![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
GASTROINTESTINAL, HEPATIC, PULMONARY, AND RENAL
Division of Gastroenterology, Rhode Island Hospital and Brown University, Providence, Rhode Island (L.C., W.C., J.B., P.B., K.M.H.); and Division of Gastroenterology, Case Western Reserve University, Cleveland, Ohio (C.F.)
Received April 12, 2006; accepted June 27, 2006.
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
(but not IL-6, prostaglandin E2, or H2O2) in mucosa, and only PAF was released into the supernatant, presumably to affect circular muscle. In circular muscle, exogenous PAF induced sequential formation of IL-6, H2O2, IL-1, and PAF. Release of PAF by the mucosa inhibits ACh release from circular muscle layer neurons and initiates sequential formation of inflammatory mediators in muscle, resulting in production of PAF by the muscle itself, possibly initiating in a self-sustaining cycle.
; Sifrim and Tutuian, 2005). Although the motor disturbances associated with reflux esophagitis are well characterized, the underlying cause-and-effect relationship with esophageal inflammation is still unclear. In patients with reflux esophagitis, a considerable number of proinflammatory molecules like cytokines, chemokines, and other mediators have been detected in esophageal mucosal biopsies and the occasional surgical tissue. Examining the role of these inflammatory mediators in esophagitis-associated motor dysfunction is difficult because human specimens are rarely available, particularly when the esophagus is affected by esophagitis, a clinical condition that never justifies its surgical removal. We have recently reported on the production of H2O2 and other inflammatory mediators in a single specimen of esophagus from an organ donor with histologically demonstrated esophagitis and examined how circular smooth muscle contractility was affected (Cheng et al., 2005d, 2006). How these products are induced is less clear, however, particularly whether primary events, such as exposure of the mucosa to HCl, can directly induce them or whether their presence is secondary in response to the damage induced by contact with HCl.
In cat, we have reported (Cao et al., 2004) that experimental esophagitis significantly reduced contraction in esophageal circular smooth muscle strips in response to electrical field stimulation (EFS) but did not affect contraction induced by ACh and that several inflammatory mediators, such as IL-1, IL-6, H2O2, platelet-activating factor (PAF), and PGE2 are present in the muscle layer (Cao et al., 2004; Cheng et al., 2005c). When applied to normal cat esophageal circular muscle these mediators reproduce esophagitis-induced changes in contraction, i.e., they inhibit contraction in response to EFS (i.e., neural) stimulation by inhibiting release of ACh but do not inhibit contraction in response to direct myogenic stimulation by direct application of ACh. Some of these inflammatory mediators, when applied to the circular muscle, induce formation of others.
We developed an in vitro model of esophagitis by separating the mucosa and muscularis propria and tying the mucosa at both ends forming a sac, with the squamous epithelium on the inside and submucosa on the outside. The sac is filled with acidified Krebs' solution, and the supernatant surrounding the sac was applied to circular muscle strips or used for measurement of inflammatory mediators (Cheng et al., 2005b).
This in vitro model of esophageal inflammation has the advantage of distinguishing the sequential activation of inflammatory events in the mucosa from events occurring in the circular muscle layer. Because the model uses normal esophageal specimens, it allows us to examine normal tissue from human organ donors, a particularly attractive feature considering that human esophagitis specimens are exceedingly rare.
We have shown in the cat that the mucosal sac, when filled with HCl, releases IL-6 and PAF into the supernatant surrounding the mucosal sac. Therefore, we examined the inflammatory mediators released by the human mucosa when exposed to HCl, their effect on contraction of circular muscle, and the inflammatory mediators produced in the muscle in response to those released by the mucosa.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
The tube was tied at both ends, forming a mucosal sac, with the squamous epithelium inside. The sac was filled with Krebs' solution (control) or with HCl, pH 4. The preparation was kept immersed in warm (37°C) oxygenated Krebs' solution at pH 7.4. After 3 h of incubation, the supernatant surrounding the sac was analyzed for content of inflammatory mediators or used in the muscle bath to examine its effect on contraction of esophageal circular muscle strips.
Measurements of Contraction. Circular muscle strips devoid of mucosa (2 mm wide) were mounted in separate 1-ml muscle chambers as described previously in detail (Cao et al., 1999). They were initially stretched to 2.5 g to bring them near conditions of optimal force development and equilibrated for 2 h while perfused continuously with oxygenated physiologic salt solution at 37°C. The physiologic salt solution contained the following: 116.6 mM NaCl, 21.9 mM NaHCO3, 1.2 mM KH2PO4, 3.4 mM KCl, 2.5 mM CaCl2, 5.4 mM glucose, and 1.2 mM MgCl2. The solution was equilibrated with a gas mixture containing 95% O2 and 5% CO2 at 37°C, pH 7.4. After equilibration, esophageal strips were stimulated with EFS consisting of 10-s trains of square wave pulses of supramaximal voltage (0.2-ms duration at 0.5-5 Hz). Similarly to opossum or cat specimens, circular muscle strips responded to electric stimulation with a variable relatively small contraction occurring at the beginning of the stimulus train ("on contraction") and a large and reproducible contraction, with some latency after the end of the stimulus ("off contraction") (Lund and Christensen, 1969; Christensen, 1970). The contractions reported in this study are off contractions. The stimuli were delivered by a stimulator (model S48; Grass Instruments, West Warwick, RI) through platinum wire electrodes placed longitudinally on either side of the strip. To study the effect of supernatant from HCl-treated mucosa or of selected cytokines on EFS-induced contraction, the strips were incubated in supernatant or in appropriate concentrations of the cytokines for 3 h before contraction in response to EFS.
