Novel voltage-dependent non-selective cation conductance in murine colonic myocytes

Authors


Corresponding author S. D. Koh: Department of Physiology and Cell Biology, University of Nevada School of Medicine, Reno, NV 89557, USA. Email: skoh@physio.unr.edu

Abstract

  • Two components of voltage-gated, inward currents were observed from murine colonic myocytes. One component had properties of L-type Ca2+ currents and was inhibited by nicardipine (5 × 10−7m). A second component did not ‘run down’ during dialysis and was resistant to nicardipine (up to 10−6m). The nicardipine-insensitive current was activated by small depolarizations above the holding potential and reversed near 0 mV.

  • This low-voltage-activated current (ILVA) was resolved with step depolarizations positive to -60 mV, and the current rapidly inactivated upon sustained depolarization. The voltage of half-inactivation was -65 mV. Inactivation and activation time constants at -45 mV were 86 and 15 ms, respectively. The half-recovery time from inactivation was 98 ms at -45 mV. ILVA peaked at -40 mV and the current reversed at 0 mV.

  • I lva was inhibited by Ni2+ (IC50= 1.4 × 10−5m), mibefradil (10−6 to 10−5m), and extracellular Ba2+. Replacement of extracellular Na+ with N-methyl-d-glucamine inhibited ILVA and shifted the reversal potential to -7 mV. Increasing extracellular Ca2+ (5 × 10−3m) increased the amplitude of ILVA and shifted the reversal potential to +22 mV. ILVA was also blocked by extracellular Cs+ (10−4m) and Gd3+ (10−6m).

  • Warming increased the rates of activation and deactivation without affecting the amplitude of the peak current.

  • We conclude that the second component of voltage-dependent inward current in murine colonic myocytes is not a ‘T-type’ Ca2+ current but rather a novel, voltage-gated non-selective cation current. Activation of this current could be important in the recovery of membrane potential following inhibitory junction potentials in gastrointestinal smooth muscle or in mediating responses to agonists.

Phasic contractions of gastrointestinal (GI) smooth muscles are timed by pacemaker potentials (slow waves) generated by interstitial cells of Cajal (ICC; see Sanders, 1996). Slow waves depolarize electrically coupled smooth muscle cells and increase the open probability of voltage-dependent ionic conductances (see Horowitz et al. 1999). The effectiveness of slow waves in initiating mechanically productive responses in smooth muscle cells is modulated by neural and hormonal inputs that regulate the ionic conductances expressed by ICC and smooth muscle cells. Various inward current pathways have been suggested as mechanisms that determine the excitability of GI muscle cells. These include high-voltage-activated (L-type) Ca2+ channels (Katzka & Morad, 1989; Rich et al. 1993; Farrugia et al. 1995; Xiong et al. 1995), Na+ conductances (Xiong et al. 1993), non-selective cation channels (Inoue & Isenberg, 1990), low-voltage-activated (T-type) Ca2+ conductances (Xiong et al. 1995; Kuga et al. 1996) and chloride conductance(s) (Crist et al. 1991). Most of the Ca2+ involved in initiation of contraction enters smooth muscle cells via L-type Ca2+ channels (Ozaki et al. 1993). Additional inward current pathways may be present to initiate or accelerate depolarization and enhance the open probability of L-type Ca2+ channels.

Low-voltage-activated (T-type) Ca2+ conductances have been reported in mammalian GI muscle cells (e.g. Xiong et al. 1995). This conductance is activated at relatively negative potentials, resistant to block by dihydro-pyridines and tetrodotoxin, impermeant to Na+, inactivated at relatively negative potentials (half-inactivation, -71 mV), and equally permeable to Ba2+ and Ca2+. These properties differ significantly from the dihydropyridine-sensitive, L-type Ca2+ channels commonly expressed by smooth muscles (Katzka & Morad, 1989; Rich et al. 1993; Farrugia et al. 1995; Xiong et al. 1995). Recently, a novel Ca2+ conductance was observed in studies of murine duodenal myocytes (Morel et al. 1997). Currents resulting from this conductance were not blocked by drugs specific for N-, L-, P-, T- or Q-type Ca2+ channels. Only mapacalcine, a dimeric peptide purified from marine sponge, inhibited these Ca2+ currents (Morel et al. 1997;Vidalenc et al. 1998). In the present study we have further investigated the inward currents of murine colonic myocytes and found a low-voltage-activated current (ILVA) that initially appeared to be a T-type conductance. However, upon further tests the properties of this conductance were found to be more like those of a non-selective cation conductance. This is a novel and possibly unique conductance expressed by GI myocytes. This conductance could contribute to excitability mechanisms in GI muscles such as action potentials or rebound depolarization following inhibitory neurotransmission.

