NC

YKa

b

a

ARRA

KMCCPC

1

ncaT1Prusic

Adh

1d

Pharmacological Research 61 (2010) 48–57

Contents lists available at ScienceDirect

Pharmacological Research

journa l homepage: www.e lsev ier .com/ locate /yphrs

eurogenic contraction of mouse rectum via the cyclooxygenase pathway:hanges of PGE2-induced contraction with dextran sulfate sodium-induced colitis

uri Hamadaa, Ikuo Murakamia, Erina Katoa, Satoshi Yamanea, Hiromichi Fujinoa,enjiro Matsumotob, Kimihito Tashimab, Shunji Horieb, Toshihiko Murayamaa,∗

Laboratory of Chemical Pharmacology, Graduate School of Pharmaceutical Sciences, Chiba University, Inohana 1-8-1, Chuo-ku, Chiba 260-8675, JapanLaboratory of Pharmacology, Faculty of Pharmaceutical Sciences, Josai International University, 1 Gumyo, Togane, Chiba 283-8555, Japan

r t i c l e i n f o

rticle history:eceived 10 July 2009eceived in revised form 24 August 2009ccepted 1 September 2009

eywords:ouse rectum

ontractionyclooxygenase

a b s t r a c t

Recent reports suggest that cyclooxygenases (COXs) including COX-2 are constitutively expressed, andprostaglandins (PGs) regulate motility and/or contraction in the colon and rectum. This study examinesthe role of COXs in the regulation of neuromuscular function in longitudinal preparations of isolated rec-tum and distal colon (Side A, close to the transverse colon; and Side B, close to the rectum) in normal miceand after the induction of colitis by dextran sulfate sodium (DSS). In control rectum, electrical stimulation(ES)-induced contractions were inhibited by atropine and by COX inhibitors, in an independent manner.PGE2 at 3 �M caused a marked contraction, but the secondary response at 20 min after the first applica-tion was 60% desensitized. In rectum from DSS-treated mice, spontaneous and ES-induced contractions

rostaglandinolitis

were significantly less intense than in the control preparations, and the response to PGE2 was abolishedbut that to 3 �M acetylcholine was not. In control distal colon, the responses to PGE2 in neither side weredesensitized by the repeated application. In DSS-treated distal colon, PGE2 response was impaired in thetwo regions, and was desensitized on Side B more than Side A. DSS treatment impaired contractions by40 mM KCl in rectum and on Side B but not Side A. DSS treatment increased COX-2 expression in rec-tum, but not in distal colon. These findings suggest that the induction of colitis by DSS affects ES- andPGE2-regulated motility in the order rectum > distal colon close to the rectum > distal colon in mice.

. Introduction

Various factors including cytokines, prostaglandins (PGs), anditric oxide are known to have physiological and pathophysiologi-al roles in the gastrointestinal (GI) tract including the distal colonnd rectum, regulating water transport, blood flow, and motility.wo isoforms of cyclooxygenase (COX) have been recognized, COX-(an endogenous type) and COX-2 (an inducible type). In general,

Gs produced via COX-2 are involved in both inflammation andegeneration whereas PGs derived from COX-1 exert immunomod-

latory, cytoprotective, and proangiogenic effects (for a review,ee Ref. [1]). It is reported that COX-1 and COX-2 are expressedn the neuromuscular compartment of normal human and mouseolon, where they modulate the cholinergic excitatory regulation

Abbreviations: PGs, prostaglandins; GI, gastrointestinal; COXs, cyclooxygenases;Ch, acetylcholine; IBD, inflammatory bowel disease; UC, ulcerative colitis; DSS,extran sulfate sodium; ES, electrical stimulation; TTX, tetrodotoxin; 5-HT, 5-ydroxytryptamine.∗ Corresponding author. Tel.: +81 43 226 2874; fax: +81 43 226 2875.

E-mail address: murayama@p.chiba-u.ac.jp (T. Murayama).

043-6618/$ – see front matter © 2010 Published by Elsevier Ltd.oi:10.1016/j.phrs.2009.09.001

© 2010 Published by Elsevier Ltd.

of colonic motility [2]. In the presence of colitis, the expressionof COX-2 was up-regulated and seemed to play a predominantrole in modulating neuromuscular function in rat colon [3]. Recep-tors for prostanoids are present on smooth muscle cells, causingcontraction or relaxation depending on the receptor subtype [4,5].Prostanoid receptors can be functionally grouped into three cate-gories. EP1-, TP- and FP-receptors are contractile receptors. EP2-,EP4-, IP- and DP-receptors are relaxant receptors. The EP3-receptoris an inhibitory receptor inhibiting muscle relaxation, although itsactivation causes the contraction of smooth muscles in several GItissues [4,5]. In addition, prostanoid receptors exist on choliner-gic neurons in the GI tract, where their activation increases ordecreases the release of acetylcholine (ACh) depending on the tis-sue and species. In longitudinal preparations of muscle-myentericplexus from guinea-pig ileum, treatment with indomethacin inhib-ited contractions and the release of ACh induced by nicotine, andthe inhibitory effects were prevented by PGE2 and PGI2 [6,7]. These

reports suggest a role for PGs in contractile responses regulated byneuronal- and non-neuronal pathways in GI tissues. Although PGE2caused the contraction of longitudinal preparations of rat rectum[8], the precise role of COX and/or PGs in contractile responses inthe rectum has not been elucidated.

logica

bip(m[fiUdDadeapthcow

aiinbrcwrritanocoitfmcstrm

2

2

Awaaffar(c

Y. Hamada et al. / Pharmaco

Inflammation of the GI tract, which occurs during inflammatoryowel disease (IBD), acute bacterial enteritis, and nematode-

nfections, markedly alters GI function [9–11]. Data obtained fromatients with IBDs such as Crohn’s disease and ulcerative colitisUC) and animal models of IBD have suggested a role for inflam-

atory effects on neurons in the symptoms associated with IBD9,12–14]. One animal model of human colitis involves dextran sul-ate sodium (DSS) being administered in drinking water to rodentsncluding mice. This model displays several characteristics of bothC and Crohn’s disease, and it has been suggested that Th-1-ominant inflammation is involved in the pathogenesis of acuteSS colitis, while the chronic DSS model is characterized by Th-1nd Th-2 cytokines [15,16]. There is evidence of increased pro-uction of prostanoids such as PGE2 and PGI2 (prostacyclin) andxpression of COXs including COX-2 in the colon of patients withctive IBD and mammals with experimental colitis [17–21]. COX-2lays a key role in IBD-related inflammation through its produc-ion of PGs [22,23], although COX-2 and/or PGs are proposed toave suppressive effects on the development and exacerbation ofolitis under several conditions [24,25]. However, the exact rolesf COXs and PGs in contractile responses in the rectum with andithout colitis have not been elucidated.

