Microsomal Prostaglandin E Synthase-1 Plays a Critical Role in Long-term Motility Dysfunction after Bowel Obstruction

Motility dysfunction is present not only during bowel obstruction (BO), but after obstruction is resolved. Previous studies found that lumen distension associated mechano-transcription of COX-2 and production of PGE2 in gut smooth muscle cells (SMC) account for motility dysfunction during obstruction. We hypothesized that PGE2 may exert autocrine effect in SMC to induce microsomal prostaglandin E synthase-1 (mPGES-1), which contributes to motility dysfunction after obstruction is resolved. Partial colon obstruction was induced in rats with an obstruction band, which was released 7 days later. Rats were further studied in the post-BO state. Circular muscle contractility of the mid colon (previously distended during obstruction) remained suppressed, and colon transit was impaired in the post-BO state. The COX-2, mPGES-1, and PGE2 levels were all increased in the distended bowel during obstruction. However, after obstruction was resolved, COX-2 expression returned to normal, whereas mPGES-1 and PGE2 levels remained increased. Expression of mPGES-1 in colon SMC was inducible by stretch or PGE2. Administration of mPGES-1 inhibitor Cay 10526 either before or after the release of obstruction normalized PGE2 levels and improved motility in the post-BO rats. In conclusion, mPGES-1 plays a critical role in the continuous suppression of motor function in the post-BO state.

is important to elucidate the underlying mechanisms of long-term motility dysfunction after obstruction, so that therapeutic targets may be identified.
Our recent studies in a rodent model of partial colon obstruction found that lumen distention during BO leads to mechanical stress-induced gene expression (mechano-transcription) of cyclo-oxygenase-2 (COX-2) in the distended oral segment, but not in the non-distended aboral segment 18 . We found that mechano-transcription of COX-2 and COX-2 -derived prostaglandins (PG) in the distended bowel plays a critical role in the suppression of motility function during the course of BO 18,19 . Lin et al. found that COX-2 inhibitor may have the therapeutic potential in the management of motility dysfunction and associated symptoms during BO 20 . However, it remains to be determined whether distention-induced COX-2 and other mechanically sensitive molecules remain up-regulated after obstruction is resolved, and whether expression of these mechanically sensitive genes exert autocrine or paracrine actions to further alter gene expression and gut functions in the previously distended bowel segment.
In the present study, we first established a rodent model of post-BO, in which partial colon obstruction was induced with an obstruction band 18 , and resolved by releasing the band 7 days later. We found that colon smooth muscle contractility remained significantly suppressed after obstruction is released (post-BO state). Colon transit rate was slower in the post-BO rats than in the sham control. The main objective of the study was to investigate the mechanisms underlying motility dysfunction in the post-BO state and to identify potential therapeutic targets to improve motility function after obstruction is resolved. We found that although mechanical stress-induced expression of COX-2 returned to normal level after obstruction is resolved, mPGES-1 expression and PGE 2 production remained elevated in the post-BO state. We tested the hypotheses that COX-2 -derived PGE 2 during obstruction further up-regulates expression of mPGES-1, and that mPGES-1 may play a critical role in the continued production of PGE 2 and prolonged motility dysfunction after obstruction is resolved.

Rodent model of post-BO and experimental design. The Institutional Animal Care and Use
Committee at University of Texas Medical Branch approved all the procedures performed on the animals. Experiments were performed in accordance to the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health, USA.
Male Sprague-Dawley rats weighing 225-275 g (Harlan Sprague Dawley, Indianapolis, IN) were used in the study. The rats were housed in a controlled environment (22 °C, 12-h:12-h light/dark cycle) and allowed food and water ad libitum. Partial colon obstruction was prepared as previously described 18,20,26 . To make post-BO model, colon obstruction was maintained with an obstruction band placed in the distal colon (3-4 cm oral to the end) until the band was released 7 days later. In brief, 7 days after initial operation, sham and obstruction rats were anesthetized with 2% isoflurane inhalation by an E-Z Anesthesia vaporizer (Palmer, PA). A midline laparotomy was made, and the obstruction band was released in the obstruction rats. The rats in post-BO state were kept for up to 6 weeks until euthanasia. The sham control rats underwent the similar surgical schedule with two laparotomy operations, except that obstruction band was installed but released immediately in the first operation.
