Digestive Diseases and Sciences, Vol. 50, No. 10 (October 2005), pp. 1898–1903 ( C© 2005)DOI: 10.1007/s10620-005-2958-1

Effects and Mechanism of Changes of LocalNeurotransmitters in Rats’ Pylorus and Bile

Reflux to the Stomach with Stress Ulcer

CHEN MIN, MD, LUO HESHENG, MD, CHEN JIHONG, MD, TONG QIAOYUN, MD, LI XIANZHEN, MD,and SHELLEY CHIREYETH, MD

Stress ulcer occurs primarily in severe conditions, with a high incidence and mortality in intensivecare units. However, studies on the association between stress ulcer and bile reflux to the stomachwith stress ulcer are still inconclusive. Therefore, our research aimed to determine whether or notbile reflux exists during stress ulcer and then to investigate the effects and mechanism of changes ofpyloric local neurotransmitters on bile reflux in such circumstances so as to provide a new pathwayfor clinical intervention. Cold water immersion was used to copy the stress ulcer model of rats.Sixty-five adult Sprague–Dawley rats of either sex were randomly divided into three groups: thenormal control group (n = 10), the stress group (n = 30), and the antagonist group (n = 25). Thegastric ulcer index, pH, and bile acid of gastric juice were measured before and after stress. RadioImmunoassay Detection Kit and Biochemic Detection Kit were used to measure local contents ofCGRP (calcitonin gene-related peptide) and nitric oxide, respectively, in rats’ pylorus. The localcontents of nitric oxide in rats’ pylorus reached a maximum at 1 hr after stress. The bile acid andpH of gastric juice peaked at 2 hr after stress and the ulcer index peaked at 4 hr after stress. Butthe local contents of CGRP in rats’ pylorus decreased to the minimum at 4 hr after stress. The bileacid and ulcer index in the L-NAME group were significantly lower than in the antagonist controlgroup. However, the bile acid in the hCGRP8-37 group was less than in the antagonist control group.Compared with hCGRP8-37 group, there was a significant reduction in bile acid in the L-NAMEgroup. There was a significant reduction in the ulcer index of the hCGRP8-37 group compared withthe L-NAME group and the antagonist control group. There was a certain kind of positive correlationbetween nitric oxide in rats’ pylorus and bile acid to the stomach, for nitric oxide could loosen thepyloric sphincter and increase the bile acid to the stomach. L-NAME might reduce the local nitricoxide contents in rats’ pylorus so that bile acid to the stomach might be decreased, obviously with alooser tight pyloric sphincter. Meanwhile, the CGRP in rats’ pylorus was negatively associated withthe ulcer index, hence CGRP might protect gastric mucosa under stress conditions.

KEY WORDS: stress ulcer; nitric oxide; calcitonin gene-related peptide; bile reflux.

Stress ulcer occurs primarily in serious circumstancessuch as severe burn, trauma, and hemorrhagic shock. Most

Manuscript received December 9, 2004; accepted February 22, 2005.From the Department of Gastroenterology, Renmin Hospital of Wuhan

University, Wuhan 430060, P.R.China.Address for reprint requests: Professor Luo HeSheng, Department

of Gastroenterology, Renmin Hospital of Wuhan University, Wuhan430060, P.R.China; luotang@public.hb.wh.cn.

patients develop stress ulcer within 72 hr after admittanceto the intensive care unit. The incidence rate of stress ul-cer in intensive care units was as high as 60% (1). Moreand more researchers have focused on its pathogenesis,treatment, and prevention. Although the concrete mech-anism underlying the pathophysiology of stress ulcer isstill unknown, recent studies suggest that the pathogene-sis of stress ulcer appears to be multifactorial. Most factors

1898 Digestive Diseases and Sciences, Vol. 50, No. 10 (October 2005)0163-2116/05/1000-1898/0 C© 2005 Springer Science+Business Media, Inc.

RAT BILE REFLUX WITH STRESS ULCER

contribute to the pathogenesis of stress ulcer by impairingthe stomach’s ability to protect itself against acid injuryrather than by increasing the amount of acid secretion.Besides acid injury, bile acid also plays an erosive roleduring stress ulcer developing. However, it is conceivablethat reflux of duodenal contents and bile salts occurs in se-riously ill patients, and bile salts are known to destroy thepermeability barrier of gastric mucosa. Bile salts not onlyincrease mucosal permeability to acid but also producedirect injury to the surface cells of the stomach and ren-der the gastric mucosa more susceptible to acid injury (2).But research on the relationship between stress ulcer andbile reflux to the stomach is still inconclusive. Therefore,this research aimed to detect the existence of bile refluxduring stress ulcer and then to investigate the effects andmechanism of changes of local neurotransmitters in rats’pylorus and bile reflux to the stomach with stress ulcer soas to provide a new pathway for clinical intervention.

