Microglial inhibition of neuroprotection by antagonists of the EP1 prostaglandin E2 receptor
© Carlson et al; licensee BioMed Central Ltd. 2009
Received: 08 December 2008
Accepted: 17 February 2009
Published: 17 February 2009
The EP1 receptor for the prostanoid PGE2 is a G-protein coupled receptor that has been shown to contribute to excitotoxic neuronal death. In this study we examined the influence of non-neuronal cells on neuroprotective properties of EP1 receptor antagonists (Ono 8711 and SC 51089).
Primary neuronal cultures systems with or without non-neuronal cells were used to examine how the neuroprotective properties of EP1 antagonists were influenced by non-neuronal cells. The influence of astrocytes or microglia were individually tested in excitotoxicity assays using a co-culture system with these cells grown on permeable transwell inserts above the neuronal-enriched cultures. The influence of microglia on PGE2 synthesis and EP1 receptor expression was examined.
EP1 antagonists were neuroprotective in neuronal-enriched cultures (> 90% neurons) but not in mixed cultures (30% neurons plus other non-neuronal cells). Co-cultures of microglia on permeable transwell inserts above neuronal-enriched cultures blocked neuroprotection by EP1 antagonists. Incubation of microglia with neuronal-enriched cultures for 48 hours prior to NMDA challenge was sufficient to block neuroprotection by EP1 antagonists. The loss of neuroprotection by EP1 antagonists was accompanied by a decrease of neuronal EP1 expression in the nucleus in cultures with microglia present.
These findings demonstrate microglial modulation of neuronal excitotoxicity through interaction with the EP1 receptor and may have important implications in vivo where microglia are associated with neuronal injury.
Cyclooxygenase-2 (COX-2), the enzyme that catalyzes the rate limiting step in the synthesis of prostanoids, contributes to neuronal death. Inhibitors of COX, termed non-steroidal anti-inflammatory drugs (NSAIDs) , can protect neurons following an assault with toxic stimuli that promote excitotoxic death; both in vitro [2, 3] and in vivo [4–7]. COX-2 knockout mice are also less susceptible to excitotoxicity following exposure to the glutamate receptor agonist N-methyl D-aspartate (NMDA) . Therefore, a loss of COX-2 activity either by inhibition of the enzyme or loss of expression is associated with increased neuronal viability. Conversely, increased COX-2 activity appears to augment neuronal death. The increased COX-2 expression in neurons observed in vivo in animal models of stroke , following stimulation with the glutamate receptor agonist kainic acid , and in vitro following NMDA stimulation [2, 3] is coincident with loss of neurons. Constitutive expression of COX-2 in neurons at high amounts in transgenic mice results in a greater loss of neurons in stroke models  and age-associated loss of neurons . In addition, constitutive COX-2 expression renders neurons more susceptible to NMDA-stimulated death .
There are two COX genes, COX-1 and COX-2 . COX catalyzes the initial steps in the conversion of arachidonic acid (AA) to one of the five prostanoids, prostacyclin (PGI2), thromboxane (TxA2), prostaglandin D2 (PGD2), prostaglandin F2α (PGF2α) and prostaglandin E2 (PGE2) [1, 12]. In addition to the generation of prostanoids, reactive oxygen species (ROS) are also generated by COX-2 in the reaction of prostanoids . It was demonstrated that the COX-2-generated prostanoids (and not ROS), are the major contributors by COX-2 towards excitotoxicity following administration of NMDA to animals .
Each of the prostanoids synthesized by COX activates at least one specific prostanoid receptor. These receptors are coupled to G-proteins and are designated IP (for PGI2), TP (for TXA2), DP1 or DP2 (for PGD2), FP (for PGF2α) and EP1-4 (for PGE2) . Recent investigations have focused on understanding how activation of specific prostanoids affects neuronal viability. In our earlier studies we identified that PGF2α and PGE2 were made in primary neuronal cortical cultures in response to stimulation with NMDA [3, 14]. An analog of PGE2, 17-phenol trinor PGE2 (17-pt-PGE2), but not PGF2α, could reverse the neuroprotective effect of a COX-2-specific inhibitor in vitro  and in vivo  following NMDA administration. These studies indicate that PGE2 production by COX-2 can contribute to the deleterious actions of COX-2 in NMDA-mediated excitotoxicity of neurons. However, in vitro studies investigating the role of PGE2 and its analogs have yielded contradictory results. PGE2 or its analogs have been reported to both increase neuronal survival following NMDA stimulation [15–19] and in some cases be neurotoxic [20, 21]. These opposing effects or PGE2 on neuronal viability are due to activation of specific EP receptors that exert either pro survival or pro death effects. In general, activation of EP1 contributes to neuronal death [21–24], while activation of EP2 [17–19] and EP4  promote neuroprotection. EP1 has been shown to contribute to NMDA-mediated neuronal death in vivo . Decreased EP1 activation by a pharmacologic antagonist or genetic knockout of the EP1 receptor decreased NMDA-stimulated neuronal death, whereas a specific EP1 receptor agonist augmented death [22–24].
Significant progress has been made in understanding how prostanoids contribute to neuronal death . EP1 receptor activation in neurons has been linked to two different intracellular mechanisms tied to excitotoxic cell death. EP1 receptor activation was initially shown to impair the Na+-Ca2+ exchanger (NCX) which subsequently causes greater increases in intracellular calcium leading to neuronal death . More recently, EP1 receptor activation has been linked to the AKT signaling pathway that affects neuronal viability .
