Abstract
Tonic contraction of the lower esophageal sphincter (LES) prevents gastroesophageal reflux. LES tone is produced both by cholinergic nerve and myogenic activities. The Ca++ sources for LES tone and carbachol-induced contraction in canine LES strips were determined from the effect on contractile activity of extracellular Ca++level modulation, Ca++ entrance blockade or enhancement with nifedipine or BayK8644 respectively, and/or inhibition of Ca++ store refilling using the sarcoplasmic reticulum Ca++ pump inhibitor, cyclopiazonic acid. LES tone disappeared when a Ca++-free physiological saline solution or nifedipine was applied. Sustained Ca++ free contractions to carbachol were prevented/abolished by nifedipine or increased Ca++ chelation and enhanced by BayK8644. Inhibition of sarcoplasmic reticulum Ca++ pumps by cyclopiazonic acid reduced Ca++ free contractions to carbachol; BayK8644 restored cyclopiazonic acid-reduced Ca++ free contractions to carbachol. Therefore, some Ca++ stores can be refilled by mechanisms not requiring activity of the sarcoplasmic reticulum Ca++ pump. A preferred pathway may exist whereby Ca++ enters stores directly through L-Ca++channels. The proposed Ca++ store refilling mechanism involves continuous Ca++ entry through L-Ca++channels from sites not equilibrated with external Ca++. Therefore, diverse Ca++ stores exist in canine LES which are dependent on Ca++ influx through L-Ca++channels.
Control of the LES tone is essential for nutrition and health. This sustained contractile state limits gastric reflux but neurally controlled transient relaxation in response to a swallow or a bolus in the esophagus allows nutrient passage into the stomach. After a relaxation, a neurally mediated (cholinergic) contraction aids in closure of the LES (Watanabe, 1992; Daniel, 1992). In some species, cholinergic nerve activity contributes to the mostly myogenic LES tone (reviewed inDaniel, 1992). When control of tone is deficient, esophagitis can ensue (Watanabe, 1992) and when neural relaxation is lost in achalasia or Chagas’ disease (Gauminitz et al., 1995), the LES becomes a site of obstruction.
The contractile state of the LES depends on the level of [Ca++]i as it does in other smooth muscles (Hartshorne and Kawamura, 1992; Jiang and Stephens, 1994). However, intracellular Ca++ is not homogeneously distributed within smooth muscle cells. Membrane delimited Ca++ stores or pools exist inside the SR and these stores are refilled by controlled mechanisms such as Ca++ pump ATPases (Mayer and Sun, 1992). Also, a region of higher cytoplasmic Ca++ between the PM and the superficial SR, maintained by Ca++ uptake and vectorial extrusion from this SR toward the PM may exist in smooth muscle cells. The superficial SR provides a buffer function, the SBB function, by taking up entering Ca++ if not fully loaded and extruding it vectorially as described. Removal of Ca++ from this region is thought to occur by Na-Ca exchange (Van Breemen, 1989; Chen and Van Breemen, 1993). Because it is superficial and does not contain the contractile proteins, changes in [Ca++]i in the SBB may not change tone. Inhomogeneity in smooth muscle [Ca++]i levels may also occur because localized Ca++ release from SR causes local elevations (Ca++ sparks) which open Ca++-activated K+ channels leading to STOCs and membrane hyperpolarization; i.e., causing relaxation rather than contraction (Nelson et al., 1995). The existence of a store of Ca++ in a pericellular region surrounding arterial smooth muscle and accessing the cytoplasm through L-Ca++channels was reported (Wheeler-Clark and Buja, 1995).
In LES muscle strips of all species studied, muscle tone is supported by on-going entrance of extracellular Ca++ through plasmalemmal Ca++ channels. Adding L-Ca++channel blockers or reducing [Ca++]e reduces LES tone: opossum (Fox and Daniel, 1979); cat (Biancani et al., 1987); dog (Allescher et al., 1988); human (Tottrup et al., 1990). However, Biancani et al.(1994) suggested that on-going activation of PLC (mechanism unknown) and production of IP3 resulted in Ca++ release from intracellular stores and potentiation of DAG to activate a PKC-dependent pathway. These results suggested that LES tone was dependent on continuous low levels of Ca++ release from intracellular stores. Although species differences may exist, there remains controversy over the sources and relative importance of Ca++ sources for LES tone.
