Original article



1Department of Physiology, Jagiellonian University Medical College, Cracow, Poland, 2Department of Physiology, Lviv National Medical University, Lviv, Ukraine, 3Department of Pathology, Lviv National Medical University, Lviv, Ukraine, 4Department of Medicine I, University of Erlangen-Nuremberg, Erlangen, Germany, 5Instiyute of Stomatology, Jagiellonian University Medical College, Cracow, Poland

  Melatonin (MT) is known to protect gastrointestinal mucosa against various types of injury but its effects on esophageal damage have not been studied. We examined the effects of MT on acute esophageal injury and the mechanism involved in the action of this indole. Acute esophageal lesions were induced by perfusion with acid-pepsin solution using tube inserted through the oral cavity into the mid of esophagus of anaesthetized rats with or without inhibition of prostaglandin (PG) generation by indomethacin (5 mg/kg/day), nitric oxide (NO) formation by NG-nitro-L-arginine (L-NNA, 20 mg/kg/day) or sensory nerves deactivation by capsaicin (125 mg/kg, sc). The esophageal injury was assessed by macroscopic score and histologic activity index. The esophageal mucosal blood flow (EBF) was determinated by H2-gas clearance method. The plasma TNF-alpha and nitrate/nitrite (NOx) levels and mucosal PGE2 contents were assessed by immunoassays. Esophageal acid-pepsin perfusion induced noticeable esophageal mucosal injury as compared to perfusion with vehicle saline. The pretreatment with MT prevented significantly esophageal injury, raised EBF and mucosal content of PGE2, while decreasing the levels of TNF-alpha. Inhibition of COX/PG and NOS/NO systems by indomethacin and L-NNA, respectively, or inactivation of sensory nerves by capsaicin, that manifested in further increase of esophageal injury, reduced the levels of EBF, markedly raised the levels TNF-alpha and reduced mucosal PGE2, but the pretreatment with MT prevented significantly esophageal injury, improved EBF and raised mucosal PGE2 contents. These studies suggest that MT can be considered as a novel esophagoprotector, acting, at least in part, through the COX/PG and NOS/NO systems and activation of sensory nerves.

Key words: melatonin, esophageal mucosa, acidic-pepsin injury, esophageal lesions, COX/PG, NOS/NO, esophageal blood flow, sensory nerves, PGE2


Recent studies have demonstrated that the acid-related diseases are currently widespread disorders with increasing prevalence of gastroesophageal reflux disease (GERD) (1). Dangerous complications of GERD including esophageal ulcer (2-7%), haemorrhage (< 2%), stricture (4-20%) and Barrett's esophagus (10-15%, BE) are linked with the damage of the esophageal mucosa by acid-pepsin or biliary secretions refluxed to esophagus (2 - 4). In the situation of persistent mucosal injury of esophagus, BE can give rise to esophageal adenocarcinoma. Feagins et al. (5) noted that esophageal cancer is one of the most deadly forms of gastrointestinal (GI) cancer with a mortality rate exceeding 90%.

Pathogenesis of GERD is still controversial, but current data suggest that the disease is multifactorial and includes epithelial, secretory and motor impairments (6, 7). The esophageal defence depends on the activation of neural (e.g. esophago-salivary and esophago-bronchial reflexes) and local mucosal mechanisms. The properties of esophageal epithelial barrier are determined by its permeability, cell ion exchange, mucosal blood flow and mucosal repair processes (8, 9). Secretory component of esophageal defence consists of overlapping actions of the salivary secretion (HCO3-, mucus, epidermal growth factor - EGF and PG) and esophageal mucosal gland secretion (HCO3-, mucus, some phospholipids, EGF) that form multilayer barrier (10 - 12) resisting corrosive and proteolytic effects of gastric acid-pepsin and biliary secretion (13 - 15). Moreover, dysmotility, including impairment of esophageal peristalsis and dysfunction of lower esophageal and pyloric sphincters, also seem to play an important role in functioning of "anti-reflux barriers" and the luminal clearance of the esophagus from various irritants (16).

The maintenance of integrity of epithelial barrier of GI tract against aggressive factors involves both neuronal and paracrine systems (17 - 19). Cyclooxygenase (COX)-PG and nitric oxide (NO) synthase (NOS)-NO systems are parts of this paracrine protection, while capsaicin-sensitive afferent neurons contribute to the mucosal protection involving the release of sensory mucoso-protective and vasodilatory neuropeptides such as calcitonin-gene related peptide (CGRP) (20). However, the involvement of COX/PG, NOS/NO systems and sensory innervations in the mechanism of esophageal integrity are still not completely understood. Particularly, little is known about the involvement of melatonin (MT), that is a potent antioxidant substance with confirmed protective activity in the gastro-duodeno-pancreatic region (21 - 26), in the acute esophageal mucosal injury and local esophageal microcirculation as well as in the paracrine protection including activation of COX/PG and NOS/NO systems. Recent studies in humans (27) revealed that plasma concentrations of MT, which increase at the night time, are lower in GERD patients than in healthy controls suggesting that this indole may exert beneficial influence on the esophageal mucosa.

The current animal models of induction of esophageal injury include surgical procedures that divert the gastroduodenal contents into the esophagus. The major limitation of these models is their failure to control the amount, concentration and composition of the components of refluxate entering the esophagus and significant postoperative stress and morbidity. Therefore, this study was designed to develop a novel, less complicated and well-controlled model of esophageal injury by perfusion of esophagus with exogenous acid-pepsin solution with or without addition of bile in anaesthetized rats without surgical intervention.

