INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

NITRIC OXIDE (NO) hyperpolarizes many types of smooth muscle and causes relaxation (14). The ionic mechanisms by which NO hyperpolarizes smooth muscle, however, are not fully understood. NO donors activate multiple types of K⁺ channel (12) and whole cell K⁺ currents in smooth muscle that are sensitive to either tetraethylammonium (TEA) (16), apamin and quinine (9), or 4-aminopyridine (4-AP) (23). Moreover, in muscle strips, the hyperpolarization elicited by NO donors, such as sodium nitroprusside, 3-morpholinosydnonimine hydrochloride, or S-nitrosothiols is partially inhibited by apamin (3, 10), TEA and charybdotoxin (16), and quinine (3). Nitrergic inhibitory junction potentials (IJPs) in the opossum esophagus, however, are not blocked by TEA (up to 20 mM) (10), apamin (3, 4, 6), glibenclamide (8), or 4-AP (3), suggesting that these slow IJPs are not generated by the opening of Ca²⁺-activated, ATP-sensitive, or delayed rectifier K⁺ channels, respectively. Quinine suppresses nitrergic IJPs (3) but only at concentrations that block cation and Cl channels (7).

In previous studies in opossum esophageal smooth muscle (4) and in the guinea pig ileum (5), we reported that the nitrergic slow IJP is caused by the suppression of a resting Cl conductance. This conclusion was based on the effects of Cl substitution and putative Cl channel blockers on the resting membrane potential and on slow IJPs in muscle strips from these tissues (4, 5). Moreover, a Ca²⁺-activated Cl current in the circular muscle cells of the opossum esophagus has recently been characterized (22). The purpose of the present study was to identify Cl channel current in esophageal circular smooth muscle cells using cell-attached patch-clamp recordings and to examine the effect of a NO donor, diethyenetriamine-NO (DETA-NO) (15), and cGMP on these currents. These studies reveal that DETA-NO and 8-bromo-cGMP (8-BrcGMP) both suppress a Ca²⁺-stimulated Cl current and provide strong support for the view that nitrergic IJPs may be mediated by closure of a Cl conductance.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Isolation of esophageal smooth muscle cells. Opossums were killed by lethal injection of pentobarbital sodium (40 mg/kg ip) in accordance with guidelines of the Animal Studies Committee, West Roxbury Veterans Affairs Medical Center. After a midline incision below the sternum, the lower esophagus was removed and placed in modified Hanks' solution containing 10 µM added Ca²⁺. Single smooth muscle cells were prepared as described previously, using collagenase and trypsin digestion of tissue pieces (22).

Patch-clamp recordings. Aliquots of smooth muscle cells were placed in the cell chamber on an inverted microscope (Olympus) and allowed to adhere to the glass surface. The cell chamber was then perfused continuously (0.5 ml/min) with high-K⁺ physiological solution of the following composition (in mM): 150 KCl, 1 MgCl₂, 10 HEPES, 5 D-glucose, 1 CaCl₂, and 1.38 EGTA. The pH was adjusted to 7.2 with 10 M KOH. The Ca²⁺ concentration ([Ca²⁺]) of this solution was calculated to be ~150 nM using a computer program and known binding constants between EGTA and Ca²⁺ (Eqcal; Biosoft). Patch pipettes were drawn from borosilicate capillary glass (Kimax 51 no. 34502; Fisher) on a programmable puller (Sutter P80; Novato) and fire polished (Narishige) to have resistances of 5-10 M when filled with the standard pipette-filling solution of the following composition (in mM): 150 NaCl, 2.5 KCl, 10 HEPES, 5 D-glucose, and 2 MgCl₂. The pH of this solution was adjusted to 7.2 with NaOH. The Ca²⁺-stimulating solution (CSS) consisted of high-K⁺ solution containing 1 µM Ca²⁺ and 10 µM A-23187. Ramp voltages (+50 to 100 mV, over 4 s) were delivered, and currents were recorded using an Axopatch 200A amplifier (Axon Instruments) and digitized using a Labmaster analog-to-digital converter coupled to a Pentium PC running pClamp 6.02 software (Axon Instruments). Currents were filtered at 1 kHz, and data were analyzed using pClamp software. Liquid junction potentials were canceled prior to seal formation. Junction potentials between the pipette and bathing solutions were less than 5 mV (as measured separately using a 3 M KCl agar bridge), and the reveral potentials (E_rev) have not been corrected for these values. Prior to averaging, ramp currents were corrected for a linear leak current, which was recorded from cell-attached patches in low-Ca²⁺ (150 nM) bathing solutions. These Ca²⁺-insensitive currents were assumed to represent nonionic current flow across the seal resistance. All recordings were obtained at room temperature (22-24°C).