In some experiments, in order to determine the possible roles of mPGES-1 and COX-2 in post-BO motility in vivo, animal groups were randomly assigned to include sham and Post-BO rats each with vehicle (DMSO) or mPGES-1 inhibitor CAY10526 (Cayman Chemical, Ann Arbor, MI) at 2.5 mg/kg i.p. daily 25,27 , or COX-2 inhibitor NS-398 at 5 mg/kg i.p. daily 18,20 , starting on the day when obstruction band was released. Rats were studied 7 days after obstruction was released (post-BO day 7).
In another experiment, mPGES-1 inhibitor CAY10526 was administered during the 7-day period of obstruction at 2.5 mg/kg i.p. daily. Obstruction was released and rats were studied 7 days on post-BO day 7.
Tissue collection. The mid colon segment (~3 cm in length) oral to the original site of obstruction was collected in fresh carbogenated Krebs buffer (in mmol/l: 118 NaCl, 4.7 KCl, 2.5 CaCl2, 1 NaH2PO4, 1.2 Mgcl2, 11 d-glucose, and 25 NaHCO3). The mucosa/submucosa and muscularis externa layers were separated by microdissection as described previously [18][19][20] . The tissue samples were processed immediately for muscle bath experiments or stored at −80 °C for molecular studies. Full thickness samples were saved in 10% buffered formalin for immunohistochemistry study.
Immunohistochemistry study. Immunohistochemistry staining of COX-2 was performed as described previously 18,26 on formalin-fixed, paraffin-embedded colon segments in sham control and in rats with obstruction and in post-BO state. Sections at 4-μm thickness were blocked with 5% normal goat serum in PBS for 20 min at room temperature, and incubated with the rabbit anti-COX-2 antibody (1:200, Cayman Chemical) and a biotin-conjugated anti-rabbit secondary antibody (Vector Laboratories, Burlingame, CA). After incubation with avidin-biotin complex (Vector kit, Vector Laboratories), the sections were stained in diaminobenzidine tetrahydrochloride with 0.03% hydrogen peroxide. As a negative control, sections of the same specimens were processed by the same method but omitting the anti-COX-2 primary antibody. RNA preparation and quantitative RT-PCR. Total RNA was extracted from tissues or cells using the Qiagen RNeasy kit (Qiagen, Valencia, CA), and reverse transcribed with SuperScript III First-Strand Synthesis System (Invitrogen, Carlsbad, CA) [18][19][20]26 . Real-time quantitative PCR was performed using the Applied Biosystems 7000 real-time PCR system (Foster City, CA) as described [18][19][20]26 . The TaqMan primers and probes for detection of rat COX-2, cPGES-1, mPGES-1, and mPGES-2 mRNAs were purchased from Invitrogen. The fold change relative to control was calculated with the comparative Ct (ΔΔCT) method with endogenous reference 18S rRNA as the normalizer.
Enzyme immunoassay of PGE 2 . Rat colonic muscularis externa tissue was homogenized in cold PBS buffer (in mmol/l 137 NaCl, 2.7 KCl, 10 Na2HPO4, KH2PO4, pH 7.4) supplemented with protease inhibitors. PGE 2 in the extraction was measured with the PGE 2 enzyme immunoassay kit purchased from Cayman Chemical by following the manufacturer's protocols 18,20,21 . Muscle bath experiments. Colon segments 1∼2 cm oral to the original site of obstruction were opened along the mesenteric border, and pinned flat in a Petri dish with Sylgard base in carbogenated Krebs buffer. The mucosa/submucosa layers were separated and discarded by microdissection. The smooth muscle strips (3 mm × 10 mm) were mounted along the circular muscle orientation in individual muscle baths (Radnoti Glass, Monrovia, CA) filled with 10 ml of carbogenated Krebs solution at 37 °C. The contractile activity was recorded as previously described 18,19,[28][29][30] with Grass isometric force transducers and amplifiers connected to Biopac data-acquisition system (Biopac Systems, Goleta, CA). The muscle strips were equilibrated in the muscle bath under 1 g tension for 60 min at 37 °C before they were tested for contractility. Muscle contractility was tested in response to acetylcholine (ACh) (10 −6 to 10 −2 M) and to KCl (62.5 mM). The strips were left to equilibrate for 15 to 20 min between additions of different concentrations of ACh or KCl. When the contractile response to KCl was tested, the concentration of KCl (62.5 mM) in the Kreb's buffer was increased by the equimolar replacement of NaCl. The contractile response was quantified as the increase in area under contractions (AUC) during 4 min after the addition of ACh or KCl over the baseline AUC during 4 min before the addition of ACh or KCl.