The aims of this study, therefore, were: (1) to investigate the morphological and functional characteristics of esophageal lesions and accompanying changes in esophageal mucosal blood flow (EBF) in rat model of acid/pepsin and bile-induced esophageal injury; (2) to study the effect of pretreatment with MT on acute esophageal lesions induced by acid-pepsin or bile and accompanied changes in EBF in rats without and with inhibition of COX/PG and NOS/NO systems; (3) to assess the effects of MT pretreatment on the plasma levels of the major proinflammatory cytokine such as tumor necrosis factor alfa (TNF-alpha), plasma content of nitrate/nitrite (NOx) and esophageal mucosal content of prostaglandin E2 (PGE2), and (4) to determine the implication of capsaicin-sensitive sensory nerves in the formation of esophageal lesions provoked by acid-pepsin perfusion and MT-induced protection of esophageal mucosa.


Male Wistar rats, weighing 180-220 g and fasted for 24 h before the study with free access to water, were used in our studies. These experimental procedures were approved by the Institutional Animal Care and Use Committee of Jagiellonian University Medical College in Cracow (Poland). All procedures followed the criteria, technical standards and rights applied to animal research. All trials were followed in accordance with the statements of the European Union regarding handling of experimental animals.

Esophageal lesions induced by acid-pepsin and bile perfusion of esophagus

Acute esophageal lesions were induced in rats by the esophageal perfusion using tube inserted through the oral cavity with the tip in the mid portion of the esophagus. The fasted rats were anaesthetized with pentobarbital (10 mg/kg i.p., Vetbutal, Pulawy, Poland) and perfused using the plastic tube (Intramedic Polyethylene Tubing, Clay Adams Division of Becton, Dickinson Company, NJ, USA) inserted through the oral cavity to the final destination of 3-4 cm below the upper esophageal sphincter. The test solution was applied daily for 7 consecutive days via the perfusion pump (Infusion Pump, 353, Medipan, Warsaw, Poland) at a rate of 2 ml/h during 2 h period. During the perfusions, rats were maintained under the mild sedation and in a 30° supine position (to prevent refluxate aspiration to the lung) under the spontaneous breathing. The temperature of the perfused solution was 37° C. Esophageal perfusion solution consisted of 0.25 M HCl with addition of pepsin (0.1 mg/ml perfusate) without or with the combination with bile (0.5 ml) collected spontaneously from previously operated rats with biliary-pancreatic fistula by inserting the plastic tube into the common bile duct. The pH of perfusate was about 0.7 as measured by pH-meter (CG 840, SCHOTT, Germany). The role of sensory afferent nerves in esophageal damage caused by acid-pepsin perfusion and in the mucosal protection by MT was tested in rats with capsaicin-induced deactivation (CD) of these nerves. For this purpose the animals were pretreated with capsaicin injected s.c. for 3 consecutive days at a dose of 25, 50 and 50 mg/kg for about 2 weeks before the esophageal perfusion experiment. All the injections of capsaicin were performed under the ether anesthesia to counteract the pain reactions and respiratory impairment associated with the injection of this agent.

Three (A, B and C) experimental series of rats with acute esophagitis were used. Series A included the following treatment groups; 1) vehicle (1 ml saline injected i.p.) followed 30 min later by the esophagus perfused with 2 ml/h of 0,9% NaCl solution; 2) vehicle (1 mL saline i.p.) followed 30 min later by esophagus perfused with acid-pepsin solution; 3) indomethacin (INDO), a non-selective inhibitor of COX/PG in dose of 5 mg/kg injected i.p. 30 min before the start of the esophageal perfusion with acid-pepsin; 4) NG-nitro-L-arginine (L-NNA), an inhibitor of NOS/NO, injected in a dose of 20 mg/kg i.p. 30 min before the start of later of esophageal perfusion with acid-pepsin and 5) capsaicin-denervated (CD) rats treated with vehicle (1 ml saline i.p.) administration followed 30 min later by acid-pepsin esophageal perfusion. In series B, the influence of MT, given i.p. in a dose of 20 mg/kg, on the esophageal damage induced by acid-pepsin perfusion was examined and the role of COX/PG and NOS/NO systems and sensory nerves in the esophagoprotective action of MT was assessed by employing rats pretreated, respectively, with INDO , L-NNA or CD. The following groups of rats were used in series B; (1) MT (20 mg/kg i.p.) followed 30 min later by vehicle saline and then 30 min later by esophageal acid-pepsin perfusion, (2) MT (20 mg/kg i.p.) followed 30 min later by indomethacin (5 mg/kg, i.p.) and finally 30 min later by esophageal acid-pepsin perfusion, (3) MT (20 mg/kg i.p.) followed 30 min later by L-NNA (20 mg/kg, i.p.) applied 30 min before acid-pepsin esophageal perfusion, and (4) MT (20 mg/kg) given i.p. to rats with CD and then 30 min later by acid-pepsin esophageal perfusion.

In series C, the influence of MT, given i.p. in a dose of 20 mg/kg, on the esophageal damage induced by acid-pepsin + bile perfusion was examined and the role of COX/PG and NOS/NO systems and sensory nerves in this action of MT was assessed by employing rats pretreated, respectively, with indomethacin, L-NNA and capsaicin. The following groups of rats were used in series C; (1) vehicle-saline (1 ml i.p.) followed 30 min later by esophageal perfusion with acid-pepsin; (2) MT (20 mg/kg i.p.) followed 30 min later by esophageal acid-pepsin + bile perfusion, (3) MT (5, 10 or 20 mg/kg i.p.) followed 30 min later by acid-pepsin perfusion, (4) MT (20 mg/kg i.p.) followed 30 min later by indomethacin (5 mg/kg, i.p.) and finally 30 min later by esophageal acid-pepsin + bile perfusion, (5) MT (20 mg/kg i.p.) followed 30 min later by L-NNA (20 mg/kg, i.p.) applied 30 min before acid-pepsin + bile esophageal perfusion, and (6) MT (20 mg/kg) given i.p. to rats with CD and then 30 min later by acid-pepsin + bile esophageal perfusion.

