The effects of prostanoids on estrogen-dominated rat myometrial longitudinal muscle in vitro

The purpose of these experiments was to characterize the contractile response of longitudinal muscle from the estrogen-dominated rat uterus to natural and synthetic prostanoids. The biological significance is 1) to provide evidence for or against a physiological role for each natural prostanoid in the regulation of myometrial activity, 2) to determine if each prostanoid has pharmacological potential for the manipulation of myometrial activity, and 3) to understand the structural requirements for prostanoid action on the myometrium. All analogs tested produced excitation of the myometrium in vitro through what appeared to be a direct action on the muscle. The order of potency of the natural prostanoids was prostaglandin (PG) F2 alpha = PGD2 = PGE2 = PGE1 greater than PGA2 = PGB2 = 6-keto-PGF1 alpha. This order of potency was not consistent with any single currently recognized prostanoid receptor. Furthermore, PGF2 alpha had an EC50 (effective concentration that produces 50% of the maximal response) of 0.5 microM, which was low in comparison to other PGF2 alpha-sensitive tissues. There were large differences in the maximum tension developed in response to the prostanoids tested, only PGF2 alpha, PGE2 and 6-keto PGF1 alpha were full agonists. Although the simplest explanation of these data was that the rat uterus contains a single novel type of prostanoid receptor, the existence of multiple receptor subtypes could not be disproved. Evidence from the effect of synthetic analogs suggested that neither thromboxane A2 nor PGI2 are physiological regulators of activity in this tissue.     

Profiling of bisenoic prostaglandins and thromboxane B2 in bronchoalveolar fluid from the lower respiratory tract of human subjects by combined capillary gas chromatography-mass spectrometry

Methods for the profiling of prostaglandin D2 (PGD2), prostaglandin E2 (PGE2), prostaglandin F2 alpha (PGF2 alpha), 15(S),9 alpha,11 beta-trihydroxyprosta-5Z,13E-dien-1-oic acid (9 alpha,11 beta-PGF2), 6-keto-prostaglandin F1 alpha (6kPGF1 alpha), and thromboxane B2 (TxB2) in bronchoalveolar lavage (BAL) fluids from human subjects by combined capillary gas chromatography-mass spectrometry are described. Aliquots (5 ml) of BAL fluid obtained using a standardized lavage protocol were extracted on octadecylsilyl silica cartridges after addition of 0.8 to 2.0 nanograms of tetradeuterated analogs of PGE2, PGF2 alpha, and 6kPGF1 alpha as internal standards. Eluted analytes and internal standards were prepared for vapor phase analysis by sequential reactions resulting in the formation of methyloxime-pentafluorobenzyl ester-trimethylsilyl ether derivatives. The derivatized analytes were detected by simultaneous monitoring of ions at six different masses characteristic for each of the derivatized prostanoids. The samples were of adequate purity for identification and quantitation of each of the prostanoids with detection limits of 0.1 to 0.2 picograms of each analyte per milliliter of BAL fluid. The time required for analysis of each sample was approximately 30 minutes. Standard curves of unlabeled species of the six prostanoids extracted after addition to BAL fluid were linear over a range from subpicogram to nanogram quantities. The differences between the amounts of prostanoid added and the amounts of prostanoid measured were typically less than 19%, and the intra-assay coefficients of variation for repeated measurements of a single sample were less than 20%. PGE2, PGD2, PGF2 alpha, and TxB2 were detectable in BAL fluids from normal subjects with levels of each of these compounds being less than 2.6 picograms/ml. BAL fluids from patients with lung disease presented qualitative and quantitative profiles of prostanoids markedly different than those from normal subjects. These analytical methods provide a basis for in vivo comparisons of prostanoid profiles in the lower respiratory tract of man and should be readily adaptable for use in a variety of clinical studies.     

Effect of alterations in the cyclopentane ring on bone resorptive activity of prostaglandin