Drugs. DETA-NO (Research Biochemicals), which has a half-life of >20 h at pH 7.4 (15), was dissolved directly in the perfusate. 8-BrcGMP (Sigma) was also dissolved in the perfusate, and tetrapentylammonium chloride (TPA; Aldrich or Sigma) was made up as an aqueous stock solution (10¹ M). Stock solutions of A-23187 (10² M; Molecular Probes), DIDS (10¹ M; Sigma), and LY-83583 (2 × 10¹ M; Calbiochem) were dissolved in pure DMSO.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Ca²⁺-stimulated currents in esophageal circular smooth muscle cells. Smooth muscle cells were perfused initially with a high-K⁺, low-Ca²⁺ (150 nM) physiological solution (PS) to null the resting potential. Currents were recorded using pipettes filled with low-K⁺ (2.5 mM) solution from cell-attached patches in response to ramp hyperpolarizations (+50 to 100 mV over 4 s) from a holding potential of 0 mV. Fifteen consecutive ramp currents, generated at 5-s intervals, were digitally averaged to obtain a mean ramp current for analysis. Contamination from K⁺ channel currents was minimized by adding TPA (200 µM) (2) to all the solutions. Under these conditions, the background current reversed at +4.2 ± 1.2 mV (n = 8) (see Fig. 2A).

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Activation of an inward "leak" current by Ca2+-stimulating solution (CSS), a high-K+ solution containing 1 µM Ca2+ and 10 µM A-23187, in a smooth muscle cell of opossum esophagus. Shown are ramp currents averaged from 15 consecutive traces at various times after application of CSS recorded from same cell. Current achieved maximal amplitude at 12 min. Note that current was linear between 50 and +50 mV but showed some rectification at more negative potentials. Cell was bathed in high-K+ physiological solution containing 150 nM Ca2+ and tetrapentylammonium chloride (200 µM) to block K+ channels. Ionic current was recorded from a cell-attached patch in response to ramp hyperpolarizations from +50 mV to 100 mV, over 4 s.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Ca2+ sensitivity of Ca2+-stimulated leak current (IL-Ca). A: ramp currents averaged from 15 consecutive traces recorded after equilibration at different levels of Ca2+ and A-23187 (10 µM) from the same cell. Note that amplitude of currents increased with increasing Ca2+ concentration in the bathing solution. Currents were recorded from the same cell-attached patch in response to ramp hyperpolarizations in the presence of CSS. B: normalized amplitude of averaged ramp currents measured at +50 mV (), 50 mV (), and 100 mV () plotted as a function of [Ca2+] in the CSS. Data points are means ± SE of 4 patches (i.e., n = 4 cells) and were fitted with the Hill equation, which yielded EC50 values of 1.6, 0.6, and 1.2 µM at +50, 50, and 100 mV, respectively, with the slope fixed at unity. Data indicate that Ca2+-stimulated current is activated by physiological levels of Ca2+ and is not appreciably voltage dependent.