Isolation of Epithelial and Muscle Cells for Rhodamine Fluorescence. Our methods for enzymatic isolation of muscle cell have been described previously (Cao et al., 1999). In brief, LES and esophageal muscle specimens are digested in HEPES-buffered physiologic solution containing collagenase. At the end of the digestion period, the tissue is rinsed, then incubated in collagenase-free HEPES buffer. The cells dissociate freely in collagenase-free solution.
Epithelial cells were similarly isolated from mucosa after carefully removing connective tissue under a microscope. Esophageal mucosa was cut into small squares, and the squares were incubated over-night at 4°C, with the epithelial surface facing downwards in phosphate-buffered saline (PBS; Sigma-Aldrich, St. Louis, MO), pH 7.4, containing 0.01 M phosphate buffer, 0.0027 M KCl, and 0.137 M NaCl. In addition, 1.50 mg/ml trypsin, 10 mM glucose, 30 mM HEPES, and 3.3 µM phenol red were added to the solution. The next day, the mucosa squares were transferred to a second Petri dish containing a similar solution without trypsin and containing 5 µg/ml soybean trypsin inhibitor. The epithelial cells were separated from the tissue by scraping away the overlaying tissue with a small forceps, and the remaining cells were pipetted multiple times to dissociate large cell clumps. The cell suspension was then centrifuged at 250g for 10 min, the cell pellet was resuspended in growth medium, and the cells were kept in culture or used for experiments, such as rhodamine fluorescence.
Measurement of IL-6 and IL-1. Esophageal tissue from the mucosal or circular smooth muscle layer (100 mg) was homogenized in 2 ml of PBS (Sigma-Aldrich), pH 7.4, containing 0.01 M phosphate buffer, 0.0027 M KCl, and 0.137 M NaCl. Homogenization was achieved with three 10- to 20-s bursts with a Tissue Tearer (Biospec, Bartlesville, OK). The homogenate was centrifuged at 2000g, 4°C, for 20 min. An aliquot of homogenate was taken for protein determination. The homogenate supernatant or the mucosal sac supernatant was frozen in liquid nitrogen for later use. The concentrations of cytokines were measured using enzyme immunoassay kits from Cayman Chemical (Ann Arbor, MI) for IL-6 and from R&D Systems (Minneapolis, MN) for IL-1.
Western Blot. Esophageal circular muscle (100 mg) was homogenized in 2 ml of PBS (Sigma-Aldrich), pH 7.4, containing 0.01 M phosphate buffer, 0.0027 M KCl, and 0.137 M NaCl. The suspension was centrifuged at 12,000g for 20 min. The supernatant was frozen in liquid nitrogen for later use. The supernatant was mixed with SDS loading buffer (Sigma-Aldrich) containing 0.125 M Tris HCl, pH 6.8, 20% glycerol, 10% 2-mercaptoethanol, 0.04% bromphenol blue, and 10% glycerol and boiled for 5 min. A prestained molecular weight marker was prepared in the same manner. After boiling, these supernatant samples were subjected to SDS-polyacrylamide gel electrophoresis (90 V, 2 h) using 15% SDS gel. The separated proteins were electrophoretically transferred to a nitrocellulose membrane (Bio-Rad, Hercules, CA) at 100 V for 1 h. Transfer of proteins to the nitrocellulose membrane was confirmed with Ponceau S staining reagent (Sigma-Aldrich). To block nonspecific binding, the nitrocellulose membrane was incubated in 5% donkey serum in phosphate-buffered saline for 2 h followed by three rinses in serum-free buffer. Samples were incubated with anti IL-1 (1:500, 2 h) (R&D Systems) or IL-6 (1:1000, overnight) (R&D Systems) with shaking followed by three washes with antibody-free phosphate-buffered saline with 0.5% Tween 20. This was followed by 60-min incubation in peroxidase-conjugated donkey anti-goat IgG antibody (1:5000) (Jackson Immuno Research Laboratories Inc., West Grove, PA). Detection was achieved with an enhanced chemiluminescence agent (GE Healthcare, Little Chalfont, Buckinghamshire, UK). Molecular weight was estimated by comparison of sample bands with prestained molecular weight marker (GE Healthcare).
Measurement of PAF. PAF was extracted from tissues by a modification of the method of Bligh and Dyer (1959). Esophageal tissue (100 mg from the circular muscle or mucosal layer) was homogenized in 2 ml of methanol. One milliliter of homogenate was transferred to a tube containing 0.5 ml of chloroform, 1 ml of methanol, and 0.4 ml of H2O for a final ratio of 1:2:0.8 chloroform/methanol/H2O (v/v). The mixture was vortexed, left at room temperature for 1 h, then centrifuged (5000g for 5 min). The supernatant was transferred to another glass tube, and the pellet was re-extracted with 3.8 ml of the chloroform/methanol/H2O solution. The mixture was centrifuged again, and the two supernatants were combined. Two milliliters of chloroform and 2 ml of 1 M NaCl were added, and the phases were separated by centrifugation (5000g, 5 min). The upper phase was aspirated and discarded, and the lower phase was washed once with 4 ml of 1 M NaCl/methanol [9:1 (v/v)]. Samples of this washed chloroform phase were dried under nitrogen and stored at -20°C. Measurement of PAF was performed within 72 h of extraction.
