Participation of Prostaglandin E2 in Contractile Activity of Inflamed Porcine Uterus

The aim of our study was to estimate the participation of prostaglandin E2 (PGE2) in the contractile activity of inflamed porcine uterus. On day 3 of the oestrous cycle, 50 ml of saline or 50 ml of Escherichia coli suspension, containing 109 colony-forming units/ml, was injected into each uterine horn in the control or experimental group, respectively. Seven days later the uteri were collected. Endometritis developed in all bacteria-inoculated gilts. Endometrium/myometrium and myometrium strips were incubated with PGE2 alone or together with PGE2 receptor (EP) subtypes (EP2, EP4, EP1 and EP3) blockers: AH 6809 (BEP2), ONO-AE2 (BEP4), ONO-AE3-240 (BEP1) and SC19220 (BEP3), respectively. In the control group, PGE2 (10 -8 and 10-7 M) increased the intensity of contractions in endometrium/myometrium, and at the higher dose in myometrium. PGE2 (10 -8 M) decreased the contraction intensity of the strips from inflamed uteri. After the use of BEP2, PGE2 (10 -7 M) increased the values of this indicator in endometrium/myometrium and myometrium from the control gilts. In these animals, PGE2 (10 -8 M) in the presence of BEP4 reduced the contraction intensity in endometrium/myometrium. In the bacterial group, PGE2 (10 -8 M) in the presence of BEP2 and BEP4 enhanced the intensity of contractions in myometrium. Similar reaction was evoked by PGE2 (10 -7 M) in endometrium/myometrium of the inflamed uteri in the presence of BEP4. The intensity of contractions in myometrium from the inflamed uteri significantly decreased after the use of BEP1 and PGE2 (10 -7 M). PGE2 (10 -7 M) administered after BEP3, significantly decreased the intensity of contractions in myometrium of the control gilts. These results show that PGE2 decreases the contraction intensity of inflamed porcine uteri. Further studies are needed to closely determine the role of PGE2 and other prostanoids in the contractile activity of inflamed uterine tissue. Contractile activity, uterus, endometritis, PGE2, PGE2 antagonists, gilts Endometritis is a common reproductive disorder in female domestic animals with consequences ranging from no effect on reproductive performance to permanent sterility. This pathological state can occur in female domestic animals after parturition as well as in animals that have not yet given birth, following artificial insemination or natural mating. It has been reported that a wide range of bacteria, mainly Escherichia coli (E. coli), Staphylococcus spp., Streptococcus spp., and in some cases Actinomyces pyogenes, Pasteurella multocida and Klebsiella pneumoniae, were isolated from uteri of sows with and without endometritis (De Winter et al. 1995; Aas et al. 1998). After labour, the uterine cavity is a target for bacterial flora characteristic of the environment in which the parturition takes place. Microorganisms from the uterus are removed with lochia in a mechanical manner (uterine involution) and/or eliminated by the immune system cells – both occurring simultaneously. However, an impairment of uterine involution and/or immunological response leads to intensive proliferation of microorganisms and consequently to the development of endometritis (Mateus et al. 2003). In the midst of endometritis we can differentiate states of mild progress and those with mucopurulent discharge from uterus and/or pyometra. Cases of endometritis with a mild course cause Address for correspondence: Dr. Barbara Jana Division of Reproductive Endocrinology and Pathophysiology Institute of Animal Reproduction and Food Research, Polish Academy of Sciences Tuwima 10, 10-747 Olsztyn, Poland Phone: +48 (89) 539-31-40 Fax: +48 (89) 535-74-21 E-mail: baja@pan.olsztyn.pl http://www.vfu.cz/acta-vet/actavet.htm ACTA VET. BRNO 2010, 79: 249-259; doi:10.2754/avb201079020249