Effect of Cl and Na⁺ substitution. The presence of fixed intracellular negative charges is expected to result in a lower intracellular than extracellular [Cl], establishing a negative Cl Nernst potential (E_Cl). Because the E_rev of I_L-Ca lies between the K⁺ equilibrium potential (E_K; ~100 mV) and 0 mV, the ionic nature of I_L-Ca may be anionic or of a mixture of anions and cations. To test whether I_L-Ca is carried by Cl, we reduced the [Cl] in the pipette solution to 21.5 mM by equimolar replacement with gluconate. Under these conditions, the E_rev of the background current recorded in 150 nM Ca²⁺ (+7 ± 4 mV, n = 4) did not differ significantly (P > 0.05) from the corresponding background current recorded under a normal Cl gradient (+4.2 ± 1.7 mV, n = 8). In the presence of reduced extracellular [Cl], however, the E_rev of the Ca²⁺ (1 µM)-stimulated I_L-Ca was shifted significantly positive to +16.6 ± 3.4 mV (n = 4) (P < 0.05) (Fig. 3, traces i and iii). The relative shift in E_rev of I_L-Ca was approximately one-half the value predicted for a Cl current from the Nernst equation. These data suggest that although I_L-Ca is carried largely by Cl, cations such as Na⁺ and K⁺ may also contribute to this current and account for the discrepancy between the predicted and actual changes in E_rev of I_L-Ca.

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Representative averaged ramp currents recorded from cell-attached patches showing effect of external Cl and Na+ substitution on IL-Ca. Reduction of external (pipette) Cl concentration to 21.5 mM by equimolar replacement with gluconate shifted the reversal potential (Erev) of IL-Ca ~30 mV positive from 14 mV (trace i) to +16 mV (trace iii). Substitution of Na+ with equimolar (150 mM) Tris+ suppressed the amplitude of IL-Ca but failed to shift Erev (trace ii), suggesting that IL-Ca is carried by Cl. Ramp currents were recorded in CSS from 3 different patches (cells), and the averaged current from 15 traces from each patch is plotted as a function of ramp potential.

Effect of DETA-NO on I_L-Ca. To investigate whether NO can inhibit I_L-Ca, we tested the effect of DETA-NO, a stable NO · donor (8, 15). After activation of I_L-Ca, DETA-NO (200 µM) was perfused continuously through the cell chamber, resulting in a marked decrease in the slope of ramp currents (Fig. 4A). I_L-Ca recovered upon washout of DETA-NO. In five cells tested, I_L-Ca was decreased by 50 ± 11% at +50 mV, 65 ± 16% at 50 mV, and 59 ± 15% at 100 mV. In the cell depicted in Fig. 4A, this inhibition was prevented by pretreatment with LY-83583 (200 µM), an inhibitor of soluble guanylate cyclase (19), suggesting that the action of NO is mediated by cGMP.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Suppression of IL-Ca by diethylenetriamine-nitric oxide adduct (DETA-NO) and cGMP. A: suppression of inward IL-Ca (control) by DETA-NO (200 µM) and inhibition of this effect by concomitant application of LY-83583 (200 µM) and DETA-NO. Ramp currents are averages of 15 consecutive traces recorded under different conditions from a cell-attached patch. B: suppression of IL-Ca by 8-bromo-cGMP (8-BrcGMP; 200 µM), in a different cell. Ramp currents were generated as in A. Addition of 8-BrcGMP (200 µM) to the bathing solution inhibited IL-Ca. Pretreatment with LY-83583 (200 µM), an inhibitor of guanylate cyclase, failed to prevent suppression of this current by concomitant addition of 8-BrcGMP.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

In the present study we have demonstrated that smooth muscle cells of the opossum esophageal circular muscle express a Ca²⁺-stimulated Cl conductance that is suppressed by NO via stimulation of guanylate cyclase. Replacement of extracellular Na⁺ with Tris⁺ failed to shift E_rev of this current significantly. Substitution of extracellular Cl with gluconate, shifted the E_rev of I_Cl-Ca positively. Although this shift was approximately half the value expected for a purely Cl-selective conductance, this is consistent with the known poor selectivity of many Cl channels, including small-conductance Ca²⁺-activated Cl channels in bovine pulmonary artery endothelial cells (18, 22).