Measurements of tissue levels of PAF or of PAF in the mucosal sac supernatant were made using the [3H]PAF scintillation proximity assay system (TRK 990; GE Healthcare). Scintillation proximity assay is a sensitive assay system that uses microscopic beads containing scintillant that emit light when radiolabeled molecules of interest bind to the surface of the bead.
Measurement of H2O2. Mucosa, mucosa supernatant, or circular muscle were collected, and H2O2 content was measured by BIOXYTECH H2O2-560 Quantitative Hydrogen Peroxide Assay Kit (OXIS International, Inc., Portland, OR). This assay is based on the oxidation of ferrous ions (Fe2+) to ferric ions (Fe3+) by hydrogen peroxide under acidic conditions. The ferric ion binds with the indicator dye xylenol orange (3,3'-bis[N,N-di(carboxymethyl)-aminomethyl]-o-cresoIsulfone-phthalein sodium salt) to form a stable colored complex that can be measured at 560 nm.
ACh Release. The release of ACh from esophageal smooth muscle strips was measured using a well established technique in which ACh stores in a circular smooth muscle preparation are previously labeled with [3H]choline (Collins et al., 1989). This technique has been used extensively to examine myenteric or submucosal plexus function of several species (Szerb, 1975; Teitelbaum et al., 1984). Muscle strips were mounted in 1-ml muscle chambers as described previously (Cao et al., 1999). Mounted strips were incubated for 1 h at 37°C in Krebs' buffer containing 0.2 µM [3H]choline (80 Ci/mM; New England Nuclear, Boston, MA) and 50 µM physostigmine. The strips were washed by changing the solution every 3 min for 1 h. After 1 h, the basal tritium release approached a plateau level. After incubation in [3H]choline, the strips were washed three times with 1 ml of Krebs' solution containing 50 µM physostigmine to inhibit ACh breakdown and 10 µM hemicholinium to inhibit choline uptake. At this time, a 3-ml sample resulting from washing the 1-ml chamber three times was collected, pooled, and used to measure basal release. To measure EFS-induced ACh release, strips were stimulated with the appropriate frequency for 30 s. After 30 s, strips were washed three times with 1 ml of Krebs' solution containing 50 µM physostigmine and 10 µM hemicholinium, and the 3-ml sample was collected for radioactivity measurement. Physostigmine and other cholinesterase inhibitors may have variable effects on ACh release, either enhancing (Dieterich et al., 1976; Testylier and Dykes, 1996) or reducing it (James and Cubeddu, 1984; Feuerstein et al., 1992; D'Agostino et al., 2000), depending on the type and function of the muscarinic receptors present in the tissue. Nevertheless, a cholinesterase inhibitor is necessary to prevent the enzymatic breakdown of ACh. Strips were allowed to rest 30 min before the next stimulation. Frequencies tested included 0.5, 1, 2, and 5 Hz. Under these experimental conditions, Collins et al. (1989) reported that 90% of the radioactivity in the superfusate was [3H]ACh as measured by high-performance liquid chromatography. A linear relationship between force developed and released counts per minute indicates that counts per minute are a measure of ACh release in our preparation (Cheng et al., 2005c).
Protein Determination. The homogenates of esophageal tissues were solubilized by addition of 6 ml of 0.1 N NaOH and heating the sample at 80°C for 30 min. The amount of protein present was determined by colorimetric analysis (Bio-Rad) according to the method of Bradford (1976).
Materials and Reagents. IL-6 and IL-1, were purchased from Pierce Endogen (Rockford, IL). Apocynin and PAF-16 were purchased from Cayman Chemical. CV3988 was purchased from Biomol (Plymouth Meeting, PA). All other reagents were purchased from Sigma-Aldrich.
Statistical Analysis. Data are expressed as mean ± S.E.M. Statistical differences between means were determined by Student's t test. Differences between multiple groups were tested using analysis of variance (ANOVA) for repeated measures and checked for significance using Scheffé's F test.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
, 2006), where both IL-6 and PAF are released by the mucosa in response to HCl. In the current investigation, however, supernatant-induced inhibition was reversed when the competitive PAF receptor antagonist CV3988 was added to the supernatant of the HCl-filled mucosal sac, but IL-6 antibodies had no effect on supernatant-induced inhibition. These data suggest that exposure of the mucosa to HCl causes production and release of PAF and not IL-6 from the mucosal layer. This result is confirmed by Fig. 2A, showing that the PAF analog PAF-16 causes complete inhibition of EFS-induced contraction of circular muscle strips. The figure shows frequency-response curves for control and PAF-16-treated circular muscle strips. For control specimens, the amplitude of contraction in response to electrical (i.e., neural) stimulation was greatest at 5 Hz. EFS-induced contraction was significantly reduced when muscle strips were incubated with PAF-16 (10-5 M) for 2 h. Esophageal muscle strips have little or no tone. Adding PAF-16 did not cause any contraction, and relaxation could not occur because resting force could not get any lower. However, when electrical stimulation was applied, PAF-16 abolished EFS-induced contraction, suggesting that PAF may affect either myogenic contractile mechanisms or, most likely, the release of contractile neurotransmitters as previously shown in cat esophagus (Cheng et al., 2005c).
|
|
We have reported previously that cat esophageal circular muscle contraction in response to EFS is largely mediated by ACh because the contraction is almost abolished by atropine (Behar et al., 1989). To confirm that ACh release in response to neural stimulation was damaged by PAF, we examined ACh release in response to EFS in control and PAF-16-treated muscle strips preincubated with [3H]choline (Fig. 2B). In control strips, [3H]ACh release, measured in the supernatant, increased with frequency of EFS compared with levels in the absence of EFS (basal). PAF-16 caused a significant reduction in ACh release from muscle strips at all stimulus frequencies. To demonstrate that the released radioactivity was due to diminished ACh release, we compared forces and counts per minute obtained at the same parameter of electrical field stimulation, as shown in Fig. 2C. The finding of a statistically significant (p < 0.01) correlation (r = 0.94) between contraction and released counts per minute supports the likelihood that the counts per minute measured may be associated with ACh release.