Endometritis is a common reproductive disorder in female domestic animals with consequences ranging from no effect on reproductive performance to permanent sterility. This pathological state can occur in female domestic animals after parturition as well as in animals that have not yet given birth, following artificial insemination or natural mating. It has been reported that a wide range of bacteria, mainly Escherichia coli (E. coli), Staphylococcus spp., Streptococcus spp., and in some cases Actinomyces pyogenes, Pasteurella multocida and Klebsiella pneumoniae, were isolated from uteri of sows with and without endometritis (De Winter et al. 1995;Aas et al. 1998). After labour, the uterine cavity is a target for bacterial flora characteristic of the environment in which the parturition takes place. Microorganisms from the uterus are removed with lochia in a mechanical manner (uterine involution) and/or eliminated by the immune system cells -both occurring simultaneously. However, an impairment of uterine involution and/or immunological response leads to intensive proliferation of microorganisms and consequently to the development of endometritis (Mateus et al. 2003). In the midst of endometritis we can differentiate states of mild progress and those with mucopurulent discharge from uterus and/or pyometra. Cases of endometritis with a mild course cause no serious disturbances of the oestrous cycle. The oestrus in these animals comes in time or with a small delay and the percentage of pregnant females is high. In endometritis with a severe progress, the uterus is filled with mucopurulent secretion and protrudes into the abdominal cavity. In these animals, particularly in cows, the uterine muscular layer is devoid of the ability to contract. In such events, intensive therapy is not always successful. The main cause of these failures is loss of the contraction ability of the uterine muscular layer (Olson et al. 1984;Hussain 1989;Hussain and Daniel 1992;De Winter et al. 1995;Ramadan et al. 1997).
Endometritis generates a considerable increase in the production of prostaglandins (PGs) in uterine tissues. It was indicated that lipopolysaccharide (LPS) released by Gram-negative bacteria in inflamed uterus plays a significant role in the increase of PG synthesis. It has been reported that intrauterine infusions of LPS in postpartum cows and spontaneous or experimental bacterial infections of uteri in heifers and mares increased the concentration of a plasma PGF 2 a metabolite, 13,14-dihydro-15-keto-prostaglandin F 2 a (PGFM; Neely et al. 1979;Stabenfeldt et al. 1981;Peter and Bosu 1987;Peter et al. 1990). High concentration of PGF 2 a even in serious endometritis conditions should cause normal uterine involution and cleaning of the inflammatory exudate from the uterus. During uterus inflammation, in addition to an increase of PGF 2 a, concentrations of PGE 2 and PGI 2 also considerably increased (Mateus et al. 2003;Myatt and Lye 2004;Jana et al. 2007). The high PGE 2 and PGI 2 concentrations in inflamed uterus probably inhibit or delimit the contractile activity of PGF 2 a in this organ. PGE 2 also has been reported to cause contraction of the uterine muscular layer through two subtypes of PGE 2 receptors (EP): EP 1 and EP 3 and relaxation through other two receptors: EP 2 and EP 4 (Myatt and Lye 2004). We previously reported that PGF 2 a may affect the contractile activity of the intact and inflamed uteri through the PGE 2 receptors ).
These findings lead us to hypothesize that in the inflamed uterus, PGE 2 may have predominant diastolic activity. To the best of our knowledge, no information is available on the contractile activity of this PG in the inflamed uterus. Therefore, the purpose of the present study was to determine the influence of PGE 2 on the contractile activity of intact and inflamed uteri of gilts. We determined the effect of PGE 2 on the intensity and frequency of the contractions of strips of endometrium/myometrium and myometrium, and the effect of PGE 2 on these indicators in the presence of PGE 2 receptor antagonists.

Materials and Methods
Twelve crossbred gilts (Large White × Landrace) of similar age (7-8 months) and body mass (BM, 100-120 kg) with two controlled subsequent oestrous cycles were used. Oestrous behaviour was detected using the boar-tester. The animals originated from a herd with no abnormal discharge or fertility disorders. The gilts were individually housed in stalls under natural light and temperature conditions. They were fed a commercial grain mixture and tap water ad libitum. We followed the principles of animal care (NIH publication No 86-23, revised in 1985) as well as the specific national law on animal protection. The experimental procedures were approved by the Local Ethics Committee, University of Warmia and Mazury in Olsztyn (Agreement No 20/N).