MATERIALS AND METHODS

Materials. Sixty-five adult Sprague–Dawley rats of eithersex weighing between 200 and 250 g were purchased fromHongshan Experimental Animal Center. The Nitric Oxide Bio-chemical Detection Kit was from Nanjing Jiancheng Bioengi-neering Institute; the CGRP Radioimmunoassay Detection Kit,from Beijing Beimian Dongya Biotech Insititute. L-NAME andhCGRP8-37, from Shenzhen Jingmei Biotec Co. Ltd., were dis-solved in 18.5 and 0.016 mM solution, respectively, with 0.9%normal saline.

Grouping. Sixty-five SD rats were randomly divided intothree groups: the control group (n = 10), the stress group (n =30), and the antagonist group (n = 25). The stress group wasfurther divided into six subgroups at 2, 4, 5, 6, 8, and 10 hr afterstress, respectively (n = 5). The antagonist group then consistedof the antagonist control group (n = 5), the L-NAME group(n = 10), and the hCGRP8-37 group (n = 10).

Model Copying. SD rats were fasted for 24 hr and allowedfree access to water until 1 hr before stress. Then the rats werefixed on a board and immersed in water at the level of the ster-nal angle. The water temperature was maintained at 20 ± 2◦Cand the room temperature in the laboratory varied from 5 to10◦C (3).

Intervention Method for the Antagonist Group. First, ratswere anesthetized with 20% urethane (4 ml/kg intraperitoneally).The stomach with the duodenum attached could be observedthrough a midline abdominal incision. Then the antrum andduodenum were identified anatomically. The pyloric region wasidentified as the region proximal to the darker-colored mucosaof the duodenum. The stenotic pyloric sphincter ring couldalso be observed between the stomach and the duodenum.Then 18.5 mM L-NAME (1 ml) and 0.016 mM hCGRP8-37(1 ml) solutions were injected, respectively, at five points aroundthe pyloric ring, each point with a 0.2-ml dilution (1-ml total).The same volume of 0.9% normal saline was injected at the sameplaces in the antagonist control group. Then the incisions weresutured and pasted on rainproof rubberized fabric. These ratswere placed in water when they recovered from the anesthesia.

Collection of Gastric Fluid and Measurement of Bile Acid.An incision was made in the abdomen to expose the stomach.Then the cardia and pylori, respectively, were ligated. A smallhole was made along the greater curvature with scissors. Thenthe gastric fluid in the stomach was drawn out through a 1-mlsyringe and the bile acid was centrifuged at 4000 rpm for 15 minat 4◦C: the upper layer was stored for later measurement (4, 5).The principle for measuring bile acid is based on the method ofenzyme-marked substrate.

Measurement of Ulcer Index and pH of Gastric Juice. Thestomach was separated from the body after the gastric fluid wascollected. Then the stomach was dissected along the greater cur-vature and immersed in 10% formaldehyde. The ulcer index wasrecorded based on Guth criteria. Guth scores are as follows: spoterosion, 1; erosion length <1 mm, 2; ≥1 mm, 3; ≥2 mm, 4; and≥4 mm, 5. The pH of gastric fluid was recorded using the exactpH test paper.

Specimen for Nitric Oxide Biochemical Detection Kit. A100-mg sample of stomach tissues with duodenum was taken,placed in 0.9% cold normal saline for eliminating blood stains,and then dried with filter paper. It was then weighed accuratelyand placed with 0.86% cold normal saline in a 5-ml beaker.(The ratio of tissue weight to physiological salt solution volumewas kept at 1:9.) The tissues were pounded into small piecesand then a tissue homogenate was made. Next the tissues werecentrifuged in a refrigerant centrifuge for 15 min at 3000 rpmand 4◦C, leaving the upper layer for later measurement. The nextsteps were carried out according to the Nitric Oxide BiochemicalDetection Kit operation manual. The principle was based on themethod for nitrate reductase.