Figure 3 shows EFS-induced contraction (0.2 ms, 5 Hz, 10 s) for control and supernatant-treated circular muscle strips. Muscle strips were treated for 2 h with mucosal supernatant from the Krebs-filled sac (control) or with mucosal supernatant from the HCl-filled sac. EFS-induced contraction was decreased by supernatant from HCl-treated mucosa. The decrease was equally reduced by the H2O2 scavenger catalase, by the PAF antagonist CV3988, or by a combination of the two, although the effect of catalase and CV3988 followed a different time course. The data suggest that the reduction in EFS-induced contraction is due to PAF and H2O2 acting through the same contractile pathway, possibly because PAF may induce production of H2O2 or H2O2 may induce production of PAF by circular muscle.
|
Inflammatory Mediators in Mucosa and Mucosa Supernatant
PAF. To confirm these initial conclusions derived from contractile data, we measured inflammatory mediators produced in mucosa, in response to HCl, and in muscle in response to HCl-treated mucosa supernatant. When the mucosal sac was filled with HCl-acidified Krebs' solution, pH 4 (3 h), the levels of PAF increased in mucosa tissue and in the supernatant surrounding the mucosal sac (Fig. 4A). Figure 4B shows that IL-6 levels in mucosa tissue or supernatant were not increased by exposure to HCl. These findings are consistent with the results illustrated in Fig. 1 and support the conclusion that the presence of HCl in the lumen of the esophageal mucosal sac induces production of PAF and not IL-6 by the mucosa and that PAF is released into the surrounding supernatant, presumably to affect the circular muscle.
|
. We have shown previously in cat that IL-1 levels increase in mucosa after exposure to HCl (Cheng et al., 2005b). Therefore, we examined IL-1 levels in mucosa tissue and mucosal supernatant of the HCl-filled mucosa sac (Fig. 4C). When the mucosal sac was filled with acidified Krebs' solution, pH 4 (3 h), the levels of IL-1 increased in mucosa tissue but not in the supernatant surrounding the mucosal sac, suggesting that, similar to the cat (Cheng et al., 2005b), IL-1 is produced in the mucosa layer after exposure to HCl, but it is not released.
H2O2. As shown in Fig. 5, A and B, there was no increase in H2O2 produced in the mucosa when the mucosal sac was filled with acidified Krebs' solution. This finding was confirmed by using dihydrorodamine (DHR 123) as a fluorescent indicator for measurement of intracellular H2O2 and other reactive oxygen species (ROS). DHR 123 enters the cells as a freely permeable dye that is oxidized directly to the nonpermeable rhodamine 123, which is excitable at 488 and detectable at 515 nm. The conversion from nonfluorescent to fluorescent molecule depends on oxidation. Figure 5B demonstrates diydrorodamine imaging of enzymatically isolated epithelial cells from esophageal mucosa. In these cells, there are no detectable levels of rhodamine 123 fluorescence, and the levels do not increase in response to HCl or PAF-16. When mucosa epithelial cells are directly exposed to H2O2, however, the fluorescence increased because H2O2 is membrane permeable and flows into the cells, confirming the validity of this method for detecting ROS.
|
Inflammatory Mediators in Circular Muscle
PAF-Induced H2O2. Because PAF was produced by the mucosal sac in response to HCl and released into the supernatant, presumably to affect circular muscle, we examined the effect of PAF on formation of H2O2 in the circular muscle. Figure 6A shows PAF-induced formation of H2O2 in esophageal circular muscle by activation of NADPH oxidase because the production of H2O2 was inhibited by the NADPH oxidase inhibitor apocynin. The panel on the top right hand of the figure illustrates the pathway suggested by these findings. The pathway will be modified in the following figures so as to remain consistent with the present data and those shown in successive experiments. Production of H2O2 by esophageal circular muscle was confirmed with dihydrorodamine imaging of enzymatically isolated esophageal circular muscle cells, as shown in Fig. 6B. The figure shows that in esophageal smooth muscle cells, no fluorescence is present in control conditions (control) or in response to supernatant of mucosal sac filled with normal buffer (buffer supernatant). Fluorescence, however, increases in response to supernatant of acidified Krebs-filled mucosal sac (HCl supernatant), and in response to PAF-16, indicating production of ROS in the muscle.
