Sl137

Not Just a Failure to Relax: Timing and Magnitude of Lower Esophageal Sphincter After-Contractions Also Contribute to Esophageal Outflow Obstruction in Achalasia Samer Gawrieh, Michael C. Jean, Benson T. Massey

BACKGROUND: Impaired esophageal emptying is a cardinal finding in achalasia and is traditionally felt to result from failed lower esophageal sphincter (LES) relaxation and absent esophageal body peristalsis, little attention has been given to the presence of deglutitive after-conrtactions (ACs) in achalasia and their possible effect on esophageal emptying. METHODS: Manomet~ac recordings from 68 achalasia patients (36 F, mean 47 115-85] yr) referred to our clinical manometry laboratory were analyzed. Studies were performed with the subjects lasting and supine, using a sleeve device to record from the LES. The following parameters were scored for three 5ml water swallows: Initial LES pressure (LESP); time (from swallow onset) and magnitude of peak esophageal body pressure (EBP) wave (located 2 cm above the top of the LES); incidence, time, and magnitude of LES deglutitive AC; maximum value by which EBP exceeded LESP during any time after the swallow; basal EflP between swallows. In addition, whether EBP waveforms were isobaric or non-isobaric was recorded. Values were compared to those in 16 subjects (8F, mean 47 [14-881 yr) with normal peristalsis and LES relaxation. RESULTS: The incidence of LES AC was similar in achalasia and controls (84% vs. 92%). As expected, basal LESP was higher (31 vs. 22 mmHg, p<.001) and peak EBP was lower (34 vs. 93 mmHg, p<.001) in achalasia. In achalasia, time of peak EPB was earlier (4.5 vs. 8.2 s, p<.O01) and peak EBP generally exceeded basal LESP (p<.001). However, LES ACs also occurred earlier in achalasia (5.8 vs. 9.8 s, p<.001) and typically exceeded peak EBP (55 mmHg vs. 34 mmHg, p<.O01). As a result, the EBP- LESP differential did not favor emptying in achalasia, whereas it did in controls (-4 vs. 13 mmHg, p<.001). The worse this differential, the higher the basal EBP was in achalasia patients (p= .008). Swallows showing non-isobaric waveforms tended to have a favorable pressure differential compared to those that were isobaric (3 vs. -4 mmHg, p = .026). In a multiple regression analysis, factors sigraficantly determining the pressure differential were: peak EBP (positively); basal LESP, time of LES AC and magnitude of LES AC (negatively). Higher peak EBP were offset by their association with higher LES AC (p<.001). CONCLU- SIONS: In achalasia, esophageal outflow obstruction is not simply a matter of the LES not relaxing. Timing and magnitude of LES ACs also contribute to this obstruction. Supported by NIH PO1 03191-03.

Sl138

Is Ineffective Peristalsis Really Ineffective? Nam Q. Nguyen, Rachael Rigda, Marcus Tippett, Adreas J. Smout, Richard H Holloway

Ineffective esophageal motility, defined as >30% pressure waves in the distal oesophagus <30 mmHg, has been suggested to be associated with impaired oesophageal clearance. However, this association has been based on either radiological or scintigraphic assessment of bolus clearance. Multiple intraluminal impedance (Mll) has been proposed as a non- radiological tool for studying esophageal volume clearance. Aims. The aim of this study was to assess the impact of peristaltic wave amplitude on oesophageal volume clearance using MI1. Methods. Concurrent perfusion manometry and MII was performed in 25 healthy asymptomatic volunteers (16M, 9F, 20-77 yr). Esophageal motility was measured at 4 sites 5-cm apart, starting 2cm above the lower esophageal sphincter (LES). MI1 was recorded at corresponding sites with electrodes incorporated into the manometric assembly. Ten, 5-ml liquid (saline) boluses were tested in each subject in the right-lateral position. Normal values for bolus presence time (BPT) at each site and total bolus transit time (TSTT) were derived from peristaltic sequences deemed to be manometrically normal (peristaltic propagation, wave amplitudes >35 mmHg). BPT and TBTT were then measured for all responses and the proportion of boluses cleared at each site as well as TBTT were correlated with wave amplitude. Results. The proportion of boluses cleared at each site was high and did not increase significantly above a threshold of 20 mmHg (Table). Peristaltic responses with at least all pressure waves > 30 mmHg were associated with normal BPT in 88% and normal TBTT in 98% of responses. When at least one wave was <30 mmHg normal BPT and TBTT were seen in 77% and 89% respectively. When the minimum amplitude was <15 mmHg, BPT and TBTT were normal in 67% and 80%. Conclusions. Esophageal clearance of a liquid bolus from the distal esophagus is effective at pressure wave amplitudes below 30 mmHg and in most regions as low as 11 mmHg. These data suggest that the current threshold for the definition of ineffective peristalsis is too high.

