1. Top of page
  2. Abstract
  6. Acknowledgements
  • 1
    The membrane potential of gastrointestinal smooth muscles determines the open probability of ion channels involved in rhythmic electrical activity. The role of Ba2+-sensitive K+ conductances in the maintenance of membrane potential was examined in canine proximal colon circular muscle.
  • 2
    Application of Ba2+ (1-100 μM) to strips of tunica muscularis produced depolarization of cells along the submucosal surface of the circular muscle layer. Significantly higher concentrations of Ba2+ were needed to depolarize preparations from which the submucosal and myenteric pacemaker regions were removed.
  • 3
    Elevation of extracellular [K+]o (from 5·9 to 12 mM) brought membrane potentials closer to EK (the Nernst potential for K+ ions), suggesting activation of a K+ conductance. This occurred at potentials much more negative than the activation range for delayed rectifier channels (Kv).
  • 4
    Forskolin (1 μM) caused hyperpolarization and a leftward shift in the dose-response relationship for Ba2+, suggesting that forskolin may activate a Ba2+-sensitive conductance.
  • 5
    Patch-clamp recordings from interstitial cells of Cajal (ICC) revealed the presence of a Ba2+-sensitive inward rectifier potassium conductance. Far less of this conductance was present in smooth muscle cells.
  • 6
    Kir2.1 was expressed in the circular muscle layer of the canine proximal colon, duodenum, jejunum and ileum. Kir2.1 mRNA was expressed in greater abundance along the submucosal surface of the circular muscle layer in the colon.
  • 7
    These results demonstrate that ICC express a Ba2+-sensitive conductance (possibly encoded by Kir2.1). This conductance contributes to the generation and maintenance of negative membrane potentials between slow waves.

Electrical slow waves are periodic oscillations in membrane potential that occur in smooth muscle cells in many regions of the gastrointestinal (GI) tract of animals and man (e.g. Tomita, 1981). Slow waves consist of a rapid upstroke depolarization, partial repolarization, a plateau phase lasting several seconds and repolarization to the resting membrane potential. The ionic currents involved in generation of slow waves have been characterized in colonic muscles (see Sanders, 1992, for review); however, the conductances responsible for the maintenance of the resting membrane potential (i.e. the maximum level of polarization between slow waves) have yet to be defined.

Membrane potentials of GI muscles determine the open probability of ion channels involved in rhythmic electrical and contractile activity. At the submucosal surface of the circular muscle layer in the canine colon, resting membrane potential is several millivolts positive to the equilibrium potential for potassium, EK (Smith et al. 1987). The membrane potential at the myenteric surface of the circular muscle layer is substantially less negative (averaging about -45 mV; Smith et al. 1987). The negative resting membrane potential in cells near the submucosal surface and the gradient in membrane potential across the circular layer has been attributed to a significant contribution from an electrogenic Na+-K+-ATPase. This is based on the observations that: (i) ouabain abolishes the membrane potential gradient (Burke et al. 1988); and (ii) differential expression of Na+-K+-ATPase mRNA between the submucosal vs. myenteric regions of the circular muscle layer (Burke et al. 1991). Barajas-Lopez & Huizinga (1992) suggested that the effects of ouabain could be due to inhibition of K+ conductances in response to increased intracellular Na+. However, the nature of such a K+ conductance was not resolved.

An inward rectifier K+ conductance could contribute to the negative resting potential and the membrane potential gradient across the circular muscle layer in the colon. Ba2+-sensitive inward rectifier K+ currents have been identified in vascular smooth muscles (Quayle et al. 1993; Knot et al. 1996). These currents are likely to be due to the expression of Kir2.1 mRNA (Bradley et al. 1999), but this conductance has not been reported in GI muscles to date. Although referred to as ‘inward rectifiers,’ these conductances carry small outward currents at potentials positive to EK, and this current contributes to the negative membrane potentials of several cell types, including cardiac muscle, mast cells and neurones (see Jan & Jan, 1997, for review).

In the present study, we examined the hypothesis that a Ba2+-sensitive inward rectifier potassium conductance contributes to the resting potentials of canine colonic muscles. Regional differences in expression of an inward rectifier conductance may contribute to the gradient in resting membrane potential within the circular muscle layer (Smith et al. 1987).


  1. Top of page
  2. Abstract
  6. Acknowledgements

Tissue dissection

Adult mongrel dogs of either sex were killed by an overdose of pentobarbital sodium (30 mg kg−1). The use and treatment of animals was approved by the Institutional Animal Use and Care Committee at the University of Nevada. The abdomens were opened along a midline incision and a segment of proximal colon removed. Segments of colon were cut along the mesenteric border and washed of faecal matter with Krebs-Ringer bicarbonate solution (KRB). Segments were placed in a dissecting dish with the serosal surface upwards. Muscle strips (1 mm thick) through the entire tunica muscularis were cut parallel to the circular muscle fibres. The submucosa and mucosa were removed by sharp dissection and the strips were pinned in cross-section in an electrophysiological recording chamber to allow impalement of cells at selected points within the circular muscle layer.

Full-thickness strips were further dissected to produce muscle strips with the myenteric and submucosal surfaces removed (i.e. ‘isolated’ circular muscle, ICM). Thin strips of muscle (50-100 μm) were dissected from the submucosal surface of the circular muscle layer (i.e. submucosal circular muscle, SCM). ICM strips were pinned in the electrophysiological chamber such that cells at either the myenteric (outer ICM) or submucosal (inner ICM) surfaces of these strips could be selectively impaled with microelectrodes.

In experiments in which the effects of Ba2+ were compared in ICM and SCM, the two muscle preparations were dissected from the same full-thickness muscle strip. This ensured that cells impaled in the two preparations were in close proximity to each other in situ. Therefore, the effects of Ba2+ could be tested on a similar population of smooth muscle cells with and without the pacemaker region attached.