However, the interaction of EP1 with other cell types in the central nervous system (CNS) is not well understood. In this study, we examined whether inhibition of neuronal EP1 contributes to neuronal viability in primary cultures with differing compositions of non neuronal CNS cells. We investigated the neuroprotective properties of two specific EP1 receptor antagonists, Ono 8711  and SC51089  along with a COX-2 inhibitor NS398. This study demonstrates that the presence of microglia in culture can offset the neuroprotective effects of EP1 antagonists. Our findings have important implications to conditions where microglia may accumulate following neuronal injury and potentially limit neuroprotection conferred by EP1 antagonists.
Reagents and antibodies
Ono 8711 was generously provided by Takayuki Maruyama Ph.D., Ono Pharmaceutical Company, (Osaka, Japan). N-[2-(cyclohexyloxy)-4-nitrophenyl]-methanesulfonamide (NS398) was obtained from Cayman Chemical Company, (Ann Arbor MI). SC51089 was obtained from Biomol, (Plymouth Meeting, PA). NMDA, MK801 and Arabinose cytosine (AraC) were obtained from Sigma Chemical Company (St. Louis, MO). Tissue culture media minimum essential media (MEM) was obtained from Sigma Chemical Company (St. Louis, MO), glutamine and trypsin were from Gibco Life Sciences, Invitrogen (Carlsbad, CA). Serum (defined fetal bovine serum and defined horse serum) was purchased from Hyclone (Logan, UT). Primary antibodies used in this study were, mouse monoclonal anti microtubule associated protein (MAP)-2 (BD Pharmingen, San Diego, CA), rabbit polyclonal EP1 (Cayman Chemical Company), rat monoclonal CD11b (Abcam, Cambridge, MA), rat monoclonal anti glial fibrillary acidic protein (GFAP)(Calbiochem, San Diego, CA) and rat monoclonal anti mouse F4/80 was produced by our laboratory. Secondary antibodies used included FITC-coupled goat anti-mouse, Cy5-coupled anti-rat and Cy5-coupled goat anti-rabbit. Secondary antibodies were obtained from Jackson Immuno-Research Laboratories Inc. (West Grove, PA). Propidium iodide (PI) was obtained from Molecular Probes (Eugene, OR).
Primary cortical cultures
Cortical cultures were prepared essentially as described previously [3, 29–31]. Cortices from E15 embryos (mouse strain CD1, from Charles River Laboratories, Wilmington MA) were removed and the cells were dissociated by digestion with trypsin (0.025%) for 30 min at 37°C, followed by physical dissociation resulting from pipeting the cells (15–20 times) through a 5-ml pipette placed in a 15-ml conical tube with a loose seal between the bottom of the tube and the pipette. The cells were then plated at a density of 1.1 × 106 trypan blue-excluding cells per well in a 12-well plate (Corning dishes pre coated with poly-D-lysine). Cells were initially plated in 1-ml of MEM with Earle's salts supplemented with 5% horse serum (heat inactivated), 5% fetal bovine serum (heat inactivated), 30 mM glucose, 2 mM glutamine and grown at 37°C in humidified chambers with 5% CO2. The cultures were fed the next day by replenishing about half the volume with fresh growth media (MEM with Earle's salts supplemented with 10% horse serum, 30 mM glucose, 2 mM glutamine) and fed every 2–3 days after that to generate mixed cortical cultures. After nine days in culture, these mixed cultures were exposed to AraC (4–8 μM) for a 24 hour period to limit the growth of mitotic cells. Each preparation of AraC differed with respect to its anti-mitotic efficacy and was tested for optimal inhibition of cell growth with minimal toxicity. These mixed culture preparations contained approximately 30% of the cells as neurons . All excitotoxicity experiments with mixed cultures were started 2 days after the last feeding and were used 15 days after plating. At this stage, the neurons exhibited the most consistent response to NMDA while still having low background death. Neuronal-enriched cultures were prepared as above except that, on the second day after plating, the cells were treated with 2–4 μM AraC for a 24-hour period and fed only once more with feeding media (MEM with Earle's salts supplemented with 10% horse serum, 30 mM glucose, 2 mM glutamine). Neuronal-enriched cultures which were incubated for 8 days in vitro (DIV), contained between 90–95% neurons as determined by immunofluorescent microscopy as the number of PI stained nuclei that were associated with cells labeled with the neuronal-specific marker MAP-2 (data not shown). Excitotoxicity experiments with these cultures were started after the cultures were maintained for 8 DIV, a time at which there is low background neuronal death and the neurons exhibit a consistent response to NMDA stimulated excitotoxicity. Excitotoxicity experiments were performed in growth media that contained 10% horse serum unless stated otherwise.
Cortical cultures were also examined in serum-free media. Cortical cultures were prepared as described and allowed to incubate in plating media (MEM with Earle's salts supplemented with 5% horse serum, 5% fetal bovine serum, 30 mM glucose, 2 mM glutamine) for 24 hours. The media was then replaced with neurobasal media containing 2% B27 supplement (Invitrogen) and, 24 hours later, the cultures were then exposed to AraC (4–8 μM). The cultures were fed 24 hours after this by replacing one half of the media with fresh Neurobasal media with 2% B27 supplement. These cultures were used in excitotoxicity experiments 8 days after the initial plating.