|
are present in cat esophageal circular muscle after induction of in vivo esophagitis (Cao et al., 2004) and affect circular muscle function. Therefore, we examined whether IL-6 or IL-1 may be present in the circular muscle after exposure to supernatant of HCl-treated mucosa, even though neither cytokine is released by the mucosa into the supernatant. In addition, we examined whether these cytokines are produced in response to PAF that is released by the mucosa, presumably to affect circular muscle function. Figure 7A demonstrates Western blot analysis of IL-6. IL-6 increased in the muscle after exposure to PAF-16. Thus, IL-6 is produced in the muscle itself and not released by the mucosa. These data were confirmed by measurements of esophageal smooth muscle IL-6 by enzyme immunoassay (Fig. 7B). The figure indicates that PAF increases production of IL-6 in circular muscle. Production of IL-6 was not affected by the H2O2 scavenger catalase, indicating that formation of IL-6 precedes formation of H2O2.
|
|
. Figure 9A demonstrates Western blot analysis of IL-1. Similarly to IL-6, IL-1 increased in the muscle after exposure to PAF-16. These findings were confirmed by enzyme immunoassay (Fig. 9B), indicating that H2O2 and PAF caused an increase in IL-1 levels. The PAF-induced increase was equally reduced by apocynin and by IL-6 antibodies. These data are consistent with Figures 7, 8, 9, indicating a sequential pathway, involving PAF-induced production of IL-6, followed by activation of NADPH oxidase and production of H2O2 that finally results in production of IL-1. Because apocynin and IL-6 antibody only partially inhibited PAF-induced IL-1, it is possible that other pathways may be involved in IL-1 production.
|
IL-6-Induced PAF. Figure 10 indicates that IL-6 may induce production of PAF through production of H2O2 because IL-6-induced production of PAF was abolished by the NADPH inhibitor apocynin. H2O2, in turn, induced production of PAF, in part through IL-1, because H2O2-induced production of PAF was partly reduced by immunoneutralization of IL-1. The partial reduction in formation of PAF by immunoneutralization of IL-1 suggests the presence of an IL-1-independent pathway, possibly mediating direct formation of PAF in response to H2O2. Figure 11 summarizes possible pathways that are consistent with these data.
|
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
), but examining the role of these inflammatory mediators in esophagitis-associated motor dysfunction is difficult because human esophageal circular muscle specimens are rarely available. We recently reported the production of H2O2 and other mediators in a single specimen of esophagus with histologically determined severe esophagitis and their effect on circular smooth muscle contractility (Cheng et al., 2005d).
We now show that HCl has no direct effect on muscle contraction but causes production of PAF by esophageal mucosa to affect the circular muscle layer. In the muscle layer, PAF has a dual effect; it inhibits circular muscle contraction by inhibiting release of ACh from intramural neurons without affecting the muscle response to ACh (Cheng et al., 2005c). In addition, PAF induces formation of additional inflammatory mediators by muscle cells. Thus, even though the intrinsic contractility of circular muscle cells is not affected, the circular muscle may respond to mucosal inflammation by releasing inflammatory mediators, possibly augmenting the inflammatory process.
In a cat model of in vivo-induced esophagitis by acid perfusion on 3 consecutive days, the proinflammatory cytokines IL-1 and IL-6 (but not tumor necrosis factor 
) are elevated in the muscle layer (Cao et al., 2004) and contribute to formation of H2O2, PAF, and PGE2 by the circular muscle (Cheng et al., 2005c). These inflammatory mediators inhibit contraction of esophageal circular muscle in response to EFS by inhibiting EFS-induced ACh release (Cheng et al., 2005c).
In humans, the presence of H2O2, PAF, and PGE2 has been confirmed in the lower esophageal sphincter circular muscle from an organ donor with histologically confirmed erosive esophagitis (Cheng et al., 2005d). Human esophagitis specimens, however, are exceedingly rare. Therefore, we developed an in vitro model of esophageal inflammation (Cheng et al., 2005b). Because the model uses normal esophageal specimens, it allows us to examine tissue from human organ donors that are sometimes available.
The present study demonstrates that exposure of human esophageal mucosa to HCl causes increased production of IL-1 and PAF in mucosa. In humans, no production of IL-6 occurs in the mucosa in response to HCl, and only PAF is released into supernatant, suggesting that PAF may be a major inflammatory mediator responsible for inducing inflammatory changes in mucosa and muscle in response to mucosal injury. The PAF-containing mucosal supernatant decreases EFS-induced contraction, most likely by inhibiting EFS-induced release of ACh from intramural neurons, similar to the cat (Cheng et al., 2005c). As expected, the reduction in contraction was reversed by PAF antagonist and not by IL-6 antibodies. In addition, the reduction in response to EFS was reversed by the H2O2 scavenger catalase, indicating a contribution of H2O2 to reducing contraction.
Because H2O2 is neither produced nor released by the mucosa in response to acid, it must be produced by the muscle in response to PAF. In addition, even though IL-1 is produced in the mucosa in response to HCl, it is not released in the supernatant, and IL-6 is neither produced nor released. Because both cytokines are found in circular muscle of cats with in vivo-induced esophagitis, it is likely that these inflammatory mediators are produced in the muscle layer in response to PAF.
Therefore, we investigated how production of these inflammatory mediators may occur. The data suggest that application of PAF to the circular muscle layer induced production of IL-6 that, in turn, induced production of H2O2 by activation of NADPH oxidase present in human circular muscle. PAF-induced production of IL-6 is consistent with other reports in the literature (Hostettler and Carlson, 2002; Ichinowatari et al., 2002), even though the precise pathway for PAF-induced production of IL-6 has not been satisfactorily elucidated.