METHODS

Preparation of isolated myocytes

Smooth muscle cells were prepared from colons removed from Balb/C mice. Mice were anaesthetized with chloroform and killed by cervical dislocation. Colons were removed from the animals through a midline abdominal incision. Protocols for handling of animals were reviewed and approved by the Institutional Animal Care and Use Committee.

Colons were cut open along the longitudinal axis, pinned out in a Sylgard-lined dish, and washed with Ca2+-free, phosphate-buffered saline (PBS) containing (mm): 125 NaCl, 5.36 KCl, 15.5 NaOH, 0.336 Na2HPO4, 0.44 KH2PO4, 10 glucose, 2.9 sucrose and 11 Hepes. Mucosa and submucosa were removed using fine-tipped forceps. Small pieces of muscle were cut and incubated for 8-12 min at 37 °C in a Ca2+-free solution containing: 4 mg ml−1 fatty acid-free bovine serum albumin, 2 mg ml−1 papain, 1 mg ml−1 collagenase and 1 mm dithiothreitol. After enzymatic treatment, the muscles were washed with Ca2+-free solution and agitated gently to create a cell suspension. The resulting myocytes came from the longitudinal and circular muscle layers, but because the circular layer is thicker than the longitudinal layer, most of the cells used in this study were probably circular muscle cells.

Dispersed smooth muscle cells were stored at 4 °C in buffer supplemented with minimum essential medium for suspension cultures (Sigma, M4767), 0.5 mm CaCl2, 0.5 mm MgCl2, 4.17 mm NaHCO3 and 10 mm Hepes. Cells were transferred from the refrigerator to the recording chamber. Drops of the cell suspension were placed on a glass coverslip forming the bottom of a 300 μl chamber mounted on an inverted microscope with a thermostatically regulated stage, and allowed to adhere to the bottom of the chamber for 5 min prior to recording.

Voltage-clamp methods

Whole-cell voltage-clamp techniques were used to record membrane currents from dissociated smooth muscle cells. Membrane currents were amplified by a List EPC-7 (List Electronics, Darmstadt, Germany) or Axopatch 1A (Axon Instruments, Foster City, CA, USA) and digitized with an A/D converter (Labmaster or Digidata, Axon Instruments). Data were collected at 4 kHz and filtered at 1 kHz via a Bessel filter, and digitized on-line using pCLAMP software (version 5.1.1, Axon Instruments). The data were analysed with the use of pCLAMP software (version 6.0.4, Axon Instruments). Pipette resistances were 1-4 MΩ, and the uncompensated series resistance was between 2 and 4 MΩ. The linear leak current was subtracted digitally. Most experiments were performed at room temperature (i.e. between 22 and 25 °C), and some experiments were performed at 31 °C to test the effect of higher temperatures on the current identified.

Solutions and reagents

The external solution used in conventional whole-cell recordings contained (mm): 135 NaCl, 5 KCl, 2 CaCl2, 1.2 MgCl2, 10 glucose and 10 Hepes, and was adjusted to pH 7.4 with Tris (CaPSS). In some experiments, NaCl was replaced by equimolar sodium isethionate (135 mm) or sodium methanesulfonate to assess the permeability of ILVA to anions. We used an agar bridge to minimize errors due to liquid junction potential for these experiments. To study the ionic permeability of ILVA, external Na+ (0, 5 and 50 mm) was replaced with equimolar N-methyl-d-glucamine (NMDG) while keeping [Ca2+]o constant at 2 mm. In other experiments, the external solution contained Ca2+ (0, 2 or 5 mm) at constant [Na+]o (135 mm). The standard internal solution contained (mm): 110 CsCl, 30 TEA-Cl, 10 BAPTA, 5 Na2ATP, 5 MgCl2 and 5 Hepes. This solution was adjusted to pH 7.2 with Tris (Cs-TEA). To assess the effect of internal Cs+, some experiments were performed using solutions in which CsCl was replaced with equimolar NMDGCl (110 mm). Mibefradil dihydrochloride was a gift from Dr Eva-Maria Gutknecht and Dr Pierre Weber (Hoffmann-La Roche Ltd, Basel, Switzerland). All other chemicals were purchased from Sigma (USA).