Electrophysiological recording

After pinning in the electrophysiological chamber, the muscle strips were allowed to equilibrate in flowing, pre-warmed, pre-oxygenated KRB solution (5 ml min−1) for 3 h. Temperature was monitored with a small thermistor probe placed near the muscle and maintained at 36.0 ± 0.5°C.

Cells were impaled with glass microelectrodes filled with 3 M KCl, with resistances of 20-60 MΩ. A sharp voltage drop of at least -55 mV was taken as an indication of a successful impalement. Changes in membrane potential were measured with a high-input impedance electrometer (Intra 767, World Precision Instruments Inc., Sarasoto, FL, USA), and the output was displayed on a digital storage oscilloscope (VC-6025A, Hitachi). Data were simultaneously recorded with a chart recorder (2200S, Gould), an FM tape recorder (420H, A.R. Vetter & Co., Rebersburg, PA, USA), and an A/D converter (MP100; Biopac Systems, Santa Barbara, CA, USA) interfaced to a PC clone microcomputer running AcqKnowledge 3.03 software (Biopac Systems).

Cell preparation for voltage-clamp experiments

For the preparation of interstitial cells of Cajal (ICC) and smooth muscle cells, SCM (containing ICC) and ICM strips were prepared (Fig. 1). Both preparations were cut into small pieces and placed into Ca2+-free Hanks collagenase solution for 23-25 min (for smooth muscle cell) or 29 min (for ICC) at 37°C. Following this incubation, the muscles were rinsed with Ca2+-free Hanks’ solution and agitated using a stirring bar for 5 min. Dispersed cells were collected and added to minimum essential medium (Sigma, USA) supplemented with 0.5 mM CaCl2, 0.5 mM MgCl2, 4.17 mM NaHCO3 and 10 mM Hepes. To enhance observation of ICC, cells were incubated for 30 min at 37°C. ICC were identified as stellate cells with several (> 3) processes from a prominent nuclear region (see Lee & Sanders, 1993). Cells were stored at 4°C and used within 8 h.


Figure 1. Schematic drawing showing preparations used

Strips of canine proximal colon were cut parallel to the circular muscle fibres. Full-thickness preparations consisted of a strip of the entire tunica muscularis. Pinning these strips in a cross-sectional orientation allowed selective impalement of smooth muscle cells near the submucosal surface of the circular muscle layer. Further dissections were performed to produce submucosal circular muscle strips (SCM) and isolated circular muscle strips (ICM). SCM strips contained the submucosal pacemaker region and a thin strip of adjacent circular muscle. ICM strips were devoid of the submucosal pacemaker region, the myenteric region and the longitudinal muscle layer. ICM strips could be pinned with either the inner or outer aspect of the muscle strip facing upward, thus permitting selective impalements of either population of cells.

Download figure to PowerPoint

Voltage-clamp experiments

Conventional dialysed whole-cell voltage-clamp experiments were performed on ICCs and smooth muscle cells at room temperature. Pipette resistance was 1-3 MΩ. Input resistance of smooth muscle cells and ICCs were 87 ± 9 and 98 ± 7 MΩ, respectively, from a 5 mV step depolarization pulse. Experiments were conducted in solutions containing either physiological K+ gradients (5 mM external/140 mM internal) or symmetrical K+ gradients (140 mM external/140 mM internal). Currents were obtained using a ramp protocol, in which cells were held at -80 mV, stepped to 0 mV for 250 ms, stepped to -140 mV for 250 ms, and ramped to 0 mV over 1 min. Currents were amplified using a List EPC-7 amplifier and digitized with a 12-bit A/D converter (TL-1, DMA interface, Axon Instruments). Data were stored on videotape or digitized on-line using pCLAMP software (version 5.5.1, Axon Instruments). Data were filtered at 1 kHz using an 8-pole Bessel filter.

Solutions and drugs

The standard KRB solution used in this study contained (mM): 120.35 NaCl, 5.9 KCl, 15.5 NaH2CO3, 11.5 D-glucose, 1.2 MgCl2, 1.2 NaH2PO4 and 2.5 CaCl2. This solution achieved a pH of 7.4 after equilibration with 97 % O2 and 3 % CO2 at 36.0 ± 0.5°C.

Ca2+-free Hanks’ solution contained (mM): 125 NaCl, 5.36 KCl, 15.5 NaHCO3, 0.336 Na2HPO4, 0.44 KH2PO4, 10 glucose, 2.9 sucrose and 5 Hepes; pH 7.4 with KOH. Ca2+-free Hanks’ collagenase solution also contained ATP (0.11 mg ml−1), collagenase type III (1 mg ml−1; Worthington Biochemical Co.), trypsin inhibitor (2 mg ml−1; Sigma), protease type XIV (0.1 mg ml−1; Sigma) and bovine serum albumin (1 mg ml−1; Sigma). Manganese physiological saline solution contained (mM): 5 KCl, 135 NaCl, 2 MnCl2, 10 glucose, 1.2 MgCl2 and 10 Hepes, adjusted to pH 7.4 with Tris. KCl at 140 mM replaced 5 mM KCl and 135 mM NaCl while recording K+ currents using symmetrical K+ gradients. The composition of the internal (pipette) solution for cell dialysis was (mM): 110 potassium gluconate, 20 KCl, 5 MgCl2, 2.7 K2ATP, 0.1 Na2GTP, 2.5 creatinine phosphate disodium, 5 Hepes and 1 EGTA; pH 7.2 with Tris.

Stock solutions of L-NAME (100 mM), atropine sulphate (10 mM) and BaCl2 (1 M) were prepared in distilled water. Stock solutions of nifedipine (10 mM) and forskolin (10 mM) were prepared in ethanol. Ouabain (10 mM) was prepared as a stock solution in dimethyl sulphoxide (DMSO). All drugs were obtained from Sigma and diluted to the desired concentration in KRB. Intracellular microelectrode experiments included nifedipine (1 μM) in the bathing solution were performed in the dark.

