Prostaglandin E2 inhibits the proliferation of human gingival fibroblasts via the EP2 receptor and Epac

Elevated levels of prostaglandins such as PGE₂ in inflamed gingiva play a significant role in the tissue destruction caused by periodontitis, partly by targeting local fibroblasts. Only very few studies have shown that PGE₂ inhibits the proliferation of a gingival fibroblast (GF) cell line, and we expanded this research by using primary human GFs (hGFs) and looking into the mechanisms of the PGE₂ effect. GFs derived from healthy human gingiva were treated with PGE₂ and proliferation was assessed by measuring cell number and DNA synthesis and potential signaling pathways were investigated using selective activators or inhibitors. PGE₂ inhibited the proliferation of hGFs dose‐dependently. The effect was mimicked by forskolin (adenylate cyclase stimulator) and augmented by IBMX (a cAMP‐breakdown inhibitor), pointing to involvement of cAMP. Indeed, PGE₂ and forskolin induced cAMP generation in these cells. Using selective EP receptor agonists we found that the anti‐proliferative effect of PGE₂ is mediated via the EP₂ receptor (which is coupled to adenylate cyclase activation). We also found that the effect of PGE₂ involved activation of Epac (exchange protein directly activated by cAMP), an intracellular cAMP sensor, and not PKA. While serum increased the amount of phospho‐ERK in hGFs by ～300%, PGE₂ decreased it by ～50%. Finally, the PGE₂ effect does not require endogenous production of prostaglandins since it was not abrogated by two COX‐inhibitors. In conclusion, in human gingival fibroblasts PGE₂ activates the EP₂—cAMP—Epac pathway, reducing ERK phosphorylation and inhibiting proliferation. This effect could hamper periodontal healing and provide further insights into the pathogenesis of inflammatory periodontal disease. J. Cell. Biochem. 108: 207–215, 2009. © 2009 Wiley‐Liss, Inc.     

Microsomal prostaglandin E synthase-2 is not essential for in vivo prostaglandin E2 biosynthesis

Prostaglandin E₂ (PGE₂) plays an important role in the normal physiology of many organ systems. Increased levels of this lipid mediator are associated with many disease states, and it potently regulates inflammatory responses. Three enzymes capable of in vitro synthesis of PGE₂ from the cyclooxygenase metabolite PGH₂ have been described. Here, we examine the contribution of one of these enzymes to PGE₂ production, mPges-2, which encodes microsomal prostaglandin synthase-2 (mPGES-2), by generating mice homozygous for the null allele of this gene. Loss of mPges-2 expression did not result in a measurable decrease in PGE₂ levels in any tissue or cell type examined from healthy mice. Taken together, analysis of the mPGES-2 deficient mouse lines does not substantiate the contention that mPGES-2 is a PGE₂ synthase.     

Effects of 9-deoxo-16, 16-dimethyl-9-methylene PGE2 on muscle contractile activity and collagen synthesis in the human cervix

The in vitro effects of a stable PGE-analogue (9-deoxo-16, 16-dimethyl-9-methylene PGE₂ (9-methylene PGE₂) on human cervical tissue was investigated. The influence of the analogue on collagen biosynthesis was studied by measuring the incorporation of ³H-proline, while smooth muscle effects were evaluated by isometric recording of contractile activity. The specimens were obtained by needle biopsy from women in early and late pregnancy and from nonpregnant women of fertile age.9-methylene PGE₂ compared with controls increased the incorporation of ³H-proline in the secretory phase and before the 9th week of pregnancy, whereas radiolabelling was decreased in the follicular phase, in the 9th–12th week and at term. With respect to incorporation of 3-H-proline, 9-methylene PGE₂ was equipotent to PGE₂. 9-methylene PGE₂ inhibited spontaneous contractile activity in early as well as in late pregnancy but increased muscular activity in nonpregnant patients. The inhibitory effects of the analogue was similar to that of PGE₂ but the natural compound was considerably more potent in this respect.     

