Estrogen and uterine sensitization for the decidual cell reaction in the rat: role of prostaglandin E2 and adenosine 3':5'-cyclic monophosphate

The possibility that estrogen affects uterine sensitization for decidualization by altering the ability of E-series prostaglandins (PGs) to increase adenosine 3':5'-cyclic monophosphate (cAMP) concentrations was investigated. To determine if increased endometrial vascular permeability, a response which precedes decidualization, could be obtained in nonsensitized uteri by treatments designed to increase endometrial intracellular cAMP concentrations, cholera toxin, an activator of adenylate cyclase, was injected into the uterine lumen of immature rats pretreated with progesterone and either 0, 0.5 or 10 micrograms estrone with indomethacin to inhibit endogenous PG synthesis. Endometrial vascular permeability, determined using 125I-labeled bovine serum albumin, was assessed 8 h later. Cholera toxin produced a dose-dependent increase in endometrial vascular permeability in all groups; the uteri of rats pretreated with the optimal hormone regimen (0.5 micrograms estrone plus 2 mg progesterone) responded to a lower dose of the toxin. As determined by uterine weights and histologic examination 5 days after the intrauterine administration of cholera toxin or its vehicle, the toxin induced decidualization in rats pretreated with progesterone and 0 or 0.5 micrograms estrone, but not in those receiving 10 micrograms estrone. Cholera toxin had no detectable effect on uterine cAMP concentrations in animals sacrificed 15 min or 3 h after intrauterine treatment. The intrauterine injection of 8-Br-cAMP, with or without 3-isobutyl-1-methyl-xanthine, did not increase endometrial vascular permeability in indomethacin-treated animals pretreated with the different hormone regimens.(ABSTRACT TRUNCATED AT 250 WORDS)     

Potentiation of bradykinin-induced vascular permeability increase by prostaglandin E2 and arachidonic acid in rabbit skin

The activity of prostaglandins (PG) in producing vascular permeability was quantitated by dye extraction method in skin of anaesthetized rabbits. PGE₁ and PGE₂ (0.01–10 μg) produced increase in vascular permeability. Activity was approximately equal to that of histamine (Hist) and $\text{1}\text{20}$ of that of bradykinin (BK) on a weight basis. The activity of PGF_1α and PGF_2α was only $\text{1}\text{20}$ of that of PGE₁ or PGE₂.In spite of the relatively low potency of PGE₁ and PGE₂ in the rabbit, near threshold doses (0.1 or 1 μg) of PGE₂ could potentiate permeability responses to bradykinin (0.1 μg) by 10 or 100-fold, respectively. Equivalent doses (0.1 or 1 μg) of histamine could not potentiate the bradykinin responses. Arachidonic acid (AA) at 1 μg, produced a 10-fold potentiation in the permeability response to bradykinin (0.1 μg). Pretreatment of the rabbits with indomethacin (20 mg/kg, i.p.) reduced the responses of BK (0.1 μg) + AA (1 μg) down to a similar magnitude of those seen with bradykinin alone. However, indomethacin did not block responses to either, BK alone, BK + PGE₂, or BK + Hist. Various doses (1, 10, 100 and 300 μg) of arachidonic acid alone also produced increase in cutaneous vascular permeability, although its potency was only $\text{1}\text{3}$–$\text{1}\text{8}$ of that of PGE₂. This activity of arachidonic acid was attributed in part to its bioconversion to PGE₂, since its activity was significantly reduced by the prostaglandin antagonist, diphloretin phosphate (DPP) (60 mg/kg, i.v.) and by indomethacin (20 mg/kg, i.p.), which blocks conversion of arachidonic acid to prostaglandins. Arachidonic acid may owe some of its permeability increaseing effects to histamine release, since its effects were also reduced by the antihistamine, pyrilamine (2.5 mg/kg, i.v.).     