In general, the rectum is thought to function as a reservoir,nd the anal canal is a sphincter area maintaining a high rest-ng tone to resist the inappropriate passage of feces and gas. Thenternal anal sphincter, which is anatomically well defined as a pro-ounced thickening of the circular smooth muscle layer separatedy a short transition zone, is of great significance in retaining thisesting tone [26]. The volume required to induce a desire to defe-ate and to cause sustained anal relaxation was lower in patientsith active UC compared with controls, and circular muscle prepa-

ations of UC rectum had increased sensitivity to carbachol [27]. Theectum possesses mechanical passive and active properties, includ-ng spontaneous resting tension and dynamic motility, comparableo those of the internal anal sphincter, but limited information isvailable concerning the contractile response in whole longitudi-al preparations of rectum. Hyperalgesia to mechanical distensionsf the rectum in patients with active UC is likely as a result ofhanges in mechanical properties of neurons in the rectum becausef edema and sustained contractions of the smooth muscle [28,29],n addition to the inflammation per se. PGE2-induced contrac-ions in rat rectum [8] and the role of nitric oxide in musclesrom human rectum [30,31] were reported. However, the precise

echanism(s) of neuronal regulation for rectal motility and for thehanges induced by UC have not been elucidated. In the presenttudy, the contractile response in longitudinal preparations fromhe rectum to electrical stimulation (ES) and to the stimulation ofeceptors by ACh and PGE2, and changes therein in rectum fromice having DSS-induced UC, were investigated.

. Experimental procedures

.1. Animals and induction of experimental colitis

Mail ddY mice were purchased from SLC Co. (Shizuoka, Japan).nimals, weighing 34–41 g, in groups of 5 or 6, were used. Theyere housed under controlled environmental conditions (temper-

ture of 24 ± 2 ◦C and lights on between 7:00 a.m. and 7:00 p.m.)nd fed commercial MF chow (Oriental Yeast Co. Ltd., Tokyo, Japan)or at least 1 week before the experiments. Before the experiments

or measuring contractile response, mice were kept individuallynd fasted for 16–18 h with free access to water. They were sac-ificed by cervical dislocation. Colitis was induced by adding DSSM.W. 5000, Wako, Osaka, Japan) to their drinking water to a finaloncentration of 2.5% as described previously [32]. Animals with

l Research 61 (2010) 48–57 49

diarrhea and bloody excrement were sacrificed from 7 to 10 daysfollowing the DSS treatment, and the decrease of body weight waswithin 15–25%, but not over 25%, of that before the treatment. Sincereactivity to DSS treatment differed slightly between individualanimals, we used those animals that met the above criteria. Hous-ing and handling of animals were performed in accordance withthe Guiding Principles for the Care and Use of Laboratory Animalsapproved by the Japanese Pharmacological Society, and the exper-iments were approved by the Experimental Animal Committee ofChiba University.

2.2. Reagents

The following drugs and chemicals were used: ACh chlo-ride, bethanechol and indomethacin (Sigma, St. Louis, MO, USA);tetrodotoxin (TTX, Wako, Osaka, Japan); atropine sulfate (Nacalai,Kyoto, Japan); PGE2 and diclofenac (Cayman, Ann Arbor, MI); andarachidonic acid (MP Biomedicals, Solon, OH). PGE2, indomethacinand diclofenac were dissolved in dimethyl sulfoxide, and the otherreagents were dissolved in distilled water prior to dilution inKrebs–Henseleit buffer. The stock solution of arachidonic acid wasdissolved in ethanol, and diluted with the buffer containing 0.001%albumin. The vehicle containing dimethyl sulfoxide or ethanol atless than 0.5% had no effect on contraction of the colon and rectumin the presence or absence of stimuli such as ACh and ES.

2.3. Preparations and measurement of contractile response

The entire length of the colon was removed in Krebs–Henseleitbuffer (NaCl, 112.0 mM; KCl, 5.9 mM; CaCl2, 2.0 mM; MgCl2,1.2 mM; NaH2PO4, 1.2 mM; NaHCO3, 25.0 mM; glucose, 11.5 mM;pH 7.4). The whole segment, 10 mm in length from the anal orifice,was used as the rectum. The segments (10 mm in length) of the dis-tal (descending) colon were taken 35–55 mm from the ileo-caecaljunction (10–30 mm from the anal orifice). In some experiments,the distal colon was separated into two parts; Side A close to thetransverse colon and usually referred to as the distal colon, and SideB close to the rectum, and sensitivity to stimuli such as PGE2 wasmeasured. The preparations (including mucosa, circular layer andneuronal plexus) were usually suspended in the longitudinal direc-tion under a 1-g load in a 5-mL organ bath containing the buffer.Since spontaneous contractions in the longitudinal preparationswere much less than those in the circular preparations, we usedthe longitudinal preparations in the present study. The bath wasmaintained at 37 ◦C and continuously bubbled with a mixture of95% O2 and 5% CO2. One end of each segment was attached to anisometric transducer (T-7-8-240, Orientic Co., Tokyo, Japan) or to anisotonic transducer (Type 45347, NEC San-ei, Tokyo), and record-ings were made with a recorder (056, Hitachi, Mito, Japan) via aDC strain-amplifier (6M92 or AS2102, NEC San-ei, Tokyo, Japan).The other end was mounted on a rigid support or on an anodalelectrode placed at the bottom of the bath. The reagents wereadded to the organ bath. At the start of each experiment, the max-imum response to 3 �M ACh was measured in each preparationto evaluate the effects of the reagents tested such as PGE2. Fortreatment with the inhibitors or receptor antagonists, the prepa-rations were incubated for the indicated period with the reagents,after which contractile response was measured in the presence ofthe respective reagents. The inhibitors or antagonists were usedat concentrations reported previously [32,33]. A stimulator (Sen-7203, Nihon Koden, Tokyo) was used for the electrical transmural

stimulation (ES) of the rectum, and the preparations were stimu-lated by platinum needle-ring (a ring was placed 20 mm above thebase of the needle 5 mm in length) electrodes. The conditions were10 V, 1 ms duration, 5 ms interval, and 10 trains, and the responseswere recorded with a 1-min interval between tests. For contraction

50 Y. Hamada et al. / Pharmacologica

Fig. 1. Effects of TTX, atropine, and COX inhibitors on ES-induced contractionsof the mouse rectum. Longitudinal preparations were pretreated with vehicle,1 �M TTX for 10 min, 1 �M atropine (Atro) for 10 min, or 20 �M COX inhibitor for30 min (indomethacin (Indo) or diclofenac (Dic)), and then the ES-induced isotoniccontraction was measured. In some cases, the preparations were pretreated withboth atropine and indomethacin. The recordings in Panel A were from experimentstypical of more than three independent experiments, and the response withdiclofenac was almost the same as that with indomethacin. The quantitative dataare shown in Panel B (peak) and in Panel C (dimension, area under the trace).

l Research 61 (2010) 48–57

of the distal colon, a stimulator (S9, Grass Instrument, Quincy, MA)was used, and the conditions were 30 V intensity, 0.1 ms duration,and 10 Hz frequency for 5 s. As an isometric response, the absolutevalues of contraction (mg per 10 mm length) were measured. Anisotonic contraction was expressed as a percentage of the contrac-tion induced by 3 �M ACh (% of ACh contraction). In some cases, thedata were normalized as percentages of the corresponding controlresponses to the indicated reagents. The use of isotonic transduc-ers allowed us to record contraction as changes in smooth muscleelongation under constant tension. In isotonic recordings, there canbe no change in length until the tension exceeds the load, thusminimizing possible individual variations of intrinsic contractileactivity of muscles [34,35]. The spontaneous contraction meansa maximal difference between the peak and the lowest values inthe presence of vehicle. In some cases, area under the trace wasexamined as dimension (Fig. 1C). The myeloperoxidase activity inthe preparations was determined as reported [20,36] with minormodifications.

