Search

Swelling-activated or volume-sensitive Cl− currents are found in numerous cell types and play a variety of roles in their function; however, molecular characterization of the channels is generally lacking. Recently, the molecular entity responsible for swelling-activated Cl−current in cardiac myocytes has been identified as ClC-3. The goal of our study was to determine whether such a channel exists in smooth muscle cells of the canine colon using both molecular biological and electrophysiological techniques and, if present, to characterize its functional and molecular properties. We hypothesized that ClC-3 is present in colonic smooth muscle and is regulated in a manner similar to the molecular entity cloned from heart. Indeed, the ClC-3 gene was expressed in colonic myocytes, as demonstrated by reverse transcriptase polymerase chain reaction performed on isolated cells. The current activated by decreasing extracellular osmolarity from 300 to 250 mosM was outwardly rectifying and dependent on the Cl− gradient. Current magnitude increased and reversed at more negative potentials when Cl− was replaced by I− or Br−. Tamoxifen ([Z]-1-[p-dimethylaminoethoxy-phenyl]-1,2-diphenyl-1-butene; 10 μM) and DIDS (100 μM) inhibited the current, whereas 25 μM niflumic acid, 10 μM nicardipine, and Ca2+ removal had no effect. Current was inhibited by 1 mM extracellular ATP in a voltage-dependent manner. Cl− current was also regulated by protein kinase C, as phorbol 12,13-dibutyrate (300 nM) decreased Cl− current magnitude, while chelerythrine chloride (30 μM) activated it under isotonic conditions. Our findings indicate that a current activated by hypotonic solution is present in colonic myocytes and is likely mediated by ClC-3. Furthermore, we suggest that the ClC-3 may be an important mechanism controlling depolarization and contraction of colonic smooth muscle under conditions that impose physical stress on the cells.

The patch-clamp technique was used to determine the ionic conductances activated by ATP in murine colonic smooth muscle cells. Extracellular ATP, UTP, and 2-methylthioadenosine 5′-triphosphate (2-MeS-ATP) increased outward currents in cells with amphotericin B-perforated patches. ATP (0.5–1 mM) did not affect whole cell currents of cells dialyzed with solutions containing ethylene glycol-bis(β-aminoethyl ether)- N, N, N′, N′-tetraacetic acid. Apamin (3 × 10−7M) reduced the outward current activated by ATP by 32 ± 5%. Single channel recordings from cell-attached patches showed that ATP, UTP, and 2-MeS-ATP increased the open probability of small-conductance, Ca2+-dependent K+ channels with a slope conductance of 5.3 ± 0.02 pS. Caffeine (500 μM) enhanced the open probability of the small-conductance K+ channels, and ATP had no effect after caffeine. Pyridoxal phosphate 6-azophenyl-2′,4′-disulfonic acid tetrasodium (PPADS, 10−4 M), a nonselective P2 receptor antagonist, prevented the increase in open probability caused by ATP and 2-MeS-ATP. PPADS had no effect on the response to caffeine. ATP-induced hyperpolarization in the murine colon may be mediated by P2y-induced release of Ca2+ from intracellular stores and activation of the 5.3-pS Ca2+-activated K+ channels.

We have determined that serum source plays a critical role in optimizing conditions for an organ culture model of canine proximal colon. Previous studies using equine serum in the medium have shown that some properties of canine colonic smooth muscle can be maintained in organ culture. However, many characteristics of the tissue were altered by the culture conditions. The aims of the present study were to determine whether serum isolated from canine blood would improve the preservation of physiological properties of canine proximal colon in organ culture. Strips of canine colonic smooth muscle were cultured in 10% canine serum medium, and electrical, mechanical, morphological, and molecular analyses were performed after 0, 3, and 6 days in culture. Unlike organ culture in equine serum, in which Na+-K+-adenosinetriphosphatase (Na+-K+-ATPase) expression declined, culture in canine serum maintained Na+-K+-ATPase expression, and resting membrane potential of smooth muscle cells along the submucosal surface of the circular muscle in cultured tissue remained unchanged during the culture period. Increased sensitivity in the contractile response to acetylcholine, previously observed with tissues cultured in equine serum, was not observed. However, the mechanical performance of the muscle (maximal contractile activity) declined over time in culture. Ultrastructural organization of cellular organelles and myofilaments remained intact in the majority of cells; however, some cells possessed regions devoid of contractile filaments. The results of these studies suggest that organ cultured strips of smooth muscle may provide a useful tool for evaluating electrical and mechanical events in conjunction with molecular analysis of functional components.

