H2O2: a mediator of esophagitis-induced damage to calcium-release mechanisms in cat lower esophageal sphincter

Weibiao Cao, Karen M. Harnett, Ling Cheng, Michael T. Kirber, Jose Behar, Piero Biancani

Abstract

We previously reported that induction of acute experimental esophagitis by repeated perfusion of HCl may affect release of intracellular Ca2+ stores. We therefore measured cytosolic Ca2+ in response to a maximally effective dose of ACh in fura 2-AM-loaded lower esophageal sphincter (LES) circular muscle cells and examined the contribution of H2O2 to the reduction in Ca2+ signal. In normal cells, the ACh-induced Ca2+ increase was the same in normal-Ca2+ and Ca2+-free medium and was abolished by the phosphatidylinositol 4,5-bisphosphate-specific phospholipase C inhibitor U-73122, confirming that the initial ACh-induced contraction depends on Ca2+ release from intracellular stores through production of inositol trisphosphate. In LES cells, the ACh-induced Ca2+ increase in normal-Ca2+ medium was significantly lower in esophagitis than in normal cells and was further reduced (∼70%) when the cells were incubated in Ca2+-free medium. This reduction was partially reversed by the H2O2 scavenger catalase. H2O2 measurements in LES circular muscle showed significantly higher levels in esophagitis than in normal cells. When normal LES cells were incubated with H2O2, the ACh-induced Ca2+ increase was significantly reduced in normal-Ca2+ and Ca2+-free medium and was similar to that observed in animals with esophagitis. The initial ACh-induced contraction was also reduced in normal cells incubated with H2O2. H2O2, when applied to cells at sufficiently high concentration, produced a visible and prolonged Ca2+ signal in normal cells. H2O2-induced cell contraction was also sensitive to depletion of stores by thapsigargin (TG); conversely, H2O2 reduced TG-induced contraction, suggesting that TG and H2O2 may operate through similar mechanisms. Ca2+-ATPase activity measurement indicates that H2O2 and TG reduced Ca2+-ATPase activity, confirming similarity of mechanism of action. We conclude that H2O2 may be at least partly responsible for impairment of Ca2+ release in acute experimental esophagitis by inhibiting Ca2+ uptake and refilling Ca2+ stores.

  • acetylcholine
  • smooth muscle

we previously showed that the initial contraction of lower esophageal sphincter (LES) circular smooth muscle cells induced by a maximally effective dose of ACh depends on Ca2+ release from intracellular stores (9) and activation of calmodulin and myosin light-chain kinase (8, 40). We also reported that induction of acute experimental esophagitis by repeated perfusion of HCl reduces in vivo LES resting pressure as well as the initial phase of ACh-induced contraction of LES smooth muscle cells (6, 7). In addition, after induction of experimental esophagitis, basal inositol 3,4,5-trisphosphate (IP3) levels in LES circular muscle are markedly reduced (7) and contraction of single LES cells in response to IP3 and thapsigargin is significantly decreased, but contraction in response to calmodulin or diacylglycerol is not affected. These data suggest that mechanisms responsible for release of Ca2+ from intracellular stores are affected by esophagitis but that contractile mechanisms downstream of Ca2+ release are not. After these changes, the signal transduction pathway mediating LES contraction in response to a maximally effective dose of ACh shifts from a Ca2+-calmodulin-dependent pathway to a PKC-dependent pathway (41).

Elevated levels of reactive oxygen species (ROS) have been reported in the esophageal mucosa from patients with esophagitis (30). ROS have been shown to consistently depress the Ca2+-ATPase responsible for uptake of Ca2+ into the endoplasmic reticulum (19–21, 28, 34). In addition to inhibiting Ca2+ uptake into the endoplasmic reticulum, ROS cause Ca2+ release from intracellular stores through ryanodine- and IP3-sensitive Ca2+ channels (26).

In the present study, we measured intracellular Ca2+ in response to a maximally effective dose of ACh in fura 2-loaded LES cells and examined the contribution of H2O2 to the reduction in Ca2+ release from intracellular stores.

METHODS

Experimental procedures were approved by the Animal Welfare Committee of Rhode Island Hospital.

Tissue dissection.

Adult cats of either gender weighing 3–5 kg were euthanized, and esophageal smooth muscle squares from the circumferential muscle layer were prepared as previously described (9). The chest and abdomen were opened with a midline incision to expose the esophagus and stomach. The esophagus and stomach were removed together and pinned onto a wax block at their in vivo dimensions and orientation. The esophagus and stomach were opened along the lesser curvature. After opening the esophagus and stomach and identifying the LES, we removed the mucosa and submucosal connective tissue by sharp dissection. The LES was excised, and a 3- to 5-mm-wide strip at the junction of the LES and esophagus was discarded to avoid overlap. The longitudinal muscle layer was also removed under a microscope, and LES circular muscles were cut into small (∼1-mm-wide) strips.

Isolation of smooth muscle cells.