Prostaglandin E2, adenosine 3':5'-cyclic monophosphate and changes in endometrial vascular permeability in rat uteri sensitized for the decidual cell reaction

The possibility that prostaglandin E2 (PGE2) increases endometrial vascular permeability and initiates decidualization in sensitized rat uteri by stimulation of adenosine 3':5'-cyclic monophosphate (cAMP) synthesis was investigated. Immature rats, pretreated so that they were sensitized for the decidual cell reaction, were used. Following the unilateral intrauterine injection of 50 microliters phosphate-buffered saline containing gelatin (PBS-G), a deciduogenic stimulus, uterine concentrations of both PGE and cAMP were elevated as early as 1 min after the intrauterine treatment. To determine if uterine stimuli which increase endometrial vascular permeability also increase uterine cAMP concentrations, rats, treated with or without indomethacin, an inhibitor of PG synthesis, received unilateral intrauterine injections of 50 microliters PBS-G with and without 10 micrograms PGE2 and were killed 15 min later. Uterine cAMP concentrations were elevated in all injected horns except in those of indomethacin-treated rats receiving PBS-G intraluminally, thus paralleling the expected changes in endometrial vascular permeability. As indicated by radioactivity levels in the stimulated horn 15 min after the i.v. injection of 125I-labeled bovine serum albumin, the intrauterine injection of dibutyryl cAMP, with or without theophylline, did not increase endometrial vascular permeability in indomethacin-treated animals. In contrast, cholera toxin, an activator of adenylate cyclase activity, markedly elevated permeability and induced decidualization. Except for the lack of a permeability response to the cAMP analogue, these data are consistent with the hypothesis that the effect of PGE2 on endometrial vascular permeability is mediated by cAMP.     

Secretion of plasminogen activator by cultured rat endometrial stromal cells from uteri differentially sensitized for the decidual cell reaction

Endometrial stromal cells from rat uteri differentially sensitized for the decidual cell reaction in vivo and which undergo differing degrees of decidualization in vitro were cultured and plasminogen activator (PA) in the medium determined. The cells were obtained by enzymatic dispersion from the uteri of ovariectomized, steroid-treated rats at the equivalent of day 4, 5, or 6 of pseudopregnancy or on day 5 from rats treated on day 4 with 0, 0.3, or 1.0 μg estradiol (low, intermediate, or high dose of estradiol, respectively) and cultured for 24, 48, or 72 hr. For cells from day 4, 5, and 6 uteri cultured under control conditions, PA activity in the medium was greatest for day 5 cells, which were from uteri maximally sensitized for decidualization both in vivo and in vitro. By contrast, for cells from low-, intermediate-, and high-estradiol uteri, PA activity in the medium was greatest for the high-estradiol cells; these cells do not undergo decidualization in vivo or in vitro to the same extent as intermediate-estradiol cells. Indomethacin, an inhibitor of prostaglandin (PG) synthesis, reduced PGE₂ accumulation to nondetectable amounts and for most cultures decreased PA activity in the medium, suggesting that endogenous PG production regulated in part PA secretion under control conditions. The addition of PGE₂ with indomethacin increased PA activities above those under control conditions, but activities were still lower for day 4 and 6 cells compared with day 5 cells, and for low- and intermediate-estradiol cells compared with high-estradiol cells. This indicates that the differences in PA secretion are not explainable by differences in PGE₂ production. Northern blot analysis of RNA from cells cultured for 72 hr under control conditions did not reveal significant differences in steady-state concentrations of mRNA for urokinase-type PA or plasminogen activator inhibitor 1, but those for tissue-type PA were lower in day 6 cells compared with day 4 and 5 cells. It is concluded that PA activity secreted by the cultured endometrial stromal cells, although controlled in part by the endocrine milieu to which they were exposed prior to culture, does not simulate decidualization in vitro and, therefore, that PA activity is not a marker for decidualization in vitro. Mol. Reprod. Dev. 49:268–276, 1998. © 1998 Wiley-Liss, Inc.     