Measurements of colon transit. The colonic transit rate in sham and post-BO rats was determined by the geometric center method as described previously 20,28 . In brief, a bolus of 1.5 ml of 1.5% methylcellulose (Fisher Scientific, Fair Lawn, NJ) containing 0.75 mg non-absorbable phenol red was injected into the colon via a catheter implanted earlier. Rats were euthanized by CO 2 inhalation 60 min later. The whole colon was removed immediately and divided into six segments of equal length. Each segment, along with its intraluminal contents, was placed in 25 ml of 0.1 N NaOH and homogenized. The homogenate was kept at room temperature for 1 h, and 1 ml of the supernatant was added to 0.1 ml of 20% trichloroacetic acid solution to precipitate proteins. After centrifugation at 10,000 g for 15 min, 1 ml of 0.5 N NaOH was added to the supernatant. The amount of phenol red was determined by measuring the absorption at 550 nm with a Biophotometer Plus (Eppendorf North America, Westbury, NY). Colon transit was calculated as the geometric center of distribution of phenol red as follows: geometric center = ∑ (counts of phenol red per segment × segment number)/∑ (counts of phenol red per segment).
Primary culture was allowed to grow for 8-10 days until cells were confluent. The cells were then seeded at 8 × 10 4 cells/well in six-well BioFlex culture plates coated with type I collagen (Flexcell International, Hillsborough, NC). Cells were allowed to grow to ∼80% confluence before being subjected to DMEM/1% FBS for overnight prior to mechanical stretch 18,26 via a FX-4000 Flexercell Tension Plus System (Flexcell International). This computer-regulated bioreactor applies multi-axial strain to cultured cells. Through vacuum pressure, cultured cells were stretched on the flexible membrane plates. Cells incubated in parallel under identical conditions but without exposure to stretch served as controls.
Statistical analysis. Data points are expressed as means ± SEM. Statistical analysis was performed by analysis of variance with non-repeated measures (by Student-Newman-Keuls test) for comparisons of multiple groups and Student's t-test for comparisons of two groups. A p value of ≤0.05 was considered statistically significant.
Data availability. The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Results
Motility dysfunction in post-BO state. Previous studies found that colonic motility was suppressed significantly during the course of obstruction in the rat model of BO 18 . To determine if motility function remains suppressed in the post-BO state, we first measured colon circular smooth muscle contractility in response to cholinergic muscarinic receptor agonist ACh in the sham and post-BO rats 7, 14, and 42 days after obstruction was released. Compared to sham, muscle contractility of the mid colon (previously distended during BO) remained  Fig. 1A,B). However, by post-BO day 42, the smooth muscle contractility recovered to the levels of the sham controls (Fig. 1C).
To determine if muscle contractility is affected in post-BO state in a receptor-independent manner, we also tested the muscle contractile response to KCl (62.5 mM), as KCl bypasses G-protein coupled receptors i.e. muscarinic receptors and induces smooth muscle contraction by depolarizing membrane potential and activating voltage-gated Ca 2+ channels 28,30 . We found that the contractile response to KCl was also significantly reduced in post-BO rats (Post-BO day 7, Fig. 1D). This data indicates that the suppressed contractile response in post-BO state may involve inhibition of the excitation-contraction coupling mechanisms downstream to receptor activation 30 .
We also measured colon transit rate in sham and post-BO rats (post-BO day 7) (Fig. 1E). It was found that the colon transit rate was significantly delayed in the post-BO rats compared to sham controls, as the geometric center was 2.87 ± 0.19 in the post-BO rats and 3.46 ± 0.18 in the sham control (N = 5 and 6 in sham and post-BO groups, respectively. p < 0.05 vs. sham). These data suggest that colon motility function was impaired significantly after obstruction was resolved.