We have cloned cDNAs encoding the alpha- and beta-subunits of a large-conductance Ca(2+)-activated K+ channel (BK channel) from canine colonic smooth muscle (cslo-alpha and cslo-beta). Nucleotide sequence homology of cslo-alpha with mslo and dslo suggests that it is the canine homologue of these genes. The carboxy-terminal end of the protein is the most diverse between species, and we have also found alternative exons in cslo-alpha in this region. We have identified a unique splice site in the carboxy-terminal region of cslo-alpha, which we term site 5. Northern analysis demonstrates expression of both alpha- and beta-subunits in all canine vascular and visceral smooth muscles tested. Expression of alpha-1 alone and alpha + beta-subunit cRNA in Xenopus oocytes results in a Ca(2+)- and voltage-dependent conductance. The activity of alpha/beta-channels, measured as either changes in the voltage of half-maximal activation (V0.5) in open probability (NP0) or in the normalized conductance (G/Cmax), was more sensitive to [Ca2+]free than channels composed of the alpha-subunit alone. Neither alpha- nor alpha/beta-channels expressed in membrane patches of Xenopus oocytes were found to be regulated by protein kinase G.

Regulation of Ca2+ currents by cyclic nucleotide-dependent mechanisms was studied in circular muscle cells isolated from canine proximal colon. Whole cell Ca2+ currents were recorded at 32 degrees C with the use of amphotericin B-perforated patches. The effects of several agents known to increase levels of adenosine 3',5'-cyclic monophosphate (cAMP) were tested. Vasoactive intestinal peptide (VIP) and isoproterenol (10(-7) to 10(-5) M) increased Ca2+ current in a concentration-dependent manner. Forskolin (10(-7) M) and dibutyryl cAMP (10(-6) to 10(-5) M) also increased Ca2+ current. Higher concentrations of forskolin (10(-6) to 10(-5) M) caused inhibition of Ca2+ current. Low concentrations (10(-5) to 10(-7) M) of dibutyryl cAMP or 8-bromo-cAMP caused concentration-dependent enhancement in Ca2+ current, and these effects were reversible on washout of the cAMP analogues. When the concentration of cAMP analogues was increased (10(-3) to 10(-4) M), we observed inhibition of Ca2+ current similar to the effects of forskolin. Membrane-permeable analogues of guanosine 3',5'-cyclic monophosphate produced exclusively inhibitory effects. The nonspecific protein kinase inhibitor H-7 (up to 60 microM) failed to block the effects of VIP, isoproterenol, and forskolin, and it produced inhibitory effects on Ca2+ current, independent of agonist stimulation. The data suggest that low levels of cAMP may, via phosphorylation by protein kinase A, enhance L-type Ca2+ current, but higher concentrations of cAMP may “cross over” and activate protein kinase G. Phosphorylation by protein kinase G appears to produce a dominant inhibition of Ca2+ current.

Intracellular Ca2+ concentration ([Ca2+]i) plays a central role in regulating tone and contractility of smooth muscle cells. In contrast to the “classic” model of electromechanical coupling where membrane potential determines [Ca2+]i, it is now well established that [Ca2+]i in turn may also affect membrane potential by modulating open probabilities of ion channels. Activation by [Ca2+]i of large-conductance K+ channels, Cl- channels, and nonselective cation channels has been described, as well as block of delayed rectifier K+ channels by [Ca2+]i and [Ca2+]i-induced inactivation of Ca2+ channels. Therefore, a network consisting of positive- and negative-feedback loops regulates [Ca2+]i as well as membrane potential. In this context, we review the properties of Ca(2+)-dependent ion channels and their functional role in vascular and visceral smooth muscles. Any alteration of the “Ca2+ sensitivity” of ion channels is expected to have a profound effect on the reciprocal relationship between membrane potential and [Ca2+]i. Already several molecular factors determining Ca2+ sensitivity of Ca(2+)-activated K+ channels have been identified. We provide a working definition for Ca2+ sensitivity.

Wortmannin, an inhibitor of myosin light-chain kinase (10-30 microM), completely and irreversibly abolished (in 75% of tissues from canine gastric antrum) phase contractions caused by slow waves with no significant effects on resting membrane potential or the frequency, amplitude, or duration of spontaneous slow waves. Responses to agents that normally cause hyperpolarization (cromakalim, sodium nitroprusside, and forskolin) were unaffected by wortmannin treatment. It was also possible to study the excitatory effects of agents and conditions that normally result in loss of intracellular impalements: 1) elevated extracellular K+ concentrations altered membrane potential close to values predicted by the Nernst equation, and 2) high concentrations of acetylcholine produced depolarization and rapid oscillations in membrane potential coincident with contractile activity. Cholinergic increases in myosin light-chain phosphorylation and contractions were partially blocked by wortmannin. In canine antrum, wortmannin inhibition of contraction was irreversible, although in other tissue types, partial recovery of contractions was observed when wortmannin was removed. Wortmannin can be a useful agent to investigate the electrophysiology of some smooth muscles when movement might lead to recording artifacts or loss of signal.