Tissue differences in vascular permeability induced by leukortriene B4 and prostaglandin E2 in the rat

The activity of synthetic LTB₄ and PGE₂, in increasing vascular permeability was tested simultaneously in seventeen different organs in the rat. Rats were injected in the aortic arch through a cannula in the carotid artery with ¹²⁵-I-albumin, ⁵¹Cr-erythrocytes, and ⁵⁷Co-EDTA. The rats were then injected through the carotid artery cannula with LTB₄, PGE₂ or a combination of LTB₄ and PGE₂. Eight minutes later the rats were killed and the activity of ¹²⁵I, ⁵¹Cr, and ⁵⁷Co measured in different organs. Changes in vascular permeability were infered from changes in the ratios of the isotope activities. LTB₄ (15 μg/kg) induced enhanced permeability in caecum, small bowel, skin, fat pad, stomach, pancreas, and aorta, but not in the heart, brain, colon, testes, diaphragm, forelimb, cremaster muscle, lung, kidney or eye. A lower dose of LTB₄, 3 μg/kg, enhanced vascular permeability in caecum, small bowel, skin, stomach, and aorta. PGE₂ (1 μg/kg) enhanced vascular permeability only in the caecum. A combination of LTB₄ (3 μg/kg) and PGE₂ (1 μg/kg) was more potent than either alone. Rats depleted of neutrophils with anti-neutrophil serum were less sensitive to LTB₄ than intact rats. These findings suggest that the vasculatures of different tissues in the rat vary markedly in their susceptibility to LTB₄ induced increases in permeability.     

The synthesis and biological activities of some 12-Aza-prostaglandin analogues

12-Aza-prostaglandin (PG) analogues containing the pyrrolidine-2,4-dione ring system have been synthesized from ,2-disubstituted glycine esters via cyclisation of their -ethoxycarbonylacetyl derivatives. 5-(6-Carboxyhexyl)-1-octylpyrrolidine-2,4-dione (5) had little or no PG-like activity on superfused intestinal or vascular smooth muscle preparations but it selectively antagonised smooth muscle responses to PGE₂, PGE₁, PGF_2α and PGA₂in vitro. At a concentration of 10⁻⁵ g/ml it reduced responses of the rat stomach strip to PGE₂ by over 80% but did not affect responses of this tissue to acetylcholine, 5-hydroxytryptamine (5-HT) or bradykinin. Polyphloretin phosphate (PPP), the known PG antagonist, had a similar effect at the same concentration (10⁻⁵ g/ml).5-(6-Carboxyhexyl)-1-(3-hydroxyoctyl)pyrrolidine-2,4-dione (12) had the same profile of activity on superfused smooth muscle preparations as PGE₂ or PGA₂. On intravenous injection into anaesthetised rats it caused dose-dependent falls in arterial blood pressure with associated tachycardias, which is typical of the response to PGE₂. The smooth muscle activity of (12) was not reduced by passage through isolated perfused guinea-pig lungs nor was its potency as a vasodepressor increased when given intra-arterially to rats. These results suggest that, unlike PGE₂, this analogue is not removed by the pulmonary circulation.     

Follicle stimulating hormone and prolactin release induced by prostaglandins in rat

Ten to 60 minutes following a single i.v. injection of PGE₂ (500 μg/rat) into male rats of 30 to 35 days of age FSH concentration in the serum was raised significantly. The rise in FSH was maintained from 10 to 60 minutes after treatment, then at 90 minutes FSH had declined and was not significantly different from that of the control before treatment. Prostaglandin E₁, E₂ or F_2α (670μg/rat) significantly increased the serum prolactin level 10 to 60 minutes after a single i.v. injection in spayed rats primed with estrogen and progesterone. And, rats primed with estrogen and progesterone. And, increases in prolactin in the serum were observed with as little as 2μg of PGE₁ or E₂, and 20μg of PGF_2α. Twenty μg of PGE₂, and 200μg of PGE₁ or F_2α gave the maximum stimulation. These results indicate that release of pituitary hormones is affected by prostaglandins.Prostaglandins (PGs) are widely distributed in mammalian tissues, and they have been reported to have an almost equally wide variety of endocrine and metabolic effects. It was recently postulated that PGs may be involved in the process of ovulation because ovulation was blocked by inhibitors of PG synthesis (1–5).     