IL-6, in turn, induced production of H2O2 by activating a phagocytic-like NADPH oxidase. Mechanisms for H2O2-dependent formation of IL-6 have been demonstrated in other experimental preparations (Haddad, 2002b; Yu et al., 2005). IL-6-induced production of H2O2 is less common and has been only recently suggested by work from our laboratory in cat esophageal circular muscle (Cheng et al., 2005c) and by others in vascular smooth muscle (Wassmann et al., 2004). In vascular smooth muscle, however, production of H2O2 is critically dependent on activation of angiotensin receptors because IL-6 alone did not cause a significant increase in H2O2. In the present investigation, however, we clearly demonstrate that IL-6 may induce production of H2O2 by activating NADPH oxidase. The mechanisms involved in this IL-6-induced activation of NADPH oxidase have not been previously explored.
Interleukin-6 signaling is mediated by a cell surface signaling assembly composed of IL-6, the IL-6-receptor, and the signal transducing receptor chain glycoprotein 130. IL-6 is first engaged by IL-6R and then presented to glycoprotein 130 in the proper geometry to facilitate a cooperative transition into a high-affinity signaling complex (Heinrich et al., 2003). Glycoprotein 130 is a shared cell surface signaling receptor for at least 10 different hematopoietic cytokines (Boulanger et al., 2003). IL-6-type cytokines, via the signal transducers glycoprotein 130, activate the Janus tyrosine kinase/signal transducer and activator of transcription and mitogen-activated protein (MAP) kinase cascades (Heinrich et al., 2003). It is likely that the MAP kinase pathway in turn activates cytosolic phospholipase A2, producing arachidonic acid. The link between MAP kinases and cytosolic phospholipase A2 is consistent with recent data on acetylcholine-induced contraction of rabbit intestinal smooth muscle (Zhou et al., 2003) and in agreement with data from several laboratories on other cell species (Zhou et al., 2003; Wu et al., 2004), including the lower esophageal sphincter (Cao et al., 2000). Arachidonic acid, in turn, promotes activation of NADPH oxidase by binding to the myeloid-related proteins S100A8/A9 (Bouzidi and Doussiere, 2004; Kerkhoff et al., 2005). These proteins have recently been shown to interact with the p47-p67 NADPH components promoting interactions between the different oxidase subunits and enabling full oxidase activation or directly affecting the function of flavocytochrome b (Foubert et al., 2002). Thus, it is possible that IL-6, produced in response to PAF, may induce production of arachidonic acid, resulting in activation of NADPH oxidase and production of H2O2. H2O2, in turn, causes production of IL-1. IL-1 is an additional proinflammatory cytokine present in esophageal circular muscle of animals with in vivo-induced esophagitis and in part responsible for reduced contraction in response to neural stimulation (Cao et al., 2004; Cheng et al., 2005c).
IL-1-induced formation of H2O2 has been reported in several experimental preparations (Brigelius-Flohe et al., 2004; Hwang et al., 2004), including cat esophageal and LES circular muscle (Cheng et al., 2005a,c). Conversely, it has been shown previously that H2O2 may induce the release of IL-1 by activating nuclear factor 
B (Lindstrom et al., 2001; Haddad, 2002b), perhaps through tyrosine phosphorylation of IB and serine phosphorylation of p65 (Takada et al., 2003), possibly inducing production of cytokines, with preference for IL-1 (Haddad, 2002a). In any case, we have clearly demonstrated H2O2-induced formation of IL-1 in human esophageal circular muscle. IL-1 production may occur either directly from H2O2 or through PAF-induced production of IL-6 and IL-6-induced activation of NADPH oxidase. H2O2, in turn, may induce production of PAF, at least in part through formation of IL-1, because H2O2-induced production of PAF is reduced by immunoneutralization of IL-1. It is worth noting that although apocynin abolishes formation of PAF, immunoneutralization of IL-1 only partially reduces H2O2-induced production of PAF, indicating an absolute requirement of H2O2 and only a partial contribution of IL-1 to production of PAF.
Therefore, the data suggest that, in humans, HCl-induced inflammation begins with mucosa producing PAF and IL-1. IL-1 remains in the mucosa, and only PAF is released, presumably to affect the circular smooth muscle layer by inhibiting release of ACh from intramural neurons. We have shown in the cat that muscle contractile mechanisms are not directly affected by PAF because the response to direct muscle stimulation by ACh remains unchanged (Cheng et al., 2005c). In the circular muscle, however, PAF released by the mucosa induces sequential production of IL-6, H2O2, and IL-1 that in turn causes production of PAF by the muscle itself closing a circle and perhaps causing inflammation to spiral into a self-aggravating cycle.
These data are consistent with the model illustrated in Fig. 11. Taken together, the data suggest that PAF, released by the mucosa, induces sequential production of IL-6, H2O2, IL-1, and PAF in the circular muscle. Once these inflammatory mediators are present, any one of them may contribute to sequential formation of the others, possibly initiating a self-sustaining cycle.
|
Footnotes |
|---|
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
ABBREVIATIONS: LES, lower esophageal sphincter; H2O2, hydrogen peroxide; EFS, electric field stimulation; ACh, acetylcholine; IL, interleukin; PAF, platelet-activating factor; PG, prostaglandin; PBS, phosphate-buffered saline; CV3988, (±)-3-(N-octadecylcarbamoyl)-2-methoxy) propyl-(2-thiazolioethyl) phosphate; ANOVA, analysis of variance; PAF-16, 2-O-methyl platelet-activating factor C-16; DHR 123, dihydrorodamine; ROS, reactive oxygen species; MAP, mitogen-activated protein.