The mechanism of tachykinin-induced excitation was studied in isolated colonic muscle cells and intact muscle strips. In whole cell voltage-clamp studies performed at 33 degrees C, neurokinin A (NKA) and substance P (SP) reduced L-type Ca2+ current. NKA and SP activated a cationic current that reversed near 0 mV. This current (INKA or ISP, respectively) had properties similar to the acetylcholine (ACh)-activated nonselective cation conductance (IACh), activated by muscarinic stimulation in other gastrointestinal smooth muscle cells. INKA and ISP were decreased when external Na+ was reduced. In contrast to IACh, INKA and ISP were not facilitated by increases in internal Ca2+, but little or no current was activated by these peptides when extracellular Ca2+ was low. INKA (10(-7) M) and ISP (10(-5) M) were blocked by Cd2+ (5 x 10(-4) M), quinine (10(-3) M), and the tachykinin-receptor antagonist [D-Pro2,D-Trp7,9]SP (10(-5) M). Current clamp recordings and intracellular recordings of intact tissues showed that NKA and SP depolarized the cell membrane, which is consistent with the activation of a nonselective cation conductance. These data suggest that a primary mechanism of the tachykinins is to activate a nonselective cation conductance that leads to depolarization. The increase in Ca2+ entry due to tachykinin stimulation appears to be secondary to the activation of the nonselective cation conductance.

Electrical rhythmicity in the gastrointestinal tract may originate in interstitial cells of Cajal (IC). Development of IC in the small intestine is linked to signaling via the tyrosine kinase receptor, c-kit. IC express c-kit protein, and disruption of c-kit signaling causes breakdown in IC networks and loss of slow waves. We tested whether mutations in steel factor, the ligand for c-kit, affect the development of IC networks. IC were found in the region of the myenteric plexus (IC-MY) in mice with steel mutations (i.e., Sl/Sld) at 5-10 days postpartum, but these cells formed an abnormal network. IC-MY were not observed in adult Sl/Sld animals. IC in the deep muscular plexus (IC-DMP) appeared normal in Sl/Sld animals. Electrical slow waves, normally present in the small intestine, were absent in Sl/Sld animals (10-30 days postpartum). Neural inputs were intact in Sl/Sld animals. Steel factor appears important for the development of certain classes of IC, and IC-MY appear to be involved in the generation of electrical rhythmicity in the small intestine.

The role of Ca(2+)-activated K+ channels (BK channels) in the canine colon was evaluated by testing the effects of charybdotoxin (ChTX) and tetraethylammonium on K+ currents of isolated myocytes and on electrical and mechanical activity of tissue strips. ChTX blocked Ca(2+)-activated outward current [IK(Ca)] in a dose- and voltage-dependent manner. No significant differences in IK(Ca) density, ChTX block, or Ca2+ sensitivity of BK channels were observed between circular and longitudinal myocytes. ChTX (100 nM) blocked 60% of current at +80 mV. Delayed rectifier current was not inhibited by 100 nM ChTX. In the absence of agonists, ChTX did not affect electrical or mechanical activity of circular muscle strips. In the presence of 10(-6) M BAY K 8644 or 10(-6) M acetylcholine, ChTX increased slow-wave duration and amplitude, induced membrane potential oscillations, and potentiated contraction. In unstimulated longitudinal muscle strips, ChTX depolarized the tissue, increased burst duration and spiking frequency, and resulted in an increase in contractions. These results indicate that BK channels are important regulators of colonic motility. In the longitudinal layer, BK channels are involved in setting membrane potential and determine excitability. In the circular layer, ChTX-sensitive channels do not participate in the in vitro basal electrical activity but limit the responses to excitatory agonists. w

The mechanism of muscarinic excitation was studied in colonic muscle strips and isolated cells. In whole cell voltage-clamp studies performed at 33 degrees C utilizing the permeabilized patch technique, acetylcholine (ACh) reduced an L-type Ca2+ current. With K+ currents blocked, depolarization to positive potentials in the presence of ACh elicited outward current. Difference currents showed that ACh activated a voltage-dependent current that reversed at about -8 mV; this current (IACh) had properties similar to the nonselective cation conductance found in other smooth muscle cells. The reversal potential of IACh shifted toward negative potentials when external Na+ was reduced, and the inward current elicited at -70 mV decreased when external Na+ was reduced. IACh was facilitated by internal Ca2+. After the current was activated at a holding potential of -70 mV, depolarizations to -30 to 0 mV elicited influx of Ca2+ via voltage-dependent Ca2+ channels. After repolarization to the holding potential, a large inward tail current was observed. IACh was blocked by Ni2+ and Cd2+ at concentrations of 100 microM or less. Quinine (0.5 mM) also blocked IACh. With the use of the sensitivity of IACh to reduced external Na+ and divalent cations, the role of IACh in responses of intact muscles to ACh was examined. When external Na+ was reduced, ACh failed to increase slow-wave duration, and Ni2+ (50 microM) reversed the depolarization caused by ACh. These data suggest an important role for IACh in the electrical responses of colonic muscles. The contribution of IACh appears to prolong slow waves, which would allow greater entry of Ca2+ and increased force development.