Previous studies have shown that the natural prostanoids, PGE2, PGE1 and PGF2 alpha are potent stimulators of bone resorption. In this study, we have examined the effects of alterations in the cyclopentane ring of these prostanoids for their effect on the resorptive response of cultured long bones from 19-day fetal rats as measured by the release of previously incorporated 45Ca. Indomethacin (10(-6)M) was added to minimize endogenous prostaglandin production. In this system PGE2 and PGE1, the 9 keto, 11 alpha hydroxy compounds, were approximately equally effective at concentrations of 10(-8) to 10(-6) M. The 9 alpha hydroxy, 11 alpha hydroxy compound, PGF2 alpha, was active at 10(-7) to 10(-5) M. In contrast, the 9 alpha hydroxy, 11-keto compound, PGD2, showed only a minimal stimulation of bone resorption at 10(-5) M. While these data suggested that the 11 alpha hydroxy group was important for bone resorbing activity, 11 beta PGE2 and 11-deoxy PGE1 were only slightly less potent than their physiologic counterparts. Both 9 beta, 11 alpha PGF2 and 9 alpha, 11 beta PGF2 were less potent than PGF2 alpha but did cause substantial stimulation of bone resorption and were equally effective at 10(-6) to 10(-5) M. 9 alpha, 11 beta PGF2 alpha is of particular interest since it is major metabolite of PGD2. These results suggest that the binding of prostanoids to the receptor which mediates bone resorption is affected by changes at the 9 and 11 positions of the pentane ring but do not support the hypothesis that the 11 alpha OH function is essential for this biological activity.     

Prostaglandin E(2) and stem cell factor can deliver opposing signals to B lymphocyte precursors

The synthetic prostanoid, 16,16-dimethyl PGE(2), suppressed B lymphopoiesis in mice and proliferation of normal B cell precursors or the F10 pro-B cell line to interleukin 7 in culture. This was not the case with two other prostanoids, PGD(2) and PGF(2alpha), or agonists for PGI(2) agonist and thromboxane A(2) agonist receptors. PGE(2), but not the related prostanoids or agonists, induced apoptosis in F10 cells. The apoptotic response was mediated by the EP2 class of PGE(2) receptors and required an increase in intracellular cyclic adenosine 3',5'-monophosphate, activation of protein kinase A, and protein synthesis. The influence of PGE(2) on F10 cells was diminished in the presence of a cloned stromal cell line or stem cell factor. These findings describe another potential regulatory circuit in bone marrow which might influence B lymphopoiesis under disease or steady-state conditions.     

Decreased prostaglandin synthesis in postmortem cerebral cortex from patients with Alzheimer's disease.

The syntheses of prostaglandin (PG) F2 alpha, E2 and D2, and thromboxane (TX) B2 from [14C]arachidonic acid were studied in frontal cortex of human control and Alzheimer's disease (AD) brains using the microsomal fractions. Under the assay conditions employed, it was found that the major metabolite of [14C]arachidonic acid was PGE2 accounting for 63% of total prostanoid production; PGF2 alpha accounted for 21.5%, TXB2 for 9%, and PGD2 for 6.5%. When AD samples were compared to control samples, microsomal PG synthesis was significantly decreased, with reduced production of PGE2, PGF2 alpha and PGD2. Such decreases in AD brain seem unrelated to age, sex, postmortem delay and, as far as could be determined, antemortem state. In both control and Alzheimer groups, a history of anti-inflammatory therapy seemed to correlate with increased PG synthesis.     

Effects of selective and non-selective COX inhibitors on antigen-induced release of prostanoid mediators and bronchoconstriction in the isolated perfused and ventilated guinea pig lung

The contribution of cycloxygenase (COX)-1 and COX-2 in antigen-induced release of mediators and ensuing bronchoconstriction was investigated in the isolated perfused guinea pig lung (IPL). Antigen challenge with ovalbumin (OVA) of lungs from actively sensitised animals induced release of thromboxane (TX)A(2), prostaglandin (PG)D(2), PGF(2)(alpha), PGI(2) and PGE(2), measured in the lung effluent as immunoreactive TXB(2), PGD(2)-MOX, PGF(2)(alpha), 6-keto PGF(1)(alpha) and PGE(2), respectively. This release was abolished by the non-selective COX inhibitor flurbiprofen (10 microM). In contrast, neither the selective COX-1 inhibitor FR122047 nor the selective COX-2 inhibitor celecoxib (10 microM each) significantly inhibited the OVA-induced bronchoconstriction or release of COX products, except for PGD(2). Another non-selective COX inhibitor, diclofenac (10 microM) also significantly inhibited antigen-induced bronchoconstriction. The data suggest that both COX isoenzymes, COX-1 and COX-2 contribute to the immediate antigen-induced generation of prostanoids in IPL and that the COX-1 and COX-2 activities are not associated with different profiles of prostanoid end products.     

Synthetic pathways of gallbladder mucosal prostanoids: the role of cyclooxygenase-1 and 2.