% Normal BPT and TB'I'T related to wave amplitude at Impedance segment

Impedance sag- 0-10 1t.20 21.30 31-40 4t.50 > 50 meat (cm above

LES) mmH o m m H g mmHg rnmH 9 m m H g mmHg

'12 cm BPT 93% 93% 89% 96% 97% 97% 12 cm TBTT 77% * 80% 100% 87% 100% 100% 1 cm BPT 13% * 88% 89% 87% 81% 95% Tom T B'l-r 25% * 100% 94% 93% 90% 98% 2 cm BPT 50%" 78% ** 91% 83% 100% 97% 2cm TBTT 70% 77% *~ 91% 83% 100% 98% ' P<O,O001 vs 11-20 mrnHg, '* p<O.01 vs 21-30 mmHg

51139

Rho Kinase and Cat Lower Esophageal Sphincter (LES) Tone Weibiao Cao, Karen M. Harnett, Ling Cheng, Jose Behar, Piero Bianeani

We have previously shown that PGF2a participates in maintenance of LES tone and that the initial PGF2a-induced contraction depends on activation of Gq and G,3. It has recently been shown (Hersh et al . Gastroenterology 2002, 122 Suppl: A-256) that the pathways mediating the initial phase of contraction in response to exogenously administered agonists

may be different from those mediating sustained contraction, which may depend on a Rho- kinase pathway. We therefore tested the proposition that sustained LES tone, which depends on elevated levels of endogenous PGF2a may be maintained through a Rho-kinase depen- dent pathway. In LES circular muscle strips the Rho-kinase inhibitor Y27632 dose-dependently decreased tone and abolished it at 10 ~ M, suggesting that LES tone is Rho-kinase- dependent. The PGF2a antagonist AL-8810 dose-dependently reduced tone. In strips treated with indomethacin to abolish production of endogenous PGF2a, exogenously administered PGF2a maintained a sustained contraction lasting more than 30 rain. The sustained, but not the initial contraction in response to PGF2a was abolished by Y27632. LES circular smooth muscle cells, isolated by enzymatic digestion were used to exanaine the intracelhilar contractile pathways mediating Rho-kinase- PGF2a-dependent sustained contraction. In cells PGF2a caused a relatively rapid contraction that achieved maximum shortening in approximately 30 seconds and persisted in excess of 20 minutes. After treatment with the Rho-kinase inhibitor Y27632 cells achieved the same maximum initial shortening as untreated ceils, but did not remain contracted, and returned to their unstimulated length within approximately 10 minutes. The inhibition of sustained contraction by Y27632 in cells mimics the inhibition of tone and of sustained PGF2a-induced contraction in strips In saponin-permeabilized cells, PGF2a-induced shortening was the same as in intact ceils. In these permeable cells G~3 and RhoA antibodies reduced and the selective RhoA inhibitor C3 (exoenzyme of Clostridium Botulinum) abolished sustained contraction induced by PGF2a, confirming that the contraction is mediated by RhoA-induced activation of Rho- kinase. The data suggest a role of RhoA- Rho-Kinase in maintenance of PGF2a induced sustained contraction and LES tone. Supported by NIH ROI-DK-28614.