Molecular study: RT-PCR

The mouse low-voltage-activated T-type Ca2+ channel α1G subunit sequence was retrieved from GenBank (NM 009783) to design primers. This sequence was used for a blast homology search against other T-type Ca2+ channels such as rat CACNA1G (AF027984), rat T-type calcium channel isoform (AF125161), human CACNA1G (AF134986), human CACNA1G isoform bcef (AF227750), human CACNA1G isoform bce (AF227749), human CACNA1G isoform bcd (AF227748), and human CACNA1G isoform bc (AF227747). The most homology was found in S4 and S5 in domain III, and in S4 in domain IV (Lee et al. 1999). Degenerate primers were designed based on this homology. Forward primer TCaF1 and reverse primer TCaR2 corresponding to 4303-4322 and 5311-5292 of the mouse α1G subunit sequence (NM009783) were 5′ CTSAAGCTGGTGGTRGAGAC 3′ and 5′ YRGCCATCTTCARYAGCTTC 3′. The nested forward primer TCaF2 and the nested reverse primer TCaR1 corresponding to 4364-4382 and 5267-5249 were 5′ TYTGCTGYGCCTTCTTCATC 3′ and 5′ CKCATGATRCGGATGATGG 3′. Total RNA was extracted from 50 mg of the frozen mouse brain tissues, 50 mg of proximal colon tissues, and 50 smooth muscle cells of colon tissues using TRIzol reagent (Gibco, BRL) and treated with DNase I (Stratagene) at a concentration of 1 U μg−1 in the reaction buffer, containing 40 mm Tris-HCl (pH 7.5), 6 mm MgCl2 and 2 mm CaCl2 for 30 min at 37 °C. The reaction was terminated by adding EDTA at a final concentration of 2.5 mm and heated to 65 °C for 15 min. First strand cDNA was synthesized by using 200 U μl−1 SuperScript II RNase H reverse transcriptase (Gibco, BRL) at 42 °C for 50 min in the presence of 5 μg total RNA in a 40 μl reaction volume. RNA was degraded by incubating with 2 U of RNase HI (Gibco, BRL) for 20 min at 37 °C. The first PCR reactions were performed using GeneAmp PCR System 2400 (Perkin-Elmer) by adding 1 U of Taq DNA polymerase, Recombinant (Gibco, BRL), 2 μl of the synthesized cDNA, 2 mm MgCl2, 1X buffer and degenerate primers, forward primer TCaF1 and reverse primer TCaR2. The amplification procedure was as follows: 1 cycle at 94 °C for 1 min; 40 cycles composed of 30 s at 94 °C, 30 s at 50 °C, and 2 min at 72 °C. The second PCR was performed by using 1 μl of the first PCR product and the nested primers, forward primer TCaF2 and reverse primer TCaR1 with an additional 30 cycles. After PCR, 5 μl of the PCR product was analysed on a 1.2 % agarose gel.

Statistical analyses

Data are reported as means ±s.e.m.n is the number of cells tested. All statistical analyses were performed using SigmaStat 2.0 software (Jandel Corporation, San Rafael, CA, USA) and GraphPad Prism (version 3.0, GraphPad Software Inc., San Diego, CA, USA). Statistical significance was evaluated by Student's t test as appropriate. P values less than 0.05 were considered significant.

RESULTS

Isolation of low-voltage-activated inward current (ILVA) in murine colonic myocytes