2.4. Western blot analysis

The protein levels of COX-1 and COX-2 were analyzed by West-ern blot using total protein extracts of rectum and distal colonfrom control and DSS-treated mice. Tissue samples were homoge-nized with an ice-cold homogenate buffer (250 mM sucrose, 10 mMEGTA, 2 mM EDTA, 1% Triton-X100, 10 �g/mL leupeptin, 10 �g/mLaprotinin, 100 �M phenylmethylsulfonyl fluoride, 20 mM Tris–HCl,pH 7.4) at 4 ◦C with 20 strokes of a glass-teflon homogenizer. Thelysates were centrifuged at 17,400 × g for 30 min at 4 ◦C, and theresulting supernatant was collected as total tissue extracts. Proteinextracts (50 �g of protein per lane) were fractionated by SDS-PAGEand transferred to PVDF membranes. The blocked membranes werethen incubated with the following antibodies (1:500–1:1000): anti-COX-1 antibody (sc-1752, Santa Cruz Biotech, Santa Cruz, CA),anti-COX-2 antibody (sc-1745, Santa Cruz Biotech), and anti-�-tubulin antibody (TUB 2.1, ICN Biomedicals, Aurora, OH).

2.5. Statistical analysis

Values are presented as means ± S.E.M. for three or more inde-pendent experiments. Each experiment was done by using differentanimals and on different times, and the number of experiments isshown as (n). The statistical significance of differences between twogroups was assessed using the two-tailed Student’s t-test. Multiplecomparisons against a single control group were made by a one-way analysis of variance followed by Dunnett’s test. P < 0.05 wasconsidered significant.

3. Results

3.1. Characteristics of ES-induced contraction of longitudinalpreparations from mouse rectum

The contractile response of the rectum was recorded isotoni-cally. Treatment with ES caused a transient contraction in mouserectum, and the response was almost completely inhibited by 1 �M

TTX (Fig. 1A). The responses (peak (Fig. 1B) and dimension (Fig. 1C,area under the trace)) induced by ES were inhibited 50% by 1 �Matropine and by COX inhibitors (20 �M indomethacin or 20 �Mdiclofenac), respectively, and the responses were almost com-pletely inhibited by co-treatment with atropine and indomethacin.

The number of animals is given in parentheses. aP < 0.05, significantly different fromthe control without the inhibitors. bP < 0.05, significantly different from the valuewith indomethacin alone.

Y. Hamada et al. / Pharmacological Research 61 (2010) 48–57 51

Fig. 2. Enhancement of ES-induced contractions by arachidonic acid. The prepara-tions were incubated with vehicle (containing 0.1% ethanol) or 10 �M arachidonicaw(i

Tt(TrasCcetipavtiie

3

crwn[tinw

Fig. 3. ES-induced contractions in rectum from DSS-treated mice. The spontaneous

cid (AA, containing 0.1% ethanol) for 5 min. The buffer was further supplementedith 0.001% albumin. Then, the ES-induced isotonic contraction was measured

Panel A). The quantitative data (peak values) are shown in Panel B. aP < 0.05, signif-cantly different from the control.

he inhibitory effects by 3 and 10 �M atropine on ES-induced con-raction were about 50%, almost same as that by 1 �M atropinedata not shown). Similar results (almost complete inhibition byTX, and 50% inhibition induced by atropine and by indometacin,espectively) were obtained under conditions of a 0.5-g loadnd by measuring the isometric contractile responses (data nothown). These data suggest that both the ACh pathway and theOX-mediated pathway are equally involved in the ES-inducedontraction in mouse rectum. Next, we investigated the effect ofxogenously added arachidonic acid, a major substrate for COX andhe precursor of eicosanoids such as PGE2, on contractions. The ES-nduced contraction was enhanced in the preparations that werereincubated with 10 �M arachidonic acid for 5 min (Fig. 2). Prob-bly because of ethanol and albumin, the responses to ES in theehicle-treated preparations were dependent on each experiment;hose were under 20% of the ACh response in some cases as shownn Fig. 2A, and those were about 40% in other cases like the responsen control (Fig. 1A). Treatment with arachidonic acid alone had noffect on spontaneous contraction (dada not shown).

.2. ES-induced responses in rectum from DSS-treated mice

It was reported that COX-2 expression was up-regulated in ratolon with colitis, and the enzyme appeared to play a predominantole in regulating colonic neuromuscular function [3]. Previously,e reported that DSS treatment caused changes predominantly inicotine-sensitive neurogenic responses in the distal colon in mice

32]. In the present study, we investigated changes in the contrac-ile response in rectum from control and DSS-treated mice by thesometric assay. The rectum from control mice exhibited sponta-eous phasic contractions, but the response from DSS-treated miceas markedly decreased (Fig. 3A), as previously reported in the dis-

and ES-induced contractions in rectum from control mice and DSS-treated micewere measured isometrically. The recordings in Panel A were from typical examples,and the quantitative data (peak values) are shown in Panel B. aP < 0.05, significantlydifferent from the control.

tal colon [32]. The ES-induced contraction significantly decreasedin the DSS-treated mice (Fig. 3B).

3.3. Effects of ACh and PGE2 on contraction in rectum fromDSS-treated mice

To examine the changes of contractile response induced byreceptor stimulation in the rectum from DSS-treated mice, weinvestigated the effects of ACh and PGE2 by bath application onisometric contraction (Fig. 4A). In order to eliminate the possiblesecondary effects induced by mediators from mucosa including 5-hydroxytryptamine (5-HT), we evaluated the first quick contractioninduced by the exogenous stimuli. Treatment with ACh induceda maximal contraction at 5–10 s after stimulation, and the con-traction continued for about 1 min with rhythmic responses. Then,the ACh response returned to the basal level within 1–2 min afterstimulation. Treatment with 3 �M PGE2 caused a contraction, andthe response continued for over 2 min with rhythmic responses.The peak contraction induced by 10 �M PGE2 was 80% of that by3 �M PGE2 (data not shown). Treatment with 1 �M atropine com-pletely inhibited the ACh-induced contraction without changingthe response to PGE2 (data not shown). Treatment with TTX orindomethacin did not change the responses to 3 �M ACh and 3 �MPGE2 in the rectum, and the response to 0.3 �M ACh was slightly(up to 20%) inhibited by TTX treatment (data not shown). Treat-ment with 0.3 �M ACh induced significantly fewer contractions inrectum from DSS-treated mice than control mice, but the maximalresponses evoked by 3 �M ACh in the preparations from controland DSS-treated mice were not significantly different (Fig. 4B andTable 1). The ED50 values of ACh were 1.5–2 �M in both control andDSS-treated samples, as in the distal colon [32]. By contrast, theresponse induced by 3 �M PGE2 was markedly inhibited by DSStreatment. Like spontaneous contractions (Fig. 3A), the rhythmic

contractions induced by receptor stimulation appeared to be lessin rectum from DSS-treated mice (Fig. 4A). Under our conditionsfor measurement of isometric contractions, the responses includ-ing the periods and the rhythmic contractions induced by ACh andPGE2 were dependent on the preparations even from the control

52 Y. Hamada et al. / Pharmacological Research 61 (2010) 48–57

Fig. 4. ACh-, PGE2-, and KCl-induced contractions in rectum from DSS-treated mice. The cba

TDt

T3iitoSn

y 40 mM and 100 mM KCl (Panels C and D) in rectum from control and DSS-treated micend the quantitative data (peak values) are shown in Panels B and D. aP < 0.05, significant

able 1own-regulation of the response to PGE2 on repeated application of PGE2 and DSS

reatment in the mouse rectum.