Specimen for CGRP Radioimmunoassay Detection Kit.A 100-mg sample of stomach tissues with the duodenum at-tached was taken and blood stains were eliminated by immer-sion in 0.9% cold normal saline. Then they were dried with filterpaper, weighed, and crushed in 1 N HAC (1 ml). Tissues wereboiled for 10 min, and a tissue homogenate was made and cen-trifuged for 15 min at 3000 rpm and 4◦C, storing the upper layerat −20◦C for later measurement. Calcitonin G-related peptide(CGRP) contents in tissues were detected according to the Ra-dioimmunoassay Detection Kit operation manual.

Statistical Analysis. All data are expressed as the mean ±SD. Statistical analysis was carried out with the software packageSPSS 11.5. A P value <0.05 was considered statistically signif-icant. Means of different groups were compared using the SNKand LSD methods. Independent-sample t-test was performed toevaluate significant differences between groups.

RESULTS

Observation of Gross Specimens. There were a fewcoffee-like substances and some residual food in the stom-ach of the stress group. After washing-out, diffuse swellingand hyperemia on most of the gastric mucosa could beobserved. Superficial mucosal erosions were scatteredsporadically. Dark coffee-colored ulcers and hemorrhagespots could also be seen on the gastric mucosa, especiallyin the gastric fundus and body. The results in the antagonistgroup showed that the volume of the stomach with obvioushyperemia in the L-NAME group was two or three times

Digestive Diseases and Sciences, Vol. 50, No. 10 (October 2005) 1899

MIN ET AL.

TABLE 1. CHANGES IN STOMACH PARAMETERS BEFORE AND AFTER STRESS (MEAN ± SE)

t (hr) Ulcer index Bile acid (µmol/L) pH of gastric juice NO (µmol/g) CGRP (ng/L)

0 2.2 ± 1.6 66.2 ± 58.9 3.1 ± 1.2 3.0 ± 1.1 1007.0 ± 395.42 8.4 ± 3.9 216.2 ± 96.7 3.6 ± 0.8 3.6 ± 0.8 853.0 ± 363.74 17.0 ± 2.9† 689.9 ± 347.3* 4.2 ± 1.3† 4.0 ± 1.3 894.8 ± 142.15 28.0 ± 12.6† 718.4 ± 436.3* 4.5 ± 1.2† 6.1 ± 0.8† 771.6 ± 182.46 30.7 ± 10.6† 844.8 ± 280.7† 5.3 ± 1.3† 5.6 ± 1.1† 839.6 ± 102.58 32.0 ± 16.1† 499.9 ± 219.1 3.9 ± 0.7 4.2 ± 1.0* 636.4 ± 186.8*

10 23.8 ± 6.7† 257.2 ± 88.1 3.2 ± 0.6 3.9 ± 0.7 1095.6 ± 253.9

*P < 0.05.†P < 0.01.

greater after stress than before stress. The stomach in thehCGRP8-37 group bulged less than that in the L-NAMEgroup. However, the stomach in the L-NAME group wascharacterized by much more obvious hyperemia, whereasthe volume of the stomach in the antagonist control groupwas significantly less enlarged than that in the former twogroups with less hyperemia.

Changes in Parameters in the Stomach Before andAfter Stress. The data for each time point are displayed inTable 1. Nitric oxide in rats’ pylorus reached a maximumat 1 hr after stress, but CGRP in rats’ pylorus decreased to aminimum at 4 hr after stress and then increased (as shownin Figure 2). Bile acid and pH of gastric juice peaked at2 hr after stress, while the ulcer index followed, at 4 hrafter stress. The differences between the control and thestress group are noted in Table 1.

Regression Analysis. Linear regression analysis be-tween local nitric oxide in rats’ pylorus and bile acid refluxto the stomach under stress ulcer showed statistical sig-nificance: R = 0.71, F = 12.55, and P = 0.017. Figure 1

Fig 1. Regression scatterplot of nitirc oxide in rats’ pyloric versus bileacid in the stomach with stress ulcer.

shows the regression scatterplot between local nitric oxidein rats’ pylorus and bile acid.

Effects of L-NAME and hCGRP8-37. The data forthe antagonist group are shown in Table 2. Comparedwith that in the antagonist control group, the ulcer indexin the L-NAME and hGRP8-37 group was significantlydecreased. There was a significant decrease in bile acidin the L-NAME group compared with the hCGRP8-37group. Pyloric contents of nitric oxide in these two groupsdropped sharply, however, the pyloric CGRP contents inthe hCGRP8-37 group also decreased significantly butthose in the L-NAME group showed no obvious change.

