- Email: [email protected]
Nitric oxide synthase inhibition negates bombesin-induced gastroprotection
Nitric oxide synthase inhibition negates bombesin-induced gastroprotection Antonio A. Castañeda, MD, Yong S. Kim, MD, Lily K. Chang, BS, Yan Cui, and David W. Mercer, MD, Houston, Tex
Background. Bombesin prevents gastric injury primarily by the release of endogenous gastrin. Gastroprotection by exogenous gastrin is negated by nitric oxide synthase inhibition, which implicates a role for nitric oxide as a protective mediator. Because both endothelial and inducible isoforms of this enzyme can play a role in mucosal defense, this study was done to examine the contrasting effects of 2 nitric oxide synthase inhibitors on bombesin-induced gastroprotection. Methods. Rats were given subcutaneous saline or bombesin (10-100 µg/kg) 30 minutes before they received a 1-mL orogastric bolus of acidified ethanol (150 mmol/L of hydrochloric acid/50% ethanol) and rats were killed 5 minutes later for assessment of macroscopic injury (mm2). Gastric mucosal blood flow was measured by laser Doppler. Endothelial, neural, and inducible nitric oxide synthase were assessed by using Western immunoblot. Results. Bombesin decreased gastric mucosal damage, and dose-dependently increased blood flow when compared with saline-treated rats. Endothelial but not neural or inducible nitric oxide synthase immunoreactivity was increased by bombesin. In additional studies, intraperitoneal administration of NG-nitro-L-arginine methyl ester (L-NAME, 5-10 mg/kg), a nonselective nitric oxide synthase inhibitor, negated bombesin-induced gastroprotection and hyperemia, whereas the selective inducible inhibitor aminoguanidine (45 mg/kg) did not. Subcutaneous (SC) L-arginine (300 mg/kg), but not D-arginine, abolished the effects of L-NAME. Conclusions. Taken together, these data suggest that nitric oxide produced by the endothelial isoform of nitric oxide synthase plays an important role in mediating the gastroprotective and hyperemic actions associated with bombesin. (Surgery 2000;128:422-8.) From the Department of Surgery, University of Texas–Houston Medical School, Houston
BOMBESIN IS A NEUROPEPTIDE localized to the gastrin- and acid-secreting portions of the stomach and other areas along the digestive tract.1 In the stomach, bombesin directly stimulates gastrin release from antral glands by binding to its receptors located on the gastrin-containing cells.2 Its effects on acid secretion vary from species to species and depending on whether it is given peripherally or centrally.3 We and others have shown that exogenous bombesin renders the stomach less susceptible to injury from a luminal irritant, an effect blunted by ablation of capsaicin-sensitive neurons.4,5 In addition, we have shown that gastroprotection by bombesin is mediated primarSupported by NIH grant DK 50445. Accepted for publication April 4, 2000. Reprint requests: David W. Mercer, MD, Chief of Surgery, University of Texas–Houston Medical School/LBJ General Hospital, 5656 Kelley, Suite 3-OS 62008, Houston, TX 77026. Copyright © 2000 by Mosby, Inc. 0039-6060/2000/$12.00 + 0 11/56/107982 doi:10.1067/msy.2000.107982
422 SURGERY
ily by the release of endogenous gastrin.6 Gastroprotection by exogenous gastrin on the other hand is negated by nitric oxide synthase (NOS) inhibition implicating a role for nitric oxide as a protective mediator.7 Furthermore, gastroprotection by means of sensory neuron activation is caused by an increase in gastric mucosal blood flow and involves a nitric oxide-mediated mechanism.8 Because some evidence suggests that both the endothelial9 and the inducible10 isoforms of NOS can play a role in gastric mucosal defense, this study was done to examine the contrasting effects of 2 NOS inhibitors on bombesin-induced gastroprotection. In addition, the effects of bombesin on gastric mucosal blood flow and NOS expression were assessed. It was our hypothesis that bombesin-induced gastroprotection would require a functioning endothelial isoform of NOS (eNOS), and consequently its protective effects would be attenuated when this isoform was competitively inhibited with a nonselective NOS inhibitor as opposed to a selective inducible NOS (iNOS) inhibitor. Moreover,
Surgery Volume 128, Number 3
this inhibition should be reversed in the presence of excess substrate with L- but not D-arginine. Thus, the first set of experiments examined the effects of the nonselective NOS inhibitor L-NAME and the selective iNOS inhibitor aminoguanidine on the ability of bombesin to prevent gastric injury caused by the luminal irritant acidified ethanol. In the second set of experiments, the effects of L- and D-arginine on L-NAME-induced reversal of bombesininduced gastroprotection were assessed. In the third set of experiments, the effect of bombesin on gastric mucosal blood flow, in the presence or in the absence of NOS inhibition, was evaluated. In a fourth experiment, the effect of bombesin on endothelial, neural, and iNOS expression was determined. In a fifth experiment, the effect of vagotomy on bombesin-induced gastroprotection was assessed. MATERIALS AND METHODS Animals. Female Sprague-Dawley rats weighing approximately 200 g were used in all studies and were housed at a constant room temperature with a 12-hour light-dark cycle. All experiments were performed in rats deprived of food for 18 to 24 hours but allowed free access to water up to the beginning of the study. On the day of experimentation, all animals were randomly assigned to one of several groups. The University of Texas–Houston Animal Welfare Committee previously approved all experimental protocols before any studies were conducted. Chemicals. Bombesin, aminoguanidine, LNAME, L-arginine, and D-arginine were of molecular biology grade and were purchased from Sigma Chemical Co (St Louis, Mo). Aminoguanidine was dissolved in 0.1 N hydrochloric acid and subsequently neutralized to pH 7.4 with 0.1 N sodium hydroxide. All other solutions were prepared in saline. The polyclonal antibodies against eNOS and neural NOS (nNOS) were purchased from Transduction Laboratory (Lexington, Ky), and the polyclonal antibody against iNOS was prepared and developed by the Trauma Research Center at the University of Texas Medical School, Houston, Tex. The efficacy of this antibody probe has been previously published.11 Effect of bombesin on gastric injury from acidified ethanol. The first set of experiments was designed to confirm exogenous bombesin could prevent acidified ethanol-induced gastric injury. Experiments were performed on conscious animals and bombesin given subcutaneously in a dose of 100 µg/kg, while control animals received saline. We have previously shown that bombesin dose-
Castañeda et al 423
dependently (10–100 µg/kg) prevents gastric injury and that the 100 µg/kg dose provides the highest level of gastroprotection.6 Because we have previously shown that bombesin also prevents acidified ethanol-induced morphologic injury to the gastric mucosa, morphologic analysis was not undertaken in these experiments.6 Thirty minutes after pretreatment with bombesin, all animals were given a 1-mL oral gastric bolus of acidified ethanol (150 mmol/L hydrochloric acid/50% ethanol). This concentration of ethanol in combination with hydrochloric acid has been previously demonstrated to reproducibly induce maximal visible lesion formation to the glandular portion of the stomach within 5 minutes of exposure.12-14 Thus rats were killed 5 minutes after receiving this luminal irritant. In additional rats receiving either a 30-minute pretreatment with bombesin or saline, the acidified ethanol was given, and the rats were killed 1 hour after receiving the damaging agent. These latter rats were included to examine whether bombesin protects the stomach from longer periods of exposure or simply delays the onset of injury. After the animals were killed, all stomachs were removed and the total area of macroscopic injury to the glandular portion of the stomach quantified by computerized planimetry. The surface areas of the individual lesions were measured in a blinded manner. Results were recorded as mean damage in square millimeters (mm2) ± standard error of the mean for each experimental group. Effect of NOS inhibition on bombesin-induced gastroprotection. To examine the role of NOS isoforms in bombesin-induced gastric mucosal protection, experiments were conducted using the NOS inhibitors aminoguanidine and NG-nitro-L-arginine methyl ester (L-NAME). Two NOS inhibitors were assessed because NOS inhibitors can have different effects depending upon which isoform of NOS is inhibited.15,16 Evidence suggests that aminoguanidine is more selective for the inducible isoform of NOS whereas L-NAME is nonselective in its ability to inhibit NOS.11,16 By using a similar protocol as described above and a sample size of 6 rats per group, animals were randomized to receive either aminoguanidine (45 mg/kg intraperitoneal [IP]) or L-NAME (5-10 mg/kg, SC) concurrently with saline or bombesin (100 µg/kg) while controls received saline. Thirty minutes later, gastric injury was induced with acidified ethanol (150 mmol HCl/50% ethanol) and macroscopic injury was determined as previously described. This dose of aminoguanidine was chosen because it almost abolishes the generation of gastric luminal nitrates and nitrites induced by the administration of endotox-