PCR analysis of Kir2 expression

Full-thickness circular muscle was isolated from the proximal colon to evaluate expression of Kir2. For comparison purposes, we also obtained segments of small intestine from duodenum, jejunum and ileum. Kir2 expression was also evaluated in specific regions of the proximal colon by preparing SCM and ICM strips, as described above, and thin strips of circular muscle adjacent to and including the myenteric plexus (myenteric circular muscle, MyCM). Samples of canine cerebellum were also collected and processed to test the fidelity of the PCR primers.

RNA isolation and reverse transcription-polymerase chain reaction (RT-PCR)

Total RNA was prepared from the tissue samples with SNAP Total RNA Isolation Kits (Invitrogen, CA, USA) following the manufacturer's instructions. Polyinosinic acid (20 μg) was added to lysates as a carrier of RNA, to aid in the isolation of RNA from the small tissue samples used (< 20 mg). First strand cDNA was synthesized from RNA preparations by the use of the Superscript II RNase H Reverse Transcriptase kit (Gibco BRL, MD, USA); RNA (1 μg) was reversed transcribed by use of an oligo (dT) 12-18 primer (500 μg μl−1). The primers used were as follows: Kir2.1 (gene accession number L48490), sense nucleotides 864-883, antisense 1170-1189; Kir2.2 (X88461): sense 419-438, antisense 1235-1254; Kir2.3 (S71382): sense 407-426, antisense 1303-1322; β-actin (V01217): sense 2383-2402, antisense 3071-3091.

PCR primers for β-actin were used to assess the viability of RNA samples and could also detect genomic DNA contamination whereby the primers were designed to span an intron in addition to two exons. Complimentary DNA (20 % of the first strand reaction) was combined with sense and antisense primers (20 μM), 1 mM deoxynucleotide triphosphates, 40 mM Tris-HCl (pH 8.3), 100 mM KCl, 3 units of TAQ (Promega, Madison, WI, USA), 1 Ampliwax Gem 100 (Perkin Elmer, Foster City, CA, USA), and RNase-free water to a final volume of 50 μl. PCR was performed in a Perkin Elmer 2400 Thermal Cycler under the following conditions: 32 cycles at 94°C for 15 s, 57°C for 20 s, 72°C for 1 min, and then incubated at 72°C for 10 min. All PCR products were separated by 2 % agarose gel electrophoresis and sequenced using an automated nucleotide sequencer (Model 310, Applied Biosystems, Foster City, CA, USA).

Quantitative PCR (Q-PCR)

Q-PCR was performed by use of the PCR MIMIC Construction Kit (Clontech, CA, USA), which is based upon a competitive PCR approach: non-homologous engineered DNA standards (referred to as PCR MIMICs) compete with target DNA for the same gene-specific primers. PCR MIMICs were constructed for Kir2.1 and β-actin. Competitive PCR was carried out by titration of sample cDNA with known amounts of the desired non-homologous PCR MIMIC constructs; serial dilutions (10-fold and then 2-fold) of these constructs were then added to PCR amplification reactions. Following PCR, products were separated by 2 % agarose gel electrophoresis and the density of the bands was quantified using Molecular Analyst software (BioRad, Hercules, CA, USA). The expression of Kir2.1 (amol μl−1) was expressed relative to the level of the housekeeping gene, β-actin. Kir2.1 expression was normalized against β-actin expression, as the latter did not vary between preparations (1 amol μl−1).

Data analysis

Slow wave parameters were analysed from the original chart recordings. Data are expressed as means ± standard error of the mean (s.e.m.) and plotted using Origin 4.0 software (Microcal, Northampton, MA, USA). Student's paired t tests were used to determine statistical significance (P < 0.05). n refers to the number of muscle strips or cells used in each experiment.

For quantitative PCR studies, RNA was isolated from submucosal, myenteric and circular layers of the proximal colon, as well as from full-thickness circular muscle of the proximal colon, duodenum, ileum and jejunum. The concentration of Kir2.1 cDNA as well as β-actin cDNA was determined on each sample; target DNA concentration was then normalized to β-actin expression. All results are expressed as means ±s.e.m., and n is the number of Q-PCR reactions performed for a given tissue type and primer. Data for quantitative PCR were analysed by one-way analysis of variance (ANOVA) using Graphpad Prism (San Diego, CA, USA) software, and the difference between tissues was illustrated by the Newman-Keuls multiple comparison test; P values less than 0.05 were significant.


  1. Top of page
  2. Abstract
  6. Acknowledgements

Effects of Ba2+ on electrical activity of circular muscle (SCM and ICM)

We examined the effects of low concentrations of Ba2+ to determine the role of inward rectifier K+ channels in colonic muscles. Under basal conditions smooth muscle cells near the submucosal surface in full-thickness strips of muscle (Fig. 1) had resting membrane potentials (RMP) of -70.0 ± 0.9 mV. Slow waves averaged 31.0 ± 3.3 mV in amplitude and 4.0 ± 0.4 cycles min−1 in frequency (n= 5). BaCl2 caused concentration-dependent depolarization of the RMP (10-100 μM) and reduced the amplitude and frequency of slow waves (50 and 100 μM) (Fig. 2). For example, at 50 μM, Ba2+ caused depolarization to -59 ± 1.4 mV (P < 0.05) and reduced slow wave amplitude to 12 ± 7 mV (P < 0.05). At 100 μM Ba2+, membranes were depolarized to -54 ± 1.4 mV (P < 0.005), and slow waves were blocked. Figure 3 summarizes the effects of Ba2+on five preparations. There were no significant differences in the responses of cells near the submucosal surface in full-thickness preparations or in SCM strips (data not shown).