LES circular smooth muscle strips were isolated by enzymatic digestion in HEPES-buffered collagenase solution as described previously (12, 38). Briefly, the collagenase solution (pH 7.2) contained 0.5 mg/ml collagenase (type F; Sigma), 1 mg/ml papain, 1 mg/ml BSA, 1 mM CaCl2, 0.25 mM (ethylenedinitrilo)-tetraacetic acid disodium (EDTA), 10 mM glucose, 10 mM HEPES (sodium salt), 4 mM KCl, 125 mM NaCl, 1 mM MgCl2, and 10 mM taurine. The tissue was kept in enzyme solution at 4°C for ∼16 h, warmed to room temperature for 30 min, and incubated in a water bath at 31°C for ∼30 min. At the end of the digestion period, the tissue was poured over a 200-μm Nitex mesh (Tetko, Elmsford, NY), rinsed in a collagenase-free HEPES-buffered solution to remove any trace of collagenase, and incubated in this solution at 31°C, gassed with 100% O2. Collagenase-free HEPES-buffered solution (pH 7.4) contained 112.5 mM NaCl, 3.1 mM KCl, 2.0 mM KH2PO4, 10.8 mM glucose, 24.0 mM HEPES (sodium salt), 1.9 mM CaCl2, 0.6 mM MgCl2, 0.3 mg/ml Eagle's basal medium amino acid supplement, and 0.08 mg/ml soybean trypsin inhibitor. Gentle agitation was used to release single cells.

Agonist-induced contraction of isolated muscle cells.

Cell contraction was induced by exposure to ACh (10−13–10−9 M) for 30 s. For thapsigargin- or H2O2-induced contraction, cells were exposed to HEPES-buffered solution without (control) or with thapsigargin (3 μM) or with H2O2 (1 μM) for 15 s, 30 s, 1 min, 5 min, 10 min, or 20 min. When H2O2 and thapsigargin were used to deplete the intracellular Ca2+ stores, normal muscle cells were preincubated in HEPES-buffered solution without (control) or with H2O2 (1 μM) or thapsigargin (3 μM) for 30 min before addition of stimuli.

After exposure to ACh, thapsigargin, or H2O2, the cells were fixed in acrolein at a 1% final concentration and kept refrigerated. For cell length measurement, a drop of the cell-containing medium was placed on a glass slide, and 30 consecutive cells from each slide were observed through a phase-contrast microscope (Carl Zeiss) and a closed-circuit television camera (model WV-CD51, Panasonic, Secaucus, NJ) connected to a Macintosh computer (Apple, Cupertino, CA). An image software program (National Institutes of Health, Bethesda, MD) was used to acquire images and measure cell length. The average length of 30 cells, measured in the absence of agonists, was taken as the “control” length and compared with the length measured after addition of test agents. Shortening was defined as the percent decrease in average length after agonists compared with the control length.

Cytosolic Ca2+ measurements.

Freshly isolated cells were loaded with 1.25 μM fura 2-AM for 40 min and placed in a 5-ml chamber mounted on the stage of an inverted microscope (Carl Zeiss). The cells were allowed to settle onto a coverslip at the bottom of the chamber. The bathing solution was collagenase-free HEPES-buffered solution (normal-Ca2+ medium) or solution without CaCl2 but with 200 μM BAPTA (Ca2+-free medium). When Ca2+-free medium was used, after settling to the bottom of the chamber, the cells were rinsed twice with Ca2+-free medium before the experiments.

A pressure ejection micropipette system was used to apply ACh (1 μM) or H2O2 (5 mM) directly to the cells. Solutions in the pressure ejection micropipettes were identical to the bathing solutions, except for the addition of ACh or H2O2.

Concentration of agents was considerably higher in the micropipette than in cell suspensions. The pipette tip was very small, and it was expected that the solution ejected from the tip may be diluted several times by the buffer surrounding the cells. Thus the concentration of the agonists reaching the cells was much lower than that in the micropipette.

Ca2+ measurements were obtained using a modified dual-excitation wavelength imaging system (IonOptix, Milton, MA). The Ca2+ concentrations were measured from the ratios of fluorescence elicited by 340-nm excitation to fluorescence elicited by 380-nm excitation using standard techniques (22). Ratiometric images were masked in the region outside the borders of the cell, because low photon counts give unreliable ratios near the edges. We developed a method for generating an adaptive mask that follows the borders of the cell as Ca2+ changes and as the cell contracts. A pseudoisobestic image (i.e., an image insensitive to Ca2+ changes) was formed in computer memory from a weighted sum of the images generated by 340- and 380-nm excitation. This image was then thresholded; i.e., values below a selected level were considered to be outside the cell and assigned a value of zero. For each ratiometric image, the outline of the cell was determined, and the generated mask was applied to the ratiometric image. This method allows the simultaneous imaging of the changes in Ca2+ and cell length. Our algorithm has been incorporated into the IonOptix software. Peak Ca2+ increase was defined as the difference between the peak and the basal value.

H2O2 measurement.