424 Castañeda et al
in.11 The doses of L-NAME have also been previously shown to reverse endotoxin-induced hypotension without much effect on serum nitrates or nitrites.17 In other rats, the reversibility of NOS inhibition was assessed by administration of either the natural substrate for NOS, L-arginine (300 mg/kg, SC), or its enantiomer, D-arginine (300 mg/kg, SC), given concurrently with the other pretreatments. This dose of L-arginine has been shown to reverse the effects of L-NAME.17 In other rats, the effect of NOS inhibition on gastric mucosal injury in the absence of acidified ethanol was assessed by giving rats L-NAME (10 mg/kg, SC) for 30 minutes and then killing them without exposure to the luminal irritant. Effect of bombesin on gastric mucosal blood flow. These experiments were conducted to assess the effect of bombesin on gastric mucosal blood flow. In these studies, rats were anesthetized with an intraperitoneal injection of 6 mg/kg xylazine and 70 mg/kg of ketamine. After induction of anesthesia, the stomach was exposed by a midline incision. A catheter was introduced through the nonglandular fore stomach to provide access for a Teflon-coated laser optic flow probe (Peri Flux PF409, standard probe, 0.25-mm fiber separation; Peri Med, Järfälla, Sweden). The flow probe was positioned to allow contact with the glandular or acid-secreting portion of the stomach. After the appropriate position of the probe was ensured, the stomach was allowed to equilibrate for 30 minutes. Mucosal blood flow to the stomach was then recorded continuously with a laser-Doppler flow monitor (Peri Flux 4001 Master; Peri Med). Blood flow was recorded for a 5-minute period as the measurement of basal gastric mucosal blood flow, followed by administration of subcutaneous saline or bombesin (10–100 ug/kg). Blood flow was monitored for an additional 30 minutes, and the animal was killed. The doses of bombesin used are known to dose-dependently prevent gastric injury.6 In addition, the effect of L-NAME (5 mg/kg, SC) on bombesin (100 µg/kg, SC)-induced changes in blood flow was assessed by giving L-NAME or saline concurrently with bombesin. L-NAME reversal was assessed with L- or D-arginine (300 mg/kg, SC). For all groups, a sample size of 5 or more animals per group was used. Blood flow is reported as percent of change from baseline determinations. Protein extraction and Western immunoblot analysis for eNOS, nNOS, and iNOS. To estimate and compare the content of gastric eNOS, nNOS, and iNOS, rats were anesthetized with an injection of 6 mg/kg IP of xylazine and 70 mg/kg IP of ketamine 30 minutes after receiving saline (n = 5) or
Surgery September 2000 bombesin (100 µg/kg, n = 5). After achieving satisfactory anesthesia, a midline laparotomy was performed, and the stomach was immediately removed. The stomach was opened along the greater curvature, and the mucosa of the glandular portion of the stomach was gently rinsed with saline. The mucosa was blotted dry, scraped off the underlying muscularis externa and serosa, snapfrozen in liquid nitrogen, and stored at –80°C before protein extraction and Western immunoblot analysis was performed for eNOS, nNOS, and iNOS. Protein in each sample was extracted by pulverizing the frozen tissue with the use of a mortar and pestle in a liquid nitrogen slurry. This sample was then added to 1 mL lysis buffer (10 mmol/L Tris, pH 8.6, 100 µmol/L phenylmethylsulfonyl fluoride, 1.5 mmol/L magnesium chloride, 100 µU/mL aprotinin, 10 µg/mL leupeptin, and 0.5% Nonidet [octylphenoxy] polyethoxy ethanol-40) and then subjected to two 30-second bursts of a Polytron (Vertishear). These samples were then transferred to microfuge tubes and centrifuged for 10 minutes at 11,000g. The supernatant was removed and added to a sample buffer (125 mmol/L Tris, pH 6.8, 2% sodium dodecyl sulfate, 5% glycerol, 1% β-mercaptoethanol, and 0.003% bromophenol blue). Protein concentrations within each homogenate were determined by using the bicinchoninic acid protein assay (Pierce, Rockford, Ill) before the sample buffer was added to the supernatant. The remaining proteins in these prepared homogenates were separated by sodium dodecyl sulfate (7.5%)-PAGE, using 40 µg protein per sample for iNOS determinations and 80 µg protein per sample for eNOS and nNOS determinations. Resultant proteins were electroblotted onto nitrocellulose paper and incubated for 1 hour at room temperature in blocking solution (5% nonfat dried milk, 0.1% Tween 20, and phosphate-buffered saline). The resultant blot was then washed twice in 0.1% Tween 20-phosphatebuffered saline followed by a 1-hour incubation with a specific polyclonal anti-iNOS antibody (1:2000 dilution) or a specific polyclonal antieNOS or anti-nNOS antibody (1:700 dilution). Blots were then washed twice and incubated with a horseradish peroxidase-conjugated donkey, antirabbit immunoglobin as a secondary antibody (1:10,000 dilution for iNOS and 1:5000 dilution for eNOS and nNOS) for 1 hour. After 2 final washes, the immune complexes were visualized with the use of enhanced chemiluminescence detection (Amersham; Arlington Heights, Ill). The resultant autoradiographs were then assessed semiquantitatively by using computer-assisted den-
Castañeda et al 425
Surgery Volume 128, Number 3
sitometry and were reported as mean relative densitometric units. Effect of vagotomy on bombesin-induced gastroprotection. To determine whether bombesininduced gastroprotection required an intact vagus nerve to exert its protective actions, additional studies were performed with bombesin in rats undergoing sub-diaphragmatic vagotomy. For these studies, rats were anesthetized with intraperitoneal xylazine (6 mg/kg) and ketamine (70 mg/kg). A midline laparotomy was performed, the vagal trunks identified and transected along the esophagus, and a pyloroplasty performed by making a longitudinal incision along the pylorus and then closing it with interrupted 5-0 silk sutures in a transverse fashion. Two sets of controls were included. In one group, rats underwent sham laparotomy alone. In the other group, rats underwent laparotomy with pyloroplasty without sub-diaphragmatic vagotomy. All rats were allowed to awaken after completion of their procedures and were placed in individual isolated cages with normal access to food and water. Seven days after the procedures, rats were fasted overnight and then randomized to receive either saline or bombesin (100 µg/kg) for 30 minutes. After the 30-minute treatment period, gastric injury was induced with orogastric acidified ethanol, and macroscopic injury was determined. Statistics. All values in the figure and text are expressed as mean plus or minus the standard error of the mean of “n” observations, where “n” is the number of animals examined. In all studies, a sample size of 5 or more animals per group was used. Statistical significance was determined by using analysis of variance followed by Fisher’s post hoc test. P < .05 was considered statistically significant. RESULTS Exogenous bombesin protects against acidified ethanol. Animals in the control group receiving saline followed by orogastric acidified ethanol had extensive macroscopic gastric injury. This injury was characterized by the presence of hemorrhagic lesions along the gastric folds and was confined primarily to the glandular portions of the stomach, as previously reported.6,13,14 Despite the severity of injury caused by acidified ethanol, subcutaneous bombesin prevented the development of these lesions (21 ± 5 vs 118 ± 22 mm2, P < .001) when compared with saline-treated rats. Bombesin was also efficacious at attenuating the extent of macroscopic damage caused by a longer exposure to acidified ethanol (ie, 1 hour) compared with saline–pretreated rats (23 ± 4 vs 110 ± 19 mm2, P < .001).
Table I. Effects of NOS inhibition on bombesin (100 µg/kg)-induced gastroprotection from acidified ethanol in the conscious rat Treatment
Injury (mm2)
Saline/saline Bombesin/saline Aminoguanidine/saline Aminoguanidine/bombesin L-NAME 5 mg/kg saline L-NAME 5 mg/kg bombesin L-NAME 10 mg/kg saline L-NAME 10 mg/kg bombesin
118 ± 22 21 ± 5* 115 ± 25 17 ± 4† 135 ± 25 110 ± 17‡ 147 ± 21* 126 ± 23‡
*P ≤.01 vs saline/saline. †P <.001 vs aminoguanidine/saline. ‡P ≤.003 vs saline/bombesin.