CONCLUSIONS

Stress ulcer is one of multiple symptoms in seriouslyill circumstances as described previously. Current studiessupport a multifactorial etiology in the development ofstress ulcer. Most factors contribute to stress ulcer by mal-adjustment of the neural and endocrine systems, impairingthe gastric mucosal defense function and enhancing injuryfactors on the gastric mucosa.

This study copied models successfully by immersingconstrained rats in water and confirmed the existence ofbile acid reflux to the stomach with stress ulcer (3, 6).The nitric oxide contents in rats’ pylorus reached a max-imum at 1 hr after stress. The bile acid and pH of gastricjuice peaked at 2 hr after stress, while the ulcer indexincreased to a maximum at 4 hr after stress. The CGRPcontents in rats’ pylorus first fluctuated, then decreasedto a minimum at 4 hr after stress, and, finally, increased(as shown in Figure 2). The changes in bile acid and pHrevealed that various injury factors (7) on the gastric mu-cosa (8) still existed and would continued, although thestress had already ended. The 4 hr after stress is the mostessential and significant period, during which all types ofprevention and treatment should be carried out as soonas possible. Otherwise, the prognosis is badly affected.However, bile acid increased to a maximum before theulcer index peaked, demonstrating that bile reflux played

1900 Digestive Diseases and Sciences, Vol. 50, No. 10 (October 2005)

RAT BILE REFLUX WITH STRESS ULCER

TABLE 2. STOMACH PARAMETERS IN THE ANTAGONIST GROUP AT 4 HR AFTER THE

START OF STRESS (MEAN ± SE)

Antagonist control L-NAME hCGRP8-37group group group

Ulcer index 23.8 ± 5.1 15.0 ± 5.7 10.6 ± 2.7†Bile acid (µmol/L) 813.9 ± 377.1 442.8 ± 154.8* 605.8 ± 277.9pH of gastric juice 3.9 ± 1.4 3.6 ± 1.1 4.2 ± 1.3NO (µmol/g) 3.5 ± 0.5 2.7 ± 0.6* 2.9 ± 1.6*CGRP (ng/L) 788.2 ± 256.5 656.4 ± 159.6 495.0 ± 103.0†

*P < 0.05.†P < 0.01.

a very important role among that many injury factors ongastric mucosa. As a result, decreasing bile acid reflux tothe stomach with stress ulcer might be a new and effectivetherapy for stress ulcer.

The nitric oxide contents in rats’ pylorus reached a max-imum at 1 hr after stress and then the bile acid of gastricjuice peaked at 2 hr after stress. Linear regression analy-sis showed that there was a positive association betweenthem. According to Figure 1, the changes in nitric oxidein rats’ pylorus had an obvious effect on bile acid reflux tothe stomach with stress ulcer. Compared with the antago-nist control group, injecting L-NAME in the pyloric ringdecreased the bile acid reflux to the stomach and the ulcerindex, which consequently might play a protective roleon gastric mucosa. Meanwhile, the volume of the stom-ach with obvious hyperemia in the L-NAME group wasenlarged by two or three times after stress compared tobefore stress. The volume of the stomach with less hyper-

Fig 2. The variable curves of CGRP in rats’ pyloric and ulcer index of gastric mucosa.

emia in the antagonist control group was significantly lessenlarged than that in the L-NAME group. Interestingly,these results showed that injecting L-NAME solution inthe pyloric ring could bulge the volume of the stomachwith stress ulcer and cause hyperemia of the gastric body.It also could decrease the nitric oxide contents in rats’ py-lorus and further weaken bile reflux from the duodenumto the stomach as well as decrease the ulcer index (9).Compared with the antagonist control group, it had nosignificant effect on the CGRP contents in rats’ pylorus.

The CGRP contents and ulcer index peaked and de-creased, respectively, at 4 hr after stress. According toFigure 2, the ulcer index increased continuously as CGRPdecreased and then changed conversely. All the above datashow that there is a close association between CGRP andmucosal protection. Injection of hCGRP8-37 decreasedthe ulcer index significantly compared with that in theL-NAME group, but the drop in bile acid reflux to the

Digestive Diseases and Sciences, Vol. 50, No. 10 (October 2005) 1901

MIN ET AL.

stomach was less than in the L-NAME group, which showsthat the drop in the ulcer index in the hCGRP8-37 groupwas caused by a mechanism other than a decrease in bileacid as in the L-NAME group. hCGRP8-37, as an antago-nist of CGRP, might cause the obvious decrease in CGRPand then lead to corresponding negative feedback whichcauses the CGRP contents to increase again. It might playa gastroprotective role on gastric mucosa with stress ul-cer and lead to the obvious decrease in the ulcer index.All the above data demonstrate that CGRP as well as itsantagonist is closely associated with mucosal protection.Furthermore, injecting hCGRP8-37 solution in the pyloricring caused a decrease in the nitric oxide contents in rats’pylorus, which revealed that changes in CGRP might in-fluence the nitric oxide contents. Several researchers haveconfirmed that CGRP in the gastrointestinal system func-tions greatly through nitric oxide and prostaglandin (PG),including by causing mucosal hyperemia and maintain-ing the defense function as well as maintaining mucosalintegrity (10).