Two types of inward current were detected in murine colonic myocytes using selective voltage-clamp protocols or pharmacological manipulations. To isolate L-type Ca2+ current in murine colonic myocytes, cells were bathed in CaPSS (see Methods for solution components) and patch pipettes were filled with Cs-TEA solution. The cells were held at either -50 or -80 mV and stepped to test potentials ranging from 90 to +25 mV in 5 mV intervals (Fig. 1). Current-voltage (I-V) relationships were plotted for peak currents generated by voltage steps from both holding potentials. In cells held at -50 mV (test potentials from -50 to +25 mV), an inward current was resolved at -40 mV and the current reached a maximum at 0 mV. From a holding potential of -80 mV (test potentials from -90 to -15 mV), an inward current was resolved at potentials as negative as -60 mV. Difference currents were obtained by digitally subtracting currents elicited from the two holding potentials (Fig. 1B). This analysis reveals two components of voltage-dependent inward current in murine colonic myocytes: a high-voltage-activated current with I-V characteristics similar to an L-type Ca2+ conductance, and a low-voltage-activated current (ILVA). ILVA was resolved by steps from negative holding potentials, but this current was inactivated by holding at -50 mV. The I-V of the difference current shows that the ILVA is an inward current that reaches a maximum at about -35 mV and reverses near 0 mV. It should also be noted that the high-voltage-activated current ran down during recordings lasting tens of minutes, but ILVA was stable for long periods of recording (i.e. up to 60 min; data not shown).

Figure 1.

Isolation of ILVA and IHVA by voltage protocols

A, representative currents elicited by test potentials from -50 to +25 mV (in 15 mV increments; Vh= -50 mV). B, representative currents elicited by test potentials from -90 to -15 mV (in 15 mV increments; Vh= -80 mV). C, current-voltage relationship for currents obtained from holding potentials of -80 mV (○), -50 mV (•) and difference currents (▪). D, representative currents were recorded with conventional whole-cell voltage clamp in response to depolarizing steps to -50, -35 and -15 mV from two different holding potentials of -50 mV (•) and -80 mV (○). The inset (voltage protocol) shows examples of currents in response to selected test potentials (V1). E, difference currents at each test potential were obtained by subtracting currents obtained from two holding potentials.

Expression of a distinct high-voltage-activated inward current was also suggested by pharmacological experiments. Nicardipine (10−8 to 10−6m) partially inhibited the inward currents elicited by depolarizations from a holding potential of -80 mV (test potentials from -90 to -15 mV; n = 7 cells). Nicardipine (up to 10−6m) had little effect on currents elicited by test potentials negative to -35 mV but inhibited currents elicited by larger depolarizations in a dose-dependent manner (Fig. 2a, B and D). Thus inward currents activated at potentials positive to -35 mV had the voltage dependence and dihydropyridine sensitivity of L-type Ca2+ currents (reversal potential ≈45 mV). We observed no significant differences in the inward currents elicited in the presence of 0.5 and 1.0 μm nicardipine. The dihydropyridine-resistant current had a peak amplitude of -36 ± 4 pA at -35 mV (n = 7, Fig. 2C) and similar I-V characteristics to ILVA isolated by voltage-clamp protocols (see above).

Figure 2.

Isolation of ILVA and IHVA by nicardipine

A, current responses to test potentials from -80 to -15 mV (5 mV increments) in control solutions were recorded (Vh= -80 mV). B, current responses recorded using the same protocol as in A in the presence of nicardipine (0.5 μm). C, average current-voltage relationship in control (○) and in the presence of 0.01 (•) and 1 μm (□) nicardipine. D, effects of nicardipine concentration on responses elicited by steps to -35 mV (○) and -15 mV (•) (n = 7 cells; Vh= -80 mV). * Significant differences.

It is possible that the reversal potentials measured with voltage-clamp protocols shown in Fig. 1 might be influenced by voltage-dependent inactivation of the ILVA channels and therefore do not indicate the true ionic selectivity of those channels. Accordingly, we reevaluated the reversal potential of ILVA using protocols to determine the reversal of instantaneous tail currents. Membrane currents were elicited by a two-step protocol in which membrane potential was depolarized from -80 to -40 mV for 100 ms to activate and inactivate ILVA, and then to 0 mV for 100 ms to activate IHVA (Fig. 3A). During the application of nicardipine (10−6m) while continuing these voltage steps, ILVA was unaffected whereas IHVA was blocked completely. Thus L-type Ca2+ channels did not mediate ILVA, and we examined the properties of currents in the presence of nicardipine in order to observe the behaviour of ILVA uncontaminated by IHVA. In the presence of 1.0 μm nicardipine, membrane potential was stepped from -80 to -40 mV for 10 ms, and then the potential was changed to various test potentials ranging from -50 to +40 mV (Fig. 3B). Membrane currents were measured after the capacity current relaxed. The instantaneous tail currents resolved by these experiments reversed at -5.2 ± 0.8 mV (n = 11, Fig. 3C). This value was not different from the reversal potential (-4.7 ± 0.6 mV) in the presence of 0.5 μm nicardipine. This value is consistent with ILVA being mediated by non-selective channels, but inconsistent with ILVA being mediated by Ca2+ channels.