Interrelationships among prostaglandins, vasopressin and cAMP in renal papillary collecting tubile cells in culture

To determine the influence of prostaglandins on cAMP metabolism in renal papillary collecting tubule (RPCT) cells, intracellular cAMP levels were measured after incubating cells with prostaglandins (PGs) alone or in combination with arginine vasopressin (AVP). PGE₁, PGE₂ and PGI₂, but not PGD₂ or PGF_2α, increased intracellular cAMP concentrations. At maximal concentrations (10⁻⁵ tthe effects of PGE₂ plus PGI₂ (or PGE₁), but not of PGI₂ plus PGE₁, were additive suggesting that at least two different PG receptors may be present in RPCT cell populations. Bradykinin treatment of RPCT cells caused an accumulation of intracellular cAMP which was blocked by aspirin and was quantitatively similar to that observed with 10⁻⁵ PGE₂. PGs, when tested at concentrations (e.g. 10⁻⁹ ) which had no independent effect on intracellular cAMP levels, did not inhibit the AVP-induced accumulation of intracellular cAMP in RPCT cells. These results indicate that PGs do not block AVP-induced accumulation of intracellular cAMP in RPCT cells at concentrations of PGs which have been shown to inhibit the hydroosmatic effect of AVP on perfused collecting tubule segments. However, at higher concentrations of PGs (e.g. 10⁻⁵ ), the effects of AVP plus PGE₁, PGE₂, PGI₂ or bradykinin on intracellular cAMP levels were not additive. Thus, under certain conditions, there is an interaction between PGs and AVP at the level of cAMP metabolism in RPCT cells.     

Studies on growth hormone secretion IX. Prostaglandins do not act like ionophores

To gain further insight on the mechanism of GH secretion in general and on the stimulation of this process by prostaglandins in particular, we compared the effects of PGE₁ and PGE₂ on hormone release and cyclic nucleotide levels with those of the ionophores A23187 and X537A under a variety of experimental conditions. All these substances (in the presence but not in the absence of calcium) enhanced GH release in incubated rat anterior pituitaries , prostaglandins being considerably more potent than ionophores. However, while PGE₂ caused a dose-dependent rise in pituitary cyclic AMP levels (from doubling at 10⁻⁷ M to a two-hundred fold increase at 10⁻⁵ M), the ionophores had no effect on the concentrations of this nucleotide. Neither PGE₂ nor the ionophores had any measurable effect on cyclic GMP levels. Exposure of tissues to ionophores for 60 minutes rendered them refractory to subsequent stimulation by PGE₁ or to ionophores themselves, whereas preincubation with PGE₁ did not diminish GH responses during a second incubation period. On the other hand, 60-minute preincubation of hemipituitaries in the presence of ionophores (10⁻⁵ M) did not suppress subsequent PGE₁-promoted cyclic AMP accumulation. Metabolic blockers inhibited PGE₂ and A23187-promoted GH-release but failed to suppress GH-response to X537A. Verapamil partially inhibited PGE₂ but not ionophore induced GH secretion. Ionophores particularly X537A, accelerated 45Ca efflux while PGE₁ did not influence this. Electronmicroscopy revealed extensive vacuolization localized chiefly at the Golgi apparatus when tissues were incubated with X537A. PGE₁ and A23187 had no such morphological effect. It is concluded that prostaglandins E and ionophores promote GH secretion by different mechanisms.     

Limited effect of anti-rheumatic treatment on 15-prostaglandin dehydrogenase in rheumatoid arthritis synovial tissue

Introduction

Rheumatoid arthritis (RA) is a chronic inflammatory disease in which prostaglandin E₂ (PGE₂) displays an important pathogenic role. The enzymes involved in its synthesis are highly expressed in the inflamed synovium, while little is known about 15- prostaglandin dehydrogenase (15-PGDH) that metabolizes PGE₂. Here we aimed to evaluate the localization of 15-PGDH in the synovial tissue of healthy individuals or patients with inflammatory arthritis and determine the influence of common RA therapy on its expression.

Methods

Synovial tissue specimens from healthy individuals, psoriatic arthritis, ostheoarthritis and RA patients were immunohistochemically stained to describe the expression pattern of 15-PGDH. In addition, the degree of enzyme staining was evaluated by computer analysis on stained synovial biopsies from two groups of RA patients, before and after RA specific treatment with either intra-articular glucocorticoids or oral methotrexate therapy. Prostaglandins derived from the cyclooxygenase (COX) pathway were determined by liquid-chromatography mass spectrometry in supernatants from interleukin (IL) 1β-activated fibroblast-like synoviocytes (FLS) treated with methotrexate.

Results

15-PGDH was present in healthy and inflamed synovial tissue, mainly in lining macrophages, fibroblasts and vessels. Intra-articular glucocorticoids showed a trend towards reduced 15-PGDH expression in RA synovium (p = 0.08) while methotrexate treatment left the PGE₂ pathway unaltered both in biopsies ex vivo and in cultured FLS.