The cell counting method for determining viable cells can be used when cultures are grown in the presence of serum (as in most of our current experiments) or when testing agents that may have antioxidant properties which interfere with assays that measure release of the enzyme lactate dehydrogenase (LDH) into the media from dead cells. For each experiment, the concentration of NMDA required to give about 80% neuronal death was pre-determined and varied between different lots of NMDA and cell preparations. Typical concentrations of NMDA required for 80% neuronal death ranged between 10–40 μM for neuronal-enriched cultures and 15–80 μM for mixed cultures. All experiments were conducted using this high amount of NMDA-induced neuronal death for convenience and reliability in scoring the survival after NMDA treatment. For neuroprotective experiments, the neuroprotective agent (NS398, Ono 8711 or SC51089) was added directly to the growth media coincident with the addition of NMDA. In all cases, the effect of each neuroprotectant treatment (or combination of treatments) on neuronal viability in the absence of NMDA was also determined by photographing the cells before and 20 hours after treatment with the agent indicated. In each experiment, 500 to 2000 neurons were counted for each treatment.
Microglial and astrocyte cultures
Microglial cultures were prepared according to the methods described . Brains were removed from 1 day old CD1 mice and the cerebral cortices were obtained. The meninges were removed and the tissue was minced and digested with trypsin (0.025%) for 30 min at 37°C, followed by physical dissociation as described earlier for neuronal cortical cultures. The cells were plated in MEM containing 10% fetal bovine serum into 25 cm2 tissue culture flasks (1 brain per flask) that were pretreated with poly-d-lysine. Cultures were grown for 7–9 days at 37°C and the microglial cells were removed by shaking on an orbital shaker at 250 rpm for 30 min. Microglial cells were then centrifuged for 5 min at 1500 g and resuspended in neurobasal medium with 2% B27 supplement and plated onto either poly-D-lysine coated12-well plates or transwells (12 mm diameter) at a density ranging from 1.5 × 105 to 3 × 105 cells/mm2. Lower cell density microglial cultures were more difficult to grow. After 8 days in culture, microglial transwells were transferred onto 5 day old neuronal-enriched cultures (in 12 well plates) and co-cultured another 48 hours prior to the start of excitotoxicity experiments. Microglial cultures contained greater than 95% microglial cells as determined by staining with microglial markers, F4/80 or CD11b.
Astrocyte cultures were prepared as described . Cells were isolated from cerebral cortices from 1 day old mice as described above and the cells were suspended in MEM containing 10% fetal bovine serum and plated onto either transwells or tissue cultures plates (1 × 105 cells/mm). These cultures were grown at 37°C for about 1–2 weeks until the cells reached confluency. Transwell astrocyte cultures were transferred to neuronal-enriched cultures (5 day old cultures), 48 hours before the start of excitotoxicity experiments.
Conditioned media experiments
Media was harvested from mixed cultures (15 DIV) and neuronal-enriched cultures (8 DIV) and centrifuged for 5 minutes (1000 × g) to eliminate any residual cells. Freshly harvested media was subsequently exchanged between the mixed and neuronal-enriched culture preparations. Since the neuronal-enriched cultures (8 DIV) did not survive a direct media exchange, the mixed conditioned media was exchanged in the 12-well plates by seven two-fold dilutions. Neuronal-enriched media was directly exchanged on the mixed cultures (15 DIV) in one exchange. Excitotoxicity experiments were performed as described above.
Mixed cortical cultures, microglia or astrocytes were plated onto the top side of the transwell inserts (0.4 μm pore size polycarbonate membrane coated with poly-D-lysine, Corning catalog number 3401, Corning, NY) at the cell density as described above. The mixed cultures on transwells were treated with AraC (on day 9) and fed as described for mixed cultures. These mixed culture transwells (15 DIV) were transferred to neuronal-enriched cultures 48 hours prior to treatment with NMDA. Neuronal-depleted mixed cultures were obtained by exposure of the mixed cultures (11 DIV) to 25 μM NMDA for 48 hours prior to transfer of the transwells to plates containing fresh media and three exchanges of media to remove NMDA. The transwell cultures were positioned approximately 2 mm above the neuronal-enriched cultures and the cells grown on the transwells were separated from the neuronal-enriched cultures by the permeable transwell membrane. The neuronal-enriched cultures were photographed prior to addition of the transwells and treatment with NMDA. The mixed culture transwells were then returned to the same wells of the neuronal-enriched cultures and subsequently treated with vehicle, SC51089, and then NMDA at a concentration that yielded approximately 20% neuronal survival. All drugs were added to the media below the transwells. After 24 hours, the mixed culture transwells were removed and the neuronal-enriched cultures were stained with trypan blue, photographed and scored for neuronal survival as described above. Viability of the mixed cultures, microglia cultures and astrocyte coultures on the transwells prior to NMDA treatment was verified using dead cell stain ethidium homodimer and the live cell stain Calcein AM green.
The data from the NMDA excitotoxicity experiments were analyzed using InStat software (GraphPad software, San Diego, CA). Data from each treatment group passed the method of Kolmogorov and Smirnov for normal distributions making it appropriate for parametric tests to assess significance between treatment groups. One way ANOVA with post-test using the Tukey-Kramer multiple comparison tests was performed as to determine significance between treatment groups at the confidence intervals indicated. Pair-wise comparisons were performed using unpaired students t-test.