Isometric contraction (mg) Control DSS-treated

ACh1st trial 1689 ± 73 (16) 1360 ± 162 (7)2nd trial 1767 ± 77 (16) 1435 ± 159 (7)Ratio (2nd/1st) 104.7 ± 1.3 (16) 106.9 ± 5.3 (7)

PGE2

1st trial 932 ± 139 (3) 86.3 ± 43.7a (4)2nd trial 525 ± 75b (3) 67.7 ± 28.9a (4)Ratio (2nd/1st) 57.5 ± 9.3b (3) Not determined

he preparations of rectum from control and DSS-treated mice were stimulated with�M ACh or 3 �M PGE2. Then, the same preparations were washed three times and

ncubated with reagent-free fresh buffer for 20 min, and the respective reagent-nduced responses were examined as the 2nd trial. The basal tone before the 2ndrial was almost the same as that before the 1st trial. The ratio of the response (peak)f the 2nd trial to that of the 1st was calculated as a parameter of desensitization.ince the PGE2-evoked responses in the DSS-treated mice were small, the ratio wasot determined. The number of animals is given in parentheses.a P < 0.05, significantly different from the control mice.b P < 0.05, significantly different from the value for the 1st trial.

ontractions induced by 0.3 �M and 3 �M ACh or by 3 �M PGE2 (Panels A and B), andwere measured. The recordings in Panels A and C were from typical experiments,

ly different from the control.

mice. Thus we could not evaluate the values of exact dimension(area under the trace). Treatment with 40 mM and 100 mM KClcaused a marked contraction in the rectum (Fig. 4C). TTX treat-ment slightly (up to 20%) inhibited the response induced by 40 mMKCl, but not by 100 mM KCl (data not shown). In the rectum fromthe DSS-treated mice, the response to 100 mM KCl was inhibitedbut not significantly, although the response to 40 mM KCl wassignificantly inhibited compared with that in the control rectum.These results suggest that DSS treatment impairs preferentially theresponse to PGE2 compared with the responses to ACh and KCl athigh concentrations. Next we examined whether the PGE2-evokedresponse in the rectum is desensitized or not (Table 1). In rectumfrom control mice, the response induced by the 2nd trial of PGE2was half that induced by the 1st trial of 3 �M PGE2, suggesting thatthe response to PGE2 was desensitized. In rectum from DSS-treatedmice, the ratio (2nd/1st) of the response to PGE2 could not be cal-culated because of invalidity of PGE2. The responses to 3 �M AChwere not changed by the repeated trials in rectum from the controland DSS-treated mice.

We further examined the change in the responses induced byES and PGE2 in the DSS-treated mice using an isotonic transducer.On minimization of possible variations in the intrinsic activity ofmuscles, the degree of 3 �M ACh-induced contractions in rectumpreparations from DSS-treated mice was almost the same as that

Y. Hamada et al. / Pharmacological Research 61 (2010) 48–57 53

Table 2Contractile responses in different parts of the distal colon from control and DSS-treated mice.

Isometric contraction (mg) Distal colon (Side A) Distal colon (Side B; close to rectum)

Control DSS-treated Control DSS-treated

Spontaneous 62.4 ± 8.6 (28) 13.9 ± 3.0a (24) 75.4 ± 8.0 (28) 13.4 ± 4.6a (24)KCl (40 mM) 954 ± 117 (9) 635 ± 111 (6) 1197 ± 152 (10) 389 ± 41a (7)ACh (3 �M) 1636 ± 222 (7) 1295 ± 181 (7) 2415 ± 93 (7) 2211 ± 111 (7)pEC50 −6.3 to −5.3 −6.2 to −4.6 −6.3 to −5.9 −6.2 to −5.6Bethanechol (30 �M) 1793 ± 303 (4) 1507 ± 68 (5) 2138 ± 154 (4) 1977 ± 221 (5)

PGE2

1st trial (3 �M) 1557 ± 183 (10) 471 ± 121a (10) 1517 ± 156 (10) 601 ± 127a (10)2nd trial (3 �M) 1353 ± 161 (10) 292 ± 51a (10) 1372 ± 149 (10) 317 ± 78.2a,b (10)Ratio (2nd/1st) 86.6 ± 4.6 70.1 ± 5.9 90.3 ± 5.8 60.0 ± 7.7a

Preparations of distal colon from control and DSS-treated mice were isolated as described in Section 2. The contractile activity was recorded isometrically in the presence oft respoa is give

fitpornEtm

3c

tcttlivsaDtAtiinbmrRwtnnv∼Dbttn

he indicated reagents. The pEC50 values of ACh are also shown. For measuring thend then stimulated with 3 �M PGE2 again as the 2nd trial. The number of animals

a P < 0.05, significantly different from the control mice.b P < 0.05, significantly different from the 1st response.

rom control mice. Both a desensitization of the response to PGE2n rectum from control mice and an impairment of the response inhe DSS-treated mice were observed under isotonic conditions; theeak response was 82.1 ± 5.5 (% of ACh, n = 3) and the ratio (2nd/1st)f the response was 73.6 ± 6.4% (n = 3) of the control value. Theesponse to PGE2 in the DSS-treated mice was 7.7 ± 4.9 (% of ACh,= 3) which was significantly lower than that in the control mice.ven taking a slight decrease in the ACh-evoked response on DSSreatment into consideration, the response to PGE2 appeared to be

arkedly modified by the treatment under isotonic conditions.

.4. Regional differences in the response to PGE2 in the distalolon and rectum, and their changes on DSS treatment

Next, we examined the difference of contractile responses in thewo parts of the longitudinal preparations from distal colon; distalolon (Side A, close to the transverse colon), and distal colon closeo the rectum (Side B). The order in magnitude of spontaneous con-raction was Side A < Side B (Table 2), and the value tended to beower in Side B than in rectum (Fig. 3B). The spontaneous responsen Side B was significantly decreased by DSS treatment, as pre-iously reported in the distal colon (Side A in Table 2, and alsoee Ref. [32]). The maximal responses in Side B induced by AChnd bethanechol were slightly, but not significantly, inhibited bySS treatment. The ED50 values of ACh were not modified by DSS

reatment, and the response induced by the 2nd trial with 3 �MCh was almost the same as in the 1st trial (data not shown) in

he two parts of the distal colon. Like in the rectum, 40 mM KCl-nduced contractions were significantly inhibited in Side B, but notn Side A, by DSS treatment. The PGE2-induced contraction wasot desensitized; the ratio in the 2nd/1st trial was about 90%, inoth parts of the distal colon. DSS treatment caused an impair-ent of the PGE2-evoked response in Side A, the 1st PGE2-induced

esponse decreased to 30–40% of the control response (Table 2 andef. [32]), and a similar impairment of the response to PGE2 by DSSas obtained in Side B. In the distal colon from DSS-treated mice,

he ratio (2nd/1st) of the PGE2-evoked response in Side B was sig-ificantly lower, and the ratio in Side A appeared to be lower, butot significantly, compared with that for control mice. The ED50alues of PGE obtained by using independent preparations were

20.3 �M in both rectum and distal colon (both sides). The values inSS-treated mice could not be examined because of a limited num-er of animals. In both sides of the distal colon from control mice,he ES-induced contractions were largely (over 60%) inhibited byreatment with 1 �M atropine but not 20 �M indomethacin (dataot shown).

nse to PGE2, the preparations were stimulated with 3 �M PGE2 (1st trial), washed,n in parentheses.