L-NAME reverses the gastroprotective effect of bombesin. The effects of NOS inhibition on bombesin-induced gastroprotection against acidified ethanol are shown in Table I. As expected, animals in the control group receiving saline developed significant macroscopic gastric injury after exposure to acidified ethanol, while pretreatment with bombesin for 30 minutes prevented gastric injury from this luminal irritant. Administration of aminoguanidine had no effect on the damage caused by acidified ethanol in saline-treated animals and did not exhibit any inhibitory effect on the ability of bombesin to prevent acidified ethanol-induced gastric damage. In contrast, administration of L-NAME almost negated the gastroprotective action of subcutaneous bombesin. Both doses of L-NAME tended to exacerbate gastric injury from acidified ethanol, although this was only significant with the 10 mg/kg dose. The higher dose of L-NAME did not cause macroscopic injury in the absence of acidified ethanol (data not shown). Consistent with its competitive inhibitory actions, concurrent treatment with L-arginine blunted the ability of L-NAME to prevent bombesin-induced gastroprotection, whereas Darginine did not (24 ± 4 vs 127 ± 20 mm2; P < .001). L-NAME attenuates bombesin-induced increases in gastric mucosal blood flow. The effects of bombesin on gastric mucosal blood flow are shown in Table II. When compared with rats receiving saline, bombesin dose-dependently increased gastric mucosal blood flow from baseline determinations. However, in the presence of the NOS inhibitor L-NAME, no augmentation in gastric mucosal blood flow occurred after the administration of bombesin, an effect reversed by L-, but not D-arginine (Figure). Bombesin enhances eNOS immunoreactivity. By using site-directed polyclon antibodies to
426 Castañeda et al
Surgery September 2000 Table II. Effects of bombesin on gastric mucosal blood flow in the absence of acidified ethanol Treatment Saline Bombesin (10 µg/kg) Bombesin (50 µg/kg) Bombesin (100 µg/kg)
Blood flow (% change) 0.5 ± 0 3±2 21 ± 4* 38 ± 6*
*P < .05 vs saline.
Figure. The effect of nitric oxide synthase inhibition with subcutaneous L-NAME (5 mg/kg) and the effect of combined treatment with intraperitoneal L-arginine (L-Arg; 300 mg/kg) or D-arginine (D-Arg; 300 mg/kg) on subcutaneous bombesin (100 µg/kg)-induced gastric hyperemia using laser Doppler in anesthetized rats. Data are expressed as percent of change from baseline determinations and represent the mean ± SEM. n ≥ 5 for all groups; asterisk, P < .001 vs saline/saline; two asterisks, P < .001 vs saline/bombesin; plus sign, P < .001 vs L-NAME/ bombesin.
eNOS, nNOS, and iNOS, the abundance of each protein in gastric mucosal homogenates after saline or bombesin treatment was examined without exposing the stomachs to acidified ethanol. When compared with saline-treated rats, a 30minute treatment with bombesin (100 µg/kg) significantly increased eNOS immunoreactivity (2.07 ± 0.03 vs 3 ± 0.01 relative densimetric units, P = .03), but not iNOS immunoreactivity (0.04 ± 0.01 vs 0.03 ± 0.007 relative densimetric units, P = .169) or nNOS immunoreactivity (3.1 ± 0.02 vs 3.3 ± 0.03 relative densitometric units, P = .62). In the Western immunoblot analysis for iNOS, the positive control from the macrophage cell line RAW264.7 was clearly detectable, and we have previously demonstrated our ability to detect enhanced iNOS immunoreactivity after LPS. 11 The neural NOS isoform was clearly detected in both groups. Thus, this study indicated that bombesin enhances expression of eNOS but not nNOS or iNOS in rat gastric mucosa. Vagotomy does not attenuate bombesininduced gastroprotection. In the studies designed to examine the role of an intact vagus nerve in bombesin-induced gastroprotection, it was demonstrated that bombesin was gastroprotective in animals that had a sham laparotomy, a laparotomy and pyloroplasty, or a vagotomy with pyloroplasty (19 ± 5 vs 24 ± 6 vs 26 ± 3 mm2, the P value was not significant). Taken together, this study indicated that bombesin-induced gastroprotec-
tion did not require an intact vagus nerve to exert its protective actions. DISCUSSION This study confirmed that bombesin is a potent protective agent against gastric injury that is caused by acidified ethanol in the conscious rat. Inhibition of NOS with L-NAME, but not aminoguanidine, negated the protective action of bombesin. The effect of L-NAME was reversed by administration of L-arginine, but not D-arginine. Bombesin also dosedependently increased gastric mucosal blood flow, an effect that was negated by L-NAME. L-arginine, in turn, reversed the actions of L-NAME. These findings indicate that bombesin-induced gastroprotection and the ability of bombesin to enhance blood flow are mediated primarily by nitric oxide produced by the endothelial isoform of NOS. Moreover, the finding that bombesin increased eNOS but not nNOS or iNOS immunoreactivity further suggested that the endothelial isoform of NOS is responsible for the gastroprotective actions associated with bombesin. A wide variety of gastric peptides have been shown to prevent gastric injury. These include but are not limited to pentagastrin, gastrin-17, cholecystokinin, somatostatin, and bombesin. We have previously shown that bombesin dose-dependently prevents gastric mucosal injury from acidified ethanol according to macroscopic and morphologic criteria.6 In addition, the gastroprotective effect of bombesin was negated by ablation of sensory neurons5 and gastrin receptor blockade.6 In the latter study, it was also shown that bombesin dosedependently increased gastrin release.6 Taken together, these studies suggested that bombesin maintained the integrity of the gastric epithelium by the release of endogenous gastrin and activation of sensory neurons. Exogenous gastrin, on the other hand, has also been shown to prevent damage to the gastric mucosa when given in either physiologic or pharmacologic doses.7,14 In these studies, the protective actions of gastrin were attenuated by the ablation of sensory neurons or NOS inhibition or both.7,14 Consequently, it seemed rea-
Surgery Volume 128, Number 3
sonable to conjecture that gastroprotection by bombesin should also be blunted by the administration of an NOS inhibitor if indeed its protective effects were caused by the release of gastrin. Our NOS inhibition studies clearly demonstrated that nitric oxide plays a role in bombesin-induced gastroprotection. However, our study also suggests that bombesin-induced gastroprotection does not require an intact vagal nerve because vagotomy failed to attenuate this action. Consequently, it is our contention that the actions of bombesin are mediated primarily through the noncholinergic, nonadrenergic population of neurons (ie, capsaicin sensitive) as previously reported.5 The importance of nitric oxide in mucosal defense has been illustrated by numerous investigators.11,15-21 It is a potent vasodilator and not surprisingly is involved with regulation of the gastric mucosal microcirculation.17 Exogenous and endogenous nitric oxide have been shown to protect the gastric mucosa against injury by ethanol, whereas inhibition of NOS has been demonstrated to increase gastric mucosal damage from a variety of luminal irritants.18,19 In addition, nitric oxide has been proposed to be a major mediator of calcitonin gene-related peptide-induced hyperemia and sensory neuron-mediated gastric protection.22 In most of these studies, the constitutive or endothelial isoform of NOS has been examined. However, recent evidence suggests that nitric oxide produced by the inducible isoform may also play a role in mucosal defense.10 Furthermore, we have shown that there is some background-iNOS present in gastric mucosa taken from control rats similar to that previously found in the ileum.11,23 Because the effects of NOS inhibition on bombesin-induced gastroprotection are unclear and because NOS inhibitors can have differential effects depending upon which isoform is blocked,11,15,16 this study was undertaken to assess whether two different NOS inhibitors would have a similar or a contrasting effect on bombesin-induced gastroprotection. The finding that aminoguanidine, a selective iNOS inhibitor,8 failed to reverse bombesin-induced gastroprotection suggested that the inducible isoform of NOS was not responsible for this effect. In comparison, the finding that the nonselective NOS inhibitor L-NAME impeded the ability of bombesin to exert its gastroprotective effect suggested that the constitutive isoform of NOS was required for bombesin to maintain gastric mucosal integrity. Furthermore, the fact that L-arginine, but not its enantiomer D-arginine reversed the effects of LNAME provided additional evidence that bombesin prevented gastric injury by the release of nitric