Cortical cultures were prepared for microscopy by first rinsing the cultures with phosphate-buffered saline (PBS), followed by fixing the cells with freshly prepared 2% paraformaldehyde in PBS. Following five rinses with PBS (minimum of 2 minutes per rinse), cells were permeabilized with 0.2% triton-X 100 in PBS, rinsed (3 rinses with PBS, minimum of 5 minutes per rinse) and blocked with Image-it FX signal enhancer solution (Molecular Probes). Primary antibodies were added at dilutions of (1:500) for rabbit polyclonal antibody for EP1 (final concentration 1 μg/ml), (1:400) mouse monoclonal antibody for MAP-2 (final concentration 1 μg/ml), (1:500) rat monoclonal GFAP (final concentration of 1 μg/ml), (1:100) rat polyclonal for F4/80 (final concentration 10 μg/ml) and (1:100) rat monoclonal antibody for CD11b (final concentration 5 μg/ml). Incubation of the anti-EP1 antibodies with the EP1 epitope blocking peptide (10:1 molar ratio of peptide: antibody) (Cayman Chemical Company) blocked binding (data not shown). Secondary antibodies were diluted (1:400) for FITC-conjugated anti-mouse (IgG, H&L) (final concentration 3 μg/ml), (1:400) FITC-conjugated anti-rat (final concentration 3 μg/ml), and (1:800) Cy5-conjugated anti-rabbit (IgG, H&L) (final concentration 1.5 μg/ml). The fluorochrome-conjugated antibodies were cross-adsorbed to immunoglobulins from other species to ensure specific multiple labeling. All antibodies were diluted in PBS with 0.025% Tween 20. Combinations of the selective primary antibodies were added to the sections and incubated overnight in a humidified chamber at 4°C. Conjugated secondary antibodies were added for 1 hour at room temperature (RT) followed by staining with PI to visualize cell nuclei. PBS washes were done (3 rinses with PBS, minimum of 5 minutes per rinse) between measures prior to mounting. Sections were mounted with ProLong Gold anti-fade permanent mounting media (Molecular Probes). To account for any background non-specific staining, negative controls with normal mouse and normal rabbit serum at concentrations comparable to primary antibodies were conducted in parallel.
Confocal laser scanning microscopy
Personal Confocal Microscope PCM-2000 (NIKON, Melville, NY) with Argon, HeNe green and HeNe red lasers were used to acquire the images. Simple Personal Confocal Image program (Simple PCI, Compix, Cranberry Township, PA) was used to acquire digital images and construct compilations. The fluorochromes were resolved into three different image channels with respective emission filters. The FITC label was detected with the Argon laser at 488 nm, Cy5 with the red HeNe laser at 633 nm and PI was visualized with the green HeNe laser at 605/32 nm narrow band pass width filter. The stained cultures were individually scanned for each respective fluorochrome and the multi-focal program (z-focus) was used to create a stereopsis image. The three separate images were merged together to acquire the final triple-colored image; the PI image was converted to blue color during merge. This allowed for comprehensive visualization of labeling throughout the entire depth of the sample. Images were stored as digital files and the final figures were created using Photoshop (Adobe, San Jose, CA). The total magnification of each image presented is equal to (magnification of the objective) × (width of the image panel in cm) × (0.93). This scale of image size was based on size calibration using an American Optical improved Neubauer hemacytometer and simple PCI linear calibration scales.
Quantification of neuronal nuclear EP1 expression in different culture preparations
The amount of neuronal expression in the nucleus was quantified from digital images of immunofluorescence confocal microscopy of cortical cultures stained with PI (for nuclei) and immuno-stained for EP1, and MAP-2. Each image was captured with the same sensitivity settings on the confocal microscope and all images were captured with a 60× objective with 640 × 480 pixels. A blinded observer (unaware of the culture preparation) counted the number of pixels over the nuclear regions of the neurons that were stained for EP1 with an enlarged image with Adobe Photoshop. Each culture preparation was examined in four independent fields and the number of neurons analyzed varied from 50–70 for each culture preparation.
PGE2 was measured using a prostaglandin enzyme-linked immunosorbent assay (ELISA) detection kit (Cayman Chemical Company) that uses a PGE2-specific antibody. For both the mixed cultures (15 DIV) and neuronal-enriched cultures (8 DIV), the media was exchanged from the standard growth media (MEM with Earle's salts/10% horse serum, 30 mM glucose, 2 mM glutamine) to HBSS+ (Mg2+-free Hanks buffered salt solution, 3 mM CaCl2, 20 mM HEPES (pH 7.55), 13.4 mM NaHCO3, 30 mM glucose and 25 μM glycine). Since the neuronal-enriched cultures did not survive a direct media exchange, the media was exchanged in the 12-well plates by seven two-fold dilutions (repeated series of addition of 1 ml of HBSS+ to 1 ml of media, followed by removal of 1 ml). This resulted in 0.78% of the original growth media remaining which had no detectable amounts of PGE2. After the media was exchanged, the cultures were exposed to a treatment of NMDA that yielded < 10% neuronal death over the 90 min of the experiment. The mixed cultures were treated with 100 μM NMDA for 15 min followed by treatment with 3 μM of the NMDA antagonist MK801. The neuronal-enriched cultures were treated with 200 μM NMDA for 20 min, followed by treatment with 3 μM MK801. The media was removed from the cultures 90 minutes after the initial NMDA treatment and immediately frozen (-80°C) for analyses at a later time. The amount of PGE2 in the media was determined by enzyme immunoassay as per the instructions from the manufacturer. This assay can detect as little as 8 pg/ml of PGE2.