LES circular smooth muscle squares (100 mg) were homogenized in PBS buffer. Homogenization consists of a 20-s burst with a tissue tearer (Biospec, Racine, WI) followed by 50 strokes with a Dounce tissue grinder (Wheaton, Melville, NJ). An aliquot of homogenate was taken for protein measurement. The homogenate was centrifuged at 15,000 rpm for 15 min at 4°C in a centrifuge (model J2-21, Beckman, Palo Alto, CA) with a fixed-angle rotor (model JA-20, Beckman), and the supernatant was collected.

H2O2 content was measured by a quantitative H2O2 assay kit (BIOXYTECH H2O2-560, OXIS International, Portland, OR). The assay is based on the oxidation of Fe2+ to Fe3+ by H2O2 under acidic conditions. Fe3+ binds with the indicator dye xylenol orange {3,3′-bis[N,N-di(carboxymethyl)-aminomethyl]-o-cresol sulfone-phthalein sodium salt} to form a stable-colored complex that can be measured at 560 nm.

Sarcoplasmic reticulum Ca2+-ATPase activity measurement.

LES circular muscle squares were treated with vehicle, 1 mM H2O2, or 3 μM thapsigargin for 45 min. Crude membranes were prepared from muscles according to Schulte et al. (35). Muscles were homogenized in a buffer containing (in mM) 10 Tris (pH 7.5), 250 sucrose, 1 EGTA, and 0.5 phenylmethylsulfonyl fluoride. The homogenate was centrifuged at 10,000 g for 20 min at 4°C, and the supernatant was withdrawn and stored on ice. The 10,000-g pellet was reextracted commencing with rehomogenization. The pooled supernatants are brought to 0.6 M KCl and incubated on ice for 1 h to solubilize any remaining myofibrillar proteins and then centrifuged at 150,000 g for 45 min at 4°C. The resulting pellet was suspended in 10 mM Tris (pH 7.5) and 0.3 M sucrose. Protein concentration was determined.

The ATPase activity of the sarcoplasmic reticulum was determined using an enzymatically coupled assay (36) that measures the oxidation of α-nicotinamide adenine dinucleotide (reduced form, NADH) at 340 nm. Briefly, the reaction mixture contains (in mM) 20 HEPES (pH 7.5), 100 KCl, 1 MgCl2, 1 EGTA, 0.5 CaCl2, 10 phosphoenolpyruvate, 0.001 Ca2+ ionophore A-23187, 0.2 NADH, 5 ATP (magnesium salt), and 18 U/ml lactate dehydrogenase and pyruvate kinase (Roche, Indianapolis, IN). The reaction was started with the addition of crude membrane protein. This represented the total or Ca2+- and Mg2+-dependent ATPase activity. The basal or Mg2+-dependent ATPase activity was determined in the absence of CaCl2. The difference between total and basal ATPase activity represented the Ca2+-dependent ATPase activity in the sample.

Drugs and chemicals.

Soybean trypsin inhibitor was obtained from Worthington Biochemicals (Freehold, NJ); fura 2-AM and BAPTA were from Molecular Probes (Eugene, OR); thapsigargin were from Calbiochem (San Diego, CA); and ACh, H2O2, collagenase type F, papain, catalase, Eagle's basal medium amino acid supplement, HEPES sodium, and other reagents were from Sigma (St. Louis, MO).

Statistical analysis.

Values are means ± SE. Statistical differences between two groups were determined by Student's t-test. Differences between multiple groups were tested using ANOVA and checked for significance using Fisher's protected least significant difference test.

RESULTS

ACh-induced Ca2+ signal in acute esophagitis.

We previously reported that the initial phase of ACh-induced contraction of normal LES cells was the same in normal-Ca2+ and Ca2+-free medium. In acute esophagitis, however, the initial phase of ACh-induced contraction was significantly reduced in Ca2+-free medium, and thapsigargin-induced contraction was also significantly reduced (33). The data suggest that the initial contraction of LES smooth muscle cells in response to ACh is mediated by release of Ca2+ from intracellular stores and that esophagitis may affect releasable intracellular Ca2+ stores (33). To confirm these data, we directly measured the ACh-induced cytsolic Ca2+ increase in fura 2-AM-loaded LES circular smooth muscle cells.

In normal muscle cells incubated in normal-Ca2+ medium, ACh (10−6 M) caused Ca2+ to increase from 93 ± 8 to 537 ± 53 nM, producing a 444.1 ± 47 nM (n = 8) increase in cytosolic Ca2+. When the muscle cells were incubated in Ca2+-free medium, ACh caused Ca2+ to increase to 423.6 ± 50.4 nM (n = 5), demonstrating no difference in ACh-induced Ca2+ increase between normal cells in normal-Ca2+ and Ca2+-free medium (Fig. 1; see Fig. 3C). The data suggest that the ACh-induced Ca2+ increase is almost exclusively due to Ca2+ release from intracellular Ca2+ stores. The phosphatidylinositol 4,5-bisphosphate (PIP2)-specific phospholipase C inhibitor U-73122 (10−6 M) abolished the ACh-induced Ca2+ spike (Fig. 2), confirming that Ca2+ was released from IP3-sensitive stores.

Fig. 1.