Castañeda et al 427
oxide from the constitutive isoform of NOS. The finding of increased eNOS immunoreactivity was somewhat surprising as most studies suggest that only the activity of the enzyme is enhanced. We are not aware that this has been reported before. Because 30 minutes of treatment with bombesin would be unlikely to alter the expression of eNOS, we speculate that it is caused by the stabilization of the protein. Mucosal prostaglandins, calcitonin gene-related peptide, gastrin, bombesin, and nitric oxide among many substances all interact and have been implicated in mucosal defense. Interestingly, most of these substances seem to induce mucosal hyperemia.7,14,20,24-26 The importance of blood flow to the stomach has been well-described.27 In addition to supplying nutrients and oxygen to the epithelium, the gastric microcirculation dilutes, neutralizes, or removes toxic substances that have diffused into the mucosa from the lumen. Moreover, it assures that those intracellular processes that underlie mucosal resistance to injury can proceed unabated. The fact that gastrin increases gastric mucosal blood flow through a nitric oxide-mediated mechanism7,14 and bombesin causes the release of endogenous gastrin6 led us to speculate that bombesin would likewise increase blood flow through a nitric oxide-mediated mechanism. Our blood flow experiments indicated that bombesin increases blood flow to the stomach in a dosedependent fashion and did so in doses previously shown to dose-dependently increase gastrin release and cause gastroprotection.6 Furthermore, a role for nitric oxide in this response was made by demonstrating that the administration of L-NAME negated bombesin-induced gastric hyperemia, an effect reversed by L-arginine. Taken together, these studies suggested that the ability of bombesin to maintain the integrity of the gastric mucosa in the face of a damaging irritant is in part caused by an augmentation in gastric mucosal blood flow caused by nitric oxide. In conclusion, this study confirmed that bombesin is a potent gastroprotective agent and demonstrated that inhibition of NOS by L-NAME, but not aminoguanidine, negated this effect. Bombesin also enhanced gastric mucosal blood flow, an effect likewise prevented by L-NAME. The fact that L-arginine but not D-arginine reversed the effects of L-NAME and that bombesin enhanced eNOS immunoreactivity suggests that nitric oxide produced by the constitutive isoform of NOS plays an important role in mediating the gastroprotective actions and hyperemic responses associated with bombesin.
428 Castañeda et al
REFERENCES 1. Greeley GH Jr, Partin M, Spannage A, Dinh T, Hill FLC, Trowbridge J, et al. Distribution of bombesin-like peptides in the alimentary canal of several vertebrae species. Regul Pept 1986;16:169-181. 2. Richelsen B, Rehfeld JF, Larsson LI. Antral gland cell column: a method for studying release of gastric hormones. Am J Physiol 1983;245:G463-9. 3. Makhlouf GM, Schubert ML. Antral bombesin: physiological regulator of gastrin. In: Y Tache, P Melchiorri, L Negri, editors. The bombesin-like peptides in health and disease, Volume 547. New York: New York Academy of Sciences; 1988. p. 225-33. 4. Evangelista S, Maggi CA, Lei A. Lack of influence of capsaicin-sensitive sensory fibers on adaptive cytoprotection in rat stomach. Dig Dis Sci 1988;33:1050-61. 5. Mercer DW, Cross JM, Castañeda AA, Gunter JA. Gastroprotective actions of bombesin, L-DOPA, and mild irritants: role of prostaglandins and sensory neurons. Surgery 1998;124:864-70. 6. Mercer DW, Cross JM, Chang LK, Lichtenberger LM. Bombesin prevents gastric injury in the rat: role of gastrin. Dig Dis Sci 1998;43:826-33. 7. Stroff T, Plate S, Respondek M, Muller KM, Peskar BM. Protection by gastrin in the rat stomach involves afferent neuron, calcitonin gene related peptide and nitric oxide. Gastroenterology 1995;109:89-97. 8. Holzer P, Peskar BM, Peskar BA, Amann R. Release of calcitonin gene-related peptide induced by capsaicin in the vascularly perfused rat stomach. Neurosci Lett 1990;108:195-200. 9. Whittle BJR, Lopez-Belmonte J, Moncada S. Regulation of gastric mucosal integrity by endogenous nitric oxide: interactions with prostanoids and sensory neuropeptides in the rat. Br J Pharmacol 1990;99:607-11. 10. Tepperman BL, Soper BD. Nitric oxide synthase induction and cytoprotection of rat gastric mucosa from injury by ethanol. Can J Physiol Pharmacol 1994;72: 1308-12. 11. Mercer DW, Castañeda AA, Denning JW, Chang LK, Russell DH. Effects of endotoxin on gastric injury from luminal irritants in rats: potential roles of nitric oxide. Am J Physiol 1998;275): G449-59. 12. Barreto JC, Smith GS, Russell DH, Miller TA. Gastric damage caused by acidified ethanol: role of molecular HCl. Am J Physiol 1993;265):G133-7. 13. Mercer DW, Klemm K, Cross JM, Smith GS, Cashman M, Miller TA. Cholecystokinin-induced protection against gastric injury is independent of endogenous somatostatin. Am J Physiol 1996;271:G692-G700.
Surgery September 2000 14. Mercer DW, Cross JM, Smith GS, Miller TA. Protective action of gastrin-17 against alcohol-induced gastric injury in the rat: role in mucosal defense. Am J Physiol 1997;273:G365-73. 15. Vos TA, Gouw ASH, Klock PA, Havinga R, Goor HV, Huitema S, et al. Differential effects of nitric oxide synthase inhibitors on endotoxin-induced liver damage in rats. Gastroenterology 1997;113:1323-33. 16. Takeuchi K, Takehara K, Kaneko T, Okabe S. Nitric oxide and prostaglandins in regulation of acid secretory response in rat stomach following injury. J Pharmacol Exp Ther 1995;272-375. 17. Whittle BJR. Neuronal and endothelium-derived mediators in the modulation of the gastric microcirculation: integrity in the balance. Br J Pharmacol 1993;110:3-17. 18. Masuda E, Kawano S, Nagano K, Tsuji S, Take Y, Tsujii M, et al. Endogenous nitric oxide modulates ethanol-induced gastric mucosal injury in rats. Gastroenterology 1995; 108:58-64. 19. Lopez-Belmonte J, Whittle BJR, Moncada S. The actions of nitric oxide donors in the prevention or induction of injury to the rat gastric mucosa. Br J Pharmacol 1993;108:73-8. 20. Konturek SJ, Brzozowski T, Bielanski W, Schally A. Role of endogenous gastrin in gastroprotection. Eur J Pharmacol 1995;278:203-12. 21. MacNaughton WK, Cirino G, Wallace JL. Endotheliumderived relaxing factor (nitric oxide) has protective actions in the stomach. Life Sci 1989;45:1869-76. 22. Lambrecht N, Burchert M, Respondek M, Muller KM, Peskar BM. Role of calcitonin gene-related peptide and nitric oxide in the gastroprotective effect of capsaicin in the rat. Gastroenterology 1993;104:1371-80. 23. Weisbrodt NW, Pressley TA, Li YF, Zembowicz MJ, Higham SC, Zembowicz A, et al. Decreased ileal smooth muscle contractility and increased nitric oxide synthase II expression induced by lipopolysaccharide. Am J Physiol 1996;271: G454-60. 24. Podolsky RS, Graboswski M, Milner R, Ritchie WP, Dempsey DT. Capsaicin-induced gastric hyperemia and protection are NO dependent. J Surg Res 1994;57:438-42. 25. Holzer P, Pabst MA, Lippe JT, Peskar B, Peskar BA, Livingston EH, et al. Afferent nerve-mediated protection against deep mucosal damage in the rat stomach. Gastroenterology 1990;98: 838-48. 26. Tepperman BL, Whittle BJ. Endogenous nitric oxide and sensory neuropeptides interact in the modulation of the rat gastric microcirculation. Br J Pharmacol 1992;105:171-5. 27. Ritchie WP, Mercer DW. Mediators of bile acid-induced alterations in gastric mucosal blood flow. Am J Surg 1991;161:126-30.