S l140

Integrin Linked Kinase (ILK) in ACh-lnduced Contract ion of Cat Esophageal Circular Smooth Muscle (ESO) Weibiao Cao, Ling Chang, Karen M. Hamett, .ling T. Dang, Jose Behar, Michael P. Walsh, Piero Biancani

We have previously shown that ACh-induced contraction of cat ESO is linked to phosphati- dylcholine metabolism, production of diacylglycerol and arachidonic acid, activation of the Ca t§ -insensitive protein kinase C-~ (PKCE) and activation of two distinct MAP kinase pathways: an ERK1/ERK2 MAP kinase-dependent pathway and an HSP27-1inked p38 kinase- dependent pathway. The sequence of events occurring between activation of MAP kinases and muscle contraction, however, has not been elucidated. A possible candidate intermediate protein in Ca2+-independent PKCt-mediated contraction may be ILK, a kinase associated with myofilaments and responsible for Ca 2 § -independent phnsphorylation of smooth muscle myosin. We therefore examined the role of ILK in ACh-induced contraction of cat ESO. ILK was identifiable by Western blot in ESO. lLK activity, measured by a filter assay in esophageal smooth muscle, increased significantly 30 and 60 seconds after ACh stimulation and was maximal at 60 seconds. The increased ILK activity at 60 seconds was inhibited by the ERK1/ERK2 MAP kinase inhibitor PD98059, but not by the p38 MAP kinase inhibitor SB203580, suggesting that ILK is activated by ERK1/ERK2, and not by p38 MAP kinase. In sapoinn-permeabilized ESO cells ILK antibody reduced (approximately 50%) ACh-induced contraction. The ILK-antibody in combination with the ERK1/ERK2 kinase inhibitor PD98059 caused the same reduction in contraction as PD98059 or the antibody alone, confimung that both ILK and ERK1/ERK2 may be in the same contractile pathway. ILK- antibody inhibition was increased by the p38 MAP kinase inhibitor SB203580, indicating that ILK and the p38 MAP kinase may be in different contractile pathways. Conversely, an HSP27 antibody partially reduced (approximately 40%) ACh-induced contraction. HSP27 inhibition was not increased by the p38 MAP kinase inhibitor SB203580, but was increased by the ERK1/ERK2 MAP kinase inhibitor PD98059, suggesting that p38 MAP kinase and HSP27 are in the same pathway. Finally, antibodies against HSP27 and ILK, when used in combination, almost abolished ACh-induced contraction. We conclude that ACh-induced contraction of cat ESO is mediated through a dual pathway: one involving ERK1/ERK2 and ILK and the other involving p38 MAP kinase and HSP27. Supported by NIH RO1-DK- 28614 and C1HR MOP-13101.

S1141

Spontaneous Tone and Cholinergic Contractions: Differences in Calcium Handling in Cat Lower Esophageal Sphincter Sling and Clasp Muscles Ahmad Mninuddin, Leifa Neshatian, Nicholas E. Diamant

Background: There are differences in length tension relationships and cholinerglc sensitivity of sling and clasp muscles in the cat lower esophageal sphincter (LES). Hypothesis: There are differences in the sources of Ca 2* utilized for contraction in sling and clasp muscles. Methods: Studies were performed on muscle strips from sling and clasp regions of the LES in the presence of tetrodotoxin. Results: 1)Tone: role of extracellular (EC) Ca 2+ . Clasp muscle developed more spontaneous tone than sling muscle (p<.05). In Ca 2§ free Krebs or in the presence of the L-type Ca 2* channel blocker infedipine, clasp exhibited a greater decrease in tone (p<.05). 2)Tone: role of Ca 2§ from internal stores. The sarcoplasmic reticuhim (SR) Ca2+-ATPase inhibitor, cyclopiazonic acid (CPA), caused greater increase in tone in sling than in clasp muscle(p<.02). Stimulation of ryanodine sensitive stores with caffeine inhibited tone in both muscles. Inhibition of ryanodine receptors with ryanodine caused a similar increase in tone in sling and clasp. The SR IP3 receptor blocker 2-APB abolished tone in clasp with no inhibition in sling. 3)Cbolinergic contractions: role of EC Ca t+. Acetylcholine- induced contractions (CC) were of greater amplitude in sling than in clasp muscle(p<.05). In Ca 2+ free Krebs, CC were similarly inhibited in a time and challenge dependent manner in both muscles. Nifedipine abolished CC in the clasp but only slightly inhibited the CC in the sling. 4)Cholinerglc contractions: role of Ca 2+ from internal stores. CPA, caffeine and ryanodine caused a similar degree of inhibition of CC in both sling and clasp muscle. 2- APB completely abolished CC in both muscles. Conclusions: In sling and clasp muscle, intracelhilar (IC) and EC Ca 2+ sources are utilized to different degrees in the generation of