Address correspondence to: Dr. Karen M. Harnett, Rhode Island Hospital Gastrointestinal Motor Function Research Laboratory, 55 Claverick Street, Room 333, Providence, RI 02903. E-mail: karen_harnett{at}brown.edu
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Behar J, Guenard V, Walsh JH, and Biancani P (1989) VIP and acetylcholine: neurotransmitters in esophageal circular smooth muscle. Am J Physiol 257: G380-G385.[Medline]
Bligh EG and Dyer WJ (1959) A rapid method of total lipid extraction and purification. Can J Biochem Physiol 37: 911-917.[Medline]
Boulanger MJ, Bankovich AJ, Kortemme T, Baker D, and Garcia KC (2003) Convergent mechanisms for recognition of divergent cytokines by the shared signaling receptor gp130. Mol Cell 12: 577-589.[CrossRef][Medline]
Bouzidi F and Doussiere J (2004) Binding of arachidonic acid to myeloid-related proteins (S100A8/A9) enhances phagocytic NADPH oxidase activation. Biochem Biophys Res Commun 325: 1060-1065.[CrossRef][Medline]
Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Ann Biochem 72: 248-254.[CrossRef]
Brigelius-Flohe R, Banning A, Kny M, and Bol GF (2004) Redox events in interleukin-1 signaling. Arch Biochem Biophys 423: 66-73.[CrossRef][Medline]
Cao W, Cheng L, Behar J, Fiocchi C, Biancani P, and Harnett KM (2004) Proinflammatory cytokines alter/reduce esophageal circular muscle contraction in experimental cat esophagitis. Am J Physiol 287: G1131-G1139.
Cao W, Cheng L, Harnett K, Fiocchi C, Behar J, and Biancani P (2000) Interleukin-1b induced reduction of LES tone. Gastroenterology 118: A709.
Cao WB, Harnett KM, Chen Q, Jain MK, Behar J, and Biancani P (1999) Group I secreted PLA2 (sPLA2) and arachidonic acid metabolites in the maintenance of cat LES tone. Am J Physiol 277: G585-G598.[Medline]
Cheng L, Cao W, Behar J, Biancani P, and Harnett KM (2005a) Inflammation induced changes in arachidonic acid metabolism in cat LES circular muscle. Am J Physiol 288: G787-G797.
Cheng L, Cao W, Fiocchi C, Behar J, Biancani P, and Harnett KM (2005b) In vitro model of acute esophagitis in the cat. Am J Physiol 289: G860-G869.
Cheng L, Cao W, Fiocchi C, Behar J, Biancani P, and Harnett KM (2005c) Platelet-activating factor and prostaglandin E2 impair esophageal ACh release in experimental esophagitis. Am J Physiol 289: G418-G428.
Cheng L, Cao W, Fiocchi C, Behar J, Biancani P, and Harnett KM (2006) HCl-induced inflammatory mediators in cat esophageal mucosa and inflammatory mediators in esophageal circular muscle in an in vitro model of esophagitis. Am J Physiol 290: G1307-G1317.
Cheng L, Harnett KM, Cao W, Liu F, Behar J, Fiocchi C, and Biancani P (2005d) Hydrogen peroxide reduces lower esophageal sphincter tone in human esophagitis. Gastroenterology 129: 1675-1685.[CrossRef][Medline]
Christensen J (1970) Patterns and origin of some esophageal responses to stretch and electrical stimulation. Gastroenterology 59: 909-916.[Medline]
Collins S, Blennerhassett P, Blennerhassett M, and Vermillion D (1989) Impaired acetylcholine release from the myenteric plexus of Trichinella-infected rats. Am J Physiol 257: G898-G903.[Medline]
D'Agostino G, Bolognesi ML, Lucchelli A, Vicini D, Balestra B, Spelta V, Melchiorre C, and Tonini M (2000) Prejunctional muscarinic inhibitory control of acetylcholine release in the human isolated detrusor: involvement of the M4 receptor subtype. Br J Pharmacol 129: 493-500.[CrossRef][Medline]
Dieterich HA, Kaffel H, Kilbinger H, and Loffelholz K (1976) The effects of physostigmine on cholinesterase activity, storage and release of acetylcholine in the isolated chicken heart. J Pharmacol Exp Ther 199: 236-246.