Experiments were performed to test the hypothesis that enteric inhibitory neurotransmission in pyloric muscles is mediated by NO. Junction potentials were recorded with intracellular microelectrodes from cells near the myenteric and submucosal surfaces of the circular muscle layer. Inhibitory junction potentials (IJPs) were apamin sensitive and were reduced by arginine analogues [NG-nitro-L-arginine-methyl ester (L-NAME) and NG-monomethyl-L-arginine (L-NMMA)]. The effects of arginine analogues were reversed by L-arginine. Inhibition of IJPs unmasked excitatory JPs (EJPs) in the myenteric region and increased excitability of cells in the submucosal region. IJPs were also reduced by oxyhemoglobin. As with arginine analogues, reduction in IJPs increased EJP amplitude. Combination of L-NAME and oxyhemoglobin completely blocked IJPs, suggesting that NO, or an NO-containing compound, mediated the enteric inhibitory nerve responses. Exogenous NO hyperpolarized membrane potential, and these responses were also reduced by apamin. The magnitude of the responses to a given dose of NO was similar in cells of the myenteric and submucosal regions, suggesting that relatively smaller IJPs in submucosal cells may be due to a lower density of enteric inhibitory innervation in the submucosal region. The effects of NO were mimicked by 8-bromoguanosine 3',5'-cyclic monophosphate (cGMP) and M & B 22948, a specific cGMP phosphodiesterase inhibitor, suggesting that the hyperpolarization response to NO may be mediated by enhanced production of cGMP. IJPs were also prolonged by M & B 22948. IJPs and NO disrupted normal electrical rhythmicity in cells in the myenteric region. This may provide a basis for inhibitory effects of enteric inhibitory nerve stimulation on sphincter pressure in pyloric canal in vivo.(ABSTRACT TRUNCATED AT 250 WORDS)

Experiments were performed to identify and characterize the types of calcium channels that regulate inward calcium current in canine colonic smooth muscle. Freshly dispersed smooth muscle cells from the circular layer of the canine proximal colon were used. Single-channel currents were measured with 80 mM Ba2+ as the charge carrier. Small-conductance (10 +/- 2 pS, EBa = 46 +/- 11 mV, n = 9) and large-conductance (21 +/- 1 pS, EBa = 52 +/- 3 mV, n = 19) single-channel currents were observed during depolarizing voltage steps positive to -30 mV. Both types of single-channel currents were inhibited by the addition of 10(-6) M nifedipine to the bath solution. The smaller current was infrequently observed and therefore was not further characterized. Open probability (P(o)) of the larger current amplitude was strongly dependent on voltage. Activation curves were well described by a Boltzmann function with half activation occurring at 4 mV, and a 5-mV increase in membrane potential resulted in an e-fold increase in P(o). BAY K 8644 (1 microM) shifted the activation curve to the left while nifedipine (1 microM) resulted in a right shift. Molecular analysis showed that only the C class of Ca2+ channel alpha 1-subunit is expressed in this tissue. Furthermore, only a single splice variant (rbc-II) was observed. The results suggest that a single class of dihydropyridine-sensitive calcium channels regulates inward calcium current in canine colonic smooth muscle cells.

The patterns of electrical activity and neural inputs to the proximal duodenum between the pyloric sphincter and the sphincter of Oddi were studied in muscles of the dog. Smooth muscle cells in the most proximal region were electrically quiescent, but slow waves were recorded in all regions distal to the first few millimeters. Electrical activity was recorded from circular muscle cells near the myenteric and submucosal surfaces of the circular layer, and slow wave activity was similar in both regions. The nature of neural inputs was also characterized. With electrical field stimulation, responses in cells near the submucosal surface were predominantly excitatory junction potentials (EJPs); near the myenteric border responses were either inhibitory junction potentials (IJPs) or biphasic responses (i.e., small EJPs followed by IJPs). EJPs were blocked by atropine. IJPs were nonadrenergic and noncholinergic (NANC), and several experiments suggested that nitric oxide (NO), or a NO-releasing compound, serves as the inhibitory neurotransmitter in this region. Exogenous NO caused hyperpolarization of membrane potential. IJPs and the hyperpolarization response to NO were sensitive to apamin. These data describe the myogenic mechanisms and neurogenic apparatus that appear to regulate motility in the most proximal region of the duodenum.

The role of phosphatidylinositol (PI) turnover in excitation-contraction coupling was investigated in canine antral smooth muscle. Acetylcholine (ACh; 0.1-1 microM) transiently increased tissue levels of inositol 1,4,5-trisphosphate [Ins(1,4,5)P3] and increased the amplitudes of the plateau phase of slow waves and associated Ca2+ transients and phasic contractions. ACh also increased basal concentrations of cytosolic Ca2+ ([Ca2+]c), but these changes were not associated with an increase in resting tension. ATP (0.3 mM) had similar effects on Ins(1,4,5)P3 levels, basal [Ca2+]c, and resting tension. However, in contrast to the effects of ACh, ATP transiently reduced the amplitude of the plateau phase of slow waves and reduced the amplitudes of associated Ca2+ transients and phasic contractions. We investigated the possibility that two products of PI turnover, diacylglycerol (DAG) and Ins(1,4,5)P3, might provide negative feedback to regulate Ca2+ entry during slow waves. 1) DAG is known to activate protein kinase C (PKC). Activation of PKC by phorbol 12,13-dibutyrate (PDBu, 0.5 microM) reduced the amplitude of the plateau phase of slow waves and corresponding Ca2+ transients and phasic contractions. Assay of PKC showed that ACh, ATP, and PDBu stimulated enzyme activity. 2) Ins(1,4,5)P3 is known to increase [Ca2+]c by release of Ca2+ from internal stores. Basal [Ca2+]c was also increased by elevated external K+, ionomycin, thapsigargin, or caffeine. Each of these compounds reduced the amplitude and duration of slow waves. Results suggest that products of PI turnover may provide negative-feedback control of Ca2+ influx during slow waves, tending to reduce the amplitude of phasic contractile activity in gastric muscles. Differences in responses to ACh and ATP can be explained by a G protein-dependent mechanism in which ACh suppresses the voltage dependence of Ca(2+)-activated K+ channels.