Figure 2. Effects of Ba2+ on electrical activity

Membrane potentials of cells near the submucosal surface of the circular muscle typically had negative membrane potentials between slow waves (A; see text for details). Ba2+ caused concentration-dependent depolarization (B-D) and reduction of (B and C) or block of slow waves (D).

Download figure to PowerPoint


Figure 3. Effects of Ba2+ on slow wave parameters

The graphs show summaries of the effects of Ba2+ on resting potential (A), amplitude (B) and frequency (C) in experiments on 5 muscles from 5 dogs. C here and in others figures, indicates control value. Data are means ±s.e.m.; *P < 0.05.

Download figure to PowerPoint

The slow wave pacemaker lies at the submucosal surface of the circular muscle layer in the canine colon (Smith et al. 1987). This region contains a population of interstitial cells of Cajal thought to be the pacemakers (Berezin et al. 1988; Langton et al. 1989). We also tested the effects of Ba2+ on preparations in which the submucosal pacemaker cells were removed by dissection (ICM). The resting potential of these preparations was significantly less negative than that of full-thickness preparations (-63.0 ± 1.7 mV; n= 7; P < 0.005), as previously reported (Liu & Huizinga, 1993). There was no difference in the resting membrane potentials of cells impaled along the inner or outer surface of ICM strips (data not shown). BaCl2 caused dose-dependent depolarization of ICM, but these preparations were significantly less sensitive to Ba2+ than full-thickness strips or SCM. Approximately 1 mM BaCl2 was needed to produce a depolarization equivalent to the effect of 100 μM Ba2+ on intact muscles. Figure 4 summarizes Ba2+-induced depolarizations recorded from ICM, and compares these data to the effects of Ba2+ on intact muscle strips. Ba2+ had equal effects on cells at the inner and outer surfaces of ICM strips (data not shown).


Figure 4. Comparison of the effects of Ba2+ on full-thickness muscle strips (•, data reproduced from Fig. 3) and ICM preparations (▪)

Resting membrane potential depolarized in ICM preparations when the submucosal pacemaker region and myenteric regions were removed. Ba2+ (1-100 μM) had no significant effect on membrane potential in ICM preparations. RMP was significantly decreased by 250-1000 μM Ba2+ (*P < 0.05). Data are means ±s.e.m. from 14 muscles from 14 dogs.

Download figure to PowerPoint

Regulation of Ba2+-sensitive conductance

The presence of Ba2+-sensitive inward rectifier conductances can be tested by measuring changes in membrane potential in response to increased extracellular potassium ([K+]o; Quayle et al. 1993; Knot et al. 1996). Under control conditions (5.9 mM K+o), cells near the submucosal surface of full-thickness preparations had an average RMP of -71.0 ± 2.3 mV (n= 5). Elevating [K+]o to 8 mM depolarized RMP to -67.0 ± 2.5 mV (P < 0.005). Slow wave amplitude was reduced from 36 ± 3 mV (control) to 33 ± 2.8 mV (P > 0.05), but there was no alteration in the frequency of slow waves (5.0 ± 0.1 and 5.0 ± 0.1 min−1, respectively, in 5.9 and 8 mM KCl). In four of the five preparations this depolarization was preceded by a small, transient hyperpolarization of 1.8 ± 0.4 mV. Further elevations in [K+]o to 12 mM resulted in depolarization to -58.0 ± 1.8 mV (P < 0.05); slow wave amplitude was reduced to 18.0 ± 3.4 mV (P < 0.05) and the frequency was reduced to 4.0 ± 0.2 min−1 (P > 0.05). Figure 5 summarizes the effects of raising [K+]o on the electrical activity of circular muscle cells near the submucosal surface. As [K+]o was elevated, RMP approached EK (calculated from the Nernst equilibrium potential for K+ ions, assuming an intracellular K+ concentration of 140 mM), suggesting an enhancement in the net membrane selectivity for K+ ions.


Figure 5. Summary of the effects of elevated external K+ ([K+]o) on resting membrane potential (RMP)

Full-thickness muscle strips were exposed to three concentrations of [K+]o (5.9-12 mM). A-C, the effects of changes in [K+]o on membrane potential and slow wave activity. D, elevated [K+]o caused significant depolarization (•; *P < 0.05) and brought membrane potential vs.[K+]o closer to the slope defined by the Nernst relationship (▾). Data are means ±s.e.m. from 5 preparations from 5 dogs.

Download figure to PowerPoint

Previous studies on cloned Kir2.1 channels expressed in Xenopus laevis oocytes have shown that this conductance can be activated by cAMP-dependent protein kinase (PKA; Fakler et al. 1994). We tested the effects of forskolin, which is known to enhance cAMP production, on the sensitivity of the resting potential to Ba2+. Forskolin (1 μM) hyperpolarized RMP of ICM strips by an average of 6.0 ± 0.7 mV (P < 0.005; n= 14). After the effects of forskolin reached steady state, the effects of Ba2+ were retested. In the presence of forskolin, we noted a leftward shift in the concentration-response relationship for BaCl2, indicating increased sensitivity to Ba2+ (Fig. 6). These results suggest that a Ba2+-sensitive conductance is expressed by circular muscle cells in ICM; however, under basal conditions this conductance is minimally activated.


Figure 6. Sensitivity to Ba2+ increases in ICM after application of forskolin

Isolated circular muscle preparations were exposed to forskolin (1 μM). Forskolin caused hyperpolarization of RMP (see text). For grapical comparison, the data were plotted as the degree of depolarization caused by Ba2+ before (▪) and after forskolin (•). In the presence of forskolin, Ba2+ depolarized ICM at lower concentrations, suggesting activation of a Ba2+-sensitive conductance. Data are means ±s.e.m. from 4-14 preparations from 4-14 dogs. * Significance compared with the membrane potentials before the addition of Ba2+ (P < 0.05).