Freshly isolated cells were loaded with fura 2-AM (1.25 μM) for 40 min. Bathing solution was collagenase-free HEPES buffered solution (normal-Ca2+ medium) or HEPES-buffered solution without CaCl2, but with 200 μM BAPTA (Ca2+-free medium). In normal muscle cells incubated in normal-Ca2+ medium, ACh (10−6 M) caused a rapid increase in cytosolic Ca2+. When muscle cells were incubated in Ca2+-free medium, ACh caused a similar increase in Ca2+, demonstrating no difference in ACh-induced Ca2+ increase between normal cells in normal-Ca2+ and Ca2+-free medium. Data suggest that ACh-induced Ca2+ increase is almost exclusively due to Ca2+ release from intracellular Ca2+ stores. Values are means ± SE; n = 8 (physiological salt solution) and 5 (Ca2+-free salt solution).

Fig. 2.

Data for normal muscle cells in normal-Ca2+ medium (control) from Fig. 1 demonstrate the effect of the phosphatidlyinositol 4,5-bisphosphate (PIP2)-specific phospholipase C inhibitor U-73122. U-73122 (10−6 M) abolished the ACh-induced increase in cytosolic Ca2+, confirming that Ca2+ was released from inositol trisphosphate (IP3)-sensitive stores. Values are means ± SE; n = 8 (control) and 6 (U-73122).

In esophagitis muscle cells in normal-Ca2+ medium, ACh caused Ca2+ to increase to 243.3 ± 27.4 nM (n = 7; Fig. 3, A and C) compared with 444.1 ± 47 nM for normal LES cells in normal-Ca2+ medium. Thus the ACh-induced Ca2+ increase in normal-Ca2+ medium was significantly lower in esophagitis than in normal cells, suggesting that Ca2+ release from intracellular Ca2+ stores may be impaired in esophagitis. In esophagitis cells in Ca2+-free medium, the Ca2+ peak was significantly lower (103.6 ± 50.8 nM) than in esophagitis cells in normal-Ca2+ medium, confirming impairment of Ca2+ release (Fig. 3, B and C).

Fig. 3.

A: data for normal muscle cells in normal-Ca2+ medium (normal) from Fig. 1 demonstrate the effect of esophagitis on the ACh-induced Ca2+ signal. ACh-induced Ca2+ increase in normal-Ca2+ medium was lower in esophagitis cells than in normal cells, suggesting that Ca2+ release from intracellular Ca2+ stores may be impaired in esophagitis. Values are means ± SE; n = 8 (normal) and 7 (esophagitis). B: data for esophagitis muscle cells in normal-Ca2+ medium (normal Ca2+) in A demonstrate the role of extracellular Ca2+ influx in ACh-induced Ca2+ signal in esophagitis cells. In esophagitis cells, peak Ca2+ is lower in Ca2+-free than in normal-Ca2+ medium, suggesting that part of the cytosolic Ca2+ increase in normal-Ca2+ medium may depend on influx of extracellular Ca2+. Values are means ± SE; n = 7 (normal Ca2+) and 9 (Ca2+-free). C: peak ACh-induced Ca2+ increases (peak level − initial level) for normal and esophagitis cells in normal physiological salt solution and Ca2+-free salt solution. *Statistically significant differences between normal and esophagitis cells in normal-Ca2+ medium (P < 0.05, ANOVA) and between esophagitis cells in normal-Ca2+ and Ca2+-free medium (P < 0.05, ANOVA).

The difference in Ca2+ signaling between normal and Ca2+-free medium in esophagitis cells suggests that part of the cytosolic Ca2+ increase in normal-Ca2+ medium may depend on influx of extracellular Ca2+, which does not occur in cells incubated in Ca2+-free medium.

The peak ACh-induced Ca2+ increases in normal and esophagitis cells in normal-Ca2+ and Ca2+-free medium are summarized in Fig. 3C, which shows statistically significant differences between normal and esophagitis cells in normal-Ca2+ medium and between esophagitis cells in normal-Ca2+ and Ca2+-free medium.

H2O2 and reduced Ca2+ increase in esophagitis.

H2O2 levels in LES circular muscle were significantly elevated in esophagitis compared with normal muscle: 3.6 ± 0.7 vs. 1.27 ± 0.08 nmol/mg protein (P < 0.05). To test whether the elevated H2O2 levels contribute to esophagitis-associated impairment of Ca2+ release from intracellular stores, esophagitis LES muscle cells were incubated with catalase (78 U/ml) for 50 min. After catalase treatment, the ACh-induced Ca2+ signal in Ca2+-free medium (i.e., in the absence of Ca2+ influx) was significantly enhanced in esophagitis muscle cells (Fig. 4) and close to that in normal cells, suggesting that H2O2 may contribute to impairment of Ca2+ release from intracellular stores.

Fig. 4.

Peak ACh-induced Ca2+ increases in normal and esophagitis cells were obtained in Ca2+-free salt solution to eliminate any influx of extracellular Ca2+. Data therefore represent only Ca2+ released from intracellular stores. Peak Ca2+ for normal and esophagitis cells in Fig. 3C allow comparison with esophagitis cells treated with catalase. In esophagitis cells incubated with catalase (78 U/ml), the ACh-induced Ca2+ signal was partially restored and was significantly greater than in untreated esophagitis cells, suggesting that H2O2 may be partly responsible for the decreased Ca2+ release. Values are means ± SE; n, number of cells. *P < 0.01 (ANOVA).