Background. Bombesin prevents gastric injury primarily by the release of endogenous gastrin. Gastroprotection by exogenous gastrin is negated by nitric oxide synthase inhibition, which implicates a role for nitric oxide as a protective mediator. Because both endothelial and inducible isoforms of this enzyme can play a role in mucosal defense, this study was done to examine the contrasting effects of 2 nitric oxide synthase inhibitors on bombesin-induced gastroprotection. Methods. Rats were given subcutaneous saline or bombesin (10-100 µg/kg) 30 minutes before they received a 1-mL orogastric bolus of acidified ethanol (150 mmol/L of hydrochloric acid/50% ethanol) and rats were killed 5 minutes later for assessment of macroscopic injury (mm2). Gastric mucosal blood flow was measured by laser Doppler. Endothelial, neural, and inducible nitric oxide synthase were assessed by using Western immunoblot. Results. Bombesin decreased gastric mucosal damage, and dose-dependently increased blood flow when compared with saline-treated rats. Endothelial but not neural or inducible nitric oxide synthase immunoreactivity was increased by bombesin. In additional studies, intraperitoneal administration of NG-nitro-L-arginine methyl ester (L-NAME, 5-10 mg/kg), a nonselective nitric oxide synthase inhibitor, negated bombesin-induced gastroprotection and hyperemia, whereas the selective inducible inhibitor aminoguanidine (45 mg/kg) did not. Subcutaneous (SC) L-arginine (300 mg/kg), but not D-arginine, abolished the effects of L-NAME. Conclusions. Taken together, these data suggest that nitric oxide produced by the endothelial isoform of nitric oxide synthase plays an important role in mediating the gastroprotective and hyperemic actions associated with bombesin. (Surgery 2000;128:422-8.) From the Department of Surgery, University of Texas–Houston Medical School, Houston
BOMBESIN IS A NEUROPEPTIDE localized to the gastrin- and acid-secreting portions of the stomach and other areas along the digestive tract.1 In the stomach, bombesin directly stimulates gastrin release from antral glands by binding to its receptors located on the gastrin-containing cells.2 Its effects on acid secretion vary from species to species and depending on whether it is given peripherally or centrally.3 We and others have shown that exogenous bombesin renders the stomach less susceptible to injury from a luminal irritant, an effect blunted by ablation of capsaicin-sensitive neurons.4,5 In addition, we have shown that gastroprotection by bombesin is mediated primarSupported by NIH grant DK 50445. Accepted for publication April 4, 2000. Reprint requests: David W. Mercer, MD, Chief of Surgery, University of Texas–Houston Medical School/LBJ General Hospital, 5656 Kelley, Suite 3-OS 62008, Houston, TX 77026. Copyright © 2000 by Mosby, Inc. 0039-6060/2000/$12.00 + 0 11/56/107982 doi:10.1067/msy.2000.107982
422 SURGERY
ily by the release of endogenous gastrin.6 Gastroprotection by exogenous gastrin on the other hand is negated by nitric oxide synthase (NOS) inhibition implicating a role for nitric oxide as a protective mediator.7 Furthermore, gastroprotection by means of sensory neuron activation is caused by an increase in gastric mucosal blood flow and involves a nitric oxide-mediated mechanism.8 Because some evidence suggests that both the endothelial9 and the inducible10 isoforms of NOS can play a role in gastric mucosal defense, this study was done to examine the contrasting effects of 2 NOS inhibitors on bombesin-induced gastroprotection. In addition, the effects of bombesin on gastric mucosal blood flow and NOS expression were assessed. It was our hypothesis that bombesin-induced gastroprotection would require a functioning endothelial isoform of NOS (eNOS), and consequently its protective effects would be attenuated when this isoform was competitively inhibited with a nonselective NOS inhibitor as opposed to a selective inducible NOS (iNOS) inhibitor. Moreover,
Surgery Volume 128, Number 3
this inhibition should be reversed in the presence of excess substrate with L- but not D-arginine. Thus, the first set of experiments examined the effects of the nonselective NOS inhibitor L-NAME and the selective iNOS inhibitor aminoguanidine on the ability of bombesin to prevent gastric injury caused by the luminal irritant acidified ethanol. In the second set of experiments, the effects of L- and D-arginine on L-NAME-induced reversal of bombesininduced gastroprotection were assessed. In the third set of experiments, the effect of bombesin on gastric mucosal blood flow, in the presence or in the absence of NOS inhibition, was evaluated. In a fourth experiment, the effect of bombesin on endothelial, neural, and iNOS expression was determined. In a fifth experiment, the effect of vagotomy on bombesin-induced gastroprotection was assessed. MATERIALS AND METHODS Animals. Female Sprague-Dawley rats weighing approximately 200 g were used in all studies and were housed at a constant room temperature with a 12-hour light-dark cycle. All experiments were performed in rats deprived of food for 18 to 24 hours but allowed free access to water up to the beginning of the study. On the day of experimentation, all animals were randomly assigned to one of several groups. The University of Texas–Houston Animal Welfare Committee previously approved all experimental protocols before any studies were conducted. Chemicals. Bombesin, aminoguanidine, LNAME, L-arginine, and D-arginine were of molecular biology grade and were purchased from Sigma Chemical Co (St Louis, Mo). Aminoguanidine was dissolved in 0.1 N hydrochloric acid and subsequently neutralized to pH 7.4 with 0.1 N sodium hydroxide. All other solutions were prepared in saline. The polyclonal antibodies against eNOS and neural NOS (nNOS) were purchased from Transduction Laboratory (Lexington, Ky), and the polyclonal antibody against iNOS was prepared and developed by the Trauma Research Center at the University of Texas Medical School, Houston, Tex. The efficacy of this antibody probe has been previously published.11 Effect of bombesin on gastric injury from acidified ethanol. The first set of experiments was designed to confirm exogenous bombesin could prevent acidified ethanol-induced gastric injury. Experiments were performed on conscious animals and bombesin given subcutaneously in a dose of 100 µg/kg, while control animals received saline. We have previously shown that bombesin dose-
Castañeda et al 423
dependently (10–100 µg/kg) prevents gastric injury and that the 100 µg/kg dose provides the highest level of gastroprotection.6 Because we have previously shown that bombesin also prevents acidified ethanol-induced morphologic injury to the gastric mucosa, morphologic analysis was not undertaken in these experiments.6 Thirty minutes after pretreatment with bombesin, all animals were given a 1-mL oral gastric bolus of acidified ethanol (150 mmol/L hydrochloric acid/50% ethanol). This concentration of ethanol in combination with hydrochloric acid has been previously demonstrated to reproducibly induce maximal visible lesion formation to the glandular portion of the stomach within 5 minutes of exposure.12-14 Thus rats were killed 5 minutes after receiving this luminal irritant. In additional rats receiving either a 30-minute pretreatment with bombesin or saline, the acidified ethanol was given, and the rats were killed 1 hour after receiving the damaging agent. These latter rats were included to examine whether bombesin protects the stomach from longer periods of exposure or simply delays the onset of injury. After the animals were killed, all stomachs were removed and the total area of macroscopic injury to the glandular portion of the stomach quantified by computerized planimetry. The surface areas of the individual lesions were measured in a blinded manner. Results were recorded as mean damage in square millimeters (mm2) ± standard error of the mean for each experimental group. Effect of NOS inhibition on bombesin-induced gastroprotection. To examine the role of NOS isoforms in bombesin-induced gastric mucosal protection, experiments were conducted using the NOS inhibitors aminoguanidine and NG-nitro-L-arginine methyl ester (L-NAME). Two NOS inhibitors were assessed because NOS inhibitors can have different effects depending upon which isoform of NOS is inhibited.15,16 Evidence suggests that aminoguanidine is more selective for the inducible isoform of NOS whereas L-NAME is nonselective in its ability to inhibit NOS.11,16 By using a similar protocol as described above and a sample size of 6 rats per group, animals were randomized to receive either aminoguanidine (45 mg/kg intraperitoneal [IP]) or L-NAME (5-10 mg/kg, SC) concurrently with saline or bombesin (100 µg/kg) while controls received saline. Thirty minutes later, gastric injury was induced with acidified ethanol (150 mmol HCl/50% ethanol) and macroscopic injury was determined as previously described. This dose of aminoguanidine was chosen because it almost abolishes the generation of gastric luminal nitrates and nitrites induced by the administration of endotox-