COX-2 expression and PGE 2 production in BO and post-BO.
Our previous study 18 found that mechanical stress-induced COX-2 expression and COX-2 derived PGE 2 production play a critical role in the suppression of muscle contractility during obstruction. Consistent with the earlier findings, we noticed that the expression of COX-2 protein and mRNA was significantly induced in colonic smooth muscle of the distended mid colon during obstruction (Figs 2 and 3A). However, COX-2 expression returned to normal levels after obstruction was resolved (Fig. 2). By day 7 of the post-BO period, the COX-2 mRNA level in the colonic muscularis externae was not significantly different from the sham control (Fig. 3A). Immunohistochemistry study revealed that COX-2 was induced in the SMC of the distended colon during obstruction (Fig. 2). However, no COX-2 expression is detectable in the post-BO state in the originally distended colon segment (Fig. 2).
Interestingly, the levels of PGE 2 in the colonic muscularis externae were increased not only during obstruction, but remained significantly higher in the post-BO rats than in the sham controls (3.6 ± 0.4 vs 1.1 ± 0.2 ng/mg. N = 5 each group. p < 0.05) (Fig. 3B).

PGES expression in BO and post-BO.
We sought to determine the mechanisms underlying the persistent increase of PGE 2 in the post-BO state. As production of PGE 2 is dependent not only on COX, but on PGES, we determined the mRNA expression of three PGES isomerases in the muscularis externae of the colon. The expressions of mPGES-2 and cPGES-1 were not altered in the distended colon during obstruction (Fig. 4A-C). However, the expression of mPGES-1 mRNA was significantly increased by 1.8 ± 0.4 -fold in the distended To determine muscle contractility, colonic circular muscle strips were prepared from the mid colon of post-sham and post-BO rats. The mid colon was taken, as it was where mechanical distention occurred during obstruction. Note that colonic circular muscle contractility to ACh was suppressed on day 7 (n = 6, A) and day 14 (n = 5, B), but recovered on day 42 (n = 5, C) in post-BO state. Muscle contractile response to KCl was also reduced in post-BO state (post-BO day 7, n = 5) (D). Colon transit was measured in a separate study (E) on different rats from those used for muscle bath. Note that colon transit rate is slower in post-BO rats (post-BO day 7, n = 6), compared to post-sham (7 day after operation, n = 5). *p < 0.05 vs Post-sham.

Regulation of mPGES-1 gene expression by PGE 2 and mechanical stretch. The underlying mech-
anisms for up-regulation of mPGES-1 during obstruction and in the post-BO state were not understood. We tested the hypothesis that both mechanical stress and PGE 2 may induce the expression of mPGES-1 in colonic SMC. In the in vivo model of obstruction, both mechanical stress and increase of PGE 2 are present during obstruction 23 . We thus chose to test the hypothesis in vitro in primary cultures of rat colon SMC, as mechanical stress and PGE 2 can be applied separately in the in vitro settings.
To test the effect of mechanical stress, primary culture of rat colonic SMC was stretched at 18% elongation for 3 h, cells were harvested for quantitation of mPGES-1 mRNA expression. We found that mechanical stretch significantly induced up-regulation of mPGES-1 mRNA expression by 2.6 ± 0.3-fold, compared to no-stretch control (p < 0.05, 4 independent experiments) (Fig. 5A).
In a separate study, primary culture of rat colonic SMC were treated with vehicle or exogenous PGE 2 (10 −7 M) for 3 h. We found that PGE 2 treatment also significantly up-regulated mPGES-1 mRNA by 1.6 ± 0.1 -fold (p < 0.05 vs. control, 4 independent experiments) (Fig. 5B). These experiments demonstrated that expression of mPGES-1 is subject to regulation by both mechanical stretch and by PGE 2 .