Acute cholecystitis is associated with increased gallbladder prostanoid formation and the inflammatory changes and prostanoid increases can be inhibited by nonsteroidal anti-inflammatory agents. Recent information indicates that prostanoids are produced by two cyclooxygenase (COX) enzymes, COX-1 and COX-2. The purpose of this study was to determine the COX enzymatic pathway in gallbladder mucosal cells involved in the production of prostanoids stimulated by inflammatory agents. Human gallbladder mucosal cells were isolated from cholecystectomy specimens and maintained in cell culture and studied in comparison with cells from a well differentiated gallbladder mucosal carcinoma cell line. COX enzymes were evaluated by Western immunoblotting and prostanoids were measured by ELISA. Unstimulated and stimulated cells were exposed to specific COX-1 and COX-2 inhibitors. In both normal and transformed cells constitutive COX-1 was evident and in gallbladder cancer cells lysophosphatidyl choline (LPC) induced the formation of constitutive COX-1 enzyme. While not detected in unstimulated normal mucosal cells and cancer cells, COX-2 protein was induced by both lipopolysaccharide (LPS) and LPC. Unstimulated gallbladder mucosal cells and cancer cells produced prostaglandin E2 (PGE2) and prostacyclin (6-keto prostaglandin F1alpha, 6-keto PGF1alpha) continuously. In freshly isolated normal gallbladder mucosal cells, continuously produced 6 keto PGF1alpha was inhibited by both COX-1 and COX-2 inhibitors while PGE2 levels were not affected. Both LPS and LPC stimulated PGE2 and 6 keto PGF1alpha formation were blocked by COX-2 inhibitors in freshly isolated, normal human gallbladder mucosal cells and in the gallbladder cancer cells. The prostanoid response of gallbladder cells stimulated by proinflammatory agents is inhibited by COX-2 inhibitors suggesting that these agents may be effective in treating the pain and inflammation of gallbladder disease.     

Adaptive cytoprotection against alcohol injury in the rat stomach is not due to increased prostanoid synthesis

This study evaluated the effects of 25% ethanol, a mild irritant, on endogenous prostanoid synthesis in the rat stomach before and after exposure to oral 100% ethanol. Rats received water or 25% ethanol orally. After 15 min, a portion of each group was sacrificed and the remaining animals treated with 100% ethanol prior to sacrifice one minute later. Microsomal membrane fractions were prepared from the glandular gastric mucosa in all groups and incubated with 14C arachidonic acid in the presence of cofactors. Endogenous mucosal prostanoid synthesis was analyzed by radiochromatography and results correlated with the presence or absence of gastric injury macroscopically. Prostanoids measured included PGI2, PGF2 alpha, PGE2, PGD2, PGA2, and thromboxane A2. Additional experiments were performed in like manner to those just described with the exception that indomethacin (5 mg/kg intraperitoneally) pretreatment was rendered. Stomachs exposed to water or 25% ethanol alone demonstrated a modest and equivalent level of synthesis of all prostanoids measured. Exposure to 100% ethanol (with and without mild irritant pretreatment) significantly increased prostanoid synthesis (especially PGI2, PGF2 alpha, and PGE2) compared with stomachs exposed to water or 25% ethanol alone; only mild irritant treated mucosa was protected from injury by 100% ethanol. Indomethacin pretreatment reversed the increased prostanoid synthesis in mucosa exposed to 100% ethanol, with or without mild irritant pretreatment, and partially reversed the protective effect of 25% ethanol. Other experiments using tissue slices in which perturbations in mucosal levels of prostanoids were measured by radioimmunoassay under identical experimental conditions exhibited similar results. These data dispute the notion that adaptive cytoprotection is mediated by increased endogenous prostanoid synthesis. The partial reversal of this process by indomethacin was most likely secondary to some other action of this agent, such as a reduction in gastric blood flow, rather than direct effects on prostanoid synthesis.     

Profile of eicosanoids produced by human saphenous vein endothelial cells and the effect of dietary fatty acids

Human saphenous vein endothelial cells (HSVECs) derived from primary cultures of adult human veins constitute an excellent in vitro model for studying human endothelial metabolism. In this study we report the (14)C-labelled prostanoid profile of HSVECs under resting and stimulated conditions and the effect of the n-3 polyunsaturated fatty acids eicosapentaenoic acid and docosahexaenoic acid on them. Results indicate that HSVECs while under resting conditions produce mainly prostaglandin F(2alpha)(PGF(2alpha)). After stimulation with calcium ionophore A23187, the cells were found to synthesise PGI(2), PGE(2)and PGF(2alpha) as major products and thromboxane B(2)and PGD(2)as minor products. Production of (14)C-labelled hydroxyeicosatetraenoic acids was not detected. Eicosapentaenoic acid was found to inhibit basal and stimulated prostanoid production whereas docosahexaenoic acid inhibited basal but strongly increased stimulated prostanoid production. These results may offer the basis for further studies aiming to investigate targets for pharmacological intervention in inflammatory conditions.     