Determination of esophageal blood flow (EBF)

Just before the termination of the last experiment with 2 h esophageal perfusion sessions, the animals were anaesthetised with pentobarbital and abdominal cavity was opened. The abdominal part of the esophagus was gently pulled below the diaphragm and exposed for the measurement of the EBF by means of H2-gas clearance method. For this purpose double electrodes of an electrolytic regional laser Doppler blood flowmeter (Biotechnical Science, Model RGF-2, Osaka, Japan) were inserted into the esophageal mucosa as described before for the gastric blood flow measurement (24). The EBF was presented as a percent change from the control value obtained in saline-perfused (vehicle-control group) rats.

The area of esophageal mucosal lesions was assessed after the termination of the experiment, when the animals were anaesthetised by pentobarbital overdose and sacrificed. Immediately after removal of the esophagus, it was opened longitudinally, then pinned out mucosal side up. It was examined for macroscopic changes and photographed, fixed in 10% formaldehyde for microscopic examination.

Determination of plasma TNF-alpha and melatonin levels and esophageal tissue PGE2 content.

Immediately after EBF measurement, a venous blood sample was withdrawn from vena cava into the EDTA containing vials and used for the determination of plasma TNFalpha concentration by the Quantikine M immunoassay kit specific for TNF-alpha (R & D System, Inc. Minneapolis MN, USA). Plasma melatonin was also determined quantitatively by specific RIA for this indole using kit purchased from DRG Instrument GmbH, Hamburg, Germany and performed in accordance with manufacturer instruction as described before (17, 22, 28). Esophageal mucosal PGE2 synthesis was measured in fresh biopsy samples obtained from mid and lower portions of esophagus that were immediately weighed and placed in Eppendorf tubes. Samples were suspended in 1 ml of 10 mM sodium phosphate buffer (pH 7.4) and were minced for 15 s, then placed in water bath at 37°C for 20 min, and finally centrifuged at 9000 x g for 30 s. The supernatant was stored at -80°C for subsequent determination of PGE2 concentrations by an immunoassay kit specific for PGE2 (R&D Systems Inc. Minneapolis, MN) (28). Plasma NOx levels were measured by generally accepted Griess reaction (28).

Macroscopic and microscopic structural examinations

For examination of macroscopic changes, the esophageal lesion score system was used from 0 to 3 and photographed. According to this macroscopic scoring, esophagus had score 0 for normal shimmering mucosa; 1 - for hyperemic or edematous mucosa with focal hemorrhagic spots; 2 - for multiple erosions with hematin attached; 3 - for ulcerations, dark necrotic spots. For the microscopic examinations, the segments of the distal, middle and proximal samples of esophagus were fixed in 10% formaldehyde for histological evaluation. In order to analyse the histological characteristics the evaluation of the haematoxylin and eosin specimens were used due to the histological activity index (HAI). HAI is based on the degree of microscopic lesions and is evaluated according to system of the epithelial loss: 0 - none, 1 - splitting and erosion, 2 - ulceration, 3 - large ulcer and necrosis; for leukocyte infiltration: 0 - none, 1 - mild, 2 - moderate, 3 - severe; regenerative epithelial changes: 0 - none, 1- basal hyperplasia, 2- mitosis, ballon cells, akantosis, 3 - parakeratosis).

Statistical analysis

The statistical analysis was performed with the program package STATISTICA for Windows 5.5 (Stat Soft, USA). The results of the evaluations according to the semiquantitative scale are expressed as means ± SEM. For the comparison of data a paired Newman-Keuls's test was used with a level of significance at P < 0,05.


In the series A of the experiments, the vehicle-treated rats with esophagus perfused only with saline did not show any macroscopical or microscopical alterations (Figs 1 and 2). Mean EBF averaged 65.2 ± 1.5 ml/min/100g tissue, which was considered as 100% (control). The results of macroscopic damage score and EBF in rats of A series are shown on Fig.1 (left panel). Esophageal perfusion with acid-pepsin solution in vehicle saline-treated rats of series A resulted in mucosal edema with focal hemorrhagic erosions observed in all animals tested and average damage score was about 1.4 ± 0.2. The developments of these acute esophageal lesions by the acid-pepsin perfusion of the esophagus resulted in a significant fall in mucosal EBF by about 24% when compared to that recorded in saline-perfused rats without addition of acid-pepsin solution. The combination of esophageal acid-pepsin perfusion with administration of indomethacin or L-NNA resulted in more severe ulcerations occurring mainly in the upper and middle parts of esophagus with lesion score in series A rats reaching, respectively, about 2.9 and 3.0 (Fig. 3 - left panel). These ulcerations were accompanied by further significant decrease in EBF, reaching, respectively, 49±5% and 52±3% of the value recorded in vehicle-saline perfused rats of this series. Similarly, acid-pepsin esophageal perfusion in rats with CD resulted in a mucosal injury with damage score and the fall of EBF reaching the values similar to those recorded in indomethacin or L-NNA pretreated animals.

Fig. 1. A. Macroscopic view of the intact esophagus observed in control saline-perfused rats; B. Esophagus of rat perfused with acid-pepsin solution with severe inflammation of the mucosa and deep penetrating ulcus in the upper portion of the esophagus; C. Esophagus of rat perfused with acid-pepsin solution pretreated with melatonin (20 mg.kg i.p.) and showing mild mucosal inflammation and small erosion in its upper portion.

Fig. 2. Histologic section of the mid portion of the esophagus perfused with vehicle saline (A), acid-pepsin perfusion (B) and acid pepsin perfusion combined with pretreatment with melatonin (20 mg/kg i.p) (C).

As shown on Fig. 3 (right panel), the pretreatment with MT given i.p. at a dose of 20 mg/kg i.p. in rats of series B, markedly reduced the area of mucosal lesions obtained with acid-pepsin perfusion as compared to that without such MT-pretreatment. MT given to rats with the combination of indomethacin, L-NNA or capsaicin denervation and esophageal acid-pepsin perfusion, the severity of esophageal lesions was significantly attenuated and this was accompanied by a significant increase in EBF as compared to the respective value without MT administration (Fig. 3 - right panel).