Conclusions

Early methotrexate therapy has little influence on the expression of 15-PGDH and on any of the PGE₂ synthesizing enzymes or COX-derived metabolites. Thus therapeutic strategies involving blocking induced PGE₂ synthesis may find a rationale in additionally reducing local inflammatory mediators.     

Absorption of prostaglandin E2 and uterine sensitivity of the non-pregnant and pregnant monkey in vivo

Radioactive (11-³H) prostaglandin E₂(PGE₂) levels in plasma of non-pregnant Rhesus and Japanese monkeys were determined by radioimmunoassay. The amounts of PGE₂ in plasma increased gradually and reached a peak 90 minutes after oral administration. Comparatively low levels were detected 24 hours after oral administration. Plasma PGE₂ levels increased rapidly and disappeared within 5 minutes when 5 μg/kg of PGE₂ was administered intravenously.Uterine contractile sensitivity to PGE₂ and F_2α was measured by the threshold of a venous dosage required to evoke an elevation of uterine contractility in non-pregnant and pre- and post-labor Japanese monkeys. Uterine sensitivity to PGE₂ in the non-pregnant monkey appear to vary in accordance with the sexual life span. At term of pregnancy, PGE₂ was much more potent in causing uterine contraction than PGF_2α. During labor and at postpartum period with lactation, effectiveness of PGE₂ appear to be less than that of PGF_2α. The non-pregnant and pregnant uterus of the third trimester are more sensitive to PGE₂ than the laboring and postpartum uterus.The long latency of the elevation of uterine contractility induced by the intravenous administration of PG suggests that the PG compounds have potent actions on the central nervous system.     

T cell recruitment from the thymus to the spleen in tumor-bearing mice

Summary After inoculation of tumor cells (methylcholanthrene-induced sarcoma), the number of Thy 1⁺ cells and PNA (peanut agglutinin) binding cells, which were shown to be different subpopulations were increased in the spleen of thymus-intact mice, in contrast this increase was not observed in adult thymectomized mice. In experiments performed concurrently with splenic cell analysis, we found that the plasma PGE₂ levels declined in parallel with the tumor growth. Prevention of such a decline of plasma PGE₂ level by replenishment with exogenous PGE₂ inhibited the splenic cell increase in tumor bearers. In the tumorbearing mice, cell traffic systems from the thymus to the periphery was ascertained by injecting fluorescein diacetate (FDA) into the thymus and observing fluorescein positive cells in the periphery. We suggest that increased recruitment of thymic cells to the periphery may be mediated by PGE₂ in the presence of a tumor.Abbreviations PNA⁺ cells, peanut agglutinin binding cells - Ig⁺ cells, surface IgM positive cells - Thy 1⁺ Thy 1.2 antigen positive cells - PGE₂ prostaglandin E₂     

Bone resorption and prostaglandin production by mouse calvaria in vitro: Response to exogenous prostaglandins and their precursor fatty acids

Mouse calvaria were maintained in organ culture for 96 h and endogenous prostaglandin production and active bone resorption (⁴⁵ Ca release) measured. After a lag phase of 12 h, active resorption increased over the 96 h period. The amounts of prostaglandins released into the culture medium (measured by radioimmunoassay) were highest in the first 24 h of culture. Unless these were removed by preculturing for 24 h, or suppressed by indomethacin, no response to exogenous PGE₂, PGF_2α or prostaglandin precursors could be demonstrated. Bone resorption was stimulated after preculture by both PGE₂ and PGF_2α in a dose-dependent manner (10^?18M – 10^?5M), with PGE₂ being the more potent. Collagen synthesis was unaffected by PGF_2α, whereas PGE₂ (10^?5M) had an inhibitory effect. Eicosatrienoic acid did not stimulate bone resorption at lower concentrations (10^?7M – 10^?5M_, but was inhibitory at 10^?4M. Arachidonic acid also inhibited resorption at 10^?4M, but at lower concentrations (10^?7M – 10^?5M0 increased active resorption. This was concomitant with a rise in PGE₂ and PGF_2α levels, PGE₂ production being significantly higher than PGF_2α. The effects of PGE₂ (10^?8M) and PGF_2α (10^∞M appeared additive: there was no evidence of synergistic or antagonistic effects when varying ratios of PGE₂ : PGF_2α were employed.     