The LES is innervated by cholinergic nerves and ACh released from nerve endings increases tone in most species (Diamant, 1989). Muscarinic agonists such as ACh or CCh may produce smooth muscle contraction after interaction with receptor by several mechanisms acting in parallel (Sims and Janssen, 1993): 1) Depolarization from opening of Na+ or Cl− channels or closure of K+ channels might open L-Ca++ channels to allow Ca++ influx; 2) Ca++ channels, either L-Ca++ or nonspecific cation channels, may be opened by receptor activation; 3) Ca++ may be released from internal stores after activation of PLC and production of IP3. The DAG concomitantly produced may activate PKC, leading to phosphorylation events which enhance the sensitivity of the contractile system to [Ca++]i.
The aim of this study was to discern the relative importance of extracellular and intracellular Ca++ sources or stores for both LES tone and agonist-induced contractions. Observations of the effects of a Ca++ manipulations singly and in combination on LES tone and agonist-induced contraction were used.
Methods
In vitro studies.
Adult mongrel dogs of either sex were euthanized with an overdose of sodium pentobarbital (100 mg kg−1), according to a protocol approved by the McMaster Animal Care Committee and according to the Guidelines of the Canadian Council on Animal Care. The GE region was excised and stored in Krebs-Ringer PSS at 25 ± 2°C equilibrated with 5% CO2 and 95% O2 and having the following composition (in mM): 115.0 NaCl, 4.6 KCl, 1.2 NaH2PO4, 1.2 MgSO4, 22.0 NaHCO3, 2.5 CaCl2 and 11.0 glucose. The GE junction or LES was opened on the gastric greater curvature side and the mucosa was removed by sharp dissection. The muscular equivalent of the LES was revealed as a thickened ring of muscle composed of clasp fibers with oblique gastric sling fibers on either side (Friedlandet al., 1971). LES muscle was consistently cut from the clasp region of the LES because functional differences have been demonstrated between the different muscle bundles of the LES in a number of species including human (Preiksaitis et al., 1994).
Circular muscle strips approximately 20 × 2 mm were dissected free of underlying muscle layers so that the long axis of each strip was parallel to the circularly oriented clasp muscle fibers. Muscle strips were placed in 10-ml organ baths, bathed in Krebs-Ringer solution (a physiological salt solution, PSS, composition given above) at 37°C, and bubbled with 95% O2 and 5% CO2. Each strip was oriented vertically with the bottom end fixed to the electrode holder by silk ligature and the top end passed through a pair of concentric platinum electrodes then fixed by a longer silk thread and ligature to a force displacement transducer (FTOC3; Grass, Quincy, MA). Two grams of tension was applied to each strip. Muscle strip tension changes were displayed on a Beckman R611 Dynograph. Active tension was taken to be the difference between the observed tension and the minimum tension reached in a Ca++-free PSS which was made as described above but with no CaCl2 added and a high level of the Ca++chelator EGTA (1 mM) present. In this study, all PSS solutions made with no Ca++ added and various levels of EGTA were referred to as Ca++-free solutions. Estimated free Ca++concentrations in all Ca++-free solutions were low, ranging from 0.7 to 7.7 nM free Ca++ with 1000 to 100 μM EGTA present, respectively, when a 10 μM level of Ca++contamination was assumed and MaxChelator software, version 6.72 (Berset al., 1994) was used to estimate Ca++concentration.
Experimental protocol.