3.5. Increased expression of COX-2 in rectum, but not in distalcolon, from DSS-treated mice

We analyzed the protein expression levels of COX-1 and COX-2in the rectum from control and DSS-treated mice (Fig. 5A and B). DSStreatment significantly increased levels of COX-2, but not COX-1,in rectum. In these experiments, the extracts were obtained fromthe preparations that were used for measurement of contractileresponse. In the preparations from the fresh tissues without con-tractive stimuli such as ES, similar results including an increase ofCOX-2 by DSS treatment were obtained (data not shown). Interest-ingly, DSS treatment did not change the levels of COX-1 and COX-2in Side B (Fig. 5C) and in Side A (data not shown) from distal colon.

4. Discussion

4.1. Role of COX-dependent pathway in ES-induced contractionsin mouse rectum

Our findings showed that a COX-dependent pathway is involvedin the ES-induced contractions in longitudinal preparations ofmouse rectum, in addition to the ACh-mediated pathway (Fig. 1).Treatment of the preparations with arachidonic acid enhanced theES-induced contraction (Fig. 2), and stimulation with PGE2 causeda marked contraction (Fig. 4). The ES-induced contraction and theresponse in the presence of arachidonic acid (data not shown) weresignificantly, almost completely, inhibited by TTX treatment. ThePGE2-induced contraction was not inhibited by TTX and atropine,and treatment with atropine and COX inhibitors showed addi-tional inhibitory effects on the ES-induced contraction (Fig. 1B andC). These results suggest that (1) ES activates TTX-sensitive cellsprobably neurons that release COX metabolites having contractileactivity, and (2) the COX-mediated pathway acts independentlyof the ACh pathway in mouse rectum. To our knowledge, this isthe first report of the involvement of the COX pathway in theES-induced contraction of a longitudinal preparation from the rec-tum. The immunohistochemical analysis of human [2] and mouse[37] colon has demonstrated the expression of COX-1 and COX-2 proteins in enteric neurons including myenteric ganglia. COX-2expression was increased in enteric neurons by interleukin-1�treatment [19] and in the myenteric plexus of patients with activeIBD [18]. As described in Section 1, the effects of prostanoids on

contractions in GI tissues were dependent on PG receptor sub-types and/or tissues [4,5]. Treatment with indomethacin inhibitedthe nicotine-induced release of ACh and resulting contraction inGI tissues including the ileum [6,7]. By contrast, treatment withindomethacin enhanced contractions in a TTX-sensitive manner

54 Y. Hamada et al. / Pharmacologica

Fig. 5. Increased expression of COX-2 in rectum from DSS-treated mice. The pro-tein levels of COX-1 and COX-2 were analyzed by Western blot using total proteinextracts of rectum (Panels A and B) and distal colon (Side B, Panel C) from controland DSS-treated mice. The data in Panel A were from a typical experiment by usingrectum preparations from a control mouse and two DSS-treated mice. In the toppanel showing COX-1, the upper band cross-reacted with anti-COX-1 antibody isprobably non-specific because of uncertain appearance depending on experiments.The quantitative data for rectum and for distal colon (Side B) are shown in PanelB and in Panel C, respectively. The ratio of COX-2 to �-tubulin was calculated, andtTf

iidCjtcair

e[

by 0.3 �M ACh and 40 mM KCl are mediated by the putative neu-

he data are normalized as fold-increases of the control without DSS treatment.he number of animals used is given in parentheses. aP < 0.05, significantly differentrom the control without DSS.

n circular preparations from mouse colon [37]. Treatment withndomethacin enhanced the ES-induced contractions in longitu-inal preparations of distal colon from humans and mice, andOXs inhibited the cholinergic response at both pre- and post-

unctional sites [2,3]. These findings including our data, which showhe diversity of neurons interacting with the COX pathway and/orontradictory effects on contraction, suggest that COX expressionnd/or activity in neurons may regulate the motility of GI tissuesncluding the rectum. The precise mechanism(s) responsible for the

elease of PGs by enteric neuronal activity remain to be elucidated.

In longitudinal muscle preparations from rat rectum, PGE2xerted both a neuronal (cholinergic) and a smooth muscle effect8]. Since treatment with neither TTX nor atropine inhibited the

l Research 61 (2010) 48–57

PGE2-induced contraction of mouse rectum in the present study,PGE2 appears to stimulate smooth muscles in the rectum directly.Different effects of PGE2 on circular vs. longitudinal smooth mus-cle preparations from the GI tract have been reported. In general,PGE2 causes a contractile response in longitudinal preparations andhas a relaxant effect on circular preparations from ileum and colon,although there are variations depending on the animal species, tis-sues, and preparations used [4,5]. The basal tone of rabbit internalanal sphincter, a specialized continuation of the circular musclelayer of the rectum, was not modified by indomethacin in opos-sum [38] and rabbit [39]. In preliminary experiments, treatmentwith PGE2 did not evoke marked responses including contrac-tion in circular muscle preparations from mouse rectum (data notshown). PGE2 may have a contractile influence on longitudinal andan inhibitory effect and/or no influence on circular muscle prepa-rations of rectum, respectively, as generally proposed in ileum andcolon [7]. The precise roles of COX and/or PGE2 in the contrac-tile responses in circular muscle preparations of rectum and distalcolon remained to be solved in future.

4.2. Effects of DSS treatment on spontaneous and ES-inducedcontractions in rectum

The expression of COX-2 was increased both in an animal modelof colitis and in human IBD [3,18–20]. In addition to COX-1, COX-2 plays a key role in IBD inflammation through its production ofPGs [22,23]. In the present study, DSS treatment of mice increasedCOX-2 expression in rectum (Fig. 5A and B). Thus, we focused onthe COX-mediated and PGE2-induced responses in rectum fromDSS-treated mice. In longitudinal preparations of the rectum fromDSS-treated mice, both the spontaneous and the ES-induced con-tractions were significantly decreased, by about half, comparedwith those in the control mice (Fig. 3). The decreases in the ES-induced response in DSS-treated mice were detected by measuringisotonic contractions. By contrast, bath applications of 3 �M AChand 100 mM KCl caused marked contractions to a similar degreein rectum from both control and DSS-treated mice. Thus, DSStreatment appeared to modify neuronal functions rather than theintrinsic activity of muscles in the rectum. Previously, we reportedthat DSS treatment impaired neurons activated by nicotine, caus-ing the relaxation of longitudinal preparations via the formation ofnitric oxide in mouse distal colon [32]. Changes were reported incolonic motility and the excitability of myenteric and submucosalneurons in distal colon of guinea-pigs with experimental colitis[14,40]. Our data and these reports suggest that enteric neuronalactivities are modified by DSS treatment in the GI tract including therectum. Interestingly, the remaining response to ES in rectum fromDSS-treated mice tended to be COX inhibitor-resistant, and wasmarkedly inhibited by 1 �M atropine, although the degree of theremaining response and the sensitivity to COX inhibitors dependedon the individual tested. It remains to be clarified which neuronalpathway, the ACh pathway or the indomethacin-sensitive pathway,is more sensitive to DSS treatment.