Differential PGE2 and cAMP production by allogeneic and syngeneic pregnant mice uteri

Prostaglandin E₂ (PGE₂) and cAMP production in allogeneic and syngeneic pregnant mice uteri, were measured in relation to the ratio of plasma estrogen/progesterone levels. PGE₂ generation by allopregnant uteri varied with the days of pregnancy. The increment of the prostanoid coincided with the increase in plasma estrogen concentration, whereas the decrement of its production was in parallel with the increment of plasma progesterone. The syngeneic pregnant uterus was unable to increase the release of PGE₂ above basal values during the whole pregnancy. The rise of PGE₂ production by the allogeneic pregnant uterus was correlated with an increase in cAMP levels. It is proposed that the pregnant mouse uterus increases its ability to release PGE₂ in response to an ovarian steroids.     

Further studies on biological activities of new peptide fragments derived from high molecular weight kininogen: An enhancement of the Vascular permeability increase of the fragments by prostaglandin E2

The activities of newly isolated peptide fragments, Fragment 1·2, Fragment 1 and Fragment 2, in vascular permeability were studied in rabbit skin using the pontamine sky blue leakage method. Fragment 1·2 and Fragment 2 showed an activity of about 1/100 of bradykinin, and almost equipotent to that of prostaglandin E₂ or histamine on a molar basis. Fragment 1 showed almost negligible activity. Mixtures of each of these fragments with bradykinin did not exceed the additive activities of either agent. However, the vascular permeability of each fragment was significantly enhanced by mixing with PGE₂, indicating that the activity was more than ten-fold of the fragment alone.     

Prostaglandin I2 as a potentiator of acute inflammation in rats

Prostaglandin I₂ potentiated the paw swelling induced by carrageenin in rats. Prostaglandin I₂ (0.1 μg) showed similar activity to PGE₁ (0.01 μg). This potentiating property disappeared in 60 minutes and was completely abolished by diphenhydramine (25 mg kg⁻¹, i.p.). In vascular permeability tests, PGI₂ itself (2.5 × 10⁻¹⁰ mol, 88 ng) caused no dye leakage reaction, but PGE₁ (2.5 × 10⁻¹⁰ mol, 88.5 ng) caused a significant dye leakage. This effect of PGE₁ was statistically significant compared with vehicle- or PGI₂-treated group (p<0.05). Prostaglandin I₂ potentiated the increased vascular permeability induced by 5-hydroxytriptamine (2.5 × 10⁻¹⁰ mol), bradykinin (5 × 10⁻¹⁰ mol) and histamine (2 × 10⁻¹⁰ to 2 × 10⁻⁸ mol). The potentiation was the most evidence in the case of histamine.     

Changes in morphine self-administration in rats induced by prostaglandin E1 and naloxone

Interactions of prostaglandin E₁ (PGE₁) with morphine have been reported in several test systems and an hypothesis has been advanced for a role of prostaglandins in morphine analgesia and physical dependence. In rats self-administering morphine intravenously, a simultaneous and continuous infusion of naloxone hydrochloride at 56 to 560 μg/kg/day caused the expected increase in injection rate for morphine. Infusion of PGE₁ by itself at 56 or 180 μg/kg/day had no effect on the rate of morphine intake. Likewise the addition of PGE₁ at 180 μg/kg/day did not potentiate the increase caused by naloxone (56 or 180 μg/kg/day) when it was added to the naloxone infusion. These results do not support a role for prostaglandins in the behavioral aspects of morphine addiction. However, larger doses of PGE₁ (1 and 1.8 mg/kg/day), which were without overt effects in normal rats, caused severe and incapacitating prostration in morphinized rats.     