As diagrammed in figure1, muscle strips were prepared and equilibrated for 30 min over which time muscle strips rapidly contracted and spontaneously developed tone, taken to be 100% for tone when stable. An initial maximal contraction to CCh (10−5M) was performed in all muscle strips in Ca++ containing PSS. This contraction was taken as 100% for CCh contraction. After 15 min, CCh was washed-out with normal PSS and tone allowed to stabilize for 30 min. The protocol was then as follows: first, Ca++-free extracellular solution with either a low-EGTA level present (100 μM) or a mid-EGTA level present (200 μM) replaced Ca++-containing PSS. Either Ca++-free solution caused complete relaxation of LES tone. Second, after 10 min in Ca++-free extracellular solution, 10−5 M CCh was added to the bath for 15 min. The addition of CCh resulted in a large initial, initial, sustained contraction, named a CFFC1. As shown in figure 1, the peak amplitude and the area under the CFCC and over the referenced baseline were measured with a computerized microplanimeter system (Laboratory Computer Systems, Cambridge, MA). Measured peaks and areas were normalized and expressed as a percentage of the initial CCh contraction in Ca++-containing PSS for each muscle strip. Third, after 15 min, the extracellular solution was replaced with fresh Ca++-free extracellular with no CCh present for 5 min. When the Ca++ was returned to the extracellular solution, LES tone was restored. The muscle strips remained in normal Ca++-containing extracellular solution (PSS) for 30 min. This protocol was repeated three times yielding CFCC2, CFCC3 and CFCC4. Fourth, during the 5-min washout period after CFCC2, a test agent such as CPA, NIF and/or BayK was added for the remainder of the experiment (i.e., during CFCC3 and CFCC4). The ability of muscle strips to redevelop tone when the Ca++-free extracellular solution was replaced with Ca++-containing PSS was compared to the baseline tension attained during the initial equilibration period. These tone measurements were named tone-in-PSSn and abbreviated as: TPSS0, TPSS1, TPSS2, TPSS3, TPSS4 (see fig. 1). Finally, at the end of each experiment, Ca++-free PSS [with 0 Ca++ and a high level of EGTA (1 mM) added] was applied causing relaxation and this level was recorded as baseline or zero active tension.
In one experimental series (shown in fig.2), CFCCs were repeatedly evoked with washouts in Ca++-free PSS (not Ca++-containing PSS), thereby preventing Ca++ store refilling from the extracellular space. Because the protocol was unlike that used in other experiments, successive CFCCs were enumerated differently from the others, with lower Roman numeral subscripts (CFCCi to CFCCv).
Statistical analysis of data.
Data are expressed as means ± S.E.M. and standardized by expressing means as percentages of initial tone or CCh contraction. The number of experiments or animals is indicated by n (because tissue from a given animal was used for only one experiment). The data sets were analyzed by one-way ANOVA or with repeated measures ANOVA when data were measured and compared for multiple procedures for the same muscle strip. Subsequent comparison of the means was only performed when the P value for the ANOVA was less than 0.05 (values not reported). The statistical significance of differences in the means was determined by Tukey-Kramer multiple comparisons test where P < 0.05 were considered significant (*P < .05; **P < .01; ***P < .001).
Drugs and solutions.
All drugs were purchased from Sigma Chemical Co., St. Louis, MO and dissolved in PSS on the day of the experiment. CPA, BayK or NIF, were first dissolved in 100% ethanol or DMSO in high concentration and then diluted with PSS to concentrations required. Drug concentrations applied were shown in previous studies to be maximally effective. The drugs were added to the bath in small volumes (ml) and any diluents used to solubilize them were tested on control strips in similar volumes to exclude any action by them.
Results
Initial effect of Ca++-free PSS.
All measured values, CFCC peak amplitude, CFCC area, and tone resumption, were affected by perfusion in low EGTA (0.1 mM)-Ca++-free solutions. The mean peak amplitude of CCh contractions in low EGTA-Ca++-free PSS was significantly smaller than that of the initial CCh contraction performed in Ca++-containing PSS. The mean CFCC1 amplitude (with no test agents present) was 71.0 ± 1.9% (for 131 muscle strips) of the amplitude of the initial CCh contraction performed in Ca++-containing PSS. The reduction in CFCC amplitude was due in part to the abolition of LES active tone in Ca++-free PSS (which occurred at all levels of EGTA, low, mid and high) so that the CFCC was no longer superimposed on baseline tone such as the CCh contraction performed in Ca++-containing PSS. The mean area under CCh contractions performed in Ca++-free PSS was also significantly smaller than that under the initial CCh contraction performed in Ca++-containing PSS, with CFCC1 area being on average 55.3 ± 2.2% of the area under the initial CCh contraction. This marked decrease in area was due in large part to the abolition of baseline tone.