Considerable evidence has recently been presented that suggests that nitric oxide (NO) is a nonadrenergic noncholinergic (NANC) neurotransmitter in gastrointestinal tissues. One of the criteria that must be satisfied before this hypothesis can be accepted is that enteric neurons must be shown to contain the enzymatic apparatus necessary to synthesize NO. Specific antibodies have been developed for NO synthase (NOS) isolated from rat cerebellum, and studies have shown that NOS copurifies and colocalizes with NADPH diaphorase activity, a commonly used neural marker. We used antibodies raised against the cerebellar NOS to determine the distribution of NOS-like immunoreactivity (NOS-LI) in enteric neurons of the canine proximal colon. We also tested whether NADPH diaphorase staining would label the population of neurons containing NOS-LI in this species. A subpopulation of neurons in myenteric and submucosal ganglia displayed NOS-LI and were colabeled with NADPH diaphorase. Labeled neurons had morphological characteristics similar to the Dogiel type I morphology. Cryostat sections showed NOS-positive nerve trunks throughout the circular and longitudinal muscle layers, but a high density of NOS-LI was observed within the submucosal pacemaker region, as predicted from physiological studies. These studies provide the first morphological support for the hypothesis that NO serves as a NANC neurotransmitter in the canine colon. The study also shows that the NADPH diaphorase reaction provides a useful method to label cells with NOS-LI.

The regulation of Ca2+ current by intracellular Ca2+ was studied in isolated myocytes from the circular layer of canine gastric antrum. Ca2+ current was measured with the whole cell patch-clamp technique, and changes in cytoplasmic Ca2+ ([Ca2+]i) were simultaneously measured with indo-1 fluorescence. Ca2+ currents were activated by depolarization and inactivated despite maintained depolarization. Ca2+ current inactivation was fit with a double exponential function. Using Ba2+ or Na+ as charge carriers removed the fast component of inactivation, whereas enhanced intracellular buffering of Ca2+ did not remove the fast component. Ca2+ currents were associated with a rise in [Ca2+]i. The decrease in [Ca2+]i following repolarization was exponential, and during the relaxation of [Ca2+]i, Ca2+ current was inactivated. The inward current recovered with a similar time course as the decrease in [Ca2+]i, suggesting that [Ca2+]i regulates the basal availability of Ca2+ channels. These data support the hypothesis that, although [Ca2+]i may influence the resting level of inactivation, it is the "submembrane" compartment of [Ca2+]i that regulates the development of inactivation.

Part of the regulation of gastrointestinal (GI) smooth muscles is provided by nonadrenergic noncholinergic (NANC) nerves. Stimulation of these nerves, either by field stimulation or via neural reflex pathways, elicits hyperpolarization of postjunctional smooth muscle membranes referred to as inhibitory junction potentials and relaxation. The transmitter(s) that mediate NANC inhibitory neural transmission have been a controversial topic for nearly 30 years. Recent evidence suggests that nitric oxide (NO) may serve as a NANC inhibitory transmitter in the GI tract. This hypothesis is supported by the following. 1) Immunohistochemical studies have shown that the enzyme necessary for NO synthesis is expressed in enteric neurons. In vitro studies of muscles from nearly all levels of GI tract have also shown that arginine analogues, which inhibit NO synthesis, reduce inhibitory effects of NANC neurotransmission. Effects of arginine analogues can be restored by addition of excess L-arginine, the substrate for NO synthesis. These data suggest that NO can be synthesized by enteric nerves. 2) Bioassays have demonstrated nerve-evoked release of a substance that has been identified as NO during NANC nerve stimulation. Oxyhemoglobin, known to bind to and sequester NO, also blocks NANC responses. These data suggest that NO is released into extracellular fluid during nerve stimulation. 3) Addition of NO causes rapid hyperpolarization of GI smooth muscle cells and relaxes muscles strips. These effects are similar to NANC nerve responses. NO and electrical field stimulation also increase tissue guanosine 3',5'-cyclic monophosphate, which may be the second messenger involved in NANC responses. 4) Removal of NO is easily accomplished by its rapid spontaneous breakdown in physiological solutions. 5) The pharmacology of NO and the NANC neurotransmitter in many preparations is similar, e.g., oxyhemoglobin blocks responses to NANC nerve stimulation and to exogenous NO. In summary, it would appear that many of the criteria necessary for NO to be considered a neurotransmitter have been satisfied.