Feuerstein TJ, Lehmann J, Sauermann W, van Velthoven V, and Jackisch R (1992) The autoinhibitory feedback control of acetylcholine release in human neocortex tissue. Brain Res 572: 64-71.[CrossRef][Medline]
Fiocchi C (1998) Inflammatory bowel disease: etiology and pathogenesis. Gastroenterology 115: 182-205.[CrossRef][Medline]
Foubert TR, Burritt JB, Taylor RM, and Jesaitis AJ (2002) Structural changes are induced in human neutrophil cytochrome b by NADPH oxidase activators, LDS, SDS, and arachidonate: intermolecular resonance energy transfer between trisulfopyrenyl-wheat germ agglutinin and cytochrome b(558). Biochim Biophys Acta 1567: 221-231.[Medline]
Haddad JJ (2002a) Recombinant human interleukin (IL)-1 beta-mediated regulation of hypoxia-inducible factor-1 alpha (HIF-1 alpha) stabilization, nuclear translocation and activation requires an antioxidant/reactive oxygen species (ROS)-sensitive mechanism. Eur Cytokine Netw 13: 250-260.[Medline]
Haddad JJ (2002b) Redox regulation of pro-inflammatory cytokines and IkappaB-alpha/NF-kappaB nuclear translocation and activation. Biochem Biophys Res Commun 296: 847-856.[CrossRef][Medline]
Heinrich PC, Behrmann I, Haan S, Hermanns HM, Muller-Newen G, and Schaper F (2003) Principles of interleukin (IL)-6-type cytokine signalling and its regulation. Biochem J 374: 1-20.[CrossRef][Medline]
Hostettler ME and Carlson SL (2002) PAF antagonist treatment reduces pro-inflammatory cytokine mRNA after spinal cord injury. Neuroreport 13: 21-24.[CrossRef][Medline]
Hwang YS, Jeong M, Park JS, Kim MH, Lee DB, Shin BA, Mukaida N, Ellis LM, Kim HR, Ahn BW, et al. (2004) Interleukin-1beta stimulates IL-8 expression through MAP kinase and ROS signaling in human gastric carcinoma cells. Oncogene 23: 6603-6611.[CrossRef][Medline]
Ichinowatari G, Yamada M, Yaginuma H, Tsuyuki K, Tanimoto A, and Ohuchi K (2002) Participation of prostaglandin E2 and platelet-activating factor in thapsigargin-induced production of interleukin-6. Eur J Pharmacol 434: 187-196.[CrossRef][Medline]
James MK and Cubeddu LX (1984) Frequency-dependent muscarinic receptor modulation of acetylcholine and dopamine release from rabbit striatum. J Pharmacol Exp Ther 229: 98-104.
Kerkhoff C, Nacken W, Benedyk M, Dagher MC, Sopalla C, and Doussiere J (2005) The arachidonic acid-binding protein S100A8/A9 promotes NADPH oxidase activation by interaction with p67phox and Rac-2. FASEB J 19: 467-469.
Lindstrom E, Lerner UH, and Hakanson R (2001) Isolated rat stomach ECL cells generate prostaglandin E(2) in response to interleukin-1 beta, tumor necrosis factor-alpha and bradykinin. Eur J Pharmacol 416: 255-263.[CrossRef][Medline]
Lund GF and Christensen J (1969) Electrical stimulation of esophageal smooth muscle and effects of antagonists. Am J Physiol 217: 1369-1374.
Penagini R, Carmagnola S, and Cantu P (2002) Review article: gastro-oesophageal reflux disease-pathophysiological issues of clinical relevance. Aliment Pharmacol Ther 16: 65-71.[Medline]
Sifrim D and Tutuian R (2005) Oesophageal intraluminal impedance can identify subtle bolus transit abnormalities in patients with mild oesophagitis. Eur J Gastroenterol Hepatol 17: 303-305.[CrossRef][Medline]
Szerb JC (1975) Endogenous acetylcholine release and labelled acetylcholine formation from [3H]choline in the myenteric plexus of the guinea-pig ileum. Can J Physiol Pharmacol 53: 566-574.[Medline]
Takada Y, Mukhopadhyay A, Kundu GC, Mahabeleshwar GH, Singh S, and Aggarwal BB (2003) Hydrogen peroxide activates NF-B through tyrosine phosphorylation of IB and serine phosphorylation of p65: evidence for the involvement of IB alpha kinase and Syk protein-tyrosine kinase. J Biol Chem 278: 24233-24241.
Teitelbaum DH, O'Dorisio TM, Perkins WE, and Gaginella TS (1984) Somatostatin modulation of peptide-induced acetylcholine release in guinea pig ileum. Am J Physiol 246: G509-G514.[Medline]
Testylier G and Dykes RW (1996) Acetylcholine release from frontal cortex in the waking rat measured by microdialysis without acetylcholinesterase inhibitors: effects of diisopropylfluorophosphate. Brain Res 740: 307-315.[CrossRef][Medline]
Wassmann S, Stumpf M, Strehlow K, Schmid A, Schieffer B, Bohm M, and Nickenig G (2004) Interleukin-6 induces oxidative stress and endothelial dysfunction by overexpression of the angiotensin II type 1 receptor. Circ Res 94: 534-541.
Wu T, Han C, and Shelhamer JH (2004) Involvement of p38 and p42/44 MAP kinases and protein kinase C in the interferon-gamma and interleukin-1alpha-induced phosphorylation of 85-kDa cytosolic phospholipase A(2) in primary human bronchial epithelial cells. Cytokine 25: 11-20.[CrossRef][Medline]
Yu JH, Lim JW, Kim H, and Kim KH (2005) NADPH oxidase mediates interleukin-6 expression in cerulein-stimulated pancreatic acinar cells. Int J Biochem Cell Biol 37: 1458-1469.[CrossRef][Medline]
Zhou H, Das S, and Murthy KS (2003) Erk1/2- and p38 MAP kinase-dependent phosphorylation and activation of cPLA2 by m3 and m2 receptors. Am J Physiol 284: G472-G480.
This article has been cited by other articles:
|
|
 |
|
  J. Si, J. Behar, J. Wands, D. G. Beer, D. Lambeth, Y. Eugene Chin, and W. Cao STAT5 mediates PAF-induced NADPH oxidase NOX5-S expression in Barrett's esophageal adenocarcinoma cells Am J Physiol Gastrointest Liver Physiol, January 1, 2008; 294(1): G174 - G183. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||