Altered prostanoid formation in human umbilical vasculature in response to variations in oxygen tension

Prostanoid formation in human umbilical vessels perfused in vitro was assessed at different oxygen tensions. At an atmosphere of 5% oxygen the production rate of prostacyclin (measured as 6-keto-PGF1 alpha) was higher, while those of thromboxane A2 (measured as TXB2), PGE2 and PGF2 alpha were lower than with 20%, 50% and 95% oxygen. The stimulatory effect of angiotensin II on prostanoid production was found to be independent on the prevailing oxygen tension. Vascular formation of prostanoids thus seems to be at least partially affected by the ambient oxygen tension. Though altered oxygen tension does not seem to affect angiotensin induced prostanoid formation, the action of other vasoactive agents influencing vascular formation of prostanoids may respond differently to hypoxia or hyperoxia.     

Prostaglandin production by phagocytic cells of the mouse thymic reticulum in culture and its modulation by indomethacin and corticosteroids

The production of prostaglandins by phagocytic cells of the thymic reticulum in culture (P-TR) was studied by using high pressure liquid chromatography and radioimmunoassay. Radioimmunologic determinations showed that thromboxane B2 (TXB2), prostaglandin E2 (PGE2), and 6-keto-prostaglandin F1 alpha (6 keto-PGF1 alpha) were the major compounds released into the culture medium, whereas prostaglandin F2 alpha (PGF2 alpha) was only a minor component. Indomethacin and dexamethasone exerted a similar pattern of differential inhibition of the secretion of prostanoids. PGE2 and 6-keto PGF1 alpha productions were markedly decreased by these anti-inflammatory drugs, whereas those of TXB2 and PGF2 alpha were not or were only slightly affected. Experiments performed with an antiglucocorticoid compound (RU 38486) showed that the steroid-induced inhibition of prostanoid secretion is a classical receptor-mediated action. These results demonstrated that phagocytic cells of the thymic reticulum, which resemble the thymic interdigitating cells, produce several types of prostaglandins. Because it has been described that P-TR regulate thymocyte proliferation in vitro via the secretion of both interleukin 1 and PGE2, these results suggest that anti-inflammatory agents may be able to modulate the thymic microenvironment and, consequently, thymocyte proliferation.     

Liquid chromatographic-mass spectrometric determination of cyclooxygenase metabolites of arachidonic acid in cultured cells

A liquid chromatographic-electrospray ionization-mass spectrometric (LC-ESI-MS) technique was developed to simultaneously determine the cyclooxygenase metabolites of arachidonic acid (6-keto-PGF(1alpha), PGD(2), PGE(2), PGF(2alpha), and PGJ(2)) produced by cultured cells. Samples were separated on a C(18) column with water-acetonitrile mobile phase, ionized by electrospray, and detected in the positive mode. Selected ion monitoring (SIM) of m/z 353, 335, 335, 319, and 317 were used for quantifying 6-keto-PGF(1alpha), PGD(2), PGE(2), PGF(2alpha), and PGJ(2), respectively. Prostaglandins were detected at concentrations as low as 1 pg (S/N=3) on the column. The method was used to determine the production of PGs from bovine coronary artery endothelial cells (ECs) and human prostate cancer cells (PC-3) with different degree of invasiveness. Bradykinin (10(-6) M) stimulated a marked increase in the production of 6-keto-PGF(1alpha), PGE(2), and PGF(2alpha) and a small increase of PGD(2) by ECs. 6-Keto-PGF(1alpha) was the major metabolite in these cells. The production of PGE(2) was threefold higher and PGD(2) was twofold higher in PC-3-S (invasive) cells than in PC-3-U (non-invasive) cells.     