Figure 3.

Measurement of reversal potential of ILVA using instantaneous tail current analysis

A, membrane currents elicited by a two-pulse protocol in which a cell was stepped from -80 to -40 mV and then from -80 to 0 mV. Repetitive stepping was performed during the application of nicardipine (1.0 μm). Nicardipine had no effect on the current elicited by steps to -40 mV (ILVA) and totally blocked the current elicited by steps to 0 mV (IHVA). B, a second voltage-clamp protocol applied to the same cell after addition of nicardipine. The cell was held at -80 mV, stepped to -40 mV for 10 ms to activate ILVA, and then stepped to potentials ranging from -50 to +40 mV (selected current responses are shown). C, summary of current-voltage data using the same protocols in experiments on 11 cells. ILVA reversed at -4.7 mV.

Voltage and time dependence of the ILVA

The voltage dependences of inactivation of IHVA and ILVA were compared in another series of experiments. Cells treated with nicardipine (5 × 10−7m) were stepped to conditioning potentials between -90 and -35 mV for 300 ms from a holding potential of -80 mV before a test step to -40 mV (Fig. 4A). Conditioning potentials of 300 ms were needed to reach steady-state inactivation. Normalized peak inward currents resulting from the test depolarization were plotted as a function of conditioning potential. The data were fitted with a Boltzmann function having a half-inactivation value of -65 ± 1 mV and a slope factor of -4 ± 1 (Fig. 4C). Similar inactivation protocols were performed on cells in the absence of dihydropyridine to characterize the inactivation of the L-type Ca2+ current. In these experiments a holding potential of -50 mV was used to inactivate ILVA, and the test potential was 0 mV. Conditioning steps (-80 to +20 mV) of 1.2 s were needed to inactivate the L-type Ca2+ current (Fig. 4B). Similar experiments were performed on six cells and fitted with Boltzmann functions. The L-type current had a half-inactivation voltage of -22 ± 1 mV and the Boltzmann function had a slope of -9 ± 1 (Fig. 4C). These studies demonstrate that the voltage- and time-dependent properties of inactivation differ for L-type Ca2+ currents and ILVA.

Figure 4.

Voltage dependence of inactivation of ILVA and IHVA

A, currents generated by inactivation protocol to isolate ILVA. Membrane potential was stepped to conditioning potentials between -90 and -35 mV (5 mV increments) for 300 ms from a holding potential of -80 mV, and then stepped to a test potential of -40 mV in the presence of 0.5 μm nicardipine. B, currents generated by inactivation protocols to isolate IHVA. Membrane potential was stepped to conditioning pulses between -80 and +20 mV from a holding potential of -50 mV for 1.2 s and then stepped to a test potential of 0 mV in the absence of nicardipine. C, the voltage dependences of inactivation of ILVA (○) and IHVA (•) are shown as a plot of normalized peak inward currents during the test steps as a function of conditioning potentials. The half-inactivation potentials (denoted by dotted lines) were determined from Boltzmann functions fitted to the data (continuous lines) (n = 6 cells).

Experiments were performed in the presence of nicardipine (5 × 10−7m) to characterize the recovery from inactivation of ILVA. Cells were held at -80 mV and inward currents were elicited by steps to -60, -55, -50, -45 and -40 mV for 300 ms (Fig. 5A). The voltage was then returned to the holding potential for various durations from 0 to 450 ms (in 50 ms intervals) before a second step to the test potential was applied. The time to half-activation (t1/2) decreased as a function of depolarization (i.e. 29 ± 5 ms at -55 mV, 18 ± 2 ms at -50 mV, 8 ± 2 ms at -45 mV, and 5 ± 2 ms at -40 mV; n = 4, Fig. 5C). The time course of inactivation was well fitted by a single exponential, and the time constants of inactivation showed no voltage dependence over the range of potentials tested (i.e. 98 ± 5 ms at -55 mV, 79 ± 3 ms at -50 mV, 86 ± 2 ms at -45 mV, and 92 ± 2 ms at -40 mV; n = 4, Fig. 5C). ILVA recovered from inactivation with a time constant of 85 ms at test potentials to -40 mV (Fig. 5B), and there was little difference observed in the rate of recovery from inactivation at the other test potentials (from -55 to -40 mV, not shown).