H2O2 and intracellular Ca2+ stores in normal LES circular smooth muscle.

To test whether prolonged exposure of normal cells to a relatively low concentration of H2O2 may result in depletion of intracellular Ca2+ stores, we examined the ACh-induced Ca2+ signal in fura 2-AM-loaded normal cells incubated with 70 μM H2O2 for 30 min. Exposure to H2O2 significantly reduced the ACh-induced Ca2+ signal in normal cells (Fig. 5A). In normal cells in Ca2+-free medium with 200 μM BAPTA (to ensure that the Ca2+ signal was produced entirely by Ca2+ release from intracellular stores), the ACh-induced Ca2+ increase was further reduced (Fig. 5B), indicating that exposure to H2O2 reproduces the changes in Ca2+ signaling associated with esophagitis. Similar to the changes in Ca2+ signaling, ACh-induced cell shortening of normal cells in normal-Ca2+ medium was significantly reduced by exposure to H2O2, and ACh-induced cell shortening in Ca2+-free medium was nearly abolished (Fig. 6).

Fig. 5.

A: data for normal muscle cells in normal-Ca2+ medium (normal) from Fig. 1 demonstrate the effect of H2O2 on the ACh-induced Ca2+ signal. Fura 2-AM-loaded normal cells in normal physiological salt solution were incubated with 70 μM H2O2 for 30 min. ACh-induced Ca2+ signal was reduced compared with cells incubated in normal (i.e., H2O2-free) buffer. Data suggest that exposure to a relatively low concentration of H2O2 may result in depletion of intracellular Ca2+ stores. Values are means ± SE; n = 8 (normal buffer) and 6 (H2O2 buffer). B: data for H2O2-treated muscle cells in normal-Ca2+ medium (normal Ca2+) from Fig. 6A demonstrate the role of extracellular Ca2+ influx in ACh-induced Ca2+ signal in H2O2-treated cells. Fura 2-AM-loaded normal cells were incubated with 70 μM H2O2 for 1 h in normal physiological salt solution (normal Ca2+) or Ca2+-free medium with 200 μM BAPTA to ensure that the Ca2+ signal was entirely produced by release of Ca2+ from intracellular stores. In Ca2+-free medium, the ACh-induced Ca2+ increase was reduced compared with cells in normal-Ca2+ medium. Values are means ± SE; n = 6.

Fig. 6.

Similar to the Ca2+ increase in Fig. 1, ACh-induced cell shortening of normal cells was the same in normal physiological salt solution (•) and Ca2+-free salt solution (□). Shortening was significantly reduced (P < 0.05, ANOVA) by exposure to H2O2 (⧫) and nearly abolished (P < 0.001, ANOVA) in Ca2+-free medium (▵). Values are means ± SE; n = 5 (physiological salt solution) and 3 (Ca2+-free salt solution, H2O2 in physiological salt solution, and H2O2 in Ca2+-free salt solution).

To examine whether H2O2 may directly induce a net Ca2+ release from intracellular stores, resulting eventually in Ca2+ depletion, a high concentration (5 mM) of H2O2 was directly applied to fura 2-AM-loaded cells through a pressure ejection micropipette. Direct application of H2O2 to normal smooth muscle cells in Ca2+-free medium caused a gradual increase in cytosolic Ca2+ (Fig. 7), demonstrating that H2O2 causes direct release of Ca2+ from intracellular stores.

Fig. 7.

A: direct application of H2O2 to normal smooth muscle cells in Ca2+-free medium caused a gradual increase in cytosolic Ca2+, demonstrating that H2O2 may cause direct release of Ca2+ from intracellular stores. B: Ca2+ in response to direct application of buffer (control) or H2O2 in cells incubated in Ca2+-free medium. Values are means ± SE; n = 5. *P < 0.02 vs. control (paired t-test).

Thapsigargin is known to inhibit Ca2+-ATPase activity, blocking Ca2+ uptake into intracellular stores and causing Ca2+ release, resulting eventually in depletion of Ca2+ stores (16). At high thapsigargin concentration, the Ca2+ release may be of sufficient magnitude to cause cell contraction (1). To test whether H2O2 depletes intracellular Ca2+ stores in a manner similar to thapsigargin, freshly isolated LES muscle cells were treated with H2O2. H2O2 caused cell contraction, reaching maximum contraction 1–2 min after application and returning to normal after 10 min. In cells pretreated with 3 μM thapsigargin for 30 min, contraction in response to H2O2 was nearly abolished (n = 3, P < 0.0001, ANOVA), suggesting that this contraction requires release of intracellular Ca2+ (Fig. 8). Similarly, if the cells were treated first with 70 μM H2O2 for 30 min and then exposed to thapsigargin, the contraction induced by thapsigargin was significantly reduced (Fig. 9), confirming that H2O2 causes depletion of releasable Ca2+ stores. The H2O2-associated reduction of thapsigargin-induced contraction was reversed by the H2O2 scavenger catalase, but not by the hydroxyl radical scavenger deferrioxamine (26) or the scavenger of singlet oxygen histidine (26) (Fig. 10). These data suggest that H2O2 may directly induce release of intracellular Ca2+ from stores, resulting in depletion of releasable Ca2+ stores.