Unlike CFCC amplitude and area, LES tone resumption was enhanced after a prior CCh contraction performed in low EGTA-Ca++-free PSS (CFCC1). The mean tone measurement for TPSS1 (i.e., tone resumption after CFCC1 obtained when Ca++-containing PSS was reperfused) was 96.6 ± 12.3% (for 131 muscle strips) larger than the baseline tone developed initially in Ca++-containing PSS without prior perfusion of the muscle strip in Ca++-free PSS or application of CCh.
Effects of repeated exposure to Ca++-free PSS.
All measured values, CFCC amplitude, CFCC area and tone resumption, decreased with time and repeated perfusion with low-EGTA (0.1 mM) Ca++-free PSS and CFCC production. With no test agents present, pooled values of CFCC amplitudes were compared (for 65 muscle strips) using repeated measures analysis. CFCC4 amplitude, 55.5%, and CFCC 3 amplitude, 57.8%, were significantly lower than CFCC1, 77.1%, and CFCC2, 64.9%, .05 > P > .001. Similarly, CFCC2 area, 40.3%, CFCC3 area, 35.5% and CFCC4 area, 34.5%, were significantly smaller than CFCC1 area, 54.4%, P < .001.
When CFCCs were repeatedly performed after wash-out with low-EGTA (0.1 mM) Ca++-free PSS but with wash-out of carbachol in Ca++-free PSS (in other words no intermediate exposure to extracellular Ca++), CFCC amplitude decreased with each repetition (fig. 2A). These decreases were significant during the fourth (CFCCiv = 16.2%) and fifth (CFCCv = 15.2%) repetitions when compared to the first CFCC (CFCCi = 37.7%).
Mean tone resumption values for each successive tone measurement after washout of CCh in PSS (TPSS0 through TPSS4), with no test agents present, were pooled and compared (for 70 muscle strips) using repeated measures analysis; TPSS0 tone, 216.5%, TPSS1 tone, 210.5% and TPSS2 tone, 212.2%, were all significantly higher than TPSS4 tone, 181.0%, .01 < P < .05. Therefore, differences in CFCC amplitude and area and LES tone resumption values, when test agents had been used, were assessed by comparing their mean value to their time-matched (control) CFCC and tone resumption values and not to previous CFCC or tone values. For example, mean CFCC3 amplitude with a test agent such as nifedipine present was compared to the mean CFCC3 amplitude with no nifedipine added (the time-matched control), see figures3, 4 and 6.
Effect of extracellular Ca++ chelation.
As shown in figures 3A and B, when an increased level of the Ca++chelator EGTA (mid-[EGTA] = 200 μM) was used in the Ca++-free extracellular solution, mean peak CFCC amplitude was less than when low-EGTA (low-[EGTA] = 100 μM) Ca++-free PSS was used, reaching significance at CFCC4, 39.3% (mid-EGTA Ca++-free PSS) vs. 66.6% (low-EGTA Ca++-free PSS), n = 4, P < .05. Thus, increased Ca++ chelation (200 instead of 100 μM EGTA) appeared to reduce entry of Ca++ from an extracellular source. In contrast, LES tone resumption values (TPSS0 through TPSS2 tone values) were not differentially affected by an increase in Ca++ chelator concentration present in the Ca++-free extracellular solution perfused during CFCC production between tone measurements (see fig. 3C). All other experimental protocols were performed in low-EGTA Ca++-free PSS.
Effect of plasmalemmal L-Ca++ channel antagonism.
Because effects of varying Ca++ chelator concentration in the extracellular solution suggested that Ca++ influx was occurring from nominally Ca++-free solutions, the effect of blocking a potential pathway for Ca2+ influx, the plasmalemmal L-Ca++ channel, with NIF (30 μM) was tested, as depicted in figure 4. When present, NIF reduced CFCC amplitude and tone resumption to values near and not significantly different from zero, respectively; i.e., CFCC3 and CFCC4 amplitudes were 8.2 and 8.8% with NIF present compared to time-matched control levels of 60.7 and 52.2% with no NIF present (P < .001, n = 4). LES tone did not recover when NIF was present compared to recovery in time-matched controls with no NIF present; 13.5vs. 192.4% for TPSS2, 5.2 vs. 181.6% for TPSS3 and 0.1 vs. 174.0% for TPSS4 (P < .001, n = 4).