Figure 5.

Kinetics of activation and inactivation of ILVA

A, membrane currents were elicited by test potentials from -60 to -40 mV for 250 ms. These responses were followed by returning voltage to the holding potential (-80 mV) for various periods (50 ms to 450 ms). Then cells were returned to the test potential for 300 ms to determine the rate of recovery from inactivation (all tests performed in the presence of nicardipine; 0.5 μm). B, summary of experiments describing recovery from inactivation (n = 4 cells). Normalized peak currents at -40 mV using the protocol described in A were plotted as a function of the interval between test depolarizations. The continuous line shows a fit of the data with the equation It/Imax= 1 - (1/Imax) exp (-t/τ), where It is the amplitude of the current after t ms of recovery, Imax is the amplitude of the fully recovered current, and t and τ are time and the time constant in milliseconds. C, membrane current recorded after a step to -40 mV. The activation and inactivation portions of the currents (dots) were well fitted by single exponentials (continuous lines).

Pharmacology and ion selectivity of the ILVA of murine colonic myocytes

Pharmacological experiments were performed to characterize the nature of the channels that mediate ILVA. Ni2+ (tested in the presence of 5 × 10−7m nicardipine) inhibited ILVA in a concentration-dependent manner (IC50 14 ± 3 μm; n = 6; Fig. 6). Inhibition of the current by Ni2+ was accompanied by a shift of the peak current to more positive potentials (Fig. 6E). A similar shift was noted in previous studies of T-type Ca2+ currents with inhibition by Ni2+ (Ertel & Ertel, 1997). Mibefradil, an organic blocker of T-type Ca2+ channels (Bezprozvanny & Tsien, 1995; Gothert & Molderings, 1997; Nilius et al. 1997; Verma et al. 1997; Viana et al. 1997; Brixius et al. 1998; Sandmann et al. 1998) also inhibited ILVA in a concentration-dependent manner (Fig. 7).

Figure 6.

Inhibition of ILVA by Ni2+

A, representative currents in response to test potentials from -40 to -10 mV (in 10 mV increments; Vh= -80 mV) in the presence of nicardipine (0.5 μm). A, control currents. B-D, currents in the presence of Ni2+ (5, 10 and 30 μm as indicated). E, the current-voltage relationships for peak currents recorded in control (○) and in solutions containing 5 (•), 10 (□), 30 μm (▪) Ni2+. F, inhibitory effects of Ni2+ concentration on currents elicited at -40 mV. The IC50 was determined from a sigmoidal function fitted to the data (continuous line). n = 7 cells.

Figure 7.

Inhibition of ILVA by mibefradil

A, representative currents elicited by test potentials from -80 to 0 mV (in 20 mV increments; Vh= -80 mV) in the presence of nicardipine (0.5 μm). B and C, currents elicited in the same cell using the same protocol after addition of mibefradil (1 and 10 μm). D, current-voltage relationships for peak currents in control (○), 1 (•) and 10 μm (□) mibefradil (n = 4).

It should be noted that after inward currents were inhibited by Ni2+ or by mibefradil an outward current was revealed at positive potentials and the reversal potential of the net current shifted to more negative potentials (see Fig. 6E and Fig. 7D). The outward currents might be due to removal of the block by these agents at positive test potentials or to an outward current carried by another channel. The pipette solution used in the studies illustrated by Fig. 6 contained Cs+ and TEA while the pipette solution used in the studies illustrated by Fig. 7 contained NMDG. In both protocols, the net current reversed near 0 mV, implying that TEA and NMDG had similar permeabilities. Furthermore, replacement of extracellular NaCl with sodium isethionate or sodium methanesulfonate had no effect on the reversal potential of currents (data not shown) suggesting that Cl, isethionate and methanesulfonate were not permeable to the conductance responsible for ILVA.