Effects of acetylcholine (ACh) and substance P on the electrical and mechanical activities of the circular muscle layer of the canine proximal colon were studied. Because this muscle layer is bordered by two different pacemaker regions, responses from segments containing either a single pacemaker region or no pacemaker region were compared with responses of the complete muscle layer. Concentration-response relationships for ACh and substance P were similar between the various segments, suggesting that receptors for these agonists are expressed throughout the layer. The dominant contractile pattern induced by ACh and substance P in each segment was a 1- to 3-cycle/min rhythm. In a like manner, these agonists also elicited an electrical pattern in which a long-duration slow wave occurred one to three times per minute between short-duration slow waves. Low concentrations of nifedipine (0.01 microM) selectively antagonized the 1- to 3-cycle/min rhythm. In circular muscles with no pacemaker region, ACh (1 microM) caused depolarization, induced oscillations in membrane potential averaging 24 +/- 5 mV in amplitude and 2.9 +/- 0.9 cycles/min in frequency, and generated rhythmic contractions at the same frequency. This "interior" circular muscle was functionally innervated by cholinergic excitatory nerves. Exposure to ACh (1 microM) did not alter the conduction of slow waves through the thickness of the circular layer. In summary, the excitatory neurotransmitters, ACh and substance P, induce a dominant electrical and contractile rhythm throughout the circular muscle layer that is different from the spontaneous rhythms produced at either the myenteric or submucosal border.

Many proteins including ion channels are regulated by phosphorylation. We tested the effect of 10 U/ml catalytic subunit protein kinase A on 260-pS Ca(2+)-activated K+ channels in excised inside-out membrane patches from freshly dispersed smooth muscle cells of the canine proximal colon. At +50 mV with 10(-7) M Ca2+ and -50 mV with 10(-6) M Ca2+, open probability of the channels was increased to 270 +/- 48% of control (n = 12). This increase was due to a shift in voltage-dependent activation by 13.9 +/- 3.2 mV (n = 3) to more negative potentials. Protein kinase A in the absence of ATP had no effect on channel activity (n = 3). Regulation by phosphorylation must be accompanied by dephosphorylation. We tested the effect of two potent inhibitors of protein phosphatases, calyculin A and okadaic acid. Application of 10(-9) to 10(-6) M of each inhibitor in the presence of protein kinase A further increased open probability by up to 250%. Calyculin A appeared to be less effective in increasing open probability than okadaic acid, suggesting that the phosphatase involved is neither type 1, 2A, nor 2B. Calyculin A in the absence of protein kinase A was ineffective. These data suggest that endogenous phosphatases are found in excised membrane patches and that a balance between phosphorylation and dephosphorylation may provide an important control of colonic motility.

The morphology and electrophysiology of the canine distal colon were studied to compare this region with the proximal colon. Many morphological characteristics were similar including the presence of interstitial cells at the submucosal surface of the circular layer. Muscle cells near the submucosal surface had resting membrane potentials (RMPs) of -79 +/- 1 mV, and slow waves were generated in this region. Slow waves had similar waveform characteristics to those of the proximal colon, but rapid oscillations were superimposed on slow waves of some preparations. RMPs and slow waves decreased with distance from the submucosal surface. The latter were not resolvable in the myenteric half of the circular layer. Cells at the myenteric border had RMPs of -49.5 +/- 2 mV and a higher frequency oscillation of 16 min-1. Acetylcholine increased slow-wave amplitude and duration and caused fast oscillations on the plateau phase of slow waves. Isolated circular myocytes were studied with the patch-clamp technique. Cells from the submucosal border displayed voltage-dependent inward and outward currents. With outward currents blocked, the inward current was composed of two components. Nifedipine (10(-6) M) blocked a portion of the inward current but left a substantial transient component. The effect of nifedipine correlated with its effects on tissues, suggesting that two components of Ca2+ current participate in slow waves. These studies describe numerous similarities in the structure and activity of the proximal and distal portions of the colon but also show some potentially important differences between these regions.

We measured free intracellular Ca2+ concentration ([Ca2+]i) and Ca2+ current (ICa) simultaneously in voltage-clamped, indo-1-loaded smooth muscle cells isolated from the circular layer of the canine antrum. Resting [Ca2+]i averaged 144 +/- 20 nM in cells held at -70 mV. Depolarization positive to -50 mV elicited ICa and increased [Ca2+]i. Peak [Ca2+]i occurred between 0 and +10 mV and averaged 372 +/- 48 nM. On repolarization, [Ca2+]i decreased slowly (time constant 2-3 s) and the rate depended on the magnitude of [Ca2+]i. Cells were also voltage clamped with protocols that mimicked the upstroke and plateau phases of slow waves. With simulated plateau potentials of -55 to -45 mV, [Ca2+]i increased transiently as a result of the small transient ICa elicited by the upstroke depolarization. Sustained ICa was of sufficient magnitude with plateau depolarizations positive to -40 mV to cause a secondary rise in [Ca2+]i throughout the plateau phase. These data suggest that at the plateau potential of slow waves in situ, ICa is sufficient to cause a sustained increase in [Ca2+]i. The resulting accumulation of Ca2+ may couple the slow wave plateau to contraction and may increase the open probability of Ca(2+)-activated K channels. The latter may provide the outward current necessary to initiate repolarization.