Fig. 8.

In cells pretreated with 3 μM thapsigargin (TG) for 30 min, contraction in response to H2O2 was abolished (P < 0.0001, ANOVA), suggesting that this contraction requires release of Ca2+ from intracellular stores. Values are means ± SE; n = 3.

Fig. 9.

In cells treated with 70 μM H2O2 for 30 min and then exposed to TG, contraction induced by thapsigargin was significantly reduced (P < 0.0001, ANOVA). Values are means ± SE; n = 4 (control) and 3 (H2O2).

Fig. 10.

H2O2 reduced TG-induced contraction. Reduction was reversed by the H2O2 scavenger catalase, but not by the hydroxyl radical scavenger deferrioxamine or the scavenger of singlet oxygen histidine. Values are means ± SE; n = 3. *P < 0.05; ** P < 0.01 (ANOVA)

To test whether H2O2 inhibits Ca2+-ATPase, we measured the Ca2+-ATPase activity in the absence (control) or presence of 1 mM H2O2 or 3 μM thapsigargin (positive control). H2O2 and thapsigargin significantly reduced Ca2+-ATPase activity from 1.85 ± 0.2 μmol NADH·h−1·mg protein−1 in control tissue to 0.5 ± 0.2 μmol NADH·h−1·mg protein−1 in H2O2-treated tissue (P < 0.01, ANOVA). Thapsigargin, used as a positive control, reduced ATPase activity to 0.09 ± 0.05 μmol NADH·h−1·mg protein−1 (P < 0.001 ANOVA). The data confirm that H2O2 depletes intracellular Ca2+ stores by inhibiting Ca2+-ATPase, similar to thapsigargin.

DISCUSSION

Impairment of Ca2+ release from intracellular stores in esophagitis.

We previously showed that the initial contraction of LES smooth muscle cells in response to ACh is mediated by release of Ca2+ from intracellular stores, because the ACh-induced initial contraction in normal LES cells was the same in normal-Ca2+ and Ca2+-free medium (1). In contrast to circular LES muscle strips, which maintain spontaneous tone, single LES cells do not maintain spontaneous contraction and contract only in response to exogenously added agonists. When LES circular muscle strips are incubated in Ca2+-free medium, tone diminishes over time (9), presumably because Ca2+ influx is needed to replenish Ca2+ stores: in Ca2+-free medium, the stores are not replenished and become depleted over time, causing a reduction in tone. In contrast, because single cells do not maintain spontaneous contraction, when normal single cells are incubated in Ca2+-free medium, nothing happens; i.e., the initial length and amplitude of contraction are the same in normal-Ca2+ and Ca2+-free medium (33), regardless of the incubation time in Ca2+-free medium. However, when single LES cells from animals with esophagitis are incubated in Ca2+-free medium, ACh-induced contraction is significantly reduced, because IP3 production is reduced (41) and because releasable Ca2+ stores are reduced, as demonstrated by a reduction in IP3-induced contraction (33).

In the present study, we confirmed that, in normal LES, similar to ACh-induced contraction, ACh-induced Ca2+ increase was the same in normal-Ca2+ and Ca2+-free medium (Fig. 1).

Release of Ca2+ from intracellular stores in smooth muscle may depend on activation of IP3 receptors or ryanodine receptors. Some muscle cells have only one type of Ca2+ store containing receptors for ryanodine and IP3 (24). Other muscle cells have multiple Ca2+ stores containing ryanodine or IP3 receptors, either singly or in combination. For example, guinea pig distal colonic circular muscle cells may have two functionally distinct Ca2+ stores: the first expressing only ryanodine receptors and the second expressing IP3 and ryanodine receptors (17). Murine bladder smooth muscle cells have three types of Ca2+ stores: one expressing ryanodine and IP3 receptors, another having only IP3 receptors, and the third containing only ryanodine receptors (42). The nature of the intracellular Ca2+ stores of cat LES circular muscle cells has not been defined. Because U-73122, a PIP2-phospholipase C inhibitor (10), abolished the ACh-induced Ca2+ increase (Fig. 2), our data suggest that ACh causes Ca2+ release mostly from stores containing IP3 receptors, consistent with our previous cell shortening data (41).

The ACh-induced Ca2+ increase was significantly lower in acute esophagitis than in normal cells (Fig. 3C), suggesting 1) esophagitis-induced damage to Ca2+-release mechanisms from intracellular stores and/or 2) depletion of releasable Ca2+ stores. Previous data from our laboratory support both outcomes (1 and 2), inasmuch as we have shown that ACh-induced IP3 production is significantly reduced in esophagitis (41) and that IP3-induced contraction is also reduced in esophagitis LES cells (33).