Prostaglandin E2 Induction during Mouse Adenovirus Type 1 Respiratory Infection Regulates Inflammatory Mediator Generation but Does Not Affect Viral Pathogenesis

Respiratory viruses cause substantial disease and are a significant healthcare burden. Virus-induced inflammation can be detrimental to the host, causing symptoms during acute infection and leading to damage that contributes to long-term residual lung disease. Prostaglandin E₂ (PGE₂) is a lipid mediator that is increased in response to many viral infections, and inhibition of PGE₂ production during respiratory viral infection often leads to a decreased inflammatory response. We tested the hypothesis that PGE₂ promotes inflammatory responses to mouse adenovirus type 1 (MAV-1) respiratory infection. Acute MAV-1 infection increased COX-2 expression and PGE₂ production in wild type mice. Deficiency of the E prostanoid 2 receptor had no apparent effect on MAV-1 pathogenesis. Virus-induced induction of PGE₂, IFN-γ, CXCL1, and CCL5 was reduced in mice deficient in microsomal PGE synthase-1 (mPGES-1^-/- mice). However, there were no differences between mPGES-1^+/+ and mPGES-1^-/- mice in viral replication, recruitment of leukocytes to airways or lung inflammation. Infection of both mPGES‑1^+/+ and mPGES-1^-/- mice led to protection against reinfection. Thus, while PGE₂ promotes the expression of a variety of cytokines in response to acute MAV-1 infection, PGE₂ synthesis does not appear to be essential for generating pulmonary immunity.     

Effect of prolonged prostaglandin exposure on prostaglandin synthesis by human lung fibroblasts

Pretreatment of human lung fibroblasts with PGE₂ but not PGF_2α enhanced synthesis of prostaglandins (PGs). The effect of the pretreatment on PG synthesis was related to the concentration of PGE₂ that was added to the culture medium. Pretreatment with PGE₂ at 5 × 10⁻¹²M did not enhance PG synthesis whereas pretreatment with PGE₂ at 5 × 10⁻⁶M induced a maximal effect. Production of PGs was increased following 1 day of pretreatment with PGE₂ and was increased further following 3 days of pretreatment. The PGE₂ treated cells showed only a slight increase in the bradykinin-induced release of radioactivity from cells prelabeled with [³H]arachidonic acid but showed a dramatic increase in the bradykinin-induced synthesis of radio-labeled PGs. The conversion of free arachidonate to PGs in both intact cells and in a cell-free preparation was increased by PGE₂ pretreatment. The presence of cyclohexamide during the pretreatment did not inhibit the PGE₂-induced activation of PG synthesis. Taken together, the results indicate that pretreatment of cells with PGE₂ increased PG synthesis by augmenting the conversion of arachidonate to PGs.     

Stimulation by prostaglandin E2 of glucagon and insulin release from isolated rat pancreas

To ascertain whether prostaglandins (PG) may play a role in the secretion of glucagon and in an attempt to elucidate the conflicting observations on the effects of PG on insulin release, the isolated intact rat pancreas was perfused with solutions containing 1.1 × 10⁻⁹ to 1.8 × 10⁻⁵M PGE₂. In the presence of 5.6 mM glucose significant increments in portal venous effluent levels of glucagon and insulin were observed in response to minimal concentrations of 2.8 × 10⁻⁸ and 1.4 × 10⁻⁷M PGE₂, respectively; a dose-response relationship was evident for both hormones at higher concentrations of PGE₂. When administered over 60 seconds, 1.4 × 10⁻⁶M PGE₂ resulted in a significant increase in glucagon levels within 24 seconds and in insulin within 48 seconds. Ten-minute perfusions of 1.4 × 10⁻⁶M PGE₂ elicited biphasic release of both islet hormones; Phase I glucagon release preceded that of insulin. Both phases of the biphasic glucagon and insulin release which occurred in response to 15-minute perfusions of 10 mM arginine were augmented by PGE₂. These observations indicate that PGE₂ can evoke glucagon and insulin release at concentrations close to those observed by others in the extracts of rat pancreas. We conclude that PG may be involved in the regulation of secretion of glucagon and insulin and may mediate and/or modify the pancreatic islet hormone response to other secretagogues.     