Neuroprotection by an EP1 receptor antagonist
Neuroprotection by EP1 antagonists is sensitive to modulation by non-neuronal cells
Media exchange between cultures
Cellular composition of mixed and neuronal-enriched cultures
Transwell co-cultures with neuronal-enriched cultures
The effects of astrocytes and microglia were then individually examined for their potential influence on the neuroprotective effect of the EP1 antagonist. Astrocyte transwell cultures appeared to have little effect on the neuroprotection conferred by SC51089. However, when cultures of microglia (90% of the cells were positive for the microglial marker, F4/80) grown on transwells were placed above the neuronal-enriched cultures, SC51089 was not neuroprotective (Figure 6). These same trends for the different combinations of transwell cultures on neuronal-enriched cultures were seen at two different levels of NMDA-induced excitotoxicity (20% survival, Figure 6a and 10% survival, Figure, 6b). None of the transwell cultures were toxic to the neurons in the neuronal-enriched cultures in the absence of NMDA (data not shown). Therefore, these experiments showed that the presence of microglia blocks neuroprotection by an EP1 antagonist. The same effects of microglia on neuroprotection by an EP1 antagonist was also observed in cultures grown in defined serum-free media (see methods). In these cultures, microglial transwells also abrogated the neuroprotective effects of SC51089 on neuronal-enriched cultures (data not shown). This indicates that the effect of microglia on neuroprotection by an EP1 antagonist was not due to serum-derived components in the growth media.
Effects of microglia on sensitivity to NMDA-induced excitotoxicity
Microglia alter neuronal response to EP1 antagonists
Neuronal expression of EP1 in mixed cultures and neuronal-enriched cultures with and without microglial transwells
NMDA stimulation of PGE2 in mixed and neuronal-enriched cultures
In this study we found that two structurally distinct EP1 receptor antagonists (Ono 8711 and SC51089) protect a significant percentage of neurons against a sustained exposure to NMDA in primary neuronal cultures devoid of microglia. The neuroprotective effect of these agents is mediated by specifically blocking the EP1 receptor since maximal neuroprotection with Ono 8711 occurred with concentrations as low at 3 nM, which is just above the Ki for the EP1 receptor (1.7 nM), (Watanabe et al., 1999). Concentrations of Ono 8711 greater than 100 nM were not directly toxic to neurons, but had no significant neuroprotective effect agonist NMDA-mediated toxicity. It is not known what accounts for this effect, but higher concentrations of Ono 8711 may have off target actions which may alter the contribution of EP1 to neuronal viability or neuronal sensitivity to excitotoxic stimuli.
The pharmacological evidence that we demonstrated with EP1 receptor antagonists is consistent with recent in vitro [21, 22] and in vivo studies demonstrating that EP1 activation contributes to neuronal death [23, 24]. The role that neuronal EP1 activation plays in augmentation of neuronal excitotoxicity is now beginning to be elucidated. EP1 is a G-protein coupled receptor linked to the inositol triphosphate (IP3) signaling cascade and causes release of intracellular calcium . Recent studies by Iadecola and colleagues demonstrate that EP1 activation impairs the Na+-Ca2+ exchanger (NCX) leading to dysregulation of intracellular calcium following NMDA receptor stimulation . The resulting increase in intracellular cytoplasmic calcium concentration is an integral component to excitotoxic death and contributes to neuronal death [35, 36]. Another recent study indicated that EP1 activation can also contribute to excitotoxic death through the AKT/PKB (protein kinase B) signaling cascade . NMDA-induced neuronal death is associated with decreased AKT excitotoxic stimuli . EP1 inhibitors increase AKT activity by inactivating the inhibitor of AKT, PTEN (phosphatase and tensin homologue deleted on chromosome 10), and subsequently inhibiting translocation of the proapoptotic protein BAD . Our finding that EP1 antagonists protected neurons in neuronal-enriched cultures is also consistent with a mechanism in which neuronal activation of EP1 leads to cell death.
Our results indicated that when microglia were present in the cultures, the neuroprotective properties of EP1 antagonists were lost. Gendron et al. demonstrated that EP1 antagonists conferred neuroprotection against excitotoxicity induced by oxygen-glucose-deprivation in mixed neuronal cultures . However, they did not characterize the cellular composition with respect to microglia. In their culture system anti-mitotic agents were added to their mixed cultures on day 4 after plating compared to our mixed cultures in which the anti-mitotic agents were added on day 9 (see methods). This difference may have limited the amounts of microglia in their system.
In our culture system, the loss of neuroprotection with the EP1 antagonist occurred when the microglia were either cultured with the neurons (as in the mixed cultures) or when the microglia were co-cultured in transwells above the neuronal-enriched cultures. Co-culturing of neuronal-enriched cultures with microglia for 48 hours was also sufficient to alter the neuroprotective effects of EP1 antagonists. This indicates that the microglial cultures were able to produce soluble factors that could diffuse through the 0.4 μm pores in the transwell membrane and influence EP1 response in the neurons in the neuronal-enriched cultures. Simple transfer of media from the mixed cultures to the neuronal-enriched cultures could not abolish neuroprotection. This indicates that the soluble component(s) from the mixed cultures that were able to affect EP1 response in the transwell cultures are either labile, not incubated for sufficient times before NMDA challenge or are produced in the mixed cultures during NMDA stimulation. Identification of these components produced by microglia and how they modulate neuronal EP1 response will require further investigation.
The mechanism by which microglia inhibit the neuroprotective effects of EP1 antagonists appeared to be more complex than just downregulation of neuronal EP1 expression or decreased NMDA-induced synthesis of PGE2. However, there was a significant difference in cellular distribution of EP1 to the nuclear region in neurons from cultures that lacked microglia. Additional studies will be required to determine whether this change in EP1 localization is important to EP1 signaling and the viability of neurons following NMDA stimulation. Nuclear localization of EP1 has been observed in neurons and brain-derived endothelial cells, where stimulation of EP1 receptors can induce changes in nuclear calcium concentrations and modulate gene expression of genes such as c-fos .