Download figure to PowerPoint

Previously, the gradient in membrane potential across the circular muscle layer has been related to a gradient in the expression of the Na+-K+-ATPase (Burke et al. 1988, 1991). In the present experiments ouabain (1 μM) produced an average depolarization of 14 mV after 10 min in full-thickness preparations at the submucosal surface (-68.0 ± 3.8 vs. -54.0 ± 4.6 mV; n= 7; P < 0.0005). Slow wave activity was completely inhibited. Ba2+ (100 μM) further reduced the membrane potential to -45.0 ± 5.8 mV (P < 0.0001) in the presence of ouabain. This protocol was also tested in reverse. Ba2+ (100 μM) reduced RMP from -69.0 ± 0.9 to -53.0 ± 1.5 mV (P < 0.005). With pre-treatment of the muscles with Ba2+, ouabain did not produce significant further depolarization (P > 0.05; Fig. 7). Na+ pump activity would be expected to be independent of membrane potential over the range of potentials recorded in colonic muscles (Rakowski et al. 1989), so it is unlikely that the reduced effects of ouabain after Ba2+ were due to a depolarization-induced inactivation of the ouabain-sensitive component of membrane potential.


Figure 7. Effects of ouabain and Ba2+ on RMP of colonic muscles

Cells near the submucosal surface of full-thickness muscle preparations were depolarized by ouabain (1 μM). Addition of Ba2+ (100 μM) caused further depolarization. Data are means ±s.e.m. from 5 muscle strips. In another series of 5 experiments the order of application of ouabain and Ba2+ was reversed. Initial application of Ba2+ caused depolarization, and further application of ouabain produced a small depolarization that did not reach statistical significance. Significance between data groups denoted by brackets; *P < 0.05; **P < 0.01; n.s., not significant).

Download figure to PowerPoint

Ba2+-sensitive inward rectifier potassium currents are present in ICC

The experiments described above suggest that the Ba2+-sensitive conductance is expressed in cells within the submucosal pacemaker region. Therefore, we recorded currents from isolated ICC and smooth muscle cells using the whole-cell patch clamp technique (see Methods). ICC exhibited delayed rectifier currents averaging 210 ± 8 pA at 0 mV (n= 3), as previously described (Lee & Sanders, 1993). Negative to -80 mV, a prominent inward current was observed in physiological K+ gradients. At -120 mV, the average amplitude of the inward current (averaged from 15 episodes) was -52 ± 4 pA (n= 3). Application of Ba2+ (50 μM) selectively inhibited the inward current (to -19 ± 3 pA, n= 3) without affecting the outward, delayed rectifier currents (Fig. 8A). Figure 8B demonstrates the Ba2+-sensitive current obtained by subtraction of currents in the presence and absence of Ba2+ (50 μM). The Ba2+-sensitive currents showed inward rectification negative to EK (-88 mV). Elevation of [K+]o to 140 mM shifted the reversal potential of the Ba2+-sensitive current to -5 ± 3 mV (EK= 0 mV) and increased the magnitude of the current to -257 ± 8 pA (n= 3) at -120 mV. Little or no Ba2+-sensitive current could be resolved in isolated smooth muscle cells (n > 100; Fig. 8C and D) in physiological K+ gradients.


Figure 8. Effects of Ba2+ on whole-cell currents of voltage-clamped ICC and smooth muscle cells

The top trace shows the voltage-ramp protocol used. Traces in A and C are averaged current responses from 15 episodes in physiological K+ gradients (5/140 mM). A, current responses recorded from ICC before (○) and after (•) application of Ba2+ (50 μM). Prominent inward rectification was observed at potentials negative to EK (-88 mV). Ba2+ inhibited the inward current. B, difference current describing the Ba2+-sensitive currents in ICC. C, a similar experiment conducted on a circular smooth muscle cell. Much less inward current was observed in these cells at negative potentials (•), and Ba2+ (50 μM; ○) had very little effect. D, Ba2+ difference currents obtained by subtraction from control.

Download figure to PowerPoint

Expression of Kir2.1 mRNA

Expression of Kir2.1 was detected in the proximal colon and in the duodenum, jejunum and ileum by amplification of a 325 bp specific product (Fig. 9). This amplification product was sequenced and confirmed as generated from Kir2.1. Kir2.2 and Kir2.3 were not detected in any region of GI muscle tested (data not shown); however, control preparations of canine cerebellum demonstrated the fidelity of the primers constructed to amplify canine Kir2.2 and Kir2.3. Quantitative PCR (Q-PCR) was used to measure relative levels of Kir2.1 expression in the various regions tested (e.g. Fig. 9A). The ratio of Kir2.1 to β-actin in isolated circular muscle (ICM) of the colon and small intestine was: duodenum, 0.066 ± 0.024 (n= 3); jejunum, 0.113 ± 0.009 (n= 3); ileum, 0.080 ± 0.025 (n= 3); and proximal colon, 0.081 ± 0.018 (n= 7). There were no significant differences in the expression of Kir2.1 (P > 0.05; Fig. 9B). In the proximal colon the levels of Kir2.1 were significantly greater in samples of SCM (0.161 ± 0.045; n= 3) vs. samples of MyCM (0.078 ± 0.012; n= 7) and ICM (0.081 ± 0.018; n= 7; both P < 0.05; Fig. 9C).


Figure 9. Quantification of Kir2.1 cDNA in the canine gastrointestinal tract

A representative gel of Q-PCR for Kir2.1 in the submucosal layer of canine colon is shown in A; 2-fold serial dilutions of mimic DNA were included in the PCR reactions while Kir2.1 cDNA remained constant. The concentration of Kir2.1 cRNA in different regions of the GI tract expressed relative to β-actin cDNA is illustrated in B. The amount of Kir2.1 cDNA in SCM, ICM and MyCM preparations is depicted in C. Results are expressed as means ±s.e.m.; * significant difference in the level of Kir2.1 transcript in ICM and MyCM compared with SCM preparations (P < 0.05).