Role of mPGES-1 in motility dysfunction in BO and post-BO states. Further functional studies
were carried out in vivo to test the hypothesis that both mechanical stress and PGE 2 may modulate the expression of mPGES-1 during obstruction, whereas PGE 2 , acting in an autocrine mode, is responsible for up-regulation  of mPGES-1 in the post-BO state. The increased mPGES-1 may contribute to the continued increase of PGE 2 and suppression of motor function after obstruction is resolved. To determine the possible role of mPGES-1 in the production of PGE 2 and suppression of motility in the post-BO state, we treated rats with daily administration of a specific mPGES-1 inhibitor Cay 10526 at 2.5 mg/kg i.p 24,25,27 . in the post-BO state starting on the day when obstruction was released. We found that Cay 10526 treatment almost completely normalized PGE 2 levels (Fig. 6A), and significantly improved colon motility in the post-BO rats (Fig. 6B,C). As shown in Fig. 6B, the values of geometric center (GC) were not statistically different between sham (N = 4) and post-BO rats (N = 5) when they were treated with Cay 10526 (Fig. 6B). Cay 10526 treatment in the post-BO state also restored smooth muscle contractility in the previously distended colon (Fig. 6C).  We also determined if administration of Cay 10526 during BO had any effect on motility dysfunction in the post-BO state. For this purpose, Cay 10526 was administered daily during the 7-day course of obstruction only, and colonic circular muscle contractility was measured 7 days after obstruction was released. As shown in Fig. 6D, Cay 10526 treatment during obstruction partially but significantly improved smooth muscle function in the post-BO state (Fig. 6D).

Role of COX-2 in motility dysfunction in BO and post-BO states.
To determine the possible role of COX-2 in motility dysfunction in obstruction and after obstruction is resolved, we administered COX-2 inhibitor NS-398 (5 mg/kg, i.p. daily) during BO and in the post-BO state in separate studies. When NS-398 was administered during BO, it significantly improved muscle contractility in the course of BO 18,21 , and also restored muscle function in the post-BO state (Fig. 7A). However, when NS-398 was administered after obstruction was released (in the post-BO state), it did not significantly improve muscle contractility in the post-BO rats (Fig. 7B).

Discussion
Motility dysfunctions occur in all stages of bowel obstruction, and may persist after obstruction is resolved 2,[9][10][11][12][13][14][15] . Immediately after obstruction, gut motor activity is increased in the segment oral to obstruction but suppressed in the segment aboral to obstruction 33 . This transient motility change may result from neuronal mechanisms similar to the physiological peristalsis reflex where food bolus (or luminal accumulation in obstruction) stimulates mechano-receptors in the gut wall to initiate neuronal regulations of ascending excitation and descending inhibition [33][34][35] . Motor activity in the obstructed bowel soon (in hours) becomes compromised with marked suppression of phasic contractions and muscle tone 18,33,36,37 . Gut motility function continues to be suppressed as long as obstruction is present. The suppressed motility may account for symptoms such as nausea, vomiting, bloating, abdominal distention and constipation 2,[12][13][14]18,30 . We discovered that lumen distention in BO leads to mechano-transcription of COX-2 and other pro-inflammatory mediators in gut smooth muscle 18,31,38,39 , and that the mechanical stress-induced COX-2 plays a critical role in the suppression of muscle contractility and motility function during the course of BO 18,20,38,39 .