Endogenous prostaglandin E2 enhances polyclonal immunoglobulin production by tonically inhibiting T suppressor cell activity

The immunoregulatory effect of endogenous prostaglandin E₂ (PGE₂) on immunoglobulin production was studied in an in vitro culture system of human peripheral blood mononuclear cells, stimulated with pokeweed mitogen (PWM). Three different cyclooxygenase inhibitors (indomethacin, carprofen, and piroxicam) suppressed Ig synthesis by ~50%. This inhibitory effect could be reversed by adding low doses of exogenous PGE₂ (3 × 10^?9 to 3 × 10^?8M). These doses are endogenously produced in PWM-stimulated cultures, and a concentration of 2 × 10^?8M is reached after 48 hr of culture. When B cells were directly stimulated with helper factor, PGE₂ did not enhance Ig production and doses of 3 × 10^?7M to 3 × 10^?6M were inhibitory. The effects of indomethacin and PGE₂ were eliminated when T cells were irradiated or treated with mitomycin prior to culture. The enhancing effects of PGE₂ were substantially reduced after OKT8(+) T cells were removed from the system. PWM-stimulated cultures of lymphocytes from healthy subjects over age 70 were more sensitive to inhibition by indomethacin and to stimulation by PGE₂ than were cultures of lymphocytes from young controls. Thus the major role of endogenous PGE in polyclonal Ig production in vitro is to tonically inhibit a radiosensitive, OKT8(+) suppressor T cell. This tonic inhibition is increased in subjects over 70, which provides one explanation for decreased suppressor cell function in elderly subjects.     

Production of leukotriene B4 and prostaglandin E2 by turbot (Scophthalmus maximus) leukocytes

Production of two eicosanoids derived from lipoxygenase and cyclooxygenase activities: leukotriene B₄ (LTB₄) and prostaglandin E₂ (PGE₂), respectively, have been simultaneously determined in turbot (Scophthalmus maximus) blood leucocyte and kidney macrophage supernatants by a reverse phase high performance liquid chromatography (HPLC) system coupled with a Diode–Array detector. Levels of LTB₄ after calcium ionophore challenge were 4.08 ng ml⁻¹ in blood leukocyte supernatants and 0.25 ng ml⁻¹ in kidney macrophage supernatants. The levels found for PGE₂ were 428.23 and 606.67 ng ml⁻¹ in blood leukocytes and kidney macrophage supernatants, respectively. When blood leukocytes were treated with the respective inhibitors for the enzymes implicated on the synthesis of both compounds an inhibition of 90.35% was observed for PGE₂ and 76.44% for LTB₄. The detection limit of the method was 0.15 ng ml⁻¹ for LTB₄ and 50 ng ml⁻¹ for PGE₂.     

Cyclic AMP inhibits synthesis of prostaglandin endoperoxide (PGG2) in human platelets.

Dibutyryl-cAMP but not dibutyryl-cGMP inhibited platelet aggregation and release of ¹⁴C-serotonin and ADP when induced by collagen and arachidonate but not when induced by the endoperoxide PGG₂^* (TXB₂) induced by addition of collagen to platelet rich plasma (PRP) was decreased by dibutyryl-cAMP and agents known to increase the concentration of cAMP (PGE₁, PGD₂, theophylline and acetyl choline).PGE₂ in concentrations known to decrease cAMP levels increased the formation of TXB₂ whereas concentrations of PGE₂ known to increase cAMP levels decreased the amount of TXB₂ formed. That this was due to an effect on the cyclooxygenase was indicated by inhibition of the transformation of arachidonic acid by DB-cAMP and by high concentrations of PGE₂. Additional support for regulation of the cyclo-oxygenase by cAMP and its relevance to platelet aggregation was obtained by demonstrating stimulation of PGG₂ induced aggregation by low concentrations of PGE₂ and the absence of this effect in the presence of a cyclo-oxygenase inhibitor.     

Uptake,transfer and metabolism of prostaglandin E2 in the isolated perfused human placental cotyledon

(³H) PGE₂ uptake and transfer in the isolated perfused human placental cotyledon was assessed by a ingle pass paired isotope dilution technique utilising (¹⁴C) sucrose as an extracellular marker. Metabolism of (³H) PGE₂ was measured by analysing maternal and fetal effluents from perfused human placental cotyledons after bolus injection of (³H) PGE₂ into ither the maternal or fetal sides. Maximal uptake of (³H) PGE₂ was greater on the maternal (81 +/- 8%) than the fetal sides (42 +/- 12%) and showed saturation with increasing concentrations of PGE₂ only on the fetal side with an apparent Km of 12 +/- 4.9 nmol/l and v_max of 1.5 +/- 0.2 pmol/min/g. Total recoveries of (³H) PGE₂ were 84.6 +/- 11.8 % and 32.6 +/- 6.3 % of the injected dose after injection on the fetal and maternal sides respectively.Transferof (³H) PGE₂ was the same in both directions being 6.4 +/- 1.2 % of the injected dose in the fetal-maternal direction and 5.8 +/- 2.7 % of the injected dose in the maternal-fetal direction. Metabolism was greater on the maternal side (35% of injected (³H) PGE₂) thanthe fetalside(18% of injected (³H) PGE₂) and was principally to the 13,14-dihydro-15-keto-PGE₂ metabolite. Metabolism of (³H) PGE₂ after passage across the placenta was the same in both directions and was of the order of approximately 60%.     