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  6. Acknowledgements

In the present study we have demonstrated the presence of a Ba2+-sensitive K+ conductance that is manifest in cells along the submucosal surface of the circular muscle layer in the canine colon. Electrophysiological observations were consistent with the expression of Kir2.1, a class of Ba2+-sensitive inward rectifying K+ channels in colonic muscles. The Ba2+-sensitive conductance in colonic muscles appears to contribute to the negative membrane potentials recorded from cells near the submucosal surface of the circular muscle layer (Smith et al. 1987) for the following reasons: (i) low concentrations of Ba2+ induced depolarization of the circular muscles; and (ii) the gradient in resting membrane potential across the circular muscle layer was abolished by Ba2+. Under basal conditions, the Ba2+-sensitive conductance, which has the properties of an inward rectifier K+ conductance, appeared to be mainly expressed by ICC, which are the pacemaker cells in the colon (Langton et al. 1989). Low doses of Ba2+ depolarized membrane potentials in intact muscle strips; however, there was far less sensitivity to Ba2+ in strips of muscle from which the pacemaker regions (containing ICC) were removed. Stimulation of isolated circular muscle preparations (ICM) with forskolin resulted in enhanced sensitivity to Ba2+, suggesting that the inward rectifier conductance may be regulated by cAMP.

This study identifies a potential role for an inward rectifier K+ conductance in GI smooth muscle. Previous studies have examined the effects of BaCl2 on GI muscles; however, these investigations failed to specifically identify contributions of inward rectifier K+ conductances to membrane potential (Liu & Huizinga, 1994; Murillo et al. 1997). This is because the effects of Ba2+ can be obscured by the activation of spike activity in GI muscles due to the increased current carried by Ba2+ through L-type Ca2+ channels (Hagiwara & Byerly, 1981; Bean, 1985). Prior studies tested the effects of much higher concentrations of Ba2+ (0.5-5 mM) than that used in the present study, and non-specific effects on other K+ conductances may have obscured the effects of this ion on inward rectifiers (Liu & Huizinga, 1994; Murillo et al. 1997).

Benham et al. (1987) described the presence of an inwardly rectifying non-selective current in isolated smooth muscle cells of the rabbit jejunum. This conductance is unlike those of the Kir gene family in that it carries both K+ and Na+ ions, and is significantly inhibited by 1 mM Cs+, and to a lesser extent by 10 mM Ba2+. This ‘anomalous rectifier’ may be similar to the currents described by DiFrancesco in sino-atrial cells and Purkinje cells in the heart (1985) and by Mayer & Westbrook (1988) in murine sensory neurones.

Little is known about the ionic conductance (s) responsible for the negative resting potentials in GI muscles, although it is apparent that the resting conductance for K+ ions is an important determinant of RMP (see Casteels, 1981, for review). In canine colon a membrane potential gradient exists across the circular muscle layer in which cells near the submucosal surface are highly polarized (i.e. -78 mV), and the resting membrane potential of cells near the myenteric surface lie at about -45 mV (Smith et al. 1987). Burke et al. (1988) proposed that a gradient in Na+-K+-ATPase activity might contribute to the gradient in membrane potential. It was suggested that the Na+ pump produced up to -30 mV of membrane potential at the submucosal surface of the circular muscle, and the contribution of the pump to resting potential decreased across the muscle layer. Barajas-Lopez & Huizinga (1992) argued that the depolarization caused by inhibition of the Na+-K+-ATPase was due to block of a K+ conductance by increased intracellular Na+, but the properties of the proposed K+ conductance were not determined.

Previous studies have shown that elevation in extracellular K+ increases the slope of the relationship between external K+ concentration and membrane potential in GI muscles, suggesting an increase in K+ conductance (Bolton, 1973; Bauer & Sanders, 1985; Hirst & van Helden, 1985; Hara & Szurszewski, 1986). The increase in Nernst-like behaviour was attributed simply to an enhancement in outward rectification when cells are depolarized by elevated K+. In the present study, we showed that membrane potential approached EK at potentials much more negative than the range in which voltage-dependent outward rectifier conductances are activated in canine colonic myocytes (Thornbury et al. 1992). We believe that the increase in apparent K+ conductance at these potentials may be due to an increase in the inward rectifier K+ conductance, which is activated at negative potentials by enhanced extracellular K+ (Kubo, 1996; Lopatin & Nichols, 1996).

In some studies of vascular smooth muscles, small increases in external K+ have been shown to produce hyperpolarization of RMP (Edwards et al. 1988; Knot et al. 1996). This apparently paradoxical behaviour has been attributed to the presence of an inward rectifier K+ conductance that is activated by elevated external K+. In the present study small elevations in external K+ caused depolarization. The difference between the vascular muscles tested and colonic muscles may be explained by the following: cerebral and coronary arteries have resting potentials of approximately -45 mV, which is at least 30 mV positive to EK (Edwards et al. 1988; Knot et al. 1996). Therefore, an increase in K+ conductance results in hyperpolarization as RMP approaches EK. In colonic myocytes we found that RMP approached EK when external K+ was enhanced. Colonic muscles have negative resting potentials, only a few millivolts positive to EK. Therefore, colonic muscles depolarized and membrane potential followed the estimated changes in EK (see Fig. 5).