Enhancement by levamisole of the contractions induced by prostaglandin E2 in the guinea-pig isolated ileum

The effect of levamisole on prostaglandin E₂ (PGE₂)-evoked contractions was studied on guinea-pig isolated ileum. Addition of levamisole (10 μg/ml) to the organ bath produced a pronounced increase in the amplitude of the PGE₂-evoked responses. Levamisole (10 μg/ml) also sensitized the guinea-pig isolated ileum to 5-hydroxytryptamine and bradykinin, but not to histamine. The effect of the levamisole was not due to stimulation of autonomic ganglia or cholinergic activity since it was unaffected by hexamethonium or atropine, but it was prevented by indomethacin.     

Prostaglandin E2 enhances uterine stromal cell alkaline phosphatase activity

Prostaglandins have been implicated in the process of uterine decidualization , but sites of action are uncertain. Since one of the earliest changes in endometrial stroma following induction of decidualization is an increase in alkaline phosphataseactivity, we have investigated the effects of PGs on stromal cell alkaline phosphatase activity . Immature rats were pretreated with hormones to sensitize their uteri for the decidual cell reaction. Endometrial stromal cells were isolated and cultured for up to 4 days with PGE₂ (0–10 μg/ml) or PGF₂ (0–10 μg/ml) Analysis of variance revealed a highly significant interaction between day of culture and concentration of PGE₂ in medium (P<0.01). Stromal cell alkaline phosphatase activity decreased significantly with increasing culture duration (P<0.01). In the presence of PGE₂, alkaline phosphatase activity was significantly higher (P<0.01) regardless of day of culture. In contrast, PGF_2α had only a small and inconsistent effect. These data indicate that PGs, and in particular PGE₂, can act directly upon stromal cells.     

Dissimilarity between prostaglandin E1 and nitric oxide donors as potentiators of plasma exudation in the rabbit skin in vivo

The ability of prostaglandin E₁ (PGE₁) and nitric oxide (NO) donor compounds such as sodium nitroprusside (SNP), glyceryl trinitrate (GTN), and 3-morpholino-sydnonimine (SIN-1) to modulate the histamine- and bradykinin-induced increase in microvascular permeability have been investigated in rabbit skin. The effect of the NO synthesis inhibitor N^ω-nitro- -arginine methyl ester ( -NAME) on the plasma exudation induced by histamine and bradykinin was also studied. Local edema formation was evaluated using [¹²⁵I]human serum albumin. New Zealand white rabbits received an intravenous injection of [¹²⁵I]human albumin followed immediately by the intradermal injection of edematogenic agents into the shaved dorsolateral skin. PGE₁ (0.1 nmol/site) significantly potentiated both histamine- and bradykinin-induced edema. In contrast, SNP (0.4–400 nmol/site), SIN-1 (0.4–400 nmol/site), and GTN (0.4–40 nmol/site) did not affect the edematogenic response induced by either histamine or bradykinin. GTN (0.4–40 nmol/site) also had no effect on the increase in plasma exudation induced by histamine and bradykinin in the presence of PGE₁. -NAME (50–400 nmol/site), but not its enantiomer -NAME, dose-dependently reduced the edema formation induced by a combination of either histamine or bradykinin with PGE₁. This inhibition was significantly reversed by SNP (4–400 nmol/site) and by high doses (2.5 μmol/site) of -arginine (but not by -arginine). Our results thus demonstrate that PGE₁, but not nitrovasodilators, can actually potentiate histamine- and bradykinin-induced edema in rabbit skin. This discrepancy cannot be explained by the lack of vasodilator activity of the nitrovasodilators since these were able to reverse the -NAME-induced inhibition of the edema provoked by histamine. Rather, this difference most likely reflects the ability of PGE₁ to modulate vascular permeability by mechanism(s) other than an increase in arterial flow.     