The effects of NIF on tone demonstrated that Ca++ influx through L-Ca++ channels was on-going and necessary for both tone maintenance and resumption. Also the ability of the tissue to respond to repeated CCh exposure with no wash-out period in Ca++-containing PSS (as shown in fig. 2A) appeared not to be due to inadequate intracellular store emptying by CCh. Instead this latter finding suggested that Ca++ entrance through L-Ca++ channels was required for adequate refilling of intracellular stores. Moreover, as demonstrated in figure5, Ca++ entrance through L-Ca++ channels was occurring during the CFCC, because nifedipine application during a CFCC also resulted in complete LES muscle strip relaxation (values not significantly different from zero), with PRE-NIF CFCC peak amplitude being 68.0% compared to post-NIF level of 4.0% (P < .001, n = 4). Thus, continuous Ca++ influx was required to maintain the CFCC.
Effect of L-Ca++ channel agonism.
When the L-Ca++ channel agonist BayK (10 μM) was present, CFCC amplitudes were enhanced and maintained (fig.6). The mean peak CFCC amplitudes were 88.7% for CFCC3 and 93.1% for CFCC4 with BayK present compared to time-matched control values of 58.0% for CFCC3 and 60.2% for CFCC4 with no BayK present (P < .05, n = 4). LES tone resumption was also significantly enhanced whenever BayK was present compared to time-matched control values, with values for TPSS2 being 240.1 vs. 133.0%, values for TPSS3 being 224.8vs. 116.5% and values for TPSS4 being 199.5 vs.98.0 (P < .05, n = 4). Although tone resumption values were found to vary in each data set collected, under control conditions, tone resumption values in this data set were smaller on average than those found overall, in other data sets. Thus as shown in figures 3, 4 and 5, when Ca++ influx was enhanced, CFCC amplitudes and LES tone were also enhanced.
Effect of SR Ca++ pump inhibition.
When the highly selective SR Ca++ pump inhibitor, CPA (30 μM) was present, mean peak CFCC amplitude was significantly reduced. With CPA present, amplitudes were 30.4% for CFCC3 and 25.6% for CFCC4 compared to time-matched control levels of 60.3% for CFCC3 and 55.5% for CFCC4 (P < .05, n = 4) (see fig.7B). These reduced values were significantly different from zero (P < .01). As shown in figure2B, CPA affected the contraction profile of the CFCC, making it more transient and usually phasic. When these changes were evaluated as the area under the CFCC, the areas under CFCC3 and CFCC4 were 9.6 and 5.5%, respectively, with CPA present, compared to 35.5 and 34.5% for time-matched control muscle strips (P < .001, n = 4) (see fig. 7C). As shown in figure 2B, CFCC amplitude was reduced but persisted with CPA present during each CFCC repetition (CFCCi through CFCCv) in Ca++-free PSS and wash-out of CCh in Ca++-free PSS. However, the amplitude of CFCCs with CPA present were only significantly different for time matched controls in CFCCi and CFCCii, i.e., 47.7%, P < .01 and 55.3%, P < .05 of time-matched controls with CPA present, respectively (n = 4). This demonstrated that Ca++ stores insensitive to CPA were present.
CPA enhanced tone resumption when the free Ca++concentration in the extracellular solution was restored (shown in fig.7D). Levels of tone with CPA present were significantly higher than time-matched control values, with the following values for TPSS2, TPSS3 and TPSS4, 290.2 vs. 180.1% (P < .01), 270.5vs. 169.5% (P < .05) and 248.5 vs. 148.3% (P < .05), respectively, n = 4.
Effect of a combination of L-Ca++ channel agonism and SR Ca++ pump inhibition.