In our culture systems, neurons were more resistant to NMDA induced excitotoxicity when microglia were present. Decreased neuronal nuclear expression of EP1 was also observed in microglia-containing cultures. It is possible that the neuroprotective effect of microglia could be in part due to a redistribution of EP1 expression.
Our data support a role for microglia in altering the neuronal EP1 response. However, we cannot rule out whether EP1 signaling in microglia can alter neuronal viability. Further studies will be needed to assess the role that EP1 signaling in microglia play in modulating neuronal viability. EP1 receptors were abundantly expressed in microglia in our cultures (data not shown) and could play some role in the interaction with neurons.
This study identified a new role for microglia as an important modulator of the neuroprotective properties of EP1 antagonists in vitro. If the same interactions were to occur in vivo, this could have important consequences with respect to microglial-mediated effects on neuronal viability. In vivo, microglia migrate to the site of neuronal injury and have been implicated to play a role in stroke and neurodegenerative diseases such as Alzheimer's disease, Parkinson's disease, multiple sclerosis, and HIV-dementia . It is presently controversial whether microglia play a harmful or beneficial role in neuronal death [39–41]. The role played by microglia in neuronal death may differ depending on the stimuli used to induce death, the cellular milieu and the presence of various factors along with the specific intracellular pathways that are activated .
There are two examples in vitro where microglia have been shown to be detrimental towards neurotoxin-induced neuronal death that is associated with COX-2. In an in vitro model of Parkinson's disease, COX-2 inhibitors showed a dramatically reduced level of neuroprotection towards dopaminergic neurons in cultures that lack microglia following MPP+ (1-methyl-4-phenylpyridinium)-induced death . Microglia were shown to be the major source of MPP+-induced production of PGE2 as well as contribute to increasing neuronal COX-2 induction. In another example, microglia are also required for lipopolysaccharide (LPS) – induced neuronal death in culture . In this case, stimulation of the EP2 receptor on microglia induced expression of COX-2 and iNOS which were required for neuronal death. These examples differ from our studies where microglia appeared to have neuroprotective effects and diminished the contribution of neuronal EP1 receptor activation to excitotoxic death.
In light of the recently identified complications of cardiovascular disease associated with COX-2 inhibitors ([25, 44, 45], it is important to identify new targets down-stream of COX-2 that contribute to neuronal death . Inhibition of EP1 represents an attractive alternative to COX-2 inhibition for neuroprotective strategies that could be used for treating stroke and other neurodegenerative diseases [23, 24, 47, 48].
This study demonstrates that microglia may limit the contribution of EP1 receptor activation towards neuronal excitotoxicity. Understanding how microglia could limit the contribution of EP1 to neuronal excitotoxicity is not only an important consideration for designing neuroprotective treatments with EP1 antagonists but may also generate new approaches in the treatment of neurodegenerative disease.
This work is supported by VA Merit grants (NGC and JWR) and NIH grant RO1 AG20650 (NGC). Partial support was also provided by a grant from the National Multiple Sclerosis Society (to JWR and NGC) and the Val A. Browning foundation of Utah and the Cumming Foundation for support of MAR. The authors thank Jacob Snow, Blair Wood and Philip Tang for there technical contributions, and to Angelynn Long for both technical contributions and assistance in manuscript preparation. We are very grateful for the comments on the manuscript by Dr Ikuo Tsunoda. We thank Dr Thomas Green for help with statistical analyses.
- Smith WL, DeWitt DL: Prostaglandin endoperoxide H synthases (cyclooxygenases)-1 and -2. Adv Immunol. 1996, 62: 167-215.View ArticlePubMedGoogle Scholar
- Hewett SJ, Uliasz TF, Vidwans AS, Hewett JA: Cyclooxygenase-2 contributes to N-methyl-D-aspartate-mediated neuronal cell death in primary cortical cell culture. J Pharmacol Exp Ther. 2000, 293: 417-425.PubMedGoogle Scholar
- Carlson NG: Neuroprotection of cultured cortical neurons mediated by the cyclooxygenase-2 inhibitor APHS can be reversed by a prostanoid. J Neurosci Res. 2003, 71: 79-88.View ArticlePubMedGoogle Scholar