Although surgical release of obstruction may be the treatment of choice for some patients, especially for those who have complete or complicated obstruction, motility dysfunction does not disappear after obstruction is resolved. Numerous clinical studies documented motility dysfunction in patients with Hirschsprung's disease or mechanical obstruction after constriction or obstruction is resolved by surgery [9][10][11][12][13][14][15] . In the present study, we found that mPGES-1 and m-PGES-1 -derived PGE 2 may play a critical role in the persistent suppression of gut motility in the post-BO state (Fig. 8). First, PGE 2 has profound inhibitory effects on gut smooth muscle contractility in the colon 18,20 . We found that PGE 2 production is increased in the smooth muscle tissue of the obstructed colon and remains increased even after obstruction is resolved. Second, after obstruction was released, mechanical stress was no longer present and the process of mechano-transcription of COX-2 stopped. However, mPGES-1 expression remained up-regulated after obstruction was resolved. More importantly, we found that administration of mPGES-1 inhibitor Cay 10526 25,27 during the post-BO state blocked the increase of PGE 2 and restored motility function in the post-BO rats. These studies suggest that mPGES-1 may represent a novel therapeutic target in the management of PGE 2 related motility dysfunction in obstructive conditions. Although PGE 2 is found increased both in obstruction and after obstruction is resolved, the therapeutic strategies to mitigate PGE 2 related motility dysfunction in BO and post-BO states may be distinctively different. We found that COX-2 inhibitor NS-398, if given during obstruction, is effective in improving motility function during BO and after obstruction is resolved. However, if COX-2 inhibitor is administered in the post-BO state, it does not significantly improve post-BO motility of the colon. To the contrary, treatment with mPGES-1 inhibitor Cay 10526 either during BO or in the post-BO state significantly improved motility function in the post-BO state. Based on the findings in the present and previous studies 23,26 , the differential expression patterns of COX-2 and mPGES-1 in obstruction and in the post-BO state may explain these effects of NS-398 and Cay 10526. Expression of COX-2 was up-regulated during obstruction [18][19][20] , whereas mPGES-1 expression is increased in both obstruction and in the post-BO state. Our study implies that inhibition of both COX-2 and mPGES-1 during obstruction may be effective as a therapeutic strategy to treat motility dysfunction during BO, and as a prophylactic measure to prevent long-term motility dysfunction occurring after obstruction is resolved. On the other hand, after obstruction is resolved, inhibition of mPGES-1, but not COX-2, may achieve a better therapeutic effect to tackle motility dysfunction (Fig. 8).
As shown previously [18][19][20]39 , mechano-transcription of COX-2 and COX-2 -derived PGE 2 in gut SMC play a critical role in motility dysfunction in obstruction. The present study suggests that the increased PGE 2 during bowel distention may further enhance the expression of mPGES-1, which contributes to the sustained increase of PGE 2 and long-term smooth muscle dysfunction even after distention is corrected. These results indicate that mechano-transcription in the originally distended segment during BO may exert secondary effects on plasticity of colonic SMC, i.e. up-regulation of mPGES-1 by PGE 2 in an autocrine or paracrine mode. The secondary effects of mechano-transcription contribute to the long-term motility dysfunction. These findings may have important implications for surgical treatment of BO, especially chronic BO such as in HD. The current surgical approach for obstruction is to release blockage or remove constriction (such as in pull-through procedures for HD) 6,9,40 , but  Mechanotranscription of COX-2 and mPGES-1 in SMC accounts for the increased production of PGE 2 in the distended bowel segment during obstruction. The increased PGE 2 during BO may further up-regulate the expression of mPGES-1 in the SMC via an autocrine or paracrine mechanism. The PGE 2 -driven up-regulation of mPGES-1 may contribute to the continued production of PGE 2 and long-term suppression of smooth muscle contractility in the post-BO state.
to leave alone the distended oral segment in the gut. However, our studies show that the distended oral segment may not be considered "normal"; rather it is the site of mechano-transcription and therefore the root cause of long-term GI dysfunction. The once distended bowel segment during the course of obstruction, may well be the reason for long-term bowel dysfunction even if blockage or constriction has been resolved.
Taken together, we found that gut motility function is suppressed during BO and after BO is surgically resolved. Mechano-transcription of COX-2 and COX-2-derived PGE 2 in the SMC of the distended bowel segment may not only play a critical role in motility dysfunction during obstruction, but exert secondary effects on SMC to contribute to gut dysfunction even after obstruction is resolved. Our study suggests that COX-2-derived PGE 2 during BO may further up-regulate the expression of mPGES-1 in an autocrine mode. The increased mPGES-1 contributes to continued production of PGE 2 and long-term suppression of motility dysfunction in the post-BO state. Thus, the once distended bowel segment during obstruction may not be considered "normal", as it is where mechano-transcription occurs. Pharmacological studies imply that inhibition of both COX-2 and mPGES-1 during obstruction may be effective as a therapeutic strategy to treat motility dysfunction during BO, and as a prophylactic strategy to prevent long-term motility dysfunction occurring after obstruction is resolved. However, after obstruction is resolved, inhibition of mPGES-1, but not COX-2, may be a more effective therapeutic approach for motility dysfunction.