As shown in figure8A and B, a combination of L-Ca++ channel agonism with BayK and SR Ca++pump inhibition with CPA resulted in increased CFCC amplitude over the effect of CPA alone, 77.1% (BayK + CPA) vs. 38.7% (CPA alone) for CFCC3, and 79.8% (BayK + CPA) vs.38.7% (CPA alone) for CFCC4, P < .05, n = 4. CFCC amplitudes with the BayK + CPA combination did not differ significantly from values for time-matched controls with no test agents present. BayK also reversed the effects of CPA alone on both the duration and the area under CFCCs (fig. 8A). In addition, BayK enhanced LES tone resumption: values for TPSS2 and TPSS3 were 340.5 and 294.7% respectively (BayK + CPA) versus 241.3 and 204.6%, respectively (CPA alone) (P < .01, n = 4) (fig. 8C).
Discussion
Ca++ dependence of LES tone.
Ca++chelation in the extracellular solution (with low or high levels of EGTA) or blockade of Ca++ entrance through L-Ca++ channels abolished LES basal active tone completely in less than 5 min. Conversely, L-Ca++ channel activation by BayK enhanced LES tone resumption. Therefore, basal canine LES tone relied on a continuous supply of Ca++ from the extracellular solution. In feline LES some tone persisted in Ca++-free extracellular solution (Biancani et al., 1987), a finding explained by continuous release of Ca++ from intracellular Ca++ stores to maintain LES tone. This was not occurring in canine LES because CPA, a SR Ca++ pump inhibitor, did not reduce LES tone as would be the case if release of Ca++ from intracellular stores supported LES tone (see fig. 2D). In fact, CPA enhanced LES tone as expected if less influxing Ca++ was sequestered in the CPA-sensitive stores, allowing higher [Ca++]i levels near the contractile proteins.
When intracellular Ca++, sequestered in membrane-delimited stores or in the general cytoplasm, was depleted, Ca++entrance and LES tone were enhanced, as predicted by the Putney Capacitative Calcium Hypothesis (Putney, 1990). Tone resumption after any CFCC (TPSS1 through TPSS4) was always larger than the initial tone developed during the equilibration period (see fig. 7D and figs. 3, 4,6 and 8, part C). This finding provided further evidence that canine LES tone is dependent upon Ca++ influx through L-Ca++ channels, the amount of which may have been affected by the status of the Ca++ stores, the Ca++gradient across the PM or by alterations in L-Ca++ channels themselves.
L-Ca++ channel function in LES.
Although this study suggests that L-Ca++ channels remain open under resting conditions in LES, the nature of continual L-Ca++channel activation was not addressed in this study. We must, however, postulate that a condition exists that allows sufficient L-Ca++ channels to remain open at a more negative membrane potential than is consistent with opening in other tissues (Catterall, 1995). The resting membrane potential in canine LES smooth muscle was −43 ± 2 mV as recorded in the isolated cell (Salapatek et al., 1998) and similar to that recorded in the muscle strip (Juryet al., 1991). L-Ca++ channels in smooth muscle are reported to pass significant inward current from a threshold depolarization of approximately −40 mV (Caterall, 1995).
The situation in LES clearly differs from that described by Nelsonet al. (1995) or by Rubart et al. (1996) in which the local release of Ca++ from SR or entrance of Ca++ through a small number of L-Ca++ channels at −40 mV produced Ca++ sparks and opening of Ca++-activated K+ channels to limit rather than enhance tension. In LES continuous Ca++ entry maintained tone.
Ca++ dependence of and sources for CCh contraction.
CCh contractions in Ca++-free PSS were always lower in amplitude than in Ca++-containing PSS (see figs. 3, 4, 6, 7 and 8, part B). Hillemeier et al. (1991)found that ACh-induced contractions in feline LES were unaffected by a reduction in extracellular Ca++ and concluded that they were mediated by Ca++ release from intracellular stores by IP3. Our results, however, suggested that extracellular Ca++ was required for CFCCs. Increasing the chelator concentration or blockade of Ca++ entrance through L-Ca++ channels reduced or abolished, respectively, the CFCC. One interpretation of this finding is that inhibited Ca++ entrance from L-Ca++ channel blockade reduced Ca++ store refilling during Ca++reperfusion in the period between CCFCs. This would have reduced Ca++ release during the next CFCC. However, this could not explain that the use of decreased external Ca++concentration (with a mid-EGTA level, see figs. 3A and B) or nifedipine addition during the CFCC (see fig. 5) reduced or abolished the CFCC, respectively. Therefore, Ca++ was entering through L-Ca++ channels during the CFCC. These results were surprising because calculated free Ca++ concentrations in low-EGTA-Ca2+-free (see figs. 2 through 7, A and B) or mid-EGTA-Ca++-free (see fig. 3A and B) solutions used during CFCC production were less than 8 nanomolar (see “Methods”). Furthermore, these solutions abolished LES tone. These results implied that the CCh interaction with the muscarinic receptor resulted in L-Ca++ channel activation and/or increased availability of extracellular Ca++, thus promoting Ca++entrance through L-Ca++ channels which supported the CFCC.