On day 3 of the oestrous cycle (day 0 of the study), the gilts were randomly assigned to one of two groups: group I, control gilts receiving saline (n = 6), and group II, treated with E. coli (n = 6). In all the gilts median laparotomy was performed under general anaesthesia induced by azaperone (1 ml/10 kg of body weight; Stresnil Janssen Pharmaceutica, Belgium) and sodium pentobarbital (30-40 ml/100 kg of body weight; Vetbutal, Biowet, Poland). Next, in the animals of group I, 50 ml of saline were injected into each uterine horn (10 ml in five places). Group II received similarly 50 ml of E. coli (strain O25:K23/a/:H1; National Veterinary Research Institute, Department of Microbiology, Puławy, Poland) suspension containing 10 9 colony-forming units (cfu)/ml at the same time. In addition, in order to evenly distribute either saline or bacterial suspension within the uterine horn, both horns were carefully massaged. The gilts were not treated with antibiotics during the whole period of the study. The animals were slaughtered seven days after treatment (expected day 10 of the oestrous cycle) and the uteri were collected. Next, the uterine horns were intersected and their macroscopic examination was performed. Fragments of the uterine horns, collected from the middle part of the horns, were transferred to ice and transported to the laboratory within 20 min and immediately processed for examination of contractile activity.
From fragments of the uterine walls two kinds of strips (3 × 5 mm) were prepared: endometrium with myometrium (ENDO/MYO), and myometrium (MYO). The strips were washed in saline and mounted between two stainless steel hooks in 5 ml organ bath (Schuler Organ bath type 809; Hugo Sachs Electronic, Germany) under conditions of resting tension of 5 mN. The strips were kept in Krebs-Ringer solution of the following composition (mM/l): NaCl, 120.3; KCl, 5.9; CaCl 2 , 2.5; MCl 2 , 1.2; NaHCO 3 , 15.5; glucose, 11.5; and pH 7.4. The solution was maintained at 37 °C and continuously saturated with a mixture 95% O 2 and 5% CO 2 . After equilibration the contractile activity of the strips was recorded for at least 60 min. Contraction intensity and frequency of the uterine tissues were measured using a Hugo Sachs Electronic force displacement transducer (HSE F30 type 372), and recorded with HSE-ACADW software for Windows 2000 (Germany). At the beginning of the experiment the strips were incubated with noradrenaline (NA, Polfa, Poland) at doses of 10 -7 and 10 -6 M as well as with acetylcholine (ACh, Sigma) at doses of 10 -5 , 10 -4 and 10 -3 M, to determine the viability of tissues and their usefulness to further study. Next, the effect of increasing (10 -8 and 10 -7 M) doses of PGE 2 (Sigma) on the contractile activity was studied. The effect of each dose of examined substances was recorded for 7 min. Additionally, the effect of PGE 2 on the contractile activity of uterus was examined in the presence of PGE 2 receptor (EP 4, EP 2 , EP 1 and EP 3 ) antagonists. Before administration of PGE 2 uterine tissues were incubated for 2 min with following receptor antagonists: ONO-AE 2 (Sigma) -BEP 4 ; AH 6809 (Cayman Chemical, USA) -BEP 2 ; ONO-AE 3 -240 (Sigma) -BEP 1 and SC19220 (Sigma) -BEP 3 . BEP 4 , BEP 2 and BEP 1 were used at a dose of 10 -6 M and BEP 3 at a dose of 10 -7 M. After 2 min incubation of the tissue with antagonists, PGE 2 was added at doses of 10 -8 and 10 -7 M. The time of PGE 2 influence lasted 7 min. After the end of every measurement the tissues were washed three times in 15 ml of phosphate buffer at 10 min intervals. In the end, to determine the viability of tissues NA and ACh were administered at doses given above. In the statistical analysis only those results were considered, for which the difference in response to the stimulation by NA and ACh at the beginning and end of the study was lower than 20%.
The numerical values of the contraction activity (intensity and frequency) of tissues before the application of biologically active substances were calculated for 7 min and accepted as 100%. The results calculated for 7-min period after treatments were expressed as a percentage (mean ± SEM) of the contraction intensity and frequency before drug administration. Bonferroni test was USED for calculating the significance of mean differences (ANOVA, InStat Graph Pad, San Diego, CA).