The CGRP in plasma of rats that received pretreatmentwith indomethacin and L-NAME increased markedly at2 hr during stress compared with that in the control group.But the effect was more significant in the L-NAME groupthan the indomethacin group (11). This suggested thatCGRP might perform its corresponding function throughnitric oxide. According to research by Lu-ding Zhang et al.(12), CGRP contents in gastric tissues of young rats de-creased markedly during the early period of stress. At thesame time they found that there was an obvious negativecorrelation between the CGRP contents in gastric tissuesat each time point and the ulcer index on gastric mucosa,which suggested that the pathogenesis of stress ulcer inyoung rats was perhaps associated with a reduction inCGRP contents in gastric tissues. All the above studiesdemonstrated again that CGRP as well as its antagonist isclosely related to gastric mucosal protection (13).

After stress the volume of the stomach in the L-NAMEgroup enlarged more obviously, while hyperemia in thegastric body in the hCGRP8-37 group was more obvi-ous. L-NAME is an inhibitor of nitric oxide synthase(NOS), which influences the effects and contents of nitricoxide (14). Nitric oxide, as a nonadrenergic, noncholin-ergic neurotransmitter, performs an inhibitory function.Some researchers (15) have reported that the final trans-mitter involved in receptive relaxation of stomach is justthis transmitter released from nerve endings. It could alsocause hyperpolarization of the cellular membrane of jeju-nal annular smooth muscle and inhibit jejunal function. Itsrelease regulates electrical excitability and contractibilityof the sphincter, pylorus, fundus, and antrum (16). Ac-cording to the report by Willis et al. (15), infusion of

sodium nitroprusside solution through the splenic arteryof rats weakened the contractibility of the pylorus in adose-dependent manner, which suggests that nitric ox-ide might loosen the pyloric sphincter. However, the in-hibitory effect of NOS could strengthen the contractibilityof the fundus, antrum, and pylorus. Inhibitory joint po-tential (17, 18), as a hyperpolarization of smooth muscleresponding to inhibitory neurotransmitter, was recordedwhile nerve–muscle strips were stimulated by an electri-cal field. And this potential was not blocked by atropine, anantagonist of the adrenaline receptor, but was augmentedby L-NAME. This could explain why L-NAME increasesstrain in the pyloric sphincter. Some researchers have re-ported that infusion of L-NAME via vein in rats delaysgastric emptying (17, 19). This research confirmed thatthe stomach bulged after injection of L-NAME solutionin rats’ pyloric ring because decreasing local nitric oxidein rats’ pylorus could impair receptive relaxation of thestomach, promote pyloric strain, and delay gastric empty-ing (17, 19). Consequently, it could cause evident dilationof the stomach volume and, to a lesser extent, hyperemiaof the gastric body compared with the hCGRP8-37 group.CGRP is a 37-amino acid neuropeptide that is encoded byalternative splicing products of the calcitonin gene whichcontributes to dilating blood vessels. In the stomach ofrats, CGRP, which is mainly contained in the peripheralendings of capsaicin-sensitive afferent nerves, is releasednot only from nerve endings but also from paracrine andendocrine cells when acid and other noxious chemicalssuch as capsaicin challenge the gastric mocosa. These en-docrine cells are mainly located in adenoepithelials of thegastric antrum, but its nerve fibers are distributed mainlyalong subepithelial tiny arteries, glands, and subepithelialintermuscular neuroplexus (11). Our experiment showedthat local CGRP contents in rats’ pylorus first droppedand then rebounded markedly. CGRP perhaps promotedhyperemia of the gastric body by dilating blood vessels.

But the exact mechanism for the changes in local neu-rotransmitters in rats’ pylorus has not been clarified yet.And the relationship between these two transmitters andthe degree of hyperemia of the stomach body is still un-certain. Both L-NAME and hCRGP8-37 deserve furtherinvestigation.