Fig. 3. Macroscopic damage score and esophageal blood flow (EBF) in series A and series B of rats with vehicle saline (veh) and acid-pepsin perfusion of the esophagus without and with pretreatment with indomethacin (INDO), L-NNA or capsaicin. Mean ± SEM of 6-8 experiments on 6-8 rats. Asterisk indicates significant change as compared to the vehicle-saline esophageal perfusion; cross indicates significant change as compared to the corresponding values recorded in rats without melatonin pretreatment.

Fig. 4 shows the macroscopic lesion score and EBF in rats of series C with the esophageal perfusion with the combination of acid-pepsin (1.5 ml/h) plus bile (0.5 ml/h). The macroscopical lesion score reached 2.8 ± 0.3 and it was significantly higher than that recorded with acid-pepsin alone without bile. Pretreatment with MT at a dose of 5 - 20 mg/kg i.p. before the acid-pepsin-bile perfusion attenuated in a dose-dependent, the lesion score and this effect was accompanied by a gradual increase in EBF. Both, the decrease in lesion score and the rise in EBF observed with MT applied in a dose of 20 mg in this series of animals were significantly smaller that those recorded in series B in animals pretreated with MT and perfused with acid-pepsin without bile.

Fig. 4. Macroscopic damage score and esophageal blood flow (EBF) in series C of rats with esophageal perfusion with acid-pepsin only or with perfusion of esophagus with the combination of acid-pepsin bile perfusion without and with pretreatment with melatonin at various doses; 5, 10 or 20 mg/kg i.p. Mean ± SEM of 6-8 experiments on 6-8 rats. Asterisk indicates significant change as compared to the acid-pepsin esophageal perfusion; cross indicates significant change as compared to the value recorded in vehicle-saline pertreated rats without melatonin pretreatment.

Plasma TNF-alpha level in vehicle-saline treated control rats of series A averaged 6.2 ± 0.9 pg/ml and esophageal acid-pepsin perfusion raised this level to about 14.2 ± 2.4 pg/ml . In rats pre-treated with indomethacin or L-NNA, the plasma levels of TNF-alpha rose significantly above that observed with acid-pepsin perfusion alone. Plasma NOx levels was also significantly increased in rats treated with melatonin, but they showed significant decrease after inhibition of NOS/NO system with L-NNA. It is of interest that esophageal mucosal content of PGE2 was almost doubled following mucosal damage by acid-pepsin perfusion but the pretreatment with indomethacin, dramatically attenuated in part the mucosal PGE2 content towards the level observed with acid-pepsin perfusion (Table 1).

Table 1. Plasma levels of TNFalpha, NOx and melatonin and esophageal mucosa contents of PGE2 in rats of series A, B and C. Means ± SEM of 6-8 experiments on 6-8 rats. Asterisk indicates significant change as compared to the value recorded in vehicle-saline-treated control. Cross indicates significant change as compared to the value recorded in acid-pepsin perfusion experiments without pretreatment with INDO, L-NNA or CAPS. Double sterisks indicate significant change as compared to similar tests but without MT administration.

Plasma immunoreactive MT, which in vehicle-saline treated rats averaged 105 ± 14 pg/ml, showed small but significant reduction in the mucosa of acid-pepsin perfused esophagus (81 ± 12 pg/ml), particularly when rats were pretreated with indomethacin (68 ± 4 pg/ml) or capsaicin (64 ± 7 pg/ml). Exogenous MT given i.p. in a dose of 20 mg/kg (series B) resulted in several fold elevation in plasma immunoreactive MT (336 ± 62 pg/ml) and these levels of MT remained elevated in rats pretreated with MT plus indomethacin (380 ± 46 pg/ml), L-NNA (419 ± 44 pg/ml) or capsaicin (420 ± 87 pg/ml). MT administration in graded doses (5-20 mg/kg i.p.) in rats of series C raised dose-dependently plasma MT levels that at the doses of 5, 10 and 20 mg/kg i.p., reached the values of 166 ± 14, 187 ± 16 and 425 ± 36 pg/ml, respectively, (Table 1).

Pretreatment with MT (20 mg/kg i.p.) in rats of series B with acid-pepsin esophageal perfusion resulted in significant attenuation of plasma levels of TNF-a from control value of 14.2 ± 2.4 to 8.1 ± 1.1 pg/ml and an increase in mucosal PGE2 contents from control value of 241 ± 44 ng/g to 325 ± 38 ng/g. In tests on rats pretreated with indomethacin in which mucosal PGE2 was significantly reduced (80 ± 17), the administration of MT failed to affect significantly this attenuated PGE2 content (98 ± 10 ng/g) (Table 1).

The NOx plasma level [measured by Griess reaction (28)] in vehicle-treated rats averaged 28.8 ± 3.1 µmol/L and it rose significantly with the acid-pepsin perfusion of esophagus to 39 ± 4.0 µmol/L and almost markedly fell (vehicle control) after application of indomethacin (27 ± 1.3 µmol/L). In contrast, the pretreatment with L-NNA significantly decreased NOx both in saline and acid-pepsin perfused esophagus. Administration of MT tended to increase NOx value, particularly in rats pretreated with indomethacin (18.4 ± 3.5 µmol/L) and significantly attenuated this value in L-NNA-pretreated animals (17 ± 1.9 µmol/L (Table 1).

The microscopic examination did not show any damage of the esophageal mucosa in rat of vehicle-control group. Different abnormalities in epithelial morphology, degrees of inflammatory cell infiltration and hyperplasia were seen in hematoxylin and eosin-stained sections from middle and distal parts of esophageus in vehicle-treated rats against acid-induced injury due to acid-pepsin perfusion in the study series A. Microscopic changes in indomethacin, L-NNA or capsaicin-pretreated rats were more pronounced than those in animals without such pretreatment. Superficial and erosive lesions were clearly evident in epithelial layer indicating cell injury, necrosis and also signs for apoptosis. Cotreatment with MT significantly reduced esophageal epithelial injury assessed by hematoxylin-eosin staining of mucosal samples obtained from rats perfused with acid-pepsin without and with administration of indomethacin, L-NNA or capsaicin (see Fig. 2). Local hemodynamic disorders presented by signs of irregular hyperemia, stasis, restricted perivascular diapedesis with hemorrhage and focal edema in underlayered stroma were significantly less severe in MT treated animals than in thoese receiving indomethacin, L-NNA or capsaicin.