Macroscopic examination of uteri
Inflammatory changes were not observed at slaughter in the ENDO of gilts receiving saline, whereas an inflammatory process involving the entire organ was always found in uteri challenged with E. coli. In such animals uterine horns were greatly enlarged and filled with a large amount of a gray-white mucosal exudate. The uterine wall was thickened. ENDO was red and swollen, with distinctly visible blood-injected blood vessels.
Influence of NA and ACh on the contractile activity of uteri NA at doses of 10 -6 and 10 -7 M decreased the intensity of contractions in ENDO/MYO (p < 0.01) and MYO (p < 0.001) of the control group as compared with the period before treatment. In the E. coli-injected gilts, NA at a dose of 10 -7 caused an increase in the intensity of contractions of ENDO/MYO (p < 0.05) and MYO (p < 0.01). The frequency of contractions in response to NA did not change significantly in two kinds of the uterine tissues from the control and E. coli-treated animals (Fig. 1).
Influence of PGE 2 on the contractile activity of uteri In the control group, PGE 2 at doses of 10 -8 and 10 -7 M enhanced the contraction intensity in ENDO/MYO (p < 0.01) and at the higher dose in MYO (p < 0.05). Lower (p < 0.05)

E. coli E. coli
intensity of contractions was found during incubation of two kinds of tissues from inflamed uteri with PGE 2 at a dose of 10 -7 M. PGE 2 at a dose of 10 -8 M increased, (p < 0.01) and at a dose of 10 -7 M decreased (p < 0.001) the frequency of contractions in ENDO/MYO of the control gilts. In control gilts, PGE 2 at two doses enhanced (p < 0.05) the contraction frequency in MYO. Similar effect (p < 0.05) was noted in ENDO/MYO and MYO of bacteria-infused animals after PGE 2 treatment at a dose of 10 -8 M. In contrast, 10 -7 M of PGE 2 lowered (p < 0.01) the frequency of contractions in two kinds of tissues from inflamed uteri (Fig. 3).
Influence of PGE 2 on the contractile activity of uteri in the presence of BEP 2 and BEP 4 The intensity and frequency of contractions in the uterine strips of both intact and inflamed uteri were similar after use of BEP 2 alone. PGE 2 at a dose of 10 -7 M in the presence of BEP 2 increased (p < 0.01) the intensity of contractions in ENDO/MYO and MYO of the control group. BEP 2 and PGE 2 at a dose of 10 -8 M enhanced (p < 0.05) the intensity of contractions in MYO from inflamed uteri. PGE 2 at a dose of 10 -8 M used in the presence of BEP 2 decreased (p < 0.01) the frequency of contractions in MYO of the control gilts. Similar effect (p < 0.01) was also observed in ENDO/MYO of inflamed uteri after the use of BEP 2 and PGE 2 at dose of 10 -8 M (Fig. 4).
BEP 4 alone did not affect significantly the intensity and frequency of contractions in ENDO/MYO and MYO from both studied groups. PGE 2 at a dose of 10 -7 M used after BEP 4 treatment decreased (p < 0.001) the intensity of contractions in ENDO/MYO of the control gilts. In the presence of BEP 4 , PGE 2 at doses of 10 -7 M and 10 -8 M increased (p < 0.05) the contraction intensity of ENDO/MYO and MYO, respectively, from inflamed uteri. In the presence of BEP 4 , 10 -7 M of PGE 2 led to a decrease in the frequency of 255 Fig. 7. Effect of PGE 2 on the intensity (A, C) and frequency (B, D) of contractions of endometrium/myometrium (A, B) and myometrium (C, D) strips from the control and E. coli-treated gilts in the presence of BEP 3 . Values (mean ± SEM; n = 6) are presented as percentage in relation to the basal (before treatment, bt) intensity and frequency of contractions. * p < 0.05, ** p < 0.01, *** p < 0.001
Influence of PGE 2 on the contractile activity of uteri in the presence of BEP 1 and BEP 3 In MYO of E. coli-treated gilts, BEP 1 alone decreased (p < 0.05) the intensity of contractions but increased (p < 0.01) the frequency of contractions. In the control group, the intensity of contractions in ENDO/MYO and MYO did not change significantly in the presence of BEP 1 and PGE 2 . A decrease (p < 0.05) in the contraction intensity in MYO of inflamed uteri was observed in response to BEP 1 and PGE 2 at a dose of 10 -7 M. In the control group in the presence of BEP 1 , PGE 2 at a dose of 10 -8 M decreased (p < 0.001) the frequency of contractions in ENDO/MYO whereas this effect (p < 0.01) in MYO was due to two doses of PGE 2 . In the presence of BEP 1 , the frequency of contractions in MYO of inflamed uteri was increased (p < 0.01) by PGE 2 at a dose of 10 -8 M (Fig. 6).