A - 1 6 1 A G A A b s t r a c t s

spontaneous tone and CC. Spontaneous tone in LES clasp is likely due to continuous release of Ca 2 ~ frorq the IP3 mediated stores of the SR, which in turn are replenished by continuous innux of EC Ca :+ via L-type Ca 2+ channels. Clasp muscle depends on both release of Ca 2+ from internal stores and influx of EC Ca 2+ for CC. Low intrinsic tone in the sling is likely a property of the contractile elements and their response to normal IC Ca 2+ levels. Sling muscles rely primarily on internal Ca 2+ stores for CC, and refilling of these internal stores by EC CaZ+ahrough a nifedipine-insensitive pathway.The role of these differences in the pathogenesis and treatment of disease states requires further study.

$1142

Motor Activity in The Isolated Jejunum of VR1 Knockout Mice Reza Rahmati, Michelle Cockerham, Alan Brunsden, Kirk Hillsley, Wendy J. Winchester, John B. Davis, Gareth A. Hicks, David Grundy

introduction: Isolated segments of mouse jejunum develop a complex pattern of contractile activity consisting of periods of phasic contractions that migrate in an aboral direction, interspersed by periods of quiescence (Abduet al., Am. J. Physiol. 2002; 282:G624-33.). The aim of this study was to examine the role of vanilloid receptor-1 (VR1) in the generation of intestinal motor activity and its reflex modulation by intrahiminal distension and acid. Methods: Experiments were perfnnned on mice in which the VR1 gene had been disrupted using standard gene targeting techniques (Davis et al. Nature. 2000; 405: 183-7) and wild- type littermates. Jejunal contractile activity was recorded from in vitro segments of jejunum 4-5cm in length. When distended to 2-3 cm H20 the segments generated regular motor complexes (MCs) recorded as changes in intraluminal pressure. Data are mean _+ SEM and analysed using one-way ANOVA and pair-wise comparisons (Dunnett's method). Results: The periodicity and amplitude of MCs was similar in the knockout and wild-type jejunum (208.5 • 18.6s and 7.5 • 0.8cmH20 vs 201.6 • 10.9s and 7.6 • 0.6cmH20 respectively, n = 10, NS). Capsaicin (1-100nM) caused a dose dependent inhibition of motility manifested as an increase in the interval between MCs in the wild-type animal only (e.g. 222 • 39s to 393.8• at 10OnM, N~5 , P~0.05), a response abolished by pre-treatment with the VR-1 antagonist capsazepine (3mM). At higher doses of capsaicin (1-100~M), the periodic MCs were replaced by tonic increases in pressure upon which were superimposed phasic contractions at the slow wave frequency. This stimulation occurred in both knockout and wild-type mice and was unaffected by pre-treatment with capsazepine. Luminal acidification (10mM HCI) disrupted the generation of MCs which were replaced by continuous phasic activity superimposed upon small changes in tone that were not different between knockout and wild-type animals (L5• and 0.5• but significantly reduced from their corresponding control values (5.8 • 0.5cmH20 and 6.0 • 0.6cmH20 respectively, P~0.05, n = 4). Conclusions: These data demonstrate that capsaicin acting on VR1 receptors, may exert an inhibitory influence on the enteric reflex circuits that control motor activity. Capsaicin also stimulates motility at higher doses via a non VR1 mediated mechanism, as seen in both knockout animal and after treatment with capsazepine. However, VRI does not contribute to ongoing motor activity triggered by distension or the inhibitory effect of luminal acid

$1143

Novel Human and Mouse Genes Encoding a Forkhead/Winged-Helix Family of Transcriptional Regulators and Its Down-Regulation in Gastric Fnndus of W/Wv Mouse Ichiro Takayama