Under our conditions, the moderate contractions of longitudi-nal preparations induced by 0.3 �M ACh and by 40 mM KCl wereinhibited in the rectum (Fig. 4) and distal colon (Table 2) from theDSS-treated mice. The responses to 0.3 �M ACh and 40 mM KCl inrectum from control mice were slightly (up to 20%) TTX-sensitive,although they were not modified by 300 �M hexamethonium, aganglionic-specific antagonist for the nicotinic ACh receptor (datanot shown). Thus, it is possible the moderate contractions induced

ronal pathways that are sensitive to DSS treatment. However, theinhibitory effects of DSS treatment on these responses were muchgreater (approximately 60–80% in Fig. 4B and D) compared withthose of TTX treatment. The pED50 value of ACh in the rectum

logica

fmT0ot[prtKtDiiitr

4r

ts3imaSTiifpdtttAtwoIcmUrmCBWfudt

us[sbncgv

Y. Hamada et al. / Pharmaco

rom DSS-treated mice was almost the same as that for controlice because of extensive variation; the values were −6.0 ± 0.3.

he responses induced by lower concentrations of ACh such as.1–0.5 �M, but not 3 �M, were inhibited by DSS treatment. Previ-usly, we reported that the pED50 values of ACh and bethanechol forhe contraction of distal colon were not changed by DSS treatment32]. These findings may suggest the involvement of non-neuronalathways, which are stimulated by 0.3 �M ACh or 40 mM KCl andegulate contractile responses in the rectum, are impaired by DSSreatment. Several reports found that the responses to ACh andCl were impaired in colon and ileum from animals with coli-

is probably because of dysfunction in smooth muscles [41–43].SS treatment resulted in changes of ion channel activities includ-

ng voltage-dependent L-type Ca2+ currents in smooth musclesn mouse distal colon [44]. Thus, DSS treatment may moderatelympair the intrinsic contractile activity of longitudinal muscles ofhe rectum, in addition to the neuronal changes and the loss of PGE2eceptor-mediated signaling (as described below).

.3. Effect of DSS treatment on PGE2-induced contractions inectum

In the control rectum, the response to PGE2 was desensi-ized; the second response to PGE2 in the same preparation wasignificantly weaker than the first response without loss of the�M ACh (Table 1) and 10 �M bethanechol (data not shown)-

nduced responses. The PGE2-induced contractions in rectum werearkedly less than those in the two regions of the distal colon,

lthough the responses to ACh were similar (Tables 1 and 2).mall numbers of functional receptors for PGE2 including EP1 andP receptors in rectum may explain the results. Otherwise, thentracellular pathways that regulate functions downstream fromntracellular messengers including cyclic AMP and Ca2+ may dif-er between the muscles from rectum and colon, as proposedreviously [45,46]. The precise mechanism responsible for theesensitization of the PGE2-evoked response in rectum remainso be elucidated. The PGE2-induced contraction of rectum fromhe DSS-treated mice was significantly less than that of the con-rol preparation (Fig. 4 and Table 1), without a change in the 3 �MCh- and 100 mM KCl-induced responses. These results suggest

hat DSS treatment preferentially impairs the PGE2-mediated path-ay probably in smooth muscles of the rectum. The up-regulation

f COX-2 has been shown in various GI tissues from patients withBD and mammals with experimental colitis [17–21,47]. The tissueoncentration of PGE2 was remarkably elevated in the intestinalucosa and in segments of the recto-sigmoid of patients withC [48,49], and mucosal T lymphocytes showed a change in EP

eceptor levels (an increase in the EP4 receptor) during inflam-ation [50]. In the present study, we confirmed the increase of

OX-2 expression in rectum from DSS-treated mice (Fig. 5A and), although the levels of PGE2 and its receptor were not analyzed.e speculate that the excess of COX-2 expression and resulting

ormation of PGE2 would be counter-balanced by the negative reg-lation of the response to PGE2 such as desensitization and/orown-regulation of receptor in the rectum of mice with coli-is.

In the present study, we examined the contractile responses bysing whole segments including mucosa. Several studies demon-trated an increase of 5-HT content in the mucosa in colons with UC51,52]. In cases of increased 5-HT availability or prolonged expo-ure, such as colitis, the peristaltic reflex is likely to be blunted

ecause of desensitization of 5-HT4 receptors located on entericeurons [53], and 5-HT regulates the colonic migrating motoromplex, one of abnormal motor patterns implicated in the patho-enesis of constipation and diarrhea [54]. The cross-talks betweenarious cell types including an interaction between mucosal cells

l Research 61 (2010) 48–57 55

and enteric neurons in GI tissues and their changes by IBD and/orUC should be elucidated in future.

4.4. Colon region-dependent PGE2 response and DSS sensitivity

In DSS-treated mice and rats, changes in inflammatory mark-ers and contractile responses were greater in the distal colon thanproximal colon [55,56]. DSS-induced colitis, evaluated by scores forinflammation and cell damage, was particularly severe in the rec-tum compared with the colon (distal > transverse colons) in mice[16,36]. To examine whether the sensitivity to PGE2 and DSS treat-ment is colon region-dependent or not, we separated the distalcolon into two (Side A, distal colon close to the transverse colon;Side B, that close to the rectum) and compared the sensitivitybetween the two sides and the rectum. Our data showed that PGE2-induced contractions were negatively regulated in the rectum, butnot in the distal colon (both regions), from control mice. The PGE2-induced contraction of distal colon from DSS-treated mice was halfthat in the control preparation (Table 2), although the responseof the rectum was largely, over 90%, inhibited by DSS treatment(Table 1). Interestingly, the response to PGE2 was desensitized onSide B of the distal colon from DSS-treated mice, and weak on SideA (Table 2). The contractile response induced by 40 mM KCl wassignificantly inhibited in the rectum and Side B, but not Side A,by DSS treatment (Fig. 4 and Table 2). These results suggest thatthe impairment of the response to PGE2 (and to 40 mM KCl) byDSS treatment was colon region-dependent: rectum > distal colonclose to rectum > distal colon. Under our conditions, DSS treatmentincreased COX-2 expression in rectum, but not in the both regionsfrom distal colon (Fig. 5). Park et al. [46] showed that there was aclear segmental heterogeneity with regard to electrogenic secre-tions in human proximal colon and rectum, and proposed thatPGE2 is responsible for the difference. In the present study, treat-ment with COX inhibitors decreased the ES-induced contractionin rectum but not in distal colon, although the conditions for ESin the rectum and colon were different. In addition to the role ofCOX, PGE2-regulated pathways including desensitization and/ordown-regulation of receptor and intracellular signaling appearedto differ depending on the region of the colon and rectum. In pre-liminary experiments, DSS treatment increased myeloperoxidaseactivity in rectum and distal colon in mice (data not shown), asreported previously in various species [20,36]. In addition to thechanges in the COX-2-prostanoids pathway, we should find theregional variation explaining the severity in rectum in DSS-treatedmice.