BEP 3 alone decreased (p < 0.001) the intensity of contraction in ENDO/MYO from intact uteri and increased (p < 0.05) the frequency of contractions in MYO of inflamed organs. In the control group, a decrease in the intensity of contractions of MYO (p < 0.05) was found after the use of BEP 3 together with 10 -7 M of PGE 2 . In two kinds of tissues from inflamed uteri the values of this indicator did not change significantly in response to BEP 3 and PGE 2 . In the control gilts, PGE 2 at doses of 10 -8 and 10 -7 M in the presence of BEP 3 increased (10 -8 M -p < 0.001, 10 -7 M -p < 0.01) the frequency of contractions in ENDO/MYO but decreased (p < 0.01) in MYO. The frequency of contractions in MYO from inflamed uteri were enhanced (p < 0.01) under the influence of PGE 2 at doses of 10 -8 and 10 -7 M in the presence of BEP 3 (Fig. 7).

Discussion
In the present study we showed that NA decreased the contraction intensity of the tissues from intact uteri and increased it in tissues from inflamed organs. Such response of porcine inflamed uterus to NA has been unknown to date. Relaxation of the uterine wall of the control gilts was anticipated. It is generally accepted that NA leads to an inhibition of uterine contractile activity that is probably connected with numerical superiority of β-adrenergic receptors in comparison with α-adrenergic receptors (Taneike et al. 1991;Kitazawa et al. 2001) and also with the lower sensitivity threshold of β-receptors (Kaneko et al. 1996). Although the mechanism responsible for the increase of contractile activity in the inflamed uterus after NA treatment is unknown, it may be a consequence of the highest expression or lower threshold excitability for α-adrenergic receptors. In our study, the changes in contraction intensity of the uterine tissues from both studied groups, in response to NA, were not connected with significant alterations in the frequency of contractions.
In turn, ACh increased the contraction intensity of the uterine tissues of both the control and E. coli-treated groups. This observation corresponds with the report of Kitazawa et al. (1999). However, the intensity of contractions in inflamed uteri was higher than in intact organs. We suggest that differences in response to ACh stimulation in intact and inflamed uteri are consequences of the changes in expression and sensitivity of muscarinic receptors during the inflammatory process embracing the uterine wall.
It is known that PGE 2 may both stimulate or inhibit the contractile activity of the uterus depending on its concentration and the physiological stage of the uterus (Crankshaw and Gaspar 1995;Popat and Crankshaw 2001;Cao et al. 2002). PGE 2 is also considered a potent vasodilator and relaxant acting on MYO and thus an important mediator of inflammation, and a factor contributing to the development of pathological process and its maintenance (Slama et al. 1991;Slama et al. 1994). Our earlier study showed that in inflamed porcine uterus, enhanced production of PGE 2 and PGF 2 a was connected with increases in the expressions of cyclooxygenase-2 ), microsomal PGE synthase-1 and 9-ketoreductase (converting PGE 2 to PGF 2 a; Jana et al. unpublished). Particularly, a high PGE 2 level (four times higher than PGF 2 a) was found on day 17 after intrauterine infusion of E. coli ). In cows with endometritis an increase in PGE 2 plasma level was observed a few days after parturition (Mateus et al. 2003). In the present study, we demonstrated that PGE 2 increased the uterine contraction intensity in the control gilts (with the exception of the lower dose in MYO). This is probably because the uterine tissue has more receptors responsible for contraction (EP 1 , EP 3 ) than for relaxation (EP 2 , EP 4 ) and/or differences in their distribution, as was earlier shown in human (Popat and Crankshaw 2001) and porcine (Cao et al. 2005) uteri. On the other hand, the intensity of contractions in two kinds of tissues from inflamed uteri decreased after PGE 2 treatment at the higher dose. This may suggest that in inflamed uterine tissues the content of EP 1 and EP 3 receptors responsible for the contractions diminished compared to representation and/ or activity of EP 2 and EP 4 receptors causing a diastola-like relaxing effect. It may be also presumed that the relaxing effect of PGE 2 on inflamed uterine tissues could be stronger in conditions of fully developed pathological process. The seven-day period between E. coli inoculation and investigation in our study could be too short to induce big changes in the content/distribution of PGE 2 receptors. Moreover, it is known that in cows MYO loses its contractile activity usually in serious and prolonged endometritis.