Interleukin-1 inhibits PGE2 binding to macrophage-like P388D1 cells by a cyclic AMP-independent process

The interaction between interleukin IL-1α and PGE₂ on P388D₂ on cells has been investigated. Preincubation of murine macrophage-like cells, P388D₁, with IL-1α (0–73 pM) reduced the binding of PGE₂ to these cells in a concentration-dependent manner. Scatchard analysis showed that IL-1α decreased the PGE₂ binding by lowering both the high and low affinity receptor binding capacities (from 0.31 ± 0.02 to 0.12 ± 0.01 fmol/10⁶ cells for the high affinity receptor binding sites and from 2.41 ± 0.12 to 1.51 ± 0.21 fmol/10⁶ cells for the low affinity receptor binding sites). However, the dissociation constants of the receptor of the IL-1α-treated cells remained unchanged. Inhibition of PGE₂ binding IL-1α did not involve changes in either protein phosphorylation or intracellular cyclic AMP levels. Our data clearly show that IL-1α inhibits the binding of PGE₂ to monocytes/macrophages and may thereby counter the immunosuppressive actions of PGE₂.     

Oxidized low-density lipoprotein-induced periodontal inflammation is associated with the up-regulation of cyclooxygenase-2 and microsomal prostaglandin synthase 1 in human gingival epithelial cells

Periodontitis is characterized by chronic gingival tissue inflammation, and inflammatory mediators such as IL-8 and prostaglandin E₂ (PGE₂) are associated with disease progression. Previously we showed that oxidatively modified low-density lipoprotein (oxLDL) was present in gingival crevicular fluid. In this study, the role of oxLDL in the gingival epithelial cell inflammatory response was further investigated using Ca9-22 cells and primary human oral keratinocytes (HOK). Treatment of Ca9-22 cells and HOK with oxLDL induced an up-regulation of IL-8 and the PGE₂-producing enzymes, cyclooxygenase-2 and microsomal PGE₂ synthase-1. These responses induced by oxLDL were significantly suppressed by a nuclear factor-kappa B (NF-κB) inhibitor. However, unlike the result in macrophages, oxLDL did not lead to an increase in CD36 expression in these two cells. These results suggest that oxLDL elicits gingival epithelial cell inflammatory responses through an activation of the NF-κB pathway. These data suggest a mechanistic link between periodontal disease and lipid metabolism-related disorders, including atherosclerosis.     

Prostaglandin E2 stimulates glutamate receptor‐dependent astrocyte neuromodulation in cultured hippocampal cells

Recent Ca²⁺ imaging studies in cell culture and in situ have shown that Ca²⁺ elevations in astrocytes stimulate glutamate release and increase neuronal Ca²⁺ levels, and that this astrocyte‐neuron signaling can be stimulated by prostaglandin E₂ (PGE₂). We investigated the electrophysiological consequences of the PGE₂‐mediated astrocyte‐neuron signaling using whole‐cell recordings on cultured rat hippocampal cells. Focal application of PGE₂ to astrocytes evoked a Ca²⁺ elevation in the stimulated cell by mobilizing internal Ca²⁺ stores, which further propagated as a Ca²⁺ wave to neighboring astrocytes. Whole‐cell recordings from neurons revealed that PGE₂ evoked a slow inward current in neurons adjacent to astrocytes. This neuronal response required the presence of an astrocyte Ca²⁺ wave and was mediated through both N‐methyl‐D ‐aspartate (NMDA) and non‐NMDA glutamate receptors. Taken together with previous studies, these data demonstrate that PGE₂‐evoked Ca²⁺ elevations in astrocyte cause the release of glutamate which activates neuronal ionotropic receptors. © 1999 John Wiley & Sons, Inc. J Neurobiol 41: 221–229, 1999