We investigated the Ca2+ permeability of ILVA by changing extracellular Ca2+ in the presence of nicardipine (5 × 10−7m). No inward currents were observed in the absence of extracellular Ca2+. Peak inward currents increased as external Ca2+ concentration was increased, and the reversal potential of the membrane current was shifted to more positive potentials (Fig. 8). Reversal potentials averaged +3 ± 2 mV with 2 mm[Ca2+]o and +22 ± 3 mV with 5 mm[Ca2+]o (n = 6). Instantaneous tail current analysis was performed to test the effects of reduced extracellular Ca2+ on ILVA. Decreasing external Ca2+ from 2.0 to 0.5 mm caused a shift in the reversal potential of the tail current I-V from -3 to -17 mV (n = 4, Fig. 8E and F). These data suggest that Ca2+ ions carry at least a portion of ILVA.

Figure 8.

Effects of external Ca2+ on ILVA

A, representative currents elicited by test potentials from -80 to -40 mV (in 10 mV increments; Vh= -80 mV) in the presence of nicardipine (0.5 μm) and 2 mm external Ca2+. B, currents elicited in the same cells using the same protocol after removal of external Ca2+ (nominally 0 Ca2+); and C, in the presence of elevated external Ca2+ (5 mm). D, current-voltage relationship for peak currents in 2 (○), 0 (▪) and 5 mm (•) Ca2+ (n = 6 cells). The dotted line in D denotes shift in reversal potential when external Ca2+ was 5 mm. E, representative currents elicited by instantaneous current-voltage protocol at -40, -20, +20 and +40 mV (see inset) in the presence of nicardipine (1 μm). The continuous line denotes zero current level. F, current-voltage relationship by tail current analysis in 2 mm (○) and 0.5 mm (•) Ca2+ (n = 4 cells).

L-type and T-type Ca2+ channels both conduct Ba2+. Previous studies have shown that the permeability of T-type channels is similar to Ca2+ and Ba2+ (cf. Ertel & Ertel, 1997). Therefore, we tested Ba2+ as a charge carrier of ILVA. Ba2+ (0.1, 0.5 and 2 mm added to the Ca2+-containing external solution) inhibited ILVA in a concentration-dependent manner (Fig. 9a and B). Complete replacement of Ca2+ with Ba2+ abolished ILVA (Fig. 9C). The effects of Ba2+ were reversible when normal [Ca2+]o was restored. Instantaneous tail analysis showed that Ba2+ blocked ILVA with no shift in the reversal potential (Fig. 9D and E). These data imply that ILVA is not mediated by low-threshold Ca2+ channels with previously described properties.

Figure 9.

Effects of external Ba2+ on ILVA

A, representative currents elicited test potentials from -80 to -40 mV (in 10 mV increments; Vh= -80 mV) in the presence of nicardipine (0.5 μm): Aa, control currents. Ab-Ae, effects of various concentrations of Ba2+ on ILVA in the presence of 2.0 mm Ca2+. B, effects of Ba2+ on peak current amplitude of ILVA in response to repetitive steps to -40 mV every 10 s (a-e in B denote the times when currents in A were recorded). The bar at the top shows changes in solutions during the time course of this experiment. C, current-voltage relationship in 2 mm Ca2+ (○) and when external Ca2+ was replaced with equimolar (2 mm) Ba2+ (n = 5 cells). D, representative currents elicited by instantaneous current-voltage protocol from -40 to 20 mV (in 20 mV increments, see inset for protocol) in the presence of nicardipine (1 μm). The continuous line denotes zero current level. E, current-voltage relationship by tail current analysis in control (○) and 2 mm (•) Ba2+ (n = 4 cells).

The effects of Ba2+ and the reversal potentials of ILVA suggested that a voltage-activated, non-selective cation conductance might mediate ILVA. We tested whether Na+ ions carry part of the current by replacing extracellular Na+ with NMDG. Replacement of extracellular Na+ reduced the inward current and shifted the reversal potential of the net current to more negative potentials (Fig. 10). Using instantaneous tail current analysis, a reduction in Na+ (to 50 mm) decreased ILVA and shifted the reversal potential from -1 to -23 mV (n = 5, Fig. 10C and D). These data indicate that Na+ ions carry a portion of ILVA.

Figure 10.

Effects of external Na+ replacement on ILVA

A, representative currents in response to test potentials of -40 mV in the presence of nicardipine (0.5 μm; Vh= -80 mV). Three currents recorded from the same cell are superimposed in which external Na+ was 135 mm (control), 50 mm and 5 mm. Na+ was replaced with equimolar NMDG. B, current-voltage relationship for peak currents elicited in 135 mm Na+ (○), 50 mm Na+ (•), and 5 mm Na+ (□) (n = 6). C, representative currents elicited by instantaneous current-voltage protocol from -40 to 20 mV (in 20 mV increments, see inset for protocol) in the presence of nicardipine (1 μm). The continuous line denotes zero current level. D, current-voltage relationship by tail current analysis in control (○) and 50 mm (•) Na+ (n = 5 cells).