PGE 2 administered in the presence of EP 2 and EP 4 receptor blockers did not evoke significant changes in the contraction intensity of the tissues from both examined groups or increase the values of this indicator except for ENDO/MYO of the control gilts (the lower dose of PGE 2 and BEP 4 ) where the contraction intensity decreased. Moreover, the changes in enhancement of the contraction intensity found in intact and inflamed uteri were dependent on the type of the blocked receptor. The application of PGE 2 (at the higher dose) in the presence of BEP 2 enhanced the contraction intensity in two kinds of tissues of the control gilts. In contrast, in inflamed uteri such a reaction occurred only in MYO after the use of BEP 2 and the lower dose of PGE 2 . Moreover, in these uteri, PGE 2 at the lower or higher dose in the presence of BEP 4 increased the contraction intensity in MYO or ENDO/MYO, respectively. The differences probably result from the various expressions/distributions of EP 2 and EP 4 receptors in intact and inflamed uteri. Based on these results, we suggest that PGE 2 causes dilatation in the intact uterus mainly by EP 2 receptors, whereas in inflamed uteri it causes dilatation mainly through EP 4 receptors, the content of which can be greater in inflamed than in intact uteri. In terms of uterine strip contraction frequency, PGE 2 after the treatment of EP 2 and EP 4 blockers did not significantly affect or decrease the number of contractions in both intact and inflamed uteri.
PGE 2 administered in the presence of EP 1 and EP 3 receptor blockers whose stimulation leads to uterine contraction, usually failed to evoke significant changes in the contraction intensity in uterine tissues in both studied groups. However, in the control gilts, PGE 2 at a higher dose in the presence of BEP 3 decreased the contraction intensity in MYO. A similar reaction was found in MYO of inflamed uteri in response to PGE 2 used also at a higher dose but after prior BEP 1 treatment. These results indicate that the concentration and/or sensitivity of EP 3 receptors was higher in intact uteri, whereas those of EP 1 receptors in inflamed uteri. It can be presumed that inflammation also changes the expression/distribution of two types of contractile receptors for PGE 2 . Moreover, it should be emphasised that the contractile activity of uteri changed after the use of only BEP 1 and BEP 3 . In the presence of BEP 1 , the contraction intensity decreased in MYO from inflamed uteri, while BEP 3 caused the same reaction in ENDO/MYO of intact uteri. In turn, both blockers increased the contraction frequency in MYO of inflamed organs. These data indicated that uterine strips produce endogenous PGE 2 , and its action is inhibited by PGE 2 receptor antagonists. PGE 2 acted differently in the presence of BEP 1 and BEP 3 on the frequency of contractions in the uterine tissues from both groups of gilts. In the intact uteri this indicator mainly decreased, while in inflamed uteri it increased in most cases.
In conclusion, the presented data demonstrate, for the first time, that PGE 2 , acting through EP 2 and EP 4 receptors, decreased the contraction intensity of endometrium/myometrium and myometrium collected from gilts on day 7 after intrauterine injection of an E. coli suspension. Our data also suggest that PGE 2 , through its effect on the contractile activity of inflamed uteri, can be important for the course and/or consequences of this pathological state in females of domestic animals. However, further studies should be performed to closely determine the role of PGE 2 and other prostanoids in the contractile activity of uterine tissues with more advanced inflammation.