These results suggest that extracellular Ca++ near the PM was higher during the CFCC than that calculated for the general solution. Wheeler-Clark and Buja (1995) showed in canine coronary artery smooth muscle that Ca++ concentrations near the membrane may be higher than in the general extracellular solution due to Ca++ binding to components of the glycocalyx such that Ca++ may dissociate rapidly from these sites in response to small decreases in local free Ca++.
The CCh response was dependent on SR function as well as on Ca++ influx. SR Ca++ pump inhibition with CPA markedly reduced CFCC amplitude and made the CFCC more transient and phasic, reflected in decreased area under CFCCs (see fig. 7C). Therefore, intracellular Ca++ stores were also involved in the CFCC, probably by releasing Ca++ through IP3-sensitive Ca++ channels. This effect of Ca++ pump inhibition suggested that the LES SR Ca++ pump sequesters Ca++ into the SR from a low cytoplasmic Ca++ concentration. The fact that significant amplitude and area remained during the CFCC after CPA treatment (even with CFCC repetition, see fig. 2B) required postulation of a CPA-insensitive Ca++ store. We hypothesize it could be filled by two mechanisms: 1) A passive filling mechanism such as a leak channel. Here, increased Ca++ influx through L-Ca++ channels would result in partial filling of CPA-insensitive stores. However, this mechanism would imply that, unlike Van Breemen’s SBB hypothesis, vectorial pumping of SR Ca++ toward the plasma membrane in canine LES is not required for SBB maintenance. PM Ca++ pumps, Ca++ exchange mechanisms, and leak channels might be maintaining the extracellular Ca++ store which could be acting as an external source of Ca++ in this tissue. 2) A specialized pathway via L-Ca++ channels could exist to these CPA-insensitive stores. Similar findings in respiratory (Montanoet al., 1996) and vascular (Low et al., 1992) smooth muscles have led other researchers to similar hypotheses. The need for one or both postulations is reinforced by the fact that L-Ca++ channel agonism with BayK overcame the effects of Ca++-pump inhibition with CPA, increasing the amplitude of CFCCs and sustaining them.
Possible role of caveolae.
The plasma membrane of smooth muscles has an inhomogeneous distribution of membrane proteins. Regions that correspond structurally to membrane caveolae (Anderson et al., 1992; Fijimoto et al., 1992, 1994; Rothberget al., 1992) have been reported to contain higher densities of the plasmalemma Ca++ pump (Fujimoto et al., 1994), proteins which are similar to 1,4,5-inositol triphosphate receptors (Fujimoto et al., 1992), a tyrosine kinase that is a member of the Src family of nonreceptor kinases (Stefanova et al., 1991) than the rest of the plasmalemma. Earlier these caveolae were suggested to be sites of Ca++ binding (Popescu et al., 1974). Caveolae are structurally very close to peripheral SR in smooth muscle (Gabella, 1971; Garfield and Somlyo, 1986). We have found that caveolin enriched membrane fractions from canine airway smooth muscle, which has the ability to refill Ca++ stores by a path requiring functioning L-Ca++ channels but not a functioning SR Ca++pump (Bourreau et al., 1991, 1993), are enriched in L-Ca++ channels, but not in IP3 receptors (Darby et al., 1997) and are immunoreactive for calsequestrin (Darby P and Daniel EE, unpublished data). The association between caveolin and calsequestrin is surprising because calsequestrin is believed to be confined to the interior of the SR vesicles and requires confirmation from coimmunoprecipitation studies, not using detergent insolubility. We suggest that caveolae and SR may be connected through L-type Ca++ channels (indirectly by way of a preferred pathway or compartment, not in diffusion equilibrium with the general cytosol). As in this study, in canine airway muscle, sustained CCh contraction in Ca++-free solution seemed to mobilize Ca++ from an extracellular site as nifedipine or higher EGTA concentrations completely abolished, but BayK enhanced, this response (Montano et al., 1996).