4.5. Summary

In the present study, we showed that the ES-induced TTX-sensitive contraction of longitudinal preparations of mouse rectumwas mediated by the COX-dependent pathway and by the ACh-mediated pathway. The ES-induced contraction, but not 3 �MACh-induced muscular response, was impaired by DSS treatment.The response to PGE2 in rectum, which was desensitized in con-trol mice, was impaired by DSS treatment. The loss of response onDSS treatment was region specific, with the following order: rec-tum > distal colon close to the rectum > distal colon. The responseto PGE2 was not desensitized in either region of the control dis-tal colon, but the response in the distal colon close to the rectumin DSS-treated mice was significantly desensitized and/or down-regulated. Actually, DSS treatment increased COX-2 expression in

rectum, but not in distal colon. Thus, the COX-2-dependent mecha-nisms in rectum may affect the contractile responses in neighboringdistal colon in DSS-treated mice. The mechanism responsible forthe regulation of COX activity in neuronal cells in the rectumremained to be elucidated, and the changes of contractile response

5 logica

ms

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

6 Y. Hamada et al. / Pharmaco

ediated by the COX- and PGE2-mediated pathways should betudied in patients with UC.

eferences

[1] Warner TD, Mitchell JA. Cyclooxygenases: new forms, new inhibitors, and les-son from the clinic. FASEB J 2004;18:790–804.

[2] Fornai M, Blandizzi C, Colucci R, Antonioli L, Bernardini N, Segnani C, et al. Roleof cyclooxygenases 1 and 2 in the modulation of neuromuscular functions inthe distal colons of humans and mice. Gut 2005;54:608–16.

[3] Fornai M, Blandizzi C, Antonioli L, Colucci R, Bernardini N, Segnani C, et al. Dif-ferential role of cyclooxygenase 1 and 2 isoforms in the modulation of colonicneuromuscular function in experimental inflammation. J Pharmacol Exp Ther2006;317:938–45.

[4] Coleman RA, Smith WL, Narumiya S. VIII. International Union of Pharmacology.Classification of prostanoid receptors: properties, distribution, and structure ofthe receptor and their subtypes. Pharmacol Rev 1994;46:205–29.

[5] Okada Y, Hara A, Ma H, Xiao CY, Takahata O, Kohgo Y, et al. Characteriza-tion of prostanoid receptors mediating contraction of the gastric funds andileum: studies using mice deficient in prostanoid receptors. Br J Pharmacol2000;131:745–55.

[6] Fukunaga Y, Mine Y, Yoshikawa S, Takeuchi T, Hata F, Yagasaki O. Role of prosta-cyclin in acetylcholine release from myenteric plexus of guinea-pig ileum. EurJ Pharmacol 1993;233:237–42.

[7] De Backer O, Leclere PG, Lefebvre RA. Pharmacological characterization of pre-and postsynaptic prostanoid receptors in pig gastric fundus. Neuropharmacol-ogy 2003;45:684–90.

[8] Vassilev P, Radomirov R. Contractile effects of prostaglandin E2 in rat rec-tum: sensitivity to the prostaglandin antagonists diphloretin phosphate andSC 19220. Prostaglandins 1992;44:471–83.

[9] Geboes K, Collins S. Structural abnormalities of the nervous systemin Crohn’s disease ulcerative colitis. Neurogastroenterol Motil 1998;10:189–202.

10] Danese S, Fiocchi C. Etiopathogenesis of inflammatory bowel diseases. World JGastroenterol 2006;12:4807–12.

11] Spiller R. Recent advances in understanding the role of serotonin in gastroin-testinal motility in functional bowel disorders: alternations in 5-HT signallingand metabolism in human disease. Neurogastroenterol Motil 2007;19:25–31.

12] De Giorgio R, Guerrini S, Barbara G, Stanghellini V, De Point F, Corinaldesi R, etal. Inflammatory neuropathies of the enteric nervous system. Gastroenterology2004;126:1872–83.

13] Lomax AE, Mawe GM, Sharkey KA. Synaptic facilitation and enhanced neuronalexcitability in the submucosal plexus during experimental colitis in guinea-pig.J Physiol 2005;564:863–75.

14] Lomax AE, O’Hara JR, Hyland NP, Mawe GM, Sharkey KA. Persistent alternationsto enteric neuronal signaling in the guinea pig colon following the resolutionof colitis. Am J Physiol Gastrointest Liver Physiol 2007;292:G482–91.

15] Melgar S, Karlsson A, Michaelsson A. Acute colitis induced by dextran sulfatesodium progresses to chronicity in C57BL/6 but not in BALB/c mice: correlationbetween symptoms and inflammation. Am J Physiol Gastrointest Liver Physiol2005;288:G1328–38.

16] Kudo T, Matsumoto T, Nakamichi I, Yada S, Esaki M, Jo Y, et al. Recombi-nant human granulocyte colony-stimulating factor reduces colonic epithelialcell apoptosis and ameliorates murine dextran sulfate sodium-induced colitis.Scand J Gastroenterol 2008;43:689–97.

17] Singer II, Kawka DW, Schloemann S, Tessner T, Riehl T, Stenson WF. Cyclooxy-genase 2 is induced in colonic epithelial cells in inflammatory bowel disease.Gastroenterology 1998;115:297–306.

18] Robert PJ, Morgan K, Miller R, Hunter JO, Middleton SJ. Neuronal COX-2 expres-sion in human myenteric plexus in active inflammatory bowel disease. Gut2001;48:468–72.

19] Tjwa ETTL, Bradley JM, Keenan CM, Kroese ABA, Sharkey KA. Interleukin-1� activates specific populations of enteric neurons and enteric glia inthe guinea pig ileum and colon. Am J Physiol Gastrointest Liver Physiol2003;285:G1268–76.

20] Martín AR, Villegas I, de la Lastra CA. The COX-2 inhibitor, refecoxib, amelio-rates dextran sulphate sodium induced colitis in mice. Inflamm Res 2005;54:145–51.

21] Jupp J, Hillier K, Elliott DH, Fine DR, Bateman AC, Johnson PA, et al. Colonicexpression of leukotriene-pathway enzymes in inflammatory bowel diseases.Inflamm Bowel Dis 2007;13:537–45.

22] Mahadevan U, Loftus EV, Tremaine WJ, Sandborn WJ. Safety of selectivecyclooxygenase-2 inhibitors in inflammatory bowel disease. Am J Gastroen-terol 2002;97:910–4.

23] Singh V, Patil CS, Jain NK, Singh A, Kulkarni SK. Effects of nimesulfide onacetic acid- and leukotriene-induced inflammatory bowel disease in rats.Prostaglandins Other Lipid Mediat 2003;71:163–75.

24] Reuter BK, Asfaha S, Buret A, Sharkey KA, Wallace JL. Exacerbationof inflammation-associated colonic injury in rat through inhibition ofcyclooxygenase-2. J Clin Invest 1996;98:2076–85.

25] Kabashima K, Saji T, Murata T, Nagamachi M, Matsuoka T, Segi E, et al. Theprostaglandin receptor EP4 suppresses colitis, mucosal damage and CD4 cellactivation in the gut. J Clin Invest 2002;109:883–93.

[

l Research 61 (2010) 48–57

26] Rattan S, Regan RF, Patel CA, De Godoy MAF. Nitric oxide not carbon monoxidemediates nonadrenergic noncholinergic relaxation in the murine internal analsphincter. Gastroenterology 2005;129:1954–66.