High glucose levels modulate eicosanoid production in uterine and placental tissue from non-insulin-dependent diabetic rats during late pregnancy

Severe uterine and placental disturbances have been described in diabetes pathology. The relative severity of these changes appears to correlate with high glucose levels in the plasma and incubating environment. In order to characterize changes in eicosanoid production we compared uterine and placental arachidonic acid conversion from control and non-insulin-dependent diabetes mellitus (NIDDM) rats on day 21 of pregnancy, into different prostanoids, namely PGE2, PGF22alpha, TXB2 (indicating the production of TXA2) and 6-keto-PGF1 (indicating the generation of PGI2). PGE2, PGF2alpha and TXB2 production was higher and 6-keto-PGF1alpha was similar in diabetic compared to control uteri. PLA2 activity was found diminished in the NIDDM uteri in comparison to control. A role for PLA2 diminution as a protective mechanism to avoid prostaglandin overproduction in uterine tissue from NIDDM rats is discussed. Placental tissues showed an increment in TXB2 generation and a decrease in 6-keto PGF1alpha level in diabetic rats when compared to control animals. Moreover, when control uterine tissue was incubated in the presence of elevated glucose concentrations (22 mM), similar generation of 6-keto PGF1alpha and elevated production of PGE2, PGF2alpha and TXB2 were found when compared to those incubated with glucose 11 mM. Placental TXB2 production was higher and 6-keto PGF1alpha was lower when control tissues were incubated in the presence of high glucose concentrations. However, high glucose was unable to modify uterine or placental prostanoid production in diabetic rats. We conclude that elevated glucose levels induced an abnormal prostanoid profile in control uteri and placenta, similar to those observed in non-insulin-dependent diabetic tissues.     

Release of eicosanoids from cultured rat aortic endothelial cells; studies with arachidonic acid and calcium ionophore A23187

The release of prostanoids from the three different vascular cell types derived from rat aortic explants has been studied in vitro. Under resting conditions and when incubated with exogenous arachidonic acid (AA, 10 microM), the endothelial cells (EC) produced the highest concentration of prostacyclin (PGI2 PGE2 PGF2 alpha TxA2). In contrast, PGE2 was the major prostanoid produced by the smooth muscle cells and fibroblasts. Pretreatment of EC with aspirin (10 microM) or indomethacin (10 microM) effectively inhibited the production of prostanoids by these cells. Incubation with the calcium ionophore A23187 (10 microM) did not stimulate production of PGI2 or leukotriene B4 (LTB4) by EC. However, treatment of EC with a combination of A23187 and AA led to production of amounts of both PGI2 and LTB4 which were greater than the summed values for the different drug treatments. These findings indicate that the concentration of substrate, AA, is a limiting factor in prostanoid formation by these cultured vascular cells but that rat EC are relatively poor in the enzymes required for leukotriene formation.     

Prostaglandin and thromboxane synthesis by tissue slices from human aortic aneurysms

Sliced portions of the walls of human aortic aneurysms were incubated with extracts of human plasma and serum to determine the profile of prostanoid production. 6-Oxo-prostaglandin (PG) F1 alpha, PGE2, PGF2 alpha, and thromboxane (TX) B2 were measured by gas chromatography/electron capture mass spectrometry. 6-Oxo-PGF1 alpha, the stable hydrolysis product of PGI2, was the major cyclooxygenase product but substantial amounts of TXB2 (the hydrolysis product of TXA2), with smaller amounts of PGE2 and PGF2 alpha were also synthesised. These prostanoids could contribute to the response of the vascular wall to injury, thereby influencing the disease process. Serum extracts stimulated PGI2 and TXA2 synthesis, probably as a result of their Ca2+ content.     

Prostaglandin D2 induces programmed cell death in Trypanosoma brucei bloodstream form

African trypanosomes produce some prostanoids, especially PGD2, PGE2 and PGF2alpha (Kubata et al. 2000, J. Exp. Med. 192: 1327-1338), probably to interfere with the host's physiological response. However, addition of prostaglandin D2 (but not PGE2 or PGF2alpha) to cultured bloodstream form trypanosomes led also to a significant inhibition of cell growth. Based on morphological alterations and specific staining methods using vital dyes, necrosis and autophagy were excluded. Here, we report that in bloodstream form trypanosomes PGD2 induces an apoptosis-like programmed cell death, which includes maintenance of plasma membrane integrity, phosphatidylserine exposure, loss of mitochondrial membrane potential, nuclear chromatin condensation and DNA degradation. The use of caspase inhibitors cannot prevent the cell death, indicating that the process is caspase-independent. Based on these results, we suggest that PGD2-induced programmed cell death is part of the population density regulation as observed in infected animals.     

Proliferative effect of PGD2 on osteoblast-like cells; independent activation of pertussis toxin-sensitive GTP-binding protein from PGE2 or PGF2 alpha.