There appears to be an important heterogeneity in the expression of the Ba2+-sensitive inward rectifier conductance in colonic muscles. ICC had significant inward rectifier currents under resting conditions, and this may contribute to the very negative potentials recorded in the submucosal pacemaker region (Smith et al. 1987), which is populated by ICC (Berezin et al. 1988). Previous studies have suggested that conductances in ICC may participate in setting the membrane potentials of GI muscles, because removal of ICC results in less polarized smooth muscle tissues (Liu & Huizinga, 1993; Ward et al. 1994; Torihashi et al. 1995). In contrast, smooth muscle cells may either have a much lower expression of these channels, or the conductance may be less available under basal conditions. Molecular studies support our electrophysiological findings with regard to heterogeneous expression of inward rectifier currents. Transcriptional expression of Kir2.1 mRNA (but not Kir2.2 or Kir2.3) was found in the circular muscle layer of the canine proximal colon. The expression of Kir2.1 was significantly higher in SCM preparations (which contained the ICC from which inward rectifier currents were measured) than in either ICM or MyCM preparations.

Our study suggests that a significant portion of the membrane potential gradient across the thickness of the circular muscle layer (see Smith et al. 1987) is due to the heterogeneous expression of Kir2.1 in colonic muscles. In the present study we reproduced the large depolarization caused by ouabain (see Burke et al. 1988), although the effects of ouabain were significantly reduced after prior application of Ba2+. This suggests that ouabain may be acting on the same conductance blocked by Ba2+.

We observed an increase in the sensitivity of isolated smooth muscle strips to Ba2+ in response to forskolin. This suggests that inward rectifier channels may be present in smooth muscle cells (also supported by molecular studies showing expression of Kir2.1 in ICM), but these channels may require activation by cAMP-dependent mechanisms before appreciable current is generated by this conductance. Regulation of the inward rectifier K+ conductance may provide a novel mechanism by which hormones and neurotransmitters could regulate membrane potential and excitability of GI muscles. Several endogenous compounds (including noradrenaline, vasoactive intestinal polypeptide, pituitary adenylyl cyclase-activating protein, prostaglandins and secretin) hyperpolarize and relax GI smooth muscles via cAMP-dependent mechanisms (see Makhlouf & Murthy, 1997, for review). cAMP-dependent mechanisms have been linked to activation of delayed rectifier currents (via 20 and 80 pS channels; Koh et al. 1996) and BKCa channels (Carl et al. 1991) in colonic myocytes. Activation of these conductances could result in membrane hyperpolarization or decreased membrane excitability, but these channels have low open probabilities at the resting potentials of colonic cells and they are insensitive to inhibition by Ba2+. Therefore, the hyperpolarization caused by forskolin was unlikely to be due to activation of delayed rectifier and BK channels.

In summary, membrane potential was sensitive to low concentrations of Ba2+ in the circular muscle of the canine colon. Depolarization caused by Ba2+ may be due to the expression of Kir2.1. This conductance may be predominantly located in interstitial cells of Cajal. We hypothesize that increased expression of Kir2.1 channels is responsible for the negative resting potentials in and near the submucosal surface of the circular muscle layer. The negative membrane potentials of ICC may influence the resting potential of electrically coupled circular muscle cells, creating the gradient in membrane potential across the circular muscle layer. Hormones and neurotransmitters coupled to effects via cAMP may enhance the open probabilities of Kir2.1 channels, leading to hyperpolarization and reduction in excitability.