A division of labor exists between different classes of interstitial cells of Cajal (ICC) in the gastrointestinal tract. Among them, intramuscular 1CC (1C-IM) in the stomach act as intermediaries in enteric motor neurotransmlssion. The muscle layers of the gastric fundns only possess IC-IM. IC-1M are absent in the gastric fundus of W/WV mutant mice and there are reduced excitatory and inhibitory motor nerve responses in this tissue. The absence of 1C-IM in W/WV mutants provides a unique opportunity to study the molecular changes that are associated with the loss of these intercalating cells. The gene expression profile of the gastric fundus of wild type and W/WV mice was assayed by using a murine microarray chip analysis displaying a total of 8734 elements. Twenty-cue queries were differentially expressed in wild type and W/WV mice. One candidate gene, encoding a novel protein homologous to members of forkhead/winged-helix family, was sigmficantly down-regulated in fed and starved W/WV mice. The full-length clone of the murine gene and its human counterpart were isolated and designated fork head-related protein like A (FKHLA). Human FKHLA cDNA encodes a protein of 680 amino acids. This gane is located at chromosome 6. FKHLA was abundantly expressed in human gastrointestinal tract, especially stomach and duodenum, whereas it was scarcely expressed in cecum and rectum. Gene analysis showed that FKHLA was differentially expressed in the gastric fundns of normal and W/ WV mice. The specific down-regulation of FKHLA in the fundus of W/WV mice, and high but relatively specific expression in human intestinal tissue suggest that the FKHLA gene could be associated with 1CC function in mice and humans.

S l144

Inhibitory Effect of Intestinal Electrical Pacing on Intestinal Motility Is Mediated by Nitrinergic But Not Cholinergic Pathway Jinsong Liu, Xiaotuan Zhao, Lijie Wang, Jiande Chen

Our previous study (Zhao et al, Neurogastroenterology and Motility, 2002;14:457) has shown an inhibitory effect of intestinal electrical pacing (IEP) on small bowel motility. The aim of this study was to investigate the neural pathway mediating such an inhibitory effect. Methods: Six female dogs (weight 20-30kg) were chronically implanted with one pair of electrodes on duodenal serosa 10cm beyond the pylorus and an intestinal cannula 35 cm beyond the pylorus. The study was performed in 3 sessions on different days. In session 1 (control session), IV saline was infused at a speed of lml/h for 35 minutes after a 20-min baseline recording. IEP with long pulses was performed via the electrodes 15-min after the

infusion for 20 minutes (pulse width: 200ms, amplitude: 4mA, frequency: 20cycle/min, cpm); small bowel motility in the area of the electrodes was measured by inserting a Manometric catheter via the cannnla for 20-min during and 20-min after IEP. The protocol of sessions 2 and 3 was the same as session 1 except the saline was replaced by L-NNA (2.5mg/kg/h) or bethanechol (0.2mg/kg/h), respectively. A motility index was defined as the area under the contractile curve and assessed by a computer program developed in our lab. The motility index in each baseline recording was defined as 100%. Results: 1) Consistent with our previous findings, 1EP significantly inhibited duodenal motility with a motility index of 31.59• (p~0.01 vs. baseline). 2) L-NNA partly abolished the inhibitory effect of IEP. With the treatment of L-NNA, the motility index during pacing increased to 72.90--.25.46%, (p~0.01 vs. IEP without L-NNA, p~0.04 vs. baseline). 3) Infusion of bethanechol during 1EP increased intestinal motility. The motility index was 75.82 • 31.02% (p~0.03 vs. IEP alone). Conclusions: 1) Intestinal electrical pacing has an inhibitory effect on small bowel motility. 2) This inhibitory effect is at least partially mediated via nitrinergic pathway hut not cholinergic pathway.

S l145

Sustained Inhibitory Effects of Intestinal Electric Stimulation on Intestinal Motility in Conscious Dogs Xiaotuan Zhao, Lijie Wang, Jiande Cben