424 Castañeda et al
in.11 The doses of L-NAME have also been previously shown to reverse endotoxin-induced hypotension without much effect on serum nitrates or nitrites.17 In other rats, the reversibility of NOS inhibition was assessed by administration of either the natural substrate for NOS, L-arginine (300 mg/kg, SC), or its enantiomer, D-arginine (300 mg/kg, SC), given concurrently with the other pretreatments. This dose of L-arginine has been shown to reverse the effects of L-NAME.17 In other rats, the effect of NOS inhibition on gastric mucosal injury in the absence of acidified ethanol was assessed by giving rats L-NAME (10 mg/kg, SC) for 30 minutes and then killing them without exposure to the luminal irritant. Effect of bombesin on gastric mucosal blood flow. These experiments were conducted to assess the effect of bombesin on gastric mucosal blood flow. In these studies, rats were anesthetized with an intraperitoneal injection of 6 mg/kg xylazine and 70 mg/kg of ketamine. After induction of anesthesia, the stomach was exposed by a midline incision. A catheter was introduced through the nonglandular fore stomach to provide access for a Teflon-coated laser optic flow probe (Peri Flux PF409, standard probe, 0.25-mm fiber separation; Peri Med, Järfälla, Sweden). The flow probe was positioned to allow contact with the glandular or acid-secreting portion of the stomach. After the appropriate position of the probe was ensured, the stomach was allowed to equilibrate for 30 minutes. Mucosal blood flow to the stomach was then recorded continuously with a laser-Doppler flow monitor (Peri Flux 4001 Master; Peri Med). Blood flow was recorded for a 5-minute period as the measurement of basal gastric mucosal blood flow, followed by administration of subcutaneous saline or bombesin (10–100 ug/kg). Blood flow was monitored for an additional 30 minutes, and the animal was killed. The doses of bombesin used are known to dose-dependently prevent gastric injury.6 In addition, the effect of L-NAME (5 mg/kg, SC) on bombesin (100 µg/kg, SC)-induced changes in blood flow was assessed by giving L-NAME or saline concurrently with bombesin. L-NAME reversal was assessed with L- or D-arginine (300 mg/kg, SC). For all groups, a sample size of 5 or more animals per group was used. Blood flow is reported as percent of change from baseline determinations. Protein extraction and Western immunoblot analysis for eNOS, nNOS, and iNOS. To estimate and compare the content of gastric eNOS, nNOS, and iNOS, rats were anesthetized with an injection of 6 mg/kg IP of xylazine and 70 mg/kg IP of ketamine 30 minutes after receiving saline (n = 5) or
Surgery September 2000 bombesin (100 µg/kg, n = 5). After achieving satisfactory anesthesia, a midline laparotomy was performed, and the stomach was immediately removed. The stomach was opened along the greater curvature, and the mucosa of the glandular portion of the stomach was gently rinsed with saline. The mucosa was blotted dry, scraped off the underlying muscularis externa and serosa, snapfrozen in liquid nitrogen, and stored at –80°C before protein extraction and Western immunoblot analysis was performed for eNOS, nNOS, and iNOS. Protein in each sample was extracted by pulverizing the frozen tissue with the use of a mortar and pestle in a liquid nitrogen slurry. This sample was then added to 1 mL lysis buffer (10 mmol/L Tris, pH 8.6, 100 µmol/L phenylmethylsulfonyl fluoride, 1.5 mmol/L magnesium chloride, 100 µU/mL aprotinin, 10 µg/mL leupeptin, and 0.5% Nonidet [octylphenoxy] polyethoxy ethanol-40) and then subjected to two 30-second bursts of a Polytron (Vertishear). These samples were then transferred to microfuge tubes and centrifuged for 10 minutes at 11,000g. The supernatant was removed and added to a sample buffer (125 mmol/L Tris, pH 6.8, 2% sodium dodecyl sulfate, 5% glycerol, 1% β-mercaptoethanol, and 0.003% bromophenol blue). Protein concentrations within each homogenate were determined by using the bicinchoninic acid protein assay (Pierce, Rockford, Ill) before the sample buffer was added to the supernatant. The remaining proteins in these prepared homogenates were separated by sodium dodecyl sulfate (7.5%)-PAGE, using 40 µg protein per sample for iNOS determinations and 80 µg protein per sample for eNOS and nNOS determinations. Resultant proteins were electroblotted onto nitrocellulose paper and incubated for 1 hour at room temperature in blocking solution (5% nonfat dried milk, 0.1% Tween 20, and phosphate-buffered saline). The resultant blot was then washed twice in 0.1% Tween 20-phosphatebuffered saline followed by a 1-hour incubation with a specific polyclonal anti-iNOS antibody (1:2000 dilution) or a specific polyclonal antieNOS or anti-nNOS antibody (1:700 dilution). Blots were then washed twice and incubated with a horseradish peroxidase-conjugated donkey, antirabbit immunoglobin as a secondary antibody (1:10,000 dilution for iNOS and 1:5000 dilution for eNOS and nNOS) for 1 hour. After 2 final washes, the immune complexes were visualized with the use of enhanced chemiluminescence detection (Amersham; Arlington Heights, Ill). The resultant autoradiographs were then assessed semiquantitatively by using computer-assisted den-
Castañeda et al 425
Surgery Volume 128, Number 3
sitometry and were reported as mean relative densitometric units. Effect of vagotomy on bombesin-induced gastroprotection. To determine whether bombesininduced gastroprotection required an intact vagus nerve to exert its protective actions, additional studies were performed with bombesin in rats undergoing sub-diaphragmatic vagotomy. For these studies, rats were anesthetized with intraperitoneal xylazine (6 mg/kg) and ketamine (70 mg/kg). A midline laparotomy was performed, the vagal trunks identified and transected along the esophagus, and a pyloroplasty performed by making a longitudinal incision along the pylorus and then closing it with interrupted 5-0 silk sutures in a transverse fashion. Two sets of controls were included. In one group, rats underwent sham laparotomy alone. In the other group, rats underwent laparotomy with pyloroplasty without sub-diaphragmatic vagotomy. All rats were allowed to awaken after completion of their procedures and were placed in individual isolated cages with normal access to food and water. Seven days after the procedures, rats were fasted overnight and then randomized to receive either saline or bombesin (100 µg/kg) for 30 minutes. After the 30-minute treatment period, gastric injury was induced with orogastric acidified ethanol, and macroscopic injury was determined. Statistics. All values in the figure and text are expressed as mean plus or minus the standard error of the mean of “n” observations, where “n” is the number of animals examined. In all studies, a sample size of 5 or more animals per group was used. Statistical significance was determined by using analysis of variance followed by Fisher’s post hoc test. P < .05 was considered statistically significant. RESULTS Exogenous bombesin protects against acidified ethanol. Animals in the control group receiving saline followed by orogastric acidified ethanol had extensive macroscopic gastric injury. This injury was characterized by the presence of hemorrhagic lesions along the gastric folds and was confined primarily to the glandular portions of the stomach, as previously reported.6,13,14 Despite the severity of injury caused by acidified ethanol, subcutaneous bombesin prevented the development of these lesions (21 ± 5 vs 118 ± 22 mm2, P < .001) when compared with saline-treated rats. Bombesin was also efficacious at attenuating the extent of macroscopic damage caused by a longer exposure to acidified ethanol (ie, 1 hour) compared with saline–pretreated rats (23 ± 4 vs 110 ± 19 mm2, P < .001).
Table I. Effects of NOS inhibition on bombesin (100 µg/kg)-induced gastroprotection from acidified ethanol in the conscious rat Treatment
Injury (mm2)
Saline/saline Bombesin/saline Aminoguanidine/saline Aminoguanidine/bombesin L-NAME 5 mg/kg saline L-NAME 5 mg/kg bombesin L-NAME 10 mg/kg saline L-NAME 10 mg/kg bombesin
118 ± 22 21 ± 5* 115 ± 25 17 ± 4† 135 ± 25 110 ± 17‡ 147 ± 21* 126 ± 23‡
*P ≤.01 vs saline/saline. †P <.001 vs aminoguanidine/saline. ‡P ≤.003 vs saline/bombesin.