This study provides an evidence that the protective mechanisms of the esophageal mucosa against major potential damaging factors such as acid-pepsin secretion or bile refluxed into the esophageal lumen involve several intergrated mechanisms including mucosal blood flow and local paracrine-neurocrine factors such as COX/PG and NOS/NO systems and sensory nerves. The present study demonstrates that the inhibition of COX/PG and NOS/NO systems and the deactivation of sensory nerves with capsaicin render the esophageal mucosa highly susceptible to the damage by the exposure to acid-pepsin or to the combination of acid, pepsin and bile solution.

Esophageal disorders in humans such as GERD, nonerosive reflux disease, erosive reflux disease, eosinophillic esophagitis, Barrett’s esophagus (BE) remain a significant management challenge because, while they are really common, the treatment options are limited and often inadequate (29). Risk and benefit of acid-suppressive drugs therapy, particularly potent class of proton pump inhibitors (PPI), are still contentious. Not surprisingly, PPI widely used in GERD and other acid-related diseases, were reported to cause deep alterations in intragastric environment with near-neutral pH and elevated plasma concentration of major gastric secretagogue hormone such as gastrin (40). In addition, PPI are usually safe but sometimes linked with unrecognized side effects such as bacterial overgrowth causing mucosal inflammation. Experimental results showed that gastric achlorhydria may lead to tumor development in animal due to increased enterochromaffin-like cell density (carcinoid) and enhanced growth of gastric fundic mucosa after prolonged PPIs administration. The gastric acid inhibitory therapy (including H2-receptor) antagonists and PPI) increase the risk of gastroenteritis and community-acquired pneumonia in GERD-affected children due to modification in the intestinal microflora composition (29, 30). Thus, there is a need of intense investigation in basic mechanism of esophagocytoprotection.

In this report, we described simple, novel not-surgical method of induction of esophageal injury by daily perfusion of the esophagus that allows for an adequate monitoring of the amount, concentration and composition of perfusion solution (acid, pepsin and bile) that mimics the condition of gastroduodenal reflux, at the same time being less stressful to the animal. The proposed combinations of esophageal perfusion (with hydrochloric acid with pepsin and bile) and time of exposure are sufficient for induction of the esophageal lesions and their pathology that is similar to the changes observed in human beings (14, 30 - 34). Previous studies demonstrated that the esophageal barrier is a one of the important protective component of esophageal wall between underlying tissues and external milieu, exposed to a range of irritants, results in frequent injury. It was found that the onset of esophageal lesions is likely to be multifactorial and the esophageal mucosa response to acid-pepsin and bile acid-trypsin-induced damage was associated with a failure in several mechanisms that include sensory nerves and endogenous NO release, an enhancement in the expression and release of cytokines, growth factors, adhesion molecules and heat shock protein and an increase in apoptosis (34 - 37). Our study showed that the ablation of capsaicin-sensitive sensory neurones increase aggressive effects of acid-pepsin induced damage, partly due to impairment of local microcirculation in esophageal mucosa that may affect trophic processes and augment mucosal destruction. Numerous studies showed that three NOS isozymes: neuronal NOS (nNOS), endothelial NOS (eNOS) and inducible NOS (iNOS) contribute to the function and pathology of the gastrointestinal tract. Small (nanomolar) quantities of NO produced by calcium-dependent nNOS play a physiological role in gastrointestinal motility. Decreased nNOS function can result in motility disorders and lead to lower esophageal sphincter (LES) failure and GERD scenario, with relapse and chronicity of the disease (38). NO produced by eNOS dilates mucosal blood vessels and prevents leukocyte aggregation, and is, therefore, essential for the maintenance of esophageal mucosal blood flow. Absence of eNOS-derived NO results in an increased susceptibility of the gastrointestinal tract to injury (39). We provided an evidence that the pretreatment with L-NNA, non-specific inhibitor of NOS is accompanied by the decrease in plasma NOx levels and the reduction in EBF greatly augments the macroscopic damage score, therefore, it is reasonable to conclude that NOS/NO-mediated response is important component of esophageal cytoprotection. It has been reported that selective NO delivery by gene therapy or NO-donating compounds may offer new therapeutic approach in the treatment of gastrointestinal dysfunctions and the prevention of mucosal injury (40 - 42).

It is widely accepted that melatonin, the most potent endogenous free radical scavenger, plays a pivotal role in NO mediated vasodilatation and contributes to the generation of endogenous mucosal PGE2, derived from COX activity (17, 21 - 25). In this study we investigated the effect of melatonin pretreatment on acid-pepsin and acid-pepsin plus bile-induced esophageal injury. Our study revealed that melatonin, while causing an increase in EBF, prevented dose-dependently the formation of mucosal lesions and afforded protection against the damage of the mucosa exposed to acid-pepsin-bile solution possibly via vasodilating effect on esophageal microcirculation. Further studies are required to prove that melatonin affecting NO system is associated with the activation of the gene and protein expression of NOS.

Another important endogenous substances involved in the esophageal mucosal defense are PG, classic arachidonate metabolites derived from the COX expression and activity. According to our results, esophageal acid-pepsin perfusion caused about 70% increase in the mucosal PGE2 generation and indomethacin, a nonselective blocker of COX/PG system, caused a dramatic (by about 75%) suppression of PG biosynthesis, resulting in significant aggravation of the acid-pepsin-induced esophageal injury. We also found that administration of melatonin has beneficial effects on esophageal mucosa possibly by enhancing PGE2 generation causing esophageal hyperemia, suggesting that these favorable effects of this indole were mediated by mucosal COX-PG system.