In cell-attached patches, I_Cl-Ca was stimulated by levels of Ca²⁺ that are achieved in intact tissue. Channel activity, however, had a tendency to decrease after patch excision despite the presence of the same high level of Ca²⁺ on the cytoplasmic surface of patches, indicating that soluble intracellular mediators may be involved in activation of I_Cl-Ca by Ca²⁺. This phenomenon has been reported previously for large-conductance Cl channels in chicken myotubes (20) and also for small-conductance Ca²⁺-activated Cl channels in vascular smooth muscle (11). In our recordings, ramp currents consisted mainly of poorly resolvable openings of small-conductance channels. However, channels with apparently large (~300 pS) conductances were also stimulated with high Ca²⁺ (>1 µM) in the same patches with less frequency. Sun et al. (21) have previously described "maxi" Cl channels in inside-out patches from rabbit colonic smooth muscle cells. The large-conductance Cl channels were insensitive to cytoplasmic [Ca²⁺] up to 0.5 mM and may represent a different population of Cl channel (see Ref. 13). Further studies are required to elucidate the single channel properties of I_Cl-Ca in the opossum esophagus.

An important finding in our present study has been that the I_Cl-Ca is suppressed by the NO · donor DETA-NO (15), and this effect is antagonized by LY-83583, a blocker of cytosolic guanylate cyclase (19). This suggests that NO acts to suppress I_Cl-Ca via intracellular accumulation of cGMP. This conclusion is further supported by our observation that 8-BrcGMP, a cell-permeable analog of cGMP, also suppressed this Cl current. Although NO has been shown to suppress L-type Ca²⁺-channel currents in esophageal cells (1), which may indirectly lead to suppression of I_Cl-Ca, intracellular [Ca²⁺] in our cells was most likely clamped close to 1 µM with Ca²⁺ ionophore. The suppression of I_Cl-Ca by DETA-NO therefore cannot be explained by inhibition of Ca²⁺-channel currents. Although the mechanism by which cGMP inhibits I_Cl-Ca was not addressed in the present study, inhibition of a similar Ca²⁺-activated current in mouse ileal myocytes by DETA-NO is blocked by pretreatment with H-7, a nonspecific kinase inhibitor (F. Vogalis and R. K. Goyal, unpublished observations). This suggests that cGMP is not the final mediator in the suppression of I_Cl-Ca.

It has been shown recently that in rat cerebral arteries, Cl channels are responsible for the maintenance of membrane potential and myogenic tone (17). In intact esophageal muscle strips, a resting Cl conductance has been previously shown to maintain the membrane potential positive of E_K (5). It is therefore possible that tonic stimulation of I_Cl-Ca by ongoing Ca²⁺ trafficking between the cell membrane and stores at physiological temperatures may contribute to the depolarized resting potential in intact esophageal smooth muscle. Suppression of this current by neurally released NO would then allow the membrane to be hyperpolarized by a resting K⁺ conductance, giving rise to slow IJPs that are recorded in the esophageal smooth muscle and in other visceral smooth muscle preparations. Consistent with this hypothesis are the observations that nitrergic IJP in the opossum esophagus is associated with an increase in membrane resistance and is inhibited by DIDS (4), a blocker of Cl channels (13). In contrast to the opossum esophagus, nerve stimulation in the circular muscle of the guinea pig ileum and the mouse stomach evokes purinergic fast IJPs as well as nitrergic slow IJPs. Fast IJPs are known to be generated by the opening of apamin-sensitive K⁺ channels, whereas slow IJPs may be generated by suppression of a resting Cl conductance (5, 8).

In summary, we have demonstrated that DETA-NO, a NO donor, inhibits a Ca²⁺-stimulated leak current carried by Cl. Inhibition of this current may be an important mechanism by which NO released from nitrergic motor nerves produces smooth muscle hyperpolarization and inhibits electromechanical coupling.