L-NAME reverses the gastroprotective effect of bombesin. The effects of NOS inhibition on bombesin-induced gastroprotection against acidified ethanol are shown in Table I. As expected, animals in the control group receiving saline developed significant macroscopic gastric injury after exposure to acidified ethanol, while pretreatment with bombesin for 30 minutes prevented gastric injury from this luminal irritant. Administration of aminoguanidine had no effect on the damage caused by acidified ethanol in saline-treated animals and did not exhibit any inhibitory effect on the ability of bombesin to prevent acidified ethanol-induced gastric damage. In contrast, administration of L-NAME almost negated the gastroprotective action of subcutaneous bombesin. Both doses of L-NAME tended to exacerbate gastric injury from acidified ethanol, although this was only significant with the 10 mg/kg dose. The higher dose of L-NAME did not cause macroscopic injury in the absence of acidified ethanol (data not shown). Consistent with its competitive inhibitory actions, concurrent treatment with L-arginine blunted the ability of L-NAME to prevent bombesin-induced gastroprotection, whereas Darginine did not (24 ± 4 vs 127 ± 20 mm2; P < .001). L-NAME attenuates bombesin-induced increases in gastric mucosal blood flow. The effects of bombesin on gastric mucosal blood flow are shown in Table II. When compared with rats receiving saline, bombesin dose-dependently increased gastric mucosal blood flow from baseline determinations. However, in the presence of the NOS inhibitor L-NAME, no augmentation in gastric mucosal blood flow occurred after the administration of bombesin, an effect reversed by L-, but not D-arginine (Figure). Bombesin enhances eNOS immunoreactivity. By using site-directed polyclon antibodies to
426 Castañeda et al
Surgery September 2000 Table II. Effects of bombesin on gastric mucosal blood flow in the absence of acidified ethanol Treatment Saline Bombesin (10 µg/kg) Bombesin (50 µg/kg) Bombesin (100 µg/kg)
Blood flow (% change) 0.5 ± 0 3±2 21 ± 4* 38 ± 6*
*P < .05 vs saline.
Figure. The effect of nitric oxide synthase inhibition with subcutaneous L-NAME (5 mg/kg) and the effect of combined treatment with intraperitoneal L-arginine (L-Arg; 300 mg/kg) or D-arginine (D-Arg; 300 mg/kg) on subcutaneous bombesin (100 µg/kg)-induced gastric hyperemia using laser Doppler in anesthetized rats. Data are expressed as percent of change from baseline determinations and represent the mean ± SEM. n ≥ 5 for all groups; asterisk, P < .001 vs saline/saline; two asterisks, P < .001 vs saline/bombesin; plus sign, P < .001 vs L-NAME/ bombesin.
eNOS, nNOS, and iNOS, the abundance of each protein in gastric mucosal homogenates after saline or bombesin treatment was examined without exposing the stomachs to acidified ethanol. When compared with saline-treated rats, a 30minute treatment with bombesin (100 µg/kg) significantly increased eNOS immunoreactivity (2.07 ± 0.03 vs 3 ± 0.01 relative densimetric units, P = .03), but not iNOS immunoreactivity (0.04 ± 0.01 vs 0.03 ± 0.007 relative densimetric units, P = .169) or nNOS immunoreactivity (3.1 ± 0.02 vs 3.3 ± 0.03 relative densitometric units, P = .62). In the Western immunoblot analysis for iNOS, the positive control from the macrophage cell line RAW264.7 was clearly detectable, and we have previously demonstrated our ability to detect enhanced iNOS immunoreactivity after LPS. 11 The neural NOS isoform was clearly detected in both groups. Thus, this study indicated that bombesin enhances expression of eNOS but not nNOS or iNOS in rat gastric mucosa. Vagotomy does not attenuate bombesininduced gastroprotection. In the studies designed to examine the role of an intact vagus nerve in bombesin-induced gastroprotection, it was demonstrated that bombesin was gastroprotective in animals that had a sham laparotomy, a laparotomy and pyloroplasty, or a vagotomy with pyloroplasty (19 ± 5 vs 24 ± 6 vs 26 ± 3 mm2, the P value was not significant). Taken together, this study indicated that bombesin-induced gastroprotec-
tion did not require an intact vagus nerve to exert its protective actions. DISCUSSION This study confirmed that bombesin is a potent protective agent against gastric injury that is caused by acidified ethanol in the conscious rat. Inhibition of NOS with L-NAME, but not aminoguanidine, negated the protective action of bombesin. The effect of L-NAME was reversed by administration of L-arginine, but not D-arginine. Bombesin also dosedependently increased gastric mucosal blood flow, an effect that was negated by L-NAME. L-arginine, in turn, reversed the actions of L-NAME. These findings indicate that bombesin-induced gastroprotection and the ability of bombesin to enhance blood flow are mediated primarily by nitric oxide produced by the endothelial isoform of NOS. Moreover, the finding that bombesin increased eNOS but not nNOS or iNOS immunoreactivity further suggested that the endothelial isoform of NOS is responsible for the gastroprotective actions associated with bombesin. A wide variety of gastric peptides have been shown to prevent gastric injury. These include but are not limited to pentagastrin, gastrin-17, cholecystokinin, somatostatin, and bombesin. We have previously shown that bombesin dose-dependently prevents gastric mucosal injury from acidified ethanol according to macroscopic and morphologic criteria.6 In addition, the gastroprotective effect of bombesin was negated by ablation of sensory neurons5 and gastrin receptor blockade.6 In the latter study, it was also shown that bombesin dosedependently increased gastrin release.6 Taken together, these studies suggested that bombesin maintained the integrity of the gastric epithelium by the release of endogenous gastrin and activation of sensory neurons. Exogenous gastrin, on the other hand, has also been shown to prevent damage to the gastric mucosa when given in either physiologic or pharmacologic doses.7,14 In these studies, the protective actions of gastrin were attenuated by the ablation of sensory neurons or NOS inhibition or both.7,14 Consequently, it seemed rea-
Surgery Volume 128, Number 3
sonable to conjecture that gastroprotection by bombesin should also be blunted by the administration of an NOS inhibitor if indeed its protective effects were caused by the release of gastrin. Our NOS inhibition studies clearly demonstrated that nitric oxide plays a role in bombesin-induced gastroprotection. However, our study also suggests that bombesin-induced gastroprotection does not require an intact vagal nerve because vagotomy failed to attenuate this action. Consequently, it is our contention that the actions of bombesin are mediated primarily through the noncholinergic, nonadrenergic population of neurons (ie, capsaicin sensitive) as previously reported.5 The importance of nitric oxide in mucosal defense has been illustrated by numerous investigators.11,15-21 It is a potent vasodilator and not surprisingly is involved with regulation of the gastric mucosal microcirculation.17 Exogenous and endogenous nitric oxide have been shown to protect the gastric mucosa against injury by ethanol, whereas inhibition of NOS has been demonstrated to increase gastric mucosal damage from a variety of luminal irritants.18,19 In addition, nitric oxide has been proposed to be a major mediator of calcitonin gene-related peptide-induced hyperemia and sensory neuron-mediated gastric protection.22 In most of these studies, the constitutive or endothelial isoform of NOS has been examined. However, recent evidence suggests that nitric oxide produced by the inducible isoform may also play a role in mucosal defense.10 Furthermore, we have shown that there is some background-iNOS present in gastric mucosa taken from control rats similar to that previously found in the ileum.11,23 Because the effects of NOS inhibition on bombesin-induced gastroprotection are unclear and because NOS inhibitors can have differential effects depending upon which isoform is blocked,11,15,16 this study was undertaken to assess whether two different NOS inhibitors would have a similar or a contrasting effect on bombesin-induced gastroprotection. The finding that aminoguanidine, a selective iNOS inhibitor,8 failed to reverse bombesin-induced gastroprotection suggested that the inducible isoform of NOS was not responsible for this effect. In comparison, the finding that the nonselective NOS inhibitor L-NAME impeded the ability of bombesin to exert its gastroprotective effect suggested that the constitutive isoform of NOS was required for bombesin to maintain gastric mucosal integrity. Furthermore, the fact that L-arginine, but not its enantiomer D-arginine reversed the effects of LNAME provided additional evidence that bombesin prevented gastric injury by the release of nitric
Castañeda et al 427
oxide from the constitutive isoform of NOS. The finding of increased eNOS immunoreactivity was somewhat surprising as most studies suggest that only the activity of the enzyme is enhanced. We are not aware that this has been reported before. Because 30 minutes of treatment with bombesin would be unlikely to alter the expression of eNOS, we speculate that it is caused by the stabilization of the protein. Mucosal prostaglandins, calcitonin gene-related peptide, gastrin, bombesin, and nitric oxide among many substances all interact and have been implicated in mucosal defense. Interestingly, most of these substances seem to induce mucosal hyperemia.7,14,20,24-26 The importance of blood flow to the stomach has been well-described.27 In addition to supplying nutrients and oxygen to the epithelium, the gastric microcirculation dilutes, neutralizes, or removes toxic substances that have diffused into the mucosa from the lumen. Moreover, it assures that those intracellular processes that underlie mucosal resistance to injury can proceed unabated. The fact that gastrin increases gastric mucosal blood flow through a nitric oxide-mediated mechanism7,14 and bombesin causes the release of endogenous gastrin6 led us to speculate that bombesin would likewise increase blood flow through a nitric oxide-mediated mechanism. Our blood flow experiments indicated that bombesin increases blood flow to the stomach in a dosedependent fashion and did so in doses previously shown to dose-dependently increase gastrin release and cause gastroprotection.6 Furthermore, a role for nitric oxide in this response was made by demonstrating that the administration of L-NAME negated bombesin-induced gastric hyperemia, an effect reversed by L-arginine. Taken together, these studies suggested that the ability of bombesin to maintain the integrity of the gastric mucosa in the face of a damaging irritant is in part caused by an augmentation in gastric mucosal blood flow caused by nitric oxide. In conclusion, this study confirmed that bombesin is a potent gastroprotective agent and demonstrated that inhibition of NOS by L-NAME, but not aminoguanidine, negated this effect. Bombesin also enhanced gastric mucosal blood flow, an effect likewise prevented by L-NAME. The fact that L-arginine but not D-arginine reversed the effects of L-NAME and that bombesin enhanced eNOS immunoreactivity suggests that nitric oxide produced by the constitutive isoform of NOS plays an important role in mediating the gastroprotective actions and hyperemic responses associated with bombesin.