Studies on the involvement of prostaglandins and their precursors in the rejection of Nippostrongylus brasiliensis from the rat

In vitro incubation of 6-day Nippo-strongylus brasiliensis in the presence of PGE₁ at 1000 ng/ml and PGE₂ at 500–10,000 ng/ml of medium did not affect worm motility nor in vivo survival of worms implanted into the small intestine of recipient rats. The intraduodenal injection of 250 and 500 μg PGE₁ or PGE₂ did not lead to expulsion of worms from infected rats. An in vitro exposure to precursor fatty acids of PGE₁ and PGE₂, dihomo-γ-linolenic acid and arachidonic acid, respectively, at concentrations of 1000–15,000 ng/ml of medium also failed to inhibit worm motility and in vivo worm survival. These results are at variance with some earlier reports and do not suggest that prostaglandins are directly involved in the immune rejection of N. brasiliensis. No prostaglandins could be demonstrated in worm homogenates.     

Prostaglandin F2α-induced stimulation of estrone and estradiol-17β secretion in cattle

Four of 5 Holstein heifers given intra-ovarian injections of 300 μg of prostaglandin F_2α (PGF_2α) showed transient, but statistically significant, depressions in plasma progesterone levels which returned to near normal levels within 24 hr. The same 4 animals also exhibited significant elevations in plasma estrone and estradiol-17β levels during the initial 24 hr. period following treatment, although no animals were observed in estrus during this time. Plasma levels of progesterone, estrone, estradiol-17β and PGF showed little change in control heifers receiving intra-ovarian injections of the buffer solution used as a vehicle for PGF_2α. It is concluded that PGF_2α stimulates estrogen secretion, presumably by follicular elements of the ovary.     

Characteristics of prostaglandin E1 potentiation of inflammatory activity of some agents

In rats pretreated with indomethacin, injection of PGE₁ (prostaglandin E₁) with carrageenan potentiated the carrageenan paw oedema. This effect of PGE₁, was maximal when it was injected together with carrageenan, there being a reduction in the action of PGE₁ if carrageenan injection was delayed after PGE₁ injection. PGE₁ induced potentiation of increase in plasma protein leakage induced by intradermal injections of bradykinin and histamine also depended on the injection of PGE₁ along with these agents. Thus oedema enhancement by PGE₁ differs from its action in pain, where PGs cause a long lasting sensitization of the injected area for the actions of other algesics. Since vasodilation may be a mechanism of oedema enhancement by PGs, the ability of adenosine and papaverine to mimic PGE₁ in paws and skins of rats were examined. Adenosine was active whereas papaverine was inactive in this respect. To clarify this difference, the vasodilatory properties of PGE₁, adenosine and papaverine were assessed by their ability to antagonize NA response in perfused rat mesenteric blood vessels. Only papaverine was effective in antagonising the NA response. Thus, PGE₁ and adenosine which potentiated the oedema inducing actions of other agents showed no vasodilatory properties and papaverine, a vasodilator, had no oedema potentiating actions.     

Interactions of prostaglandin E1, bradykinin and histamine and the increase of plasma fibrinogen in rats

Considering that tissue injury caused by laparotomy significantly increases the liver synthesis of plasma fibrinogen, and that PGE₁, bradykinini and histamine are released into the injured tissues, the effect of above mentioned inflammatory agents and of adrenal medulla on plasma fibrinogen levels in rats was studied. The subcutaneous administration of PGE₁, bradykinin or histamine does not modify plasma fibrinogen levels acting independently comparing with non-injected animals or injected with the drug vehicle. Bradykinin + histamine did not modify plasma fibrinogen levels either. However the administration of prostaglandin E₁ + bradykinin + histamine reproduced the increase of fibrinogen characteristics of laparotomy. This increase was partially but significantly inhibited in rats that had undergone bilateral removal of the adrenal medulla or administration of PGE₁ + bradykinin + histamine + bupivacaine (a local anesthetic), but it was not modified when the adrenal medullectomy was unilateral. It is concluded that plasma fibrinogen increase is obtained only when PGE₁ acts in presence of bradykinin or histamine and the adrenal medulla should be partially responsible for said increase.     