- Nogawa S, Zhang F, Ross ME, Iadecola C: Cyclo-oxygenase-2 gene expression in neurons contributes to ischemic brain damage. J Neurosci. 1997, 17: 2746-2755.PubMedGoogle Scholar
- Nagayama M, Niwa K, Nagayama T, Ross ME, Iadecola C: The cyclooxygenase-2 inhibitor NS-398 ameliorates ischemic brain injury in wild-type mice but not in mice with deletion of the inducible nitric oxide synthase gene. J Cereb Blood Flow Metab. 1999, 19: 1213-1219.View ArticlePubMedGoogle Scholar
- Kunz T, Oliw E: The selective cyclooxygenase-2 inhibitor rofecoxib reduces kainate-induced cell death in the rat hippocampus. Eur J Neurosci. 2001, 13: 569-575.View ArticlePubMedGoogle Scholar
- Kawaguchi K, Hickey RW, Rose ME, Zhu L, Chen J, Graham S: Cyclooxygenase-2 expression is induced in rat brain after kainate-induced seizures and promotes neuronal death in CA3 hippocampus. Brain Res. 2005, 1050: 130-137.View ArticlePubMedGoogle Scholar
- Iadecola C, Niwa K, Nogawa S, Zhao X, Nagayama M, Araki E, Morham S, Ross ME: Reduced susceptibility to ischemic brain injury and N-methyl-D-aspartate-mediated neurotoxicity in cyclooxygenase-2-deficient mice. Proc Natl Acad Sci USA. 2001, 98: 1294-1299.PubMed CentralView ArticlePubMedGoogle Scholar
- Dore S, Otsuka T, Mito T, Sugo N, Hand T, Wu L, Hurn PD: Neuronal overexpression of cyclooxygenase-2 increases cerebral infarction. Ann Neurol. 2003, 54: 155-162.View ArticlePubMedGoogle Scholar
- Andreasson KI, Savonenko A, Vidensky S, Goellner JJ, Shang Y, Shaffer A, Kaufmann WE, Worley PF, Isakson P, Markowska AL: Age-dependent cognitive deficits and neuronal apoptosis in cyclooxygenase-2 transgenic mice. J Neurosci. 2001, 21: 8198-8209.PubMedGoogle Scholar
- Kelley KA, Ho L, Winger D, Freire-Moar J, Borelli CB, Aisen PS, Pasinetti GM: Potentiation of excitotoxicity in transgenic mice overexpressing neuronal cyclooxygenase-2. Am J Pathol. 1999, 155: 995-1004.PubMed CentralView ArticlePubMedGoogle Scholar
- Hata AN, Breyer RM: Pharmacology and signaling of prostaglandin receptors: multiple roles in inflammation and immune modulation. Pharmacol Ther. 2004, 103: 147-166.View ArticlePubMedGoogle Scholar
- Manabe Y, Anrather J, Kawano T, Niwa K, Zhou P, Ross ME, Iadecola C: Prostanoids, not reactive oxygen species, mediate COX-2-dependent neurotoxicity. Ann Neurol. 2004, 55: 668-675.View ArticlePubMedGoogle Scholar
- Lerea LS, Carlson NG, Simonato M, Morrow JD, Roberts JL, McNamara JO: Prostaglandin F2α is required for NMDA receptor-mediated induction of c-fos mRNA in dentate gyrus neurons. J Neurosci. 1997, 17: 117-124.PubMedGoogle Scholar
- Cazevieille C, Muller A, Meynier F, Dutrait N, Bonne C: Protection by prostaglandins from glutamate toxicity in cortical neurons. Neurochem Int. 1994, 24: 395-398.View ArticlePubMedGoogle Scholar
- 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.View ArticlePubMedGoogle Scholar
- McCullough L, Wu L, Haughey N, Liang X, Hand T, Wang Q, Breyer RM, Andreasson K: Neuroprotective function of PGE2 EP2 receptor in cerebral ischemia. J Neurosci. 2004, 24 (1): 257-68.View ArticlePubMedGoogle Scholar
- Ahmad AS, Ahmad M, de Brum-Fernandes AJ, Dore S: Prostagland EP4 receptor agonist protects against acute neurotoxicity. Brain Res. 2005, 1066: 71-77.View ArticlePubMedGoogle Scholar
- Ahmad AS, Zhuang H, Echeveriia V, Dore S: Stimulation of prostaglandin EP2 receptors prevents NMDA-induced excitotoxicity. J Neurotrauma. 2006, 23: 1895-1903.View ArticlePubMedGoogle Scholar
- Takadera T, Shiraishi Y, Ohyashiki T: Prostaglandin E2 induced caspase-dependent apoptosis possibly through activation of EP2 receptors in cultured hippocampal neurons. Neurochem Int. 2004, 45: 713-719.View ArticlePubMedGoogle Scholar
- Carrasco E, Casper C, Werner P: PGE2 receptor EP1 renders dopaminergic neurons selectively vulnerable to low-level oxidative stress and direct PGE2 neurotoxicity. J Neurosci Res. 2007, 85: 3109-3117.View ArticlePubMedGoogle Scholar
- Gendron TF, Brunette E, Tauskela JS, Morley P: The dual role of prostaglandin E2 in excitotoxicity and preconditioning-induced neuroprotection. Eur J Pharm. 2005, 517: 12-27.View ArticleGoogle Scholar
- Kawano T, Anrather J, Zhou P, Park L, Wang G, Frys KA, Kunz A, Cho S, Orio M, Iadecola C: Prostaglandin E(2) EP1 receptors: downstream effectors of COX-2 neurotoxicity. Nat Med. 2006, 12: 225-229.View ArticlePubMedGoogle Scholar
- Ahmad AS, Saleem S, Ahmad M, Dore S: Prostaglandin EP1 receptor contributes to excitotoxicity and focal ischemic brain damage. Toxicological Sci. 2006, 89: 265-270.View ArticleGoogle Scholar
- Iadecola C, Gorelick PB: The Janus face of cyclooxygenase-2 in ischemic stroke: shifting toward downstream targets. Stroke. 2005, 36: 182-185.View ArticlePubMedGoogle Scholar