Electrical slow waves decay in amplitude as they conduct from the myenteric to the submucosal regions of the circular muscle layer in the canine pyloric sphincter. We used the partitioned chamber method to study the passive and active properties of pyloric muscles, and we found that length constants of circular muscles of myenteric region were significantly longer than muscles near the submucosal surface. These data suggested differences in either membrane resistance, junctional resistance, or cytoplasmic resistance. The first parameter was evaluated by measuring time constants in intact tissues and single cells isolated from the submucosal and myenteric regions. Membrane time constants were not different in the two regions, nor were differences found in the input resistances of isolated cells. Morphological studies failed to demonstrate differences in cell diameters in the two regions suggesting that cytoplasmic resistances are similar. These findings suggest that the different cable properties in the two regions may be due to differences in electrical coupling. Morphological examination revealed similar numbers of gap junctions between cells in the two regions, but large differences were noted in the size of muscular bundles. Muscles of the myenteric region were arranged into large, tightly packed bundles, whereas muscles of the submucosal region consisted of small bundles with an extensive extracellular space filled with connective tissue. We suggest that the difference in cable properties may be due to differences in electrical coupling between bundles. These data suggest that submucosal muscles function more like a multiunit smooth muscle, whereas myenteric muscles behave as a single unit.

The effects of cromakalim (BRL 34915) and its optical isomer lemakalim (BRL 38227) were investigated in intact tissue and freshly dispersed circular muscle cells from canine proximal colon. Cromakalim and lemakalim hyperpolarized resting membrane potential, shortened the duration of slow waves by abolishing the plateau phase, and decreased the frequency of slow waves. Glyburide, a K channel blocker, prevented the effect of cromakalim on slow-wave activity. The mechanisms of these alterations in slow-wave activity were studied in isolated myocytes under voltage-clamp conditions. Cromakalim and lemakalim increased the magnitude of a time-independent outward K current, but cromakalim also reduced the peak outward K current. Glyburide inhibited lemakalim stimulation of the time-independent background current. Nisoldipine also reduced the peak outward current, and in the presence of nisoldipine, cromakalim did not affect the peak outward component of current. This suggested that cromakalim may block a Ca-dependent component of the outward current. Lemakalim did not affect the peak outward current. We tested whether the effects of cromakalim on outward current might be indirect due to an effect on inward Ca current. Cromakalim, but not lemakalim, was found to inhibit L-type Ca channels; however, glyburide did not alter cromakalim inhibition of inward Ca current. We conclude that the effects of cromakalim and lemakalim on membrane potential and slow waves in colonic smooth muscle appear to result primarily from stimulation of a time-independent background K conductance. The effects of these compounds on channel activity may explain the inhibitory effect of these compounds on contractile activity.

Previous studies have suggested that the membrane potential gradient across the circular muscle layer of the canine proximal colon is due to a gradient in the contribution of the Na(+)-K(+)-ATPase. Cells at the submucosal border generate approximately 35 mV of pump potential, whereas at the myenteric border the pump contributes very little to resting potential. Results from experiments in intact muscles in which the pump is blocked are somewhat difficult to interpret because of possible effects of pump inhibitors on membrane conductances. Therefore, we studied isolated colonic myocytes to test the effects of ouabain on passive membrane properties and voltage-dependent currents. Ouabain (10(-5) M) depolarized cells and decreased input resistance from 0.487 +/- 0.060 to 0.292 +/- 0.040 G omega. The decrease in resistance was attributed to an increase in K+ conductance. Studies were also performed to measure the ouabain-dependent current. At 37 degrees C, in cells dialyzed with 19 mM intracellular Na+ concentration [( Na+]i), ouabain caused an inward current averaging 71.06 +/- 7.49 pA, which was attributed to blockade of pump current. At 24 degrees C or in cells dialyzed with low [Na+]i (11 mM), ouabain caused little change in holding current. With the input resistance of colonic cells, pump current appears capable of generating at least 35 mV. Thus an electrogenic Na+ pump could contribute significantly to membrane potential.