Potential applications of gastrointestinal electrical stimulation for gastroparesis and obesity have recently been explored. While there are variations in the methodology, all of them use a method of continuous stimulation which is known to possibly induce muscle fatigue and/ or adaptation. The aim of this study was to study whether the effect of intestinal electrical stimulation (IES) is sustained and therefore whether stimulation could be performed intermit- tently. Methods: Five hound dogs were equipped with a duodenal cannula (20 cm from pylorus) and 1 pair of serosal electrodes 15 cm from the cannula. All experiments were performed immediately after an ingestion of a standard solid meal. A manometric catheter was inserted into the jejunum 5 cm distal to the stimulation electrodes to measure intestinal tone and contractile activity. After a lO-min baseline recording, 1ES was performed with different stimulation parameters (frequency: 20 cpm, pulse width: 50 ms, 100 ms, 200 ms, and 300 ms, at each pulse width, the pulse amplitude increased from 2.5 mA to 5 mA, 7.5 mA, and 10 mA). Between consecutive IES, there was a period without IES for a sufficient recovery of intestinal motility. An index for tone or phasic contractile activity was defined and assessed as the area under the tone or contractions. The tone at the baseline was defined as 0. The contractile index at each IES period was computed as the reduction in percentage against the baseline value. Results: 1) IES energy-dependently inhibited intestinal contractile activity (r-0.77, p~0.O01). The contractile index was 9.1 _-4-2.0 during IES of the lowest energy and 1.7 • 0.2 (p~O.001, one-way ANOVA) during IES of the highest energy. 2) IES energy-dependently reduced intestinal tone (r-0.74, p~O.001). The tonic index was 0.0 • during IES of the lowest energy and -19.2 • 11.4 (p~0.001, One-way ANOVA) during IES of the highest energy. 3) IES energy-dependently increased the duration of the sustained inhibitory effect (r=0.78, p~0.O01). The sustained duration was 0 sec with 1ES of the lowest energy and 436sec with IES of the highest energy. Conclusions: IES energy-depen- dently inhibits intestinal tone and contractions. The duration of the sustained effect is dependent on stimulation energy, suggesting a feasibility of intermittent stimulation.

$1146

Contribution of Sodium Ions to the Human Jejunum Circular Smooth Muscle Slow Wave Lei Sha, Joseph H. Szurszewski, Gianrico Farrugia

Background: Slow waves in gastrointestinal smooth muscle originate from interstitial cells of Cajal (ICC) and they control gastrointestinal motility. A tetrodotoxin (TTX)-resistant mechanosensitive Na + channel is expressed in human small intestinal smooth muscle and ICC. Aim~ The aim of this study was to determine the contribution of sodium ions to the slow wave recorded from human jejunum circular smooth muscle cells. Methods: Human jejunal tissue (12 mm x 2 ram) was placed in a recording chamber superfused with oxygenized normal Krebs solution containing atropine, propranolol and phentolamine (1 ~M each) at 35 ~ Intracelhilar recordings of circular smooth muscle cells were made via glass microelec- trodes filled with 3M KCI. Results: The resting membrane potential of circular jejunal circular smooth muscle cells was - 58.9 • 1.6 mV (n=30). The slow wave frequency was 7.9 • 0.12/rain (n = 30). Superfnsion of low Na + Krebs solution (5 mM Na +, Na + replaced with Tris) resulted in an immediate hyperpolarization from -60 _+ 2 mV to -66 -+ 3 mV (n = 6, P ~ 0.01). Slow waves were abolished about 240 sec after superfusion of the low Na + solution. Lidocaine (200 ~M, 10 min), a Na + channel blocker, applied in the presence of TTX (1 ~M) to block neuronal TTX-sensitive Na + channels hyperpolarized smooth muscle membrane potential from -58 -+ 4 mV to -68 -+ 7 mV and slowed the slow wave frequency from 8.1 -+ 0.2 per minute to 7.6 4"- 0.4 per minute (p ~ 0.05, n=4) . In the presence of lidocaine and TTX the slow wave duration increased from 6.5 --4" 0.2 sec to 7.1 • 0.3 sec (P< 0.01) and the time constant of the rising phase of the slow wave increased from 1.2 -+ 0.3 sec to 1.5 • 0.2 sec (P < 0.05). No change in slow wave amplitude was noted. In view of the mechanosensitivity of the Na + channel, the effect of tension was also tested. At 0 gram tension, (n = 7) the slow wave frequency was 7.1 • 0A Hz (n=7). At 2 gram tension, the slow wave frequency in the same ceils increased to 7.3 • 0.2 Hz (P < 0.05). Conclusion: The data suggest that mechanosensitive Na+channels play a role in the control of human intestinal circular muscle layer membrane potential and slow wave frequency. Supported by DK 17238, DK 52766 and DK 57061

AGA Abstracts A-162