The IP3-dependent contribution to the Ca2+ signal in esophagitis can be estimated by the Ca2+ increase in Ca2+-free medium (Fig. 3C). The difference between the Ca2+ signals in normal-Ca2+ and Ca2+-free medium must be due to Ca2+ influx, whereas the Ca2+ signal in Ca2+-free medium must be due to Ca2+ release. Figure 3C shows that the contribution of released Ca2+ in esophagitis is about one-third of the signal observed in the same cells in normal-Ca2+ medium and about one-fourth of the signal of normal cells. The 75% reduction in Ca2+ released in esophagitis compared with normal cells may be due to a reduction in IP3 formation and/or a reduction in releasable Ca2+ stores. The specific contribution of either factor cannot be estimated at this time; however, although contraction of normal LES cells is entirely supported by release of Ca2+ from intracellular stores, in esophagitis cells the contraction depends to a large extent on influx of extracellular Ca2+.

The Ca2+ influx in esophagitis may be due to activation of capacitative Ca2+ channels induced by esophagitis-associated depletion of Ca2+ stores (11, 23, 39, 45). Depletion of intracellular Ca2+ stores is known to activate store-operated (or capacitative) Ca2+ (SOC) channels in a variety of cells (4, 5, 15, 31, 32). The idea of capacitative Ca2+ entry was initially proposed for smooth muscle cells. It has been shown that capacitative Ca2+ entry and contraction can be activated in smooth muscle by passive depletion of intracellular Ca2+ stores, even without activation of receptor-dependent cascades (18).

It is thought that inasmuch as Ca2+ stores are depleted, at least two channels are activated: the Ca2+ release-activated Ca2+ channel, which is selective for Ca2+, and a 3-pS channel, which is a nonselective cation channel and is present in excitable and nonexcitable cells. A Ca2+ influx factor (CIF) has been partially purified from human platelets or from yeast with depleted Ca2+ stores (45) and has been shown to activate the 3-pS channel.

SOC channels, including the 3-pS channel, are voltage independent, inasmuch as they are not directly gated by membrane voltage changes when Ca2+ stores are full. However, once activated in response to store depletion, the SOC channels are thought to be “exquisitely sensitive to membrane potential changes” (31) and, thus, may open, allowing Ca2+ influx and contraction. Although CIF has not been identified, a mechanism for CIF-induced activation of SOC has been recently proposed (39). According to this mechanism, CIF displaces inhibitory calmodulin from Ca2+-insensitive cytosolic phospholipase A2 (iPLA2), resulting in activation of iPLA2 and generation of arachidonic acid and lysophospholipids, which in turn activate SOC channels and induce capacitative Ca2+ influx. Upon refilling of the stores and termination of CIF production, calmodulin rebinds to iPLA2 and inhibits it, and the activity of SOC channels and capacitative Ca2+ influx are terminated (39).

This model nicely explains our data: when agonists (e.g., ACh) are applied to normal LES cells with full Ca2+ stores, if phospholipase C is not inhibited, ACh induces production of IP3, release of Ca2+, and contraction. After Ca2+ is released, CIF is produced, SOC channels open, and the stores are refilled. If, however, formation of IP3 is inhibited by phospholipase C antagonists such as U-73122, ACh does not cause formation of IP3 and Ca2+ release, CIF is not produced, iPLA2 remains inhibited, SOC channels do not open, no Ca2+ signal occurs (Fig. 2), and contraction does not occur.

In the presence of H2O2, which inhibits Ca2+-ATPase, Ca2+ uptake is blocked, resulting in depletion of Ca2+ stores and production of CIF. CIF displaces inhibitory calmodulin from iPLA2, resulting in activation of iPLA2 and generation of arachidonic acid and lysophospholipids, which in turn activate SOC channels. SOC channels are then free to open in response to ACh, which causes membrane depolarization (37). Thus, in esophagitis cells, which contain excess H2O2, ACh causes influx of extracellular Ca2+. This ACh-induced influx of Ca2+ in esophagitis is unlike normal LES, where ACh-induced contraction is entirely supported by Ca2+ release.

H2O2 and impairment of Ca2+ release in esophagitis.

H2O2 levels were significantly elevated in esophagitis in LES circular muscle, as well as in isolated LES muscle cells (14), compared with normal muscle and are most likely related to increased cytokine levels after acid perfusion (14). Therefore, we examined the role of H2O2 in the impairment of ACh-induced Ca2+ release in esophagitis.

Catalase is a known H2O2 scavenger that converts H2O2 to water. In human sigmoid colon, circular muscle catalase had no effect of its own, but it restored contraction of sigmoid cells from patients with ulcerative colitis (13). Similarly, catalase treatment did not affect resting tone of normal LES (14). In esophagitis LES muscle cells, however, catalase significantly enhanced the ACh-induced Ca2+ signal in Ca2+-free medium (Fig. 4), indicating that H2O2 is present in the enzymatically isolated LES smooth muscle cells, as shown by Cheng et al. (14), and contributes to esophagitis-associated impairment of Ca2+ release from intracellular stores. The role of H2O2 in the impairment of Ca2+ release from intracellular stores in esophagitis is further supported by our finding that pretreatment of normal LES circular muscle cells with H2O2 significantly decreased ACh-induced release of intracellular Ca2+ (Fig. 5A) and cell shortening (Fig. 6).