Studies were performed to characterize the intrinsic innervation of the circular muscle layer of the canine pylorus. Cross-sectional strips of muscle were studied with intracellular recording techniques, and junction potentials were elicited with transmural nerve stimulation. Neurally mediated responses were recorded from cells at several points through the thickness of the circular layer. Excitatory junction potentials (EJPs) increased and inhibitory junction potentials (IJPs) decreased in amplitude with distance from the myenteric border of the circular muscle. Atropine blocked EJPs throughout the circular layer, demonstrating that excitatory inputs are primarily cholinergic. The gradient in IJP amplitude persisted after blockade of EJPs. Three components of IJPs were identified: 1) a fast, apamin-sensitive component that reached a peak and decayed within approximately 1 s; 2) a slower, apamin-insensitive component that reached a peak within 800 ms but decayed slowly over 5 s; and 3) a very slow component that reached a maximum in 7-10 s. Junctional potentials affected the pattern of myogenic electrical activity. Transmural stimulation could evoke premature slow waves in the myenteric portion of the circular layer but when excitatory inputs were blocked, IJPs greatly reduced the amplitude of slow waves. EJPs elicited action potentials in submucosal portion of circular muscles, and IJPs hyperpolarized these cells. The influence of intrinsic nerves on contractile patterns of pyloric muscles was also characterized. These data demonstrate that a neuromuscular apparatus exists within the gastroduodenal junction for 1) local regulation of slow waves and 2) independent control of the myenteric and submucosal regions of the circular layer.

By using isolated gastric muscle cells of Bufo marinus, others have studied the mechanism of action of muscarinic agonists and substance P (SP). To compare responses of isolated cells with those of intact muscles, we have studied the effects of acetylcholine (ACh) and SP on membrane potentials of circular muscle cells in strips of intact muscle from the toad gastric corpus region. These cells had average resting potentials of -69 +/- 0.7 mV. Membrane potential rhythmically depolarized, producing slow waves at an average frequency of 1/min and average amplitude of 25 +/- 2.2 mV. The major effect of ACh (10(-7) to 10(-4) M) was chronotropic; the frequency of slow waves was increased by 88 +/- 11% by 10(-6) M ACh. The amplitudes and rates of rise of slow waves were decreased by ACh. SP had effects similar to ACh; its major effect was chronotropic. The data suggest that ACh and SP primarily affect the pacemaker mechanism in gastric muscles. Since rhythmicity is apparently not expressed in isolated gastric myocytes, it is possible that this effect of these agonists may have been missed in studies of dispersed cells. Our data suggest that the excitatory effects of ACh and SP on contractions may be due to summation of Ca2+ signals, a partial tetanus-like effect.

The effects of membrane potential on the waveforms and propagation of slow waves were tested using circular muscles of the canine colon. Studies were conducted with intracellular recording techniques on cross-sectional strips of canine proximal colon. Circular muscle cells near the submucosa generated slow waves that decayed in amplitude as they spread through the circular layer. The membrane potentials of cells were less negative as a function of distance from the submucosal border. Cells near the submucosa were depolarized with elevated external K+ and electrical pulses using the partitioned chamber technique. The waveforms of depolarized submucosal cells were compared with events recorded from cells in the bulk of the circular layer. The waveform changes caused by experimental depolarization were different from the changes in waveform that occur during propagation, suggesting the latter are due to a different mechanism than depolarization. The effects of the membrane potential on syncytial input resistance and length constant were also evaluated. The results of these studies are consistent with the hypothesis that slow-wave propagation across the circular layer in canine proximal colon occurs passively.

Acetylcholine (ACh) increases the amplitude and duration of colonic electrical slow waves. This suggests that ACh either increases an inward current or suppresses an outward current. The latter hypothesis was tested in whole cell voltage-clamp experiments performed on freshly dispersed smooth muscle cells from canine proximal colon. Addition of ACh (10(-5) M) to solutions bathing cells reduced time-dependent outward currents elicited by depolarizing test pulses in the range of -45 to +30 mV. Analysis of tail currents showed that ACh caused a 10- to 15-mV positive shift in voltage-dependent activation. When cells were pretreated with 10(-6) M nifedipine to abolish the Ca2+-dependent component of the outward current, the reduction of outward current by ACh was blocked. Single-channel experiments were performed to determine whether ACh had a direct effect on Ca2+-activated K channels. ACh, 10(-5) M, added to bath and pipette solutions caused a positive shift in voltage-dependent activation in on-cell experiments. This effect of ACh on Ca2+-activated K channels provides a mechanism for the effects of muscarinic, excitatory stimulation of circular muscle of the colon.

Mathematical models based on relaxation oscillators have heavily influenced the terminology and experimental designs of investigations in gastrointestinal motility for nearly two decades. Relaxation oscillator equations have been used to stimulate the electrical activities of the esophagus, stomach, small intestine, colon, and rectosigmoid region. It has been suggested that many attributes of gastrointestinal electrical activity cannot be adequately explained by classic "core-conductor" or "cable" models of excitation and conduction. This article critically reviews the relaxation oscillator model and provides an explanation for each of the putative inadequacies of core-conductor theory. Furthermore, we question whether relaxation oscillator equations are able to simulate the waveforms of gastrointestinal slow waves, alterations in waveform in response to drugs or electrical stimulation, patterns of slow-wave activity when stimulated at physiological frequencies, prolonged periods of constant resting membrane potential between gastric slow waves and electrotonic spread into inactive regions. We conclude that the relaxation oscillator equations do not fully describe gastrointestinal electrical activity; excitation and propagation can be modeled by a theory that provides for morphological features, ionic conductances, and other elements included in the cable equations.