428 Castañeda et al
REFERENCES 1. Greeley GH Jr, Partin M, Spannage A, Dinh T, Hill FLC, Trowbridge J, et al. Distribution of bombesin-like peptides in the alimentary canal of several vertebrae species. Regul Pept 1986;16:169-181. 2. Richelsen B, Rehfeld JF, Larsson LI. Antral gland cell column: a method for studying release of gastric hormones. Am J Physiol 1983;245:G463-9. 3. Makhlouf GM, Schubert ML. Antral bombesin: physiological regulator of gastrin. In: Y Tache, P Melchiorri, L Negri, editors. The bombesin-like peptides in health and disease, Volume 547. New York: New York Academy of Sciences; 1988. p. 225-33. 4. Evangelista S, Maggi CA, Lei A. Lack of influence of capsaicin-sensitive sensory fibers on adaptive cytoprotection in rat stomach. Dig Dis Sci 1988;33:1050-61. 5. Mercer DW, Cross JM, Castañeda AA, Gunter JA. Gastroprotective actions of bombesin, L-DOPA, and mild irritants: role of prostaglandins and sensory neurons. Surgery 1998;124:864-70. 6. Mercer DW, Cross JM, Chang LK, Lichtenberger LM. Bombesin prevents gastric injury in the rat: role of gastrin. Dig Dis Sci 1998;43:826-33. 7. Stroff T, Plate S, Respondek M, Muller KM, Peskar BM. Protection by gastrin in the rat stomach involves afferent neuron, calcitonin gene related peptide and nitric oxide. Gastroenterology 1995;109:89-97. 8. Holzer P, Peskar BM, Peskar BA, Amann R. Release of calcitonin gene-related peptide induced by capsaicin in the vascularly perfused rat stomach. Neurosci Lett 1990;108:195-200. 9. Whittle BJR, Lopez-Belmonte J, Moncada S. Regulation of gastric mucosal integrity by endogenous nitric oxide: interactions with prostanoids and sensory neuropeptides in the rat. Br J Pharmacol 1990;99:607-11. 10. Tepperman BL, Soper BD. Nitric oxide synthase induction and cytoprotection of rat gastric mucosa from injury by ethanol. Can J Physiol Pharmacol 1994;72: 1308-12. 11. Mercer DW, Castañeda AA, Denning JW, Chang LK, Russell DH. Effects of endotoxin on gastric injury from luminal irritants in rats: potential roles of nitric oxide. Am J Physiol 1998;275): G449-59. 12. Barreto JC, Smith GS, Russell DH, Miller TA. Gastric damage caused by acidified ethanol: role of molecular HCl. Am J Physiol 1993;265):G133-7. 13. Mercer DW, Klemm K, Cross JM, Smith GS, Cashman M, Miller TA. Cholecystokinin-induced protection against gastric injury is independent of endogenous somatostatin. Am J Physiol 1996;271:G692-G700.
Surgery September 2000 14. Mercer DW, Cross JM, Smith GS, Miller TA. Protective action of gastrin-17 against alcohol-induced gastric injury in the rat: role in mucosal defense. Am J Physiol 1997;273:G365-73. 15. Vos TA, Gouw ASH, Klock PA, Havinga R, Goor HV, Huitema S, et al. Differential effects of nitric oxide synthase inhibitors on endotoxin-induced liver damage in rats. Gastroenterology 1997;113:1323-33. 16. Takeuchi K, Takehara K, Kaneko T, Okabe S. Nitric oxide and prostaglandins in regulation of acid secretory response in rat stomach following injury. J Pharmacol Exp Ther 1995;272-375. 17. Whittle BJR. Neuronal and endothelium-derived mediators in the modulation of the gastric microcirculation: integrity in the balance. Br J Pharmacol 1993;110:3-17. 18. Masuda E, Kawano S, Nagano K, Tsuji S, Take Y, Tsujii M, et al. Endogenous nitric oxide modulates ethanol-induced gastric mucosal injury in rats. Gastroenterology 1995; 108:58-64. 19. Lopez-Belmonte J, Whittle BJR, Moncada S. The actions of nitric oxide donors in the prevention or induction of injury to the rat gastric mucosa. Br J Pharmacol 1993;108:73-8. 20. Konturek SJ, Brzozowski T, Bielanski W, Schally A. Role of endogenous gastrin in gastroprotection. Eur J Pharmacol 1995;278:203-12. 21. MacNaughton WK, Cirino G, Wallace JL. Endotheliumderived relaxing factor (nitric oxide) has protective actions in the stomach. Life Sci 1989;45:1869-76. 22. Lambrecht N, Burchert M, Respondek M, Muller KM, Peskar BM. Role of calcitonin gene-related peptide and nitric oxide in the gastroprotective effect of capsaicin in the rat. Gastroenterology 1993;104:1371-80. 23. Weisbrodt NW, Pressley TA, Li YF, Zembowicz MJ, Higham SC, Zembowicz A, et al. Decreased ileal smooth muscle contractility and increased nitric oxide synthase II expression induced by lipopolysaccharide. Am J Physiol 1996;271: G454-60. 24. Podolsky RS, Graboswski M, Milner R, Ritchie WP, Dempsey DT. Capsaicin-induced gastric hyperemia and protection are NO dependent. J Surg Res 1994;57:438-42. 25. Holzer P, Pabst MA, Lippe JT, Peskar B, Peskar BA, Livingston EH, et al. Afferent nerve-mediated protection against deep mucosal damage in the rat stomach. Gastroenterology 1990;98: 838-48. 26. Tepperman BL, Whittle BJ. Endogenous nitric oxide and sensory neuropeptides interact in the modulation of the rat gastric microcirculation. Br J Pharmacol 1992;105:171-5. 27. Ritchie WP, Mercer DW. Mediators of bile acid-induced alterations in gastric mucosal blood flow. Am J Surg 1991;161:126-30.