  • Barajas-Lopez, C. & Huizinga, J. D. (1992). Ouabain-induced excitation of colonic smooth muscle due to block of K+ conductance by intracellular Na+ ions. European Journal of Pharmacology 221, 5158.DOI: 10.1016/0014-2999(92)90771-U
  • Bauer, A. J. & Sanders, K. M. (1985). Gradient in excitation-contraction coupling in canine gastric antral circular muscle. The Journal of Physiology 369, 283294.
  • Bean, B. P. (1985). Two kinds of calcium channels in canine atrial cells. Differences in kinetics, selectivity, and pharmacology. Journal of General Physiology 86, 130.
  • Benham, C. D., Bolton, T. B., Denbigh, J. S. & Lang, R. J. (1987). Inward rectification in freshly isolated single smooth muscle cells of the rabbit jejunum. The Journal of Physiology 383, 461476.
  • Berezin, I., Huizinga, J. D. & Daniel, E. E. (1988). Interstitial cells of Cajal in the canine colon: a special communication network at the inner border of the circular muscle. Journal of Comparative Neurology 273, 4251.
  • Bolton, T. B. (1973). Effects of electrogenic sodium pumping on the membrane potential of longitudinal smooth muscle from terminal ileum of guinea-pig. The Journal of Physiology 228, 693712.
  • Bradley, K. K., Jaggar, J. H., Bonev, A. D., Heppner, T. J., Flynn, E. R., Nelson, M. T. & Horowitz, B. (1999). Kir2.1 encodes the inward rectifier potassium channel in rat arterial smooth muscle cells. The Journal of Physiology 515, 639651.
  • Burke, E. P., Reed, J. B. & Sanders, K. M. (1988). Role of sodium pump in membrane potential gradient of canine proximal colon. American Journal of Physiology 254, C475483.
  • Burke, E. P., Sanders, K. M. & Horowitz, B. (1991). Sodium pump isozymes are differentially expressed in electrically dissimilar regions of colonic circular smooth muscle. Proceedings of the National Academy of Sciences of the USA 88, 23702374.
  • Carl, A., Kenyon, J. L., Uemura, D., Fuestani, N. & Sanders, K. M. (1991). Regulation of Ca2+-activated K+ channels by protein kinase A and phosphatase inhibitors. American Journal of Physiology 264, C387392.
  • Casteels, R. (1981). Membrane potential in smooth muscle cells. In Smooth Muscle: An Assessment of Current Knowledge, ed. Bulbring, E., Brading, A. F., Jones, A. W. & Tomita, T., pp. 105126. University of Texas Press, Austin , TX , USA .
  • Difrancesco, D. (1985). The cardiac hyperpolarising current, If. Origins and developments. Progress in Biophysics and Molecular Biology 46, 163183.DOI: 10.1016/0079-6107(85)90008-2
  • Edwards, F. R., Hirst, G. D. S. & Silverberg, G. D. (1988). Inward rectification in rat cerebral arterioles; involvement of potassium ions in autoregulation. The Journal of Physiology 404, 455466.
  • Fakler, B., Brandle, U., Glowatzki, E., Zenner, H. P. & Ruppersberg, J. P. (1994). Kir2.1 inward rectifier K+ channels are regulated independently by protein kinases and ATP hydrolysis. Neuron 13, 14131420.DOI: 10.1016/0896-6273(94)90426-X
  • Hagiwara, S. & Byerly, L. (1981). Calcium channel. Annual Review of Neuroscience 4, 69125.DOI: 10.1146/
  • Hara, Y. & Szurszewski, J. H. (1986). Effect of potassium and acetylcholine on canine intestinal smooth muscle. The Journal of Physiology 372, 521537.
  • Hirst, G. D. & van Helden, D. F. (1983). Ionic basis of the resting potential of submucosal arterioles in the ileum of the guinea-pig. The Journal of Physiology 333, 5367.
  • Jan, L. Y. & Jan Y. N. (1997). Annual Review Prize Lecture: Voltage-gated and inwardly rectifying potassium channels The Journal of Physiology 505, 267282.
  • Knot, H. J., Zimmermann, P. A. & Nelson, M. T. (1996). Extracellular K+-induced hyperpolarization and dilatations of rat coronary and cerebral arteries involve inward rectifier K+ channels. The Journal of Physiology 492, 419430.
  • Koh, S. D., Sanders, K. M. & Carl, A. (1996). Regulation of smooth muscle delayed rectifier K+ channels by protein kinase A. Pflügers Archiv 432, 401412.
  • Kubo Y. (1996). Two aspects of the inward rectification mechanism. Effects of cytoplasmic blockers and extracellular K+ on the inward rectifier K+ channel. Japanese Heart Journal 37, 631641.
  • Langton, P., Ward, S. M., Carl, A., Norell, M. A. & Sanders, K. M. (1989). Spontaneous electrical activity of interstitial cells of Cajal isolated from canine proximal colon. Proceedings of the National Academy of Sciences of the USA 86, 72807284.
  • Lee, H. K. & Sanders, K. M. (1993). Comparison of ionic currents from interstitial cells and smooth muscle cells of canine colon. The Journal of Physiology 460, 135152.
  • Liu, L. W. & Huizinga, J. D. (1993). Electrical coupling of circular muscle to longitudinal muscle and interstitial cells of Cajal in canine colon. The Journal of Physiology 470, 445461.
  • Liu, L. W. & Huizinga, J. D. (1994). Canine colonic circular muscle generates action potentials without the pacemaker component. Canadian The Journal of Physiology and Pharmacology 72, 7081.
  • Lopatin, A. N. & Nichols, C. G. (1996). [K+] dependence of open-channel conductance in cloned inward rectifier potassium channels (IRK1, Kir2.1). Biophysical Journal 71, 682694.
  • Makhlouf, G. M. & Murthy, K. S. (1997). Signal transaction in gastrointestinal smooth muscle. Cell Signal 9, 269276.
  • Mayer, M. L. & Westbrook, G. L. (1988). A voltage-clamp analysis of inward (anomalous) rectification in mouse spinal sensory ganglion neurons. The Journal of Physiology 340, 1945.
  • Murillo, M. D., Plaza, M. A. & Arruebo, M. P. (1997). The effect of Mn2+, Zn2+, Ba2+ and Ca2+ on spontaneous motility in sheep duodenum in vitro. General Pharmacology 28, 6571.
  • Quayle, J. M., Mccarron, J. G., Brayden, J. E. & Nelson, M. T. (1993). Inward rectifier K+ currents in smooth muscle cells from rat resistance-sized arteries. American Journal of Physiology 265, C13631370.
  • Rakowski, R. F., Gadsby, D. C. & De Weer, P. (1989). Stiochiometry and voltage dependence of the sodium pump in voltage-clamped, internally dialyzes squid giant axon. Journal of General Physiology 93, 903941.
  • Sanders, K. M. (1992). Ionic mechanisms of electrical rhythmicity in gastrointestinal smooth muscles. Annual Review of Physiology 54, 439453.
  • Smith, T. K., Reed, J. B. & Sanders, K. M. (1987). Origin and propagation of electrical slow waves in circular muscle of canine proximal colon. American Journal of Physiology 252, C215224.
  • Thornbury, K. D., Ward, S. M. & Sanders, K. M. (1992). Participation of fast-activating, voltage-dependent K currents in electrical slow waves of colonic circular muscle. American Journal of Physiology 263, C226236.
  • Tomita, T. (1981). Electrical activity (spikes and slow waves) in gastrointestinal smooth muscle. In Smooth Muscle: An Assessment of Current Knowledge, ed. Bulbring, E., Brading, A. F., Jones, A. W. & Tomita, T., pp. 127156. University of Texas Press, Austin , TX , USA .
  • Torihashi, S., Ward, S. M., Nishikawa, S., Nishi, K., Kobayashi, S. & Sanders, K. M. (1995). c-kit-dependent development of interstitial cells and electrical activity in the murine gastrointestinal tract. Cell Tissue Research 280, 97111.
  • Ward, S. M., Burns, A. J., Torihashi, S. & Sanders, K. M. (1994). Mutation of the proto-oncogene c-kit blocks development of interstitial cells and electrical rhythmicity in murine intestine. The Journal of Physiology 480, 9197.


  1. Top of page
  2. Abstract
  6. Acknowledgements

This study was supported by a Program Project Grant from the National Institutes of Health, DK 41315.