed] [Google Scholar]
  53. Théry C, Dobbertin A, Mallat M. Downregulation of in vitro neurotoxicity of brain macrophages by prostaglandin E2 and a β-adrenergic agonist. Glia. 1994;11:383–386. doi: 10.1002/glia.440110411. [DOI] [PubMed] [Google Scholar]
  54. Van Dam AM, Brouns M, Man-A-Hing W, Berkenbosch F. Immunocytochemical detection of prostaglandin E2 in microvasculature and in neurons of rat brain after administration of bacterial endotoxin. Brain Res. 1993;613:331–336. doi: 10.1016/0006-8993(93)90922-a. [DOI] [PubMed] [Google Scholar]
  55. Visentin S, Agresti C, Patrizio M, Levi G. Ion channels in rat microglia and their different sensitivity to lipopolysaccharide and interferon-gamma. J Neurosci Res. 1995;42:439–451. doi: 10.1002/jnr.490420402. [DOI] [PubMed] [Google Scholar]
  56. Williams TE, Ayala A, Chaudry IH. Inducible macrophage apoptosis following sepsis is mediated by cysteine protease activation and nitric oxide release. J Surg Res. 1997;70:113–118. doi: 10.1006/jsre.1997.5117. [DOI] [PubMed] [Google Scholar]
  57. Wong GH, Bartlett PF, Clark-Lewis I, Battye F, Schrader JW. Inducible expression of H-2 and Ia antigens on brain cells. Nature. 1984;310:688–691. doi: 10.1038/310688a0. [DOI] [PubMed] [Google Scholar]
  58. Woodroofe MN, Hayes GM, Cuzner ML. Fc receptor density, MHC antigen expression and superoxide production are increased in interferon-gamma-treated microglia isolated from adult rat brain. Immunology. 1989;68:421–426. [PMC free article] [PubMed] [Google Scholar]
  59. Yamagata K, Andreasson KI, Kaufmann WE, Bames CA, Worley PF. Expression of a mitogen-inducible cyclooxygenase in brain neurons: regulation by synaptic activity and glucocorticoids. Neuron. 1993;11:371–386. doi: 10.1016/0896-6273(93)90192-t. [DOI] [PubMed] [Google Scholar]
  60. Yokochi T, Kato Y, Sugiyama T, Koide N, Morikawa A, Jiang GZ, Kawai M, Yoshida T, Fukada M, Takahashi K. Lipopolysaccharide induces apoptotic cell death of B memory cells and regulates B cell memory in antigen-nonspecific manner. FEMS Immunol Med Microbiol. 1996;15:1–8. doi: 10.1111/j.1574-695X.1996.tb00351.x. [DOI] [PubMed] [Google Scholar]
  61. Young JD, Unkeless JC, Young TM, Mauro A, Cohn ZA. Role for mouse macrophage IgG Fc receptor as ligand-dependent ion channel. Nature. 1983;306:186–189. doi: 10.1038/306186a0. [DOI] [PubMed] [Google Scholar]
  62. Zielasek J, Tausch M, Toyka KV, Hartung HP. Production of nitrite by neonatal rat microglial cells/brain macrophages. Cell Immunol. 1992;141:111–120. doi: 10.1016/0008-8749(92)90131-8. [DOI] [PubMed] [Google Scholar]

RESOURCES