27] Cook TA, Brading AF, Mortensen NJM. Abnormal contractile properties ofrectal smooth muscle in chronic ulcerative colitis. Aliment Pharmacol Ther2000;14:1287–94.

28] Drewes AM, Frøkjær JB, Larsen E, Reddy H, Arendt-Nielsen L, Gregersen H. Painand mechanical properties of the rectum in patients with active ulcerativecolitis. Inflamm Bowel Dis 2006;12:294–303.

29] Qin C, Malykhina AP, Akbarali HI, Van Meerveld BG, Foreman RD. Acute colitisenhances responsiveness of lumbosacral spinal neurons to colorectal disten-sion in rats. Dig Dis Sci 2008;53:141–8.

30] Glavind EB, Forman A, Madsen G, Tøttrup A. Effects of transmural field stimu-lation in isolated smooth muscle of human rectum and internal anal sphincter.Am J Physiol Gastrointest Liver Physiol 1997;272:G1075–82.

31] Stebbing JF, Brading AF, Mortensen NJ. Role of nitric oxide in relaxation ofthe longitudinal layer of rectal smooth muscle. Dis Colon Rectum 1997;40:706–10.

32] Murakami I, Hamada Y, Yamane S, Fujino H, Horie S, Murayama T. Nicotine-induced neurogenic relaxation in the mouse colon: its changes with dextransodium sulfate-induced colitis. J Pharmacol Sci 2009;109:128–38.

33] Horie S, Yasuda S, Tsurumaki Y, Someya A, Saito T, Okuma Y, et al. Contrac-tion of isolated guinea-pig ileum by urotensin II via activation of ganglioniccholinergic neurons and acetylcholine release. Neuropharmacology 2003;45:1019–27.

34] Barlow RB, Bond SM, Grant C, McQueen DS, Yaqoob Z. A comparison of effectsmeasured with isotonic and isometric recording. I. Concentration-effect curvesfor agonists. Br J Pharmacol 2001;133:1081–6.

35] Barlow RB, Bond SM, Grant C, McQueen DS, Yaqoob Z. A comparison ofeffects measured with isotonic and isometric recording. II. Concentration-effect curves for physiological antagonists. Br J Pharmacol 2001;133:1087–95.

36] Sasaki S, Hirata I, Maemura K, Hamamoto N, Murano M, Toshina K, et al.Prostaglandin E2 inhibits lesion formation in dextran sodium sulphate-inducedcolitis in rats and reduces the levels of mucosal inflammatory cytokines. ScandJ Immunol 2000;51:23–8.

37] Porcher C, Horowitz B, Ward SM, Sanders KM. Constitutive and functionalexpression of cyclooxygenase 2 in the murine proximal colon. Neurogastroen-terol Motil 2004;16:785–99.

38] Chakder S, Rattan S. Mechanisms and sites of action of endothelins 1 and 2 onthe opossum internal anal sphincter smooth muscle tone in vitro. J PharmcolExpTher 1999;288:239–46.

39] Koyuncu A, Yildirim S, Aydin C, Bagcivan I, Sarac B. Relaxation effects ofadrenomedullin in carbachol-precontracted rabbit internal anal sphincter. JGastroenterol 2003;38:460–4.

40] Linden DR, Sharkey KA, Ho W, Mawe GM. Cyclooxygenase-2 contributes todysmotility and enhanced excitability of AH neurons in the inflamed guineapig distal colon. J Physiol 2004;557:191–205.

41] Gonzalez A, Sarna SK. Different types of contractions in rat colon and theirmodulation by oxidative stress. Am J Physiol Gastrointest Liver Physiol2001;280:G546–54.

42] Sato K, Ninomiya H, Ohkura S, Ozaki H, Nasu T. Impairment of PAR-2-mediatedrelaxation system in colonic smooth muscle after intestinal inflammation. Br JPharmacol 2006;148:200–7.

43] Sato K, Ohkura S, Kitahara Y, Ohama T, Hori M, Sato M, et al. Involvement ofCPI-17 downregulation in the dysmotility of the colon from dextran sodiumsulphate-induced experimental colitis in a mouse model. NeurogastroenterolMotil 2007;19:504–14.

44] Akbarali HI, Pothoulakis C, Castagliuolo I. Altered ion channel activity in murinecolonic smooth muscle myocytes in an experimental colitis model. BiochemBiophy Res Commun 2000;275:637–42.

45] Nobles M, Diener M, Rummel W. Segment-specific effects of the heat-stableenterotoxin of E. coli on electrolyte transport in the rat colon. Eur J Pharmacol1991;202:201–11.

46] Park JH, Rhee PL, Lee JH, Kim JJ, Rhee JC, Kim SJ, et al. Segmental heterogeneity ofelectrogenic secretions in human ascending colon and rectum. Int J ColorectalDis 2006;21:357–64.

47] Okayama M, Hayashi S, Aoi Y, Nishio H, Kato S, Takeuchi K. Aggravation byselective COX-1 and COX-2 inhibitors of dextran sulfate sodium (DSS)-inducedcolon lesions in rats. Dig Dis Sci 2007;52:2095–103.

48] Raab Y, Sundberg C, Hällgren R, Knutson L, Gerdin B. Mucosal synthesis andrelease of PGE2 from activated eosinophils and macrophages in ulcerative col-itis. Am J Gastroenterol 1995;90:614–20.

49] Wiercinska-Drapalo A, Flosiak R, Prokopowicz D. Effects of ulcerative colitisactivity on plasma and mucosal prostaglandin E2 concentration. ProstaglandinsOther Lipid Mediat 1999;58:159–65.

50] Cosme R, Lublin D, Takafuji V, Lynch K, Roche JK. Prostanoids in human colonicmucosa: effects of inflammation on PGE2 receptor expression. Hum Immunol2000:61684–96.

51] Linden DR, Chen JX, Gershon MD, Sharkey KA, Mawe GM. Serotonin availability

is increased in mucosa of guinea pigs with TNBS-induced colitis. Am J PhysiolGastrointest Liver Physiol 2003;285:G207–16.

52] O’Hara JR, Ho W, Linden DR, Mawe GM, Sharkey KA. Enteroendocrinecells and 5-HT availability are altered in mucosa of guinea pigswith TNBS colitis. Am J Physiol Gastrointest Liver Physiol 2004;287:G998–1007.

logica

[

[

Y. Hamada et al. / Pharmaco

53] Grider JR. Desensitization of the peristaltic reflex induced by mucosal stimula-tion with the selective 5-HT4 agonist tagaserod. Am J Physiol Gastrointest LiverPhysiol 2006;290:G319–27.

54] Heredia DJ, Dickson EJ, Bayguinov PO, Henning GW, Smith TK. Localized releaseof serotonin (5-hydroxytryptamine) by a fecal pellet regulates migrating motorcomplexes in murine colon. Gastroenterology 2009;136:1328–38.

[

[

l Research 61 (2010) 48–57 57

55] Van Meeteren ME, Van Bergeijk JD, Van Dijk AP, Meijssen MA, Zijlstra FJ.Intestinal permeability and contractility in murine colitis. Mediators Inflamm1998;7:163–8.

56] Mizuta Y, Isomoto H, Takahashi T. Impaired nitrergic innervation in ratcolitis induced by dextran sulfate sodium. Gastroenterology 2000;118:714–23.