PGD2 stimulated DNA synthesis and decreased alkaline phosphatase activity dose-dependently between 10 nM and 10 microM in osteoblast-like MC3T3-E1 cells. PGD2 had little effect on cAMP production, but caused very rapid enhancement of phosphoinositide (PI) hydrolysis dose-dependently between 10 nM and 10 microM. The formation of inositol trisphosphate (IP3) induced by PGD2 reached the peak within 1 min and decreased thereafter, which is more rapid than that induced by PGE2 or PGF2 alpha and both PGE2 and PGF2 alpha affected PGD2-induced IP3 formation additively. Pertussis toxin (PTX) inhibited both PGD2-induced formation of inositol phosphates and DNA synthesis. The degree of these PTX (1 micrograms/ml)-induced inhibitions was similar. In addition, neomycin, a phospholipase C inhibitor, inhibited PGD2-induced DNA synthesis as well as the formation of IP3, and the patterns of both inhibitions were similar. In the cell membranes, PTX-catalyzed ADP-ribosylation of a 40-kDa protein was significantly attenuated by pretreatment of PGD2. Time course of the attenuation of PTX-catalyzed ADP-ribosylation by PGD2 was apparently different from that by PGE2 or PGF2 alpha. These results indicate that PGD2 activates PTX-sensitive GTP-binding protein independently from PGE2 or PGF2 alpha and stimulates PI hydrolysis resulting in proliferation of osteoblast-like cells.     

Prostanoid synthesis by vascular slices and cultured vascular cells of piglet aorta

Biosynthesis of prostanoids was studies in vascular slices of human umbilical arteries, piglet aorta and vena cava as well as in cultured vascular cells of piglet aorta. After preincubation with radioactive labeled arachidonic acid, prostanoids in the incubation media of slices or cultured cells were measured by radioimmunoassay or by radioactivity determination of labeled compounds following separation on reserved-phase high performance liquid chromatography. In all vascular slices 6-keto-PGF1α was the main prostanoid found, followed by PGE2. Thromboxane B2 and PGF2α were also formed, but only in trace amounts. In cultured cells taken from the three layers of the vascular wall, the prostanoid profiles differed markedly from those obtained from vascular slices. Each cell strain showed a specific prostanoid pattern. Endothelial cells synthesized predominantly 6-keto-PGF1α and PGF2α. In smooth muscle cells no 6-keto-PGF1α could be detected; here the predominant prostanoid was PGE2. PGF2α was formed in smaller quantities. Fibroblasts synthesized all prostanoids (PGE2, PGF2α, TXB2, 6-keto-PGF1α), PGE2 and PGF2α being the major products. In vascular slices and in cultured endothelial cells, the predominant prostacyclin derivative detected was 6-keto-PGF1α; the enzymatic PGI2-metabolite, 6,15-diketo-PGF1α, could be detected only in piglet vena cava slices in small amounts.     

Role of COX-1 and COX-2 on skin PGs biosynthesis by mechanical scratching in mice

We examined the involvement of cyclooxygenase (COX)-1 and COX-2 on mechanical scratching-induced prostaglandins (PGs) production in the skin of mice. The dorsal regions of mice were scratched using a stainless brush. COXs expressions in the skin were analyzed using real-time PCR and Western blotting. The effect of acetylsalicylic acid (ASA) on the ability of PGs production were determined based on skin PGs level induced by arachidonic acid (AA) application. Mechanical scratching increased PGD2, PGE2, PGI2 and PGF(2 alpha). COX-1 was constitutively expressed and COX-2 expression was enhanced by scratching. Intravenous administration of ASA inhibited PGs biosynthesis in the normal skin. PGs levels of the skin 6h after ASA administration (ASA 6 h) were almost equal to those of the skin 10 min after ASA administration (ASA 10 min). In the scratched skin, AA-induced PGE2 and PGI2 of ASA 6 h were significantly higher than those of ASA 10 min. The skin PGD2 and PGF(2 alpha) of ASA 10 min were almost same to those of ASA 6 h. In the normal skin of COX-1-deficient mice, skin PGD2 level was lower than that of wild-type mice, although PGE2, PGI2 and PGF(2 alpha) levels were almost equal to those of wild type. In the scratched skin of COX-1-deficient mice, PGD2, PGE2, PGI2 and PGF(2 alpha) levels were lower than those of wild-type mice. These results suggested that cutaneous PGD2 could be mainly produced by COX-1, and PGE2 and PGI2 could be produced by COX-1 and COX-2, respectively, in mice.