REFERENCES

1. Daley RJ, Rebuck JA, Welage LS, Rogers FB: Prevention of stress ul-ceration: current trends in critical care. Crit Care Med 32(10):2008–2013, 2004

2. Fukuhara K, Osugi H, Takada N, Takemura M, Lee S, Taguchi S,Kaneko M, Tanaka Y, Fujiwara Y, Nishizawa S, Kinoshita H: Cor-relation between duodenogastric reflux and remnant gastritis afterdistal gastrectomy. Hepatogastroenterology 51:1241–1244, 2004

1902 Digestive Diseases and Sciences, Vol. 50, No. 10 (October 2005)

RAT BILE REFLUX WITH STRESS ULCER

3. Li X, Yu AG, Deng ZP, Liu YS, Li YJ, Yang CH, Lin JT, Hu YZ:Study on pathogenesis and prevention of stress ulcer of rats. ChinBasic Clin J Gen Surg 7:138–140, 2000

4. Zhu AY, Xu GM, Li ZS, Yin N, Zou DW: Monitoring of bile refluxfor patients with chronic gastritis. Chin J Dig 23:530–531, 2002

5. Tan DY, Yao HC: Monitoring of bile reflux for patients withfunctional dyspepsia and gastric motility. Chin J Dig 32:313–314,2002

6. Qui BS, Mei QB, Liu L, Tchou-Wong KM: Effects of nitric oxideon gastric ulceration induced by nicotine and cold-restraint stress.W J Gastroenterol 10:594–597, 2004

7. Chen SZ, Zhao H, Wu CY, Fu WH, Chen XC: Gallbladder emp-tying function in patients with bile reflux gastritis. World ChinJ Gastroenterol 6:427–429, 1998

8. Zhang J, Li DZ, Hu HB, Wen XD, Wu QP: Experimental and clinicalresearch about the treatment of Talcid on bile reflux gastritis. ChinJ Clin Med 11:75–77, 2002

9. Cui ZM, Li ZS , Xu GM, Zhan XB: Influence of L-NAME and L-argon gastric mucosal tolerant cytoprotection under stress. Chin J Dig11:219–222, 2002

10. Bayguinov O, Sanders KM: Role of nitric oxide as an inhibitoryneurotransmitter in the canine pyloric sphincter. Am J Physiol264(5, Pt 1):G975–G983, 1993

11. Zhang LD, Gao GH, Su X, Jin ZJ, Bao J: Possible relationshipbetween protective effect of calcitonin gene-related peptide on gas-tric stress ulcer and nitric oxide content in stomach. J Clin Paediatr19:308–309, 2001

12. Zhang LD, Su X, Jin ZJ, Bao J, Xu T: The protective effect ofcalcitonin gene-related peptide on gastric stress ulcer of young rats.Chin J Dig 21:187–188, 2001

13. Dong XY, Wang LX, Zhou LY, Lin SR: The effect and their mech-anism of gastric mucosal drugs. Chin J Dig 22:526–529, 2002

14. Wang X, Zhong YX, Zhang ZY, Lu J, Lan M, Miao JY, Guo XG,Shi YQ, Zhao YQ, Ding J, Wu KC, Pan BR, Fan DM: Effect ofL-NAME on nitric oxide and gastrointestinal motility alterations incirrhotic rats. W J Gastroenterol 8:328–332, 2002

15. Qu SY, Li W, Zheng TZ: The role of nitric oxide in gastrointestinalmotility and pathogenesis of digestive diseases. Rec Dev Physiol26:77–79, 1995

16. Fan YP, Chakder S, Gao F, Rattan S: Inducible and neuronal nitricoxide synthase involvement in lipopolysaccharide-induced sphinc-teric dysfunction. Am J Physiol Gastrointest Liver Physiol 280:32–42, 2001

17. Mashimo H, Kjellin A, Goyal RK: Gastric stasis in neuronalnitric oxide synthase-deficient knockout mice. Gastroenterology119:766–773, 2001

18. Hang CH, Shi JX, Li JS, Wu W, Li WQ, Yin HX: Levels of vasoac-tive intestinal peptide, cholecycstokinin and calcitonin gene-relatedpeptide in plasma and jejunum of rats following traumatic braininjury and underlying significance in gastrointestinal dysfunction.W J Gastroenterol 10:875–880, 2004

19. Grisoni E, Dusleag D, Super D: Nitric oxide synthesis inhibition:the effect on rabbit pyloric muscle. J Pediatr Surg 31(6):800–804,1996

Digestive Diseases and Sciences, Vol. 50, No. 10 (October 2005) 1903