Harmful acid-pepsin action due to the impaired ability of esophageal epithelium to neutralize acid load may weaken the esophageal barrier leading to the failure of the esophageal defense system and decrease in the esophageal resistance to the action of gastroduodenal content. Esophageal squamous epithelium covered by mucus is a multilayered aqueous gel produced by submucosal glands (43) and surface mucous cells, continuously linines the entire luminal surface of the esophagus. Mucin, in vivo, is stored in a dense cluster of apical granules, in mucous cell and secretion can occur constitutively and numerous injurious chemical agents stimulate mucus secretion, as do mechanical stimuli (44). Duct and serous cells of esophageal glands play a role in HCO3- secretion and its transport in these cells is dependent on cytosolic and serosal membrane-bound carbonic anhydrase (43) and COX/PG system (45). Esophageal epithelial cell-to-cell contact is more "leaky" than in stomach, therefore, paracellular pathway is shorter and more open, thereby, luminal contents more easy enter to submucosa that was confirmed by clinical investigation of GERD (46). It seems likely that esophageal protection does not require only the secreting mucus cells, but that release of mucus is liberated by Ca2+- depended process from its cytoplasmic, membrane-bound state. Two sources of extracellular Ca2+ are evident in the esophagus. One is saliva, which is rich in Ca2+ and, for this reason, represents a constitutive source of extracellular Ca2+ for surface epithelial cells. A second source, available apically as a signal for mucus-producing cells, but only upon injury, is are Ca2+ ions present in the extracellular environment (47, 48). In animals with sialoadenectomy esophageal lesions were markedly worsened due to weakening mucus-bicarbonate barrier, hypocalcaemia, affected epithelial surface permeability, dilatation of intercellular spaces, lack of EGF and other salivatory bioregulators and endothelial dysfunction (49). Based on our present results, we cannot exclude the possibility, that impaired production of mucus due to an excessive formation of peroxynitrate or excessive luminal NO formed from nitrites of saliva could contribute to the damage induced by esophageal refuxate as reported previously (50, 51). Indeed, plasma levels of NOx in our study was significantly altered by simple acid-pepsin esophageal perfusion and further changed following pretreatment with indomethacin or capsaicin. After application of L-NNA, a potent non-specific suppressor of NOS/NO system, the plasma level of NOx tended to decline in all tests but it was not accompanied by any beneficial influence on the esophageal mucosa possibly due to the decrease in mucosal microcirculation resulting from the deficiency of mucosal generation on NO that is potent local vasodilator. It should be mentioned that the deficiency of mucosal generation of melatonin has been suggested to contribute to the pathogenesis of reflux esophagitis in humans as demonstrated previously showing reduced plasma levels of melatonin in GERD patients (27). Dietary supplementation with melatonin, which was shown to inhibit gastric emptying (52), and its precursor, L-tryptophan, in patients with GERD, caused complete remission of symptoms of this disease, indicating that melatonin may be of practical use in treatment of GERD in humans. Further studies should explained the involvement of heat shock proteins (HSP) in esophagoprotection induced by melatonin, since recent evidence indicate an important role of HSP expression in the mechanism of gastrointestinal integrity (53, 54). It should be emphasized that esophageal mucosa is just a continuation of the mucosal oral lining. It is, therefore, not unexpected that the damage or the mucosal in oral cavity raises the formation of reactive oxygen (ROS) and reactive nitrogen species (RNS), while reducing local generation and salivary production of MT. As interstingly demonstrated most recently (56), exogenous MT applied to oral cavity immediately after tooth extraction in animals, counteracted effectively the oxidative stress damage. Our recent, as yet unpublished studies on rats with multiple ulcers induced by acetic acid applied on oral mucosa in rats, demonstrated a remarkable increase in the generation of ROS and RNS after such oral ulcerogenesis and this could be augmented by clockade or MT2 receptors with luzindole and suppressed by orally administered MT (57).

We conclude that the oral and esophageal defence against acid, pepsin and bile-induced injury as well as other damaging substances such as ulcer-inducing acetic acid, in our new experimental models, depend upon a balance of factors that promote injury of oral and esophageal epithelial barrier, impairment of mucin formation and secretion, oro-esophageal microcirculation, and activation esophageal defence mechanisms which lead to limitation of the progression of aggressive processes. The mucosal integrity of the oro-esophagus is maintained by paracrine systems such as COX/PG and NOS/NO, and sensory capsaicin-sensitive afferent nerves also involved in gastro-, pancreatic- and hepatoprotective responses (55-57). Exogenous MT possesses protective action mediated by PG and vasodilatory effects of NO/NOS system as well as a reduction in oxidative stress induced upper gastrointestinal tube.

Acknowledgments: This study was supported in part by research project from KBN 2PO5B 06130 and by LEK-AM, Warsaw, Poland.