Refilling of Ca++ stores.
Therefore, the sustained phase of the CFCC or CCh response involved Ca++ release from Ca++ stores normally refilled by Ca++pumps. However, even if SR Ca++ pumps were inhibited, L-Ca++ channel opening enhancement restored Ca++ refilling of CPA-insensitive stores. Ca++from these stores was presumably released by CCh-initiated IP3 formation and contributed to the sustained phase of the CFCC. In canine LES, Ca++ entrance through L-Ca++ channels and release of Ca++ from intracellular Ca++ stores in LES could not be given independent responsibilities for the initial and sustained phases of CFCCs because the Ca++ influx and Ca++ release were both contributing and interacting during both phases;i.e., enhanced Ca++ entrance, enhanced Ca++ release and enhanced release enhanced Ca++entry.
Conclusions.
In canine LES, several potential Ca++sources may support agonist-induced contraction: 1) a general extracellular source, free in the extracellular space; 2) a special extracellular source located near the PM from which Ca++enters through L-Ca++ channels during a CFCC, 3) an intracellular Ca++ store refilled by Ca++ pumps and 4) an intracellular Ca++ store not requiring filling by Ca++ pumps but dependent on sufficient Ca++influx through L-Ca++ channels from extracellular source 2 (above). Our data allow the possibility that 3) and 4) involve the same physical compartment filled by different mechanisms under different conditions. LES spontaneous tone, however, is solely dependent on Ca++ entry from the general extracellular solution. Contraction to CCh ultimately requires Ca++ entrance through L-Ca++ channels from an extracellular site to refill Ca++ stores.
Footnotes
-
Send reprint requests to: Dr. E. E. Daniel, Professor Emeritus, Department of Biomedical Sciences, McMaster University, Faculty of Health Sciences, 1200 Main St. W., Hamilton, Ontario L8N 3Z5, Canada.
-
↵1 This study was supported by a grant from the Medical Research Council of Canada.
-
↵2 Current address: Playfair Neurosciences Unit, The Toronto Hospital (Western Division), 399 Bathurst St., Toronto, Ontario M5T 2S8, Canada.
-
A.M.S. was a recipient of a Research/Travel Awards for abstracts of this work which were supported by the AGA Foundation at Digestive Diseases Week (New Orleans, LA, 1994) and the XIIth International Congress of Pharmacology (Montreal, Canada, 1994). A.L. was supported by the AGA Foundation as summer Student for part of this work.
- Abbreviations:
- ACh
- acetylcholine
- BayK
- BayK 8644
- KCa
- Ca++-activated K+ channel
- CFCC
- Ca++ free contractions to carbachol
- CCh
- carbachol
- CPA
- cyclopiazonic acid
- DAG
- diacyl glycerol
- DMSO
- dimethylsulfoxide
- GE
- gastroesophageal
- L-Ca++
- L-type Ca++
- LES
- lower esophageal sphincter
- NIF
- nifedipine
- CPLC
- phospholipase
- PSS
- physiological saline solution
- PM
- plasma membrane
- C-PKC
- protein kinase C
- SR
- sarcoplasmic reticulum
- SBB
- superficial buffer barrier
- TPSS
- tone in physiological saline solution
- [Ca++]i
- intracellular Ca++
- STOC
- spontaneous transient outward currents
- IP3
- inositol 1,4,5-trisphosphate
- PLC
- phospholipase C
- EGTA
- ethylene ethylene glycol-bis (b-aminoethyl ether)-N,N,N′,N′-tetraacetic acid
- ANOVA
- analyses of variance
- Received December 9, 1997.
- Accepted May 28, 1998.
- The American Society for Pharmacology and Experimental Therapeutics