- Zhou P, Qian L, Chou T, Iadecola C: Neuroprotection by PGE2 receptor EP1 inhibition involves the PTEN/AKT pathway. Neurobiol Dis. 2008, 29: 543-551.PubMed CentralView ArticlePubMedGoogle Scholar
- Watanabe Y, Matsumura K, Takechi H, Kato K, Morii H, Bjorkman M, Langstrom B, Noyori R, Suzuki M, Watanabe Y: A novel subtype of prostacyclin receptor in the central nervous system. J Neurochem. 1999, 72: 2583-2592.View ArticlePubMedGoogle Scholar
- Hallinan EA, Hagen TJ, Tsymbalov S, Husa RK, Lee AC, Stapelfeld A, Savage MA: Aminoacetyl moiety as a potential surrogate for diacylhydrazine group of SC-5 a potent PGE2 antagonist, and its analogs. J Med Chem. 1996, 39 (2): 609-13.View ArticlePubMedGoogle Scholar
- Carlson NG, Bacchi A, Rogers SW, Gahring LC: Nicotine blocks TNF-α-mediated neuroprotection to NMDA by an α-bungarotoxin-sensitive pathway. J Neurobiol. 1998, 35: 29-36.View ArticlePubMedGoogle Scholar
- Carlson NG, Wieggel WA, Chen J, Bacchi A, Rogers SW, Gahring LC: Inflammatory cytokines IL-1 alpha, IL-1 beta, IL-6, and TNF-alpha impart neuroprotection to an excitotoxin through distinct pathways. J Immunol. 1999, 163: 3963-3968.PubMedGoogle Scholar
- Gahring LC, Carlson NG, Wieggel WA, Howard J, Rogers SW: Alcohol blocks TNFalpha but not other cytokine-mediated neuroprotection to NMDA. Alcohol Clin Exp Res. 1999, 23: 1571-1579.PubMedGoogle Scholar
- Bruce-Keller AJ, Geddes JW, Knapp PE, McFall RW, Keller JN, Holtsberg FW, Parthasarathy S, Steiner SM, Mattson MP: Anti-death properties of TNF against metabolic poisoning: mitochondrial stabilization by MnSOD. J Neuroimmunol. 1999, 93: 53-71.View ArticlePubMedGoogle Scholar
- Boie Y, Stocco R, Sawyer N, Slipetz DM, Ungrin MD, Neuschäfer-Rube F, Püschel GP, Metters KM, Abramovitz M: Molecular cloning and characterization of the four rat prostaglandin E2 prostanoid receptor subtypes. Eur J Pharmacol. 1997, 340: 227-241.View ArticlePubMedGoogle Scholar
- Katoh H, Watabe A, Sugimoto Y, Ichikawa A, Negishi M: Characterization of the signal transduction of prostaglandin E receptor EP1 subtype in cDNA-transfected Chinese hamster ovary cells. Biochim Biophys Acta. 1995, 1244: 41-48.View ArticlePubMedGoogle Scholar
- Choi DW: Cellular defences destroyed. Nature. 2005, 433: 696-698.View ArticlePubMedGoogle Scholar
- Sattler R, Tymianski M: Molecular mechanisms of calcium-dependent excitotoxicity. J Mol Med. 2000, 78: 3-13.View ArticlePubMedGoogle Scholar
- Luo HR, Hattori H, Hossanin MA, Hester L, Huang Y, Lee-Kwon W, Donowitz M, Nagata E, Snyder SH: Akt as a mediator of cell death. Proc Natl Acad Sci USA. 2003, 100: 11712-11717.PubMed CentralView ArticlePubMedGoogle Scholar
- Bhattacharya M, Peri KG, Almazan G, Ribeiro-da-Silva A, Shichi H, Durocher Y, Abramovitz M, Hou X, Varma DR, Chemtob S: Nuclear localization of prostaglandin E2 receptors. Proc Natl Acad Sci USA. 1998, 95: 15792-15797.PubMed CentralView ArticlePubMedGoogle Scholar
- Dheen ST, Kaur C, Ling EA: Microglial activation and its implications in the brain diseases. Curr Med Chem. 2007, 14: 1189-1197.View ArticlePubMedGoogle Scholar
- del Zoppo GJ, Becker KJ, Hallenbeck JM: Inflammation after stroke: is it harmful?. Arch Neurol. 2001, 58: 669-672.View ArticlePubMedGoogle Scholar
- Feuerstein GZ, Wang X: Inflammation and stroke: benefits without harm?. Arch Neurol. 2001, 58: 672-674.View ArticlePubMedGoogle Scholar
- Wang T, Pei Z, Zhang W, Liu B, Langenbach R, Lee C, Wilson B, Reece JM, Miller DS, Hong JS: MPP+-induced COX-2 activation and subsequent dopaminergic neurodegeneration. FASEB J. 2005, 19: 1134-1136.View ArticlePubMedGoogle Scholar
- Shie FS, Montine KS, Breyer RM, Montine TJ: Microglial EP2 is critical to neurotoxicity from activated cerebral innate immunity. Glia. 2005, 52: 70-77.View ArticlePubMedGoogle Scholar
- Fitzgerald GA: Coxibs and cardiovascular disease. N Engl J Med. 2004, 351: 1709-1711.View ArticlePubMedGoogle Scholar
- Topol EJ: Failing the public health-rofecoxib, Merck, and the FDA. N Engl J Med. 2004, 351: 1707-1709.View ArticlePubMedGoogle Scholar
- Hewett SJ, Bell SC, Hewett JA: Contributions of cyclooxygenase-2 to neuroplasticity and neuropathology of the central nervous system. Pharmacol Ther. 2006, 112: 335-357.View ArticlePubMedGoogle Scholar
- Dore S: GPCR antagonists as an alternative to COX-2 inhibitors: a case for PGE2 EP1 receptor. Trends Pharmacol Sci. 2006, 27 (9): 458-60.View ArticlePubMedGoogle Scholar
- Saleem S, Li RC, Wei G, Dore S: Effects of EP1 receptor on cerebral blood flow in the middle cerebral artery occlusion model of stroke in mice. J Neurosci Res. 2007, 85: 2433-2440.PubMed CentralView ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.