  1. Dent J, El-Serag HB, Wallander M-A, Johansson S. Epidemiology of gastro-oesophageal reflux disease: a systemic review. Gut 2005; 54 (7): 710-717.
  2. Hattori T, Mukaisho K, Miwa K. Pathogenesis of Barrett's esophagus--new findings in the experimental studies of duodenal reflux models. Nippon Rinsho 2005; 63 (8): 1341-1349.
  3. Nishijima K, Miwa K, Miyashita T, et al. Impact of the biliary diversion procedure on carcinogenesis in Barrett's esophagus surgically induced by duodenoesophageal reflux in rats. Ann Surg 2004; 240 (1): 57-67.
  4. Farhadi A., Fields J, Banan A, Kashavarzian A. Reactive oxygen species: are the involved in the pathogenesis of GERD, Barrett's Esophagus, and latter's progression toward esophageal cancer. Am J Gastroenterol 2002; 97 (1): 22-26.
  5. Feagins LA, Souza RF. Molecular targets for treatment of Barrett's esophagus. Dis Esophagus 2005; 18(2): 75-86.
  6. Kahrilas PJ, Lee TJ. Pathophysiology of gastroesophageal reflux disease. Thorac Surg Clin 2005; 15(3): 323-333.
  7. Urita Y, Ishihara S, Arai K, et al. Salivatory gland scintigraphy in gastro-esophageal reflux disease. Gastroenterology 2006; 130(4): Suppl 2: A-398.
  8. Szentpali K, Kaszaki J, Eros G, et al. Changes in microcirculation of esophageal mucosa during acute experimental reflux. Pathogenic role of the biliary component. Magy Seb 2003; 56(2): 61-67.
  9. Fujiwara Y, Higuchi K, Takashima T, et al. Roles of epidermal growth factor and Na+/H+ exchanger-1 in esophageal epithelial defense against acid-induced injury. Am J Physiol Gastrointest Liver Physiol 2006; 290(4): G665-673.
  10. Abdulnour-Nakhoul S, Nakhoul NL, et al. HCO3- secretion in the esophageal submucosal glands. Am J Physiol Gastrointest Liver Physiol 2006; 288(4): G736-744.
  11. Eckley CA, Michelsohn N, Rizzo LV, et al. Salivary epidermal growth factor concentration in adults with reflux laryngitis. Otolaryngol Head Neck Surg 2004; 131(4): 401-406.
  12. White RJ, Morris GP, Moyes CD, et al. Analysis of mucosal stress response in acid-induced esophagitis in opossum. Dig Dis Sci 2002; 47(8): 1873-1885.
  13. Lanas A, Royo Y, Ortego J, Molina M, Sainz R. Experimental esophagitis induced by acid and pepsin in rabbits mimicking human reflux esophagitis. Gastroenterology 1999; 116: 997-1107.
  14. Li Y, Wo JM, Su RR, Ray MB, Jones W, Martin RC. Esophageal injury with external esophageal perfusion. J Surg Res 2005; 129 (1): 107-113.
  15. Manning JJ, Wirz A, McColl KEL. Chemicals formed from the gastric acidification of salivary nitrite influence oesophageal and gastric function. J Pharmacol Sci 2006; 100: Suppl I: A 109.
  16. Bove M, Ruth M, Lundell L, Ny L. Epithelial barrier integrity and intraluminal nitric oxide production in response to acid perfusion of the ferret oesophagus. Acta Physiol Scand 2005; 183 (2): 211-218.
  17. Brzozowska I, Konturek PC, Brzozowski T, et al. Role of prostaglandins, nitric oxide, sensory nerves and gastrin in acceleration of ulcer healing by melatonin and its precursor, L-tryptophan. J Pineal Res 2002; 3: 149-162.
  18. Bhat YM, Bielefeldt K. Capsaicin receptor (TRPV1) and non-erosive reflux disease. Eur J Gastroenterol Hepatol 2006; 18 (3): 263-270.
  19. Reiter RJ, Tan DX, Mayo JC, et al. Neurally-mediated and neurally-independent beneficial actions of melatonin in the gastrointestinal tract. J Physiol Pharmacol 2003; 54 (4): 113-125.
  20. Patterson LM, Zheng H, Ward SM, Berthoud HR. Vanilloid receptor (VR1) expression in vagal afferent neurons innervating the gastrointestinal tract. Cell Tissue Res 2003; 311: 277-287.
  21. Jaworek J, Brzozowski T, Konturek SJ. Melatonin as an organoprotector in the stomach and the pancreas. J Pineal Res 2005; 38(2): 73-83.
  22. Konturek PC, Konturek SJ, Majka J. Melatonin afford protection against gastric lesions induced by ischemia-reperfusion due to its antioxidant and mucosal microcirculatory effect. Eur J Pharmacol 1997; 122: 73-77.
  23. Konturek SJ, Konturek PC, Brzozowski T. Melatonin in gastroprotection against stress-induced acute gastric lesions and healing of chronic gastroduodenal ulcers. J Physiol Pharmacol 2006; 57 (Suppl 5): 51-66.
  24. Jaworek J. Ghrelin and melatonin in the regulation of pancreatic exocrine secretion and maintaining of integrity. J Physiol Pharmacol 2006; 56 (Suppl 5): 83-96.
  25. Messner M, Huether G, Lorf T, et al. Presence of melatonin in the human hepatobiliary -gastrointestinal tract. Life Sci 2001; 69(5): 543-551.
  26. Bubenik GA. Gastrointestinal melatonin localization, function and clinical relevance. Dig Dis Sci 2002; 47: 2336-2346.
  27. Klupinska GA, Wisniowska-Jarosinska M, Haeasiuk, et al. Nocturnal secretion of melatonin in patients with upper digestive tract disorders. J Physiol Pharmacol 2006; 57 (Suppl 5): 41-50.
  28. Konturek SJ, Konturek PC, Brzozowski T. Role of prostaglandins in ulcer healing. J Physiol Pharmacol 2005; 56 (Suppl 5): 33-56.
  29. Fullard M, Kang JY, Neild P, Poullis A, Maxwel JD. Systematic review: does gastro-oesophageal reflux disease progress? Aliment Pharmacol Ther 2006; 24(1): 33-45.
  30. Canani RB, Cirillo P, Roggero P, et al. Therapy with gastric acidity inhibitors increases the risk of acute gastroenteritis and community-acquired pneumonia in children. Pediatrics 2006; 117 (5): 817-820.
  31. Fujimura T, Ohta T, Oyama K, Miyashita T, Miwa K. Role of cyclooxygenase-2 in the carcinogenesis of gastrointestinal tract cancers: A review and report of personal experience. World J Gastroenterol 2006; 12(9): 1336-1345.