The influence of dose and duration of estrogen treatment on corpus luteum regression in ewes

Estrogen-induced regression of corpora lutea (CL) was studied in two experiments using 190 cycling ewes. In an experiment with a 3 × 5 factorial design, the minimum amounts of estradiol-17β (E₂), estrone (E₁) and diethylstilbestrol (DES) required to induce CL regression by intramuscular injection were determined. Injections of either 0, 100, 250, 500 or 1,000 μg of each estrogen were administered on days 10 and 11 of the estrous cycle. Each dose level of estrogen significantly reduced CL weight by day 14, and the 250 μg and higher dosages significantly reduced CL progesterone content. The luteolytic potencies of the three estrogens did not differ significantly.In the second experiment, E₂ was infused into the jugular vein of ewes on day 10 of the estrous cycle at a rate of 1.3 to 41.6 μg/hr for either 12, 24, or 48 hours. Infusion of E₂ for 12 hr did not significantly reduce either the weight or progesterone content of CL, even when as much as 500 μg of E₂ (41.6 μg/hr) was administered. In contrast, a total of 62 μg of E₂ infused over a 24-hr period (2.6 μg/hr) significantly reduced CL weight and CL progesterone. Therefore, CL regression induced by infusion of E₂ on day 10 of the cycle was dependent on the duration of the E₂ treatment as well as on the amount of E₂ infused.     

Estrogen metabolites in the release of inflammatory mediators from human amnion-derived cells

AimsHuman amnion-derived cells have been used as in vitro models to test the release of inflammatory mediators, such as arachidonic acid (AA) and prostaglandin E₂ (PGE₂). We compared estrogen metabolites for their ability to induce AA release, to influence PGE₂ production and to interact toward intracellular estrogen receptors (ERs).Main methodsMetabolite effects on AA and PGE₂ release were examined by radiolabelled substrate incorporation and by colorimetric enzyme immunoassays, respectively. [³H]17-β-estradiol binding displacements were performed on Ro-20-1724 treated whole cells.Key findingsIn WISH cells, estrone, 2-hydroxy-estrone and estriol induced a rapid dose dependent release of AA that was not inhibited by cycloheximide. Estrone and 2-hydroxy-estrone showed biphasic dose–response curves of PGE₂, whereas estriol and 16-α-hydroxy-estrone increased PGE₂ levels at high concentrations. 2-methoxy-estrone, 4-hydroxy-estradiol and 4-hydroxy-estrone did not significantly affect PGE₂ release. 2-methoxy-estradiol and 2-hydroxy-estradiol decreased the PGE₂ release. Effects of metabolites on PGE₂ were inhibited by cycloheximide and by the ER antagonist tamoxifen. In AV3 cells PGE₂ production was poorly detectable. On Ro-20-1724 treated WISH cells the K_i of 17-β-estradiol was 29.2 ± 5.4 nM. Estrone, 2-methoxy-estrone and 2-methoxy-estradiol showed similar affinity values. The hydroxyl substituent at position 2, 4 and 16 decreased or markedly increased the affinity for estradiol or estrone derivatives, respectively.SignificanceThe estrogen metabolites induced nongenomic effects on AA release from WISH cells. The influence on PGE₂ release was detectable only on WISH cells. These effects appeared genomic and mediated by intracellular ERs, whose properties seemed strongly dependent on intracellular cAMP levels.