H2O2 may be derived from reduction of the superoxide anion (O2·) (44) as part of the mitochondrial respiratory process. O2· in aqueous solution is short-lived and is rapidly reduced to the much more stable molecule H2O2. In this process, O2· is produced from mitochondrial respiratory chain complex I (NADH-ubiquinone oxidoreductase) and complex III (ubiquinol-cytochrome c oxidoreductase). Alternatively, in cells such as macrophages and neutrophils, the function of which is to attack invaders by producing H2O2 through an oxidative burst, NADPH oxidase is responsible for H2O2 production. Isoforms of this enzyme have been found in nonphagocytic cells (3, 43), suggesting that nonphagocytic cells may also produce H2O2 under appropriate conditions. The nonphagocytic NADPH oxidases are similar in structure to the neutrophil NADPH oxidase, which consists of two transmembrane subunits, p22phox and gp91phox (which comprise flavocytochrome b558), and at least three cytosolic subunits, p47phox, p67phox, and Rac2 (2). The source of excess H2O2 in the LES circular muscle layer of esophagitis specimens, whether of mitochondrial origin or generated by NADPH oxidases, has not been established.

The findings that intracellular H2O2 is present at higher levels in esophagitis than in normal cells (14) and that catalase restores the ACh-induced Ca2+ signal in esophagitis cells support the presence of H2O2 in LES circular muscle, as has been previously reported in rat aortic smooth muscle cells (25) and in circular smooth muscle from the human sigmoid (which, similar to LES circular muscle, utilizes release of intracellular Ca2+ to contract in response to the endogenous agonist neurokinin A). In the rectosigmoid, catalase did not affect contraction of normal cells, but it restored contraction of cells from ulcerative colitis patients (13), supporting H2O2 as a possible mediator of dysmotility in gastrointestinal inflammation, at least when contraction is mediated by Ca2+ release from intracellular stores.

The mechanism of removal of H2O2 by exogenously added catalase is not entirely clear. If H2O2 is formed outside the cells by NADPH oxidase (27), it is easily accessible to the H2O2 scavenger, which does not diffuse across the cell membrane. Alternatively, if H2O2 is formed inside the muscle cells, it can diffuse across biological membranes (44), because the molecule is not electrically charged, and removal of extracellular H2O2 by catalase may facilitate diffusion of intracellular H2O2 into the extracellular medium. This would result, eventually, in removal of intracellular and extracellular H2O2 through a process similar to the neutralization of intracellular H2O2 in adipocytes and Hep G2 cells (29).

In any case, catalase somehow neutralizes H2O2, allowing the stores to refill. Once the stores are refilled, esophagitis cells will contract again in Ca2+-free medium by releasing stored Ca2+. The implications of these data are that LES circular muscle cells in acute esophagitis are similar to normal cells, except that Ca2+ stores are reduced by the presence of H2O2, which prevents Ca2+ uptake into intracellular storage sites.

ROS have been shown to consistently depress the Ca2+-ATPase responsible for uptake of Ca2+ into the endoplasmic reticulum (19–21, 28, 34). We confirm that, in LES circular muscle, H2O2 inhibits Ca2+-ATPase activity (Fig. 11). In addition to inhibiting Ca2+ uptake into the endoplasmic reticulum, ROS cause release of Ca2+ from intracellular stores through ryanodine- and IP3-sensitive Ca2+ channels (26). Application of a high concentration H2O2 to enzymatically isolated smooth muscle cells caused a gradual cytosolic Ca2+ increase in cells kept in Ca2+-free medium (Fig. 7). Because H2O2 inhibits Ca2+-ATPase, it shifts the balance between Ca2+ uptake and Ca2+ release, causing a net Ca2+ release, similar to the mode of action of thapsigargin. Similar to thapsigargin, H2O2 caused cell contraction followed by relaxation. As expected, thapsigargin abolished H2O2-induced contraction (Fig. 8), and H2O2 abolished thapsigargin-induced contraction (Fig. 9), consistent with the similarity of mechanisms of action.

Fig. 11.

Ca2+-ATPase activity in the absence (control) or presence of 1 mM H2O2 or 3 μM TG (positive control). H2O2 and TG significantly reduced Ca2+-ATPase activity. Values are means ± SE; n = 3. *P < 0.01; **P < 0.001 vs. control (ANOVA).

We conclude that H2O2 may be at least partly responsible for the impairment of Ca2+ release from intracellular stores in acute experimental esophagitis and that neutralization of H2O2 may result in improvement of esophagitis-associated dysmotility.

GRANTS

This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant RO1-DK-57030.

Acknowledgments

These data were presented in part at the 102nd Annual Meeting of the American Gastroenterological Association, Atlanta, GA, May 2001.

Footnotes

  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

REFERENCES

View Abstract