  32. Naito Y, Uchiyama K, Kuroda M, et al. Role of pancreatic trypsin in chronic esophagitis induced by gastroduodenal reflux in rats. J Gastroenterol 2006; 41 (3): 198-208.
  33. Oh DS, Hagen JA, Fein M, et al. The impact of reflux composition on mucosal injury and esophageal function. J Gastrointest Surg 2006; 10(6): 787-797.
  34. Poplawski C, Sosnowski D, Szaflarska-Poplawska A, Sarosiek J, McCallum R, Bartuzi Z. Role of bile acids, prostaglandins and COX inhibitors in chronic esophagitis in a mouse model. World J Gastroenterol 2006; 12 (11): 1739-1742.
  35. Bhat YM, Bielefeldt K. Capsaicin receptor (TRPV1) and non-erosive reflux disease. Eur J Gastroenterol Hepatol 2006; 18(3): 263-270.
  36. Hamaguchi M, Fujiwara Y, Takashima T, et al. Increased expression of cytokines and adhesion molecules in rat chronic esophagitis. Digestion 2003; 68(4): 189-197.
  37. Chen T, Hwang H, Rose ME, Nines RG, Stoner GD. Chemopreventive properties of blackrRaspberries in N-nitrosomethylbenzylamine-induced rat esophageal tumorigenesis: Down-regulation of cyclooxygenase-2, inducible nitric oxide synthase, and c-Jun. Cancer Res 2006; 66 (5): 2853-2859.
  38. Yamaguchi T, Yoshida N, Tomatsuri N, et al. Cytokine-induced neutrophil accumulation in the pathogenesis of acute reflux esophagitis in rats. Int J Mol Med 2005; 16 (1): 71-77.
  39. Modlin IM, Sachs G. Acid related diseases: biology and treatment. Philadelphia: Lippincott Williams & Wilkins 2005, pp. 359-362.
  40. Brzozowski T, Konturek SJ, Sliwowski Z, et al. Role of capsaicin-sensitive sensory nerves in gastroprotection against acid-independent and acid-dependent ulcerogens. Digestion 1996; 57: 424-432.
  41. Dijkstra G, van Goor H, Jansen PL, Moshage H. Targeting nitric oxide in the gastrointestinal tract. J Physiol Pharmacol 2004; 55(4): 773-790.
  42. Vallance BA, Dijkstra G, Qiu B, et al. Relative contributions of NOS isoforms during experimental colitis: endothelial-derived NOS maintains mucosal integrity. Am J Physiol Gastrointest Liver Physiol 2004; 287(4): G865-874.
  43. Kumagai H, Mukaisho K, Sugihara H, et al. Cell kinetic study on histogenesis of Barrett's esophagus using rat reflux model. Scand J Gastroenterol 2003; 38(7): 687-692.
  44. Abdulnour-Nakhoul S, Nakhoul NL, Wheeler SA, Wang P, Swenson ER, Orlando RC. HCO3- secretion in the esophageal submucosal glands. Am J Physiol Gastrointest Liver Physiol 2005; 288 (4): G736-744.
  45. Takeuchi K, Aihara E, Sasaki Y, Nomura Y, Ise F. Involvement of cyclooxygenase-1, prostaglandin E2 and EP1 receptors in acid-induced HCO3-secretion in stomach. J Physiol Pharmacol 2006; 57: 4: 661-676.
  46. Miyake K, Tanaka T, McNeil PL. Disruption-induced mucus secretion: repair and protection. PLoS Biology/www.plosbiology.org 2006; 4(9): E276: 0001-0007.
  47. Tobey NA, Hosseini SS, Argote CM, et al. Dilated intercellular spaces and shunt permeability in nonerosive acid-damaged esophageal epithelium. Am J Gastroenterol 2004; 9: 13-22.
  48. Jaworski T, Majewski M, Sarosiek I, et al. Significant enhancement of salivary protective components in males with chronic heartburn symptomatology: a new insight into the balance between aggressive factors and protective mechanisms. Gastroenterology 2006; 30 (4) Suppl 2: A-399.
  49. Kongara K, Soffer EE. Saliva and esophageal protection. Am J Gastroenterol 1999; 94 (6): 1446-1452.
  50. Fujiwara Y, Higuchi K, Takashima T, et al. Roles of epidermal growth factor and Na+/H+ exchanger-1 in esophageal epithelial defense against acid-induced injury. Am J Physiol Gastrointest Liver Physiol 2006; 290 (4): G665-673.
  51. de Souza Pereira R. Regression of gastroduodenal reflux disease symptoms using dietary supplementation with melatonin, vitamins and aminoacids: comparison with omeprazole. J Pineal Res 2006; 41: 195-200.
  52. Kasimay O, Cakir E, Devseren E, Yegen B.C. Exogenous melatonin delays gastric emptying rate in rats: role of CCK2 and 5-HT3 receptors. J Physiol Pharmacol 2005; 56: 543-553.
  53. Targosz A, Pierzchalski P, Krawiec A, et al. Helicobacter pylori inhibits expression of heat shock protein 70 (HSP70) in human epithelial cell line. Importance of CAG A protein. J Physiol Pharmacol 2006; 57: 265-278.
  54. Bonior J, Jaworek J, Konturek SJ, Pawlik WW. Increase of heat shock protein gene expression by melatonin in AR42J cells. J Physiol Pharmacol 2005; 56: 471-481.
  55. Biernat J, Pawlik WW, Sendur R, Dembinski A, Brzozowski T, Konturek SJ. Role of afferent nerves and sensory peptides in the mediation of hepatic artery buffer response. J Physiol Pharmacol 2005; 56: 133-145
  56. Cutando A, Arana C, Gomez-Moreno G, et al. Local application of melatonin into alveolar sockets of beagle dogs reduces tooth removal-induced oxidative stress. J Periodontol 2007; 78: 576-583.
  57. Czesnikiewicz-Guzik M, Konturek SJ, Loster B et al., Oral oxidative stress lesions activate immune system that can be attenuated by topical application of exogenous melatonin or tryptophan. (in press)

R e c e i v e d : January 8, 2007
A c c e p t e d : February 2, 2007

Author’s address: Prof. Stanislaw J. Konturek, M.D., Department of Physiology Jagiellonian University Medical College, 16, Grzegorzecka St., 31-531 Krakow, Poland; phone (+4812-4211006; fax (+4812-4211578);
e-mail: mpkontur@cyf-kr.edu.pl