We also examined the effects of Cs+, an inhibitor of the hyperpolarization-activated cationic current (If). Four cells were stepped from -80 mV to test potentials ranging from -80 to +60 mV (5 mV increments) in the presence of 0.5 μm nicardipine to record ILVA. Cs+ (0.1 and 0.2 mm) inhibited ILVA (Fig. 11a and B) without shifting the reversal potential (Fig. 11D, E and F).

Figure 11.

Effects of external Cs+ and Gd3+ on ILVA

A, currents elicited by a test potential of -40 mV in the presence of nicardipine (0.5 μm). Superimposed currents are before and after addition of 0.2 mm Cs+. B, current-voltage relationship showing currents in control (○), 0.1 (•) and 0.2 mm (□) Cs+. C, current-voltage relationship for the Cs+-sensitive (0.2 mm) current. D and E, representative currents elicited by instantaneous current-voltage protocol from -40 to 20 mV (in 20 mV increments, see inset for protocol) in control and in the presence of Cs+ (1 mm), respectively. The continuous line denotes zero current level. F, current-voltage relationship by tail current analysis in control (○) and 1 mm (•) Cs+ (n = 4 cells). G, currents elicited by a test potential of -40 mV in the presence of nicardipine (0.5 μm). Superimposed currents are before and after addition of 2 μm Gd3+. H, current-voltage relationship showing currents in control (○) and 2 μm Gd3+ (•). I, current-voltage relationship for the Gd3+-sensitive current.

We also tested the ability of Gd3+, a blocker of a variety of non-selective cation conductances, to inhibit ILVA. Gd3+ (2 × 10−6m) rapidly and reversibly blocked ILVA (Fig. 11D). Results from four cells are summarized in Fig. 10E in which the I-V relationship of ILVA is plotted in the presence and absence of Gd3+.

It should be noted that Cs+ and Gd3+ inhibited the inward current attributed to ILVA, but neither ion blocked outward currents (see Fig. 11C and F). These studies were conducted with pipettes containing TEA (30 mm) and NMDG (110 mm). Additional experiments using TEA (30 mm) and Cs+ (110 mm) internal solutions showed that Gd3+ blocked both the inward and outward currents (not shown). The results suggest that ILVA is mediated by voltage-activated non-selective cation channels that are inhibited by Gd3+. The outward current observed in cells dialysed with TEA (30 mm) and Cs+ (110 mm) is mediated by cation fluxes through these channels (probably carried by Cs+). Residual outward currents in cells dialysed with TEA and NMDG are not likely to be mediated by the ILVA channels, and these currents are due to an independent membrane conductance of an unknown nature.

Temperature dependence of ILVA

We investigated the effect of warming on the amplitude and kinetics of ILVA in order to learn about the function of this current under more physiological conditions. In preliminary experiments we found that warming dissociated murine colonic myocytes to temperatures above 32 °C caused them to spontaneously contract. Thus we limited temperature studies to 31 °C and below. ILVA was recorded in the presence of nicardipine (1.0 μm) from four cells at room temperature, and then the cells were warmed to 31°C and ILVA was recorded (Fig. 12). Warming increased the rates of activation (Fig. 12C) and deactivation (Fig. 12D), but had no significant effect on the amplitude of the peak current. These findings imply that our measurements of ILVA at room temperature are a reasonable indicator of the contribution of this current at physiological temperatures.

Figure 12.

Effects of temperature on ILVA

A, ILVA elicited in the same cell by steps from -80 to -40 mV at room temperature (RT) and 31°C. The currents are averages of currents elicited by 16 steps. The activation and inactivation of ILVA were increased by warming. Note there was little or no change in the amplitude of the peak current as a function of temperature. B-D, summary of the comparisons of peak current (B), activation (C) and inactivation (D) time constants for four cells studied with the same protocol.

Molecular evidence

Since the kinetics of ILVA are similar to those T-type Ca2+ channels, we performed RT-PCR to test whether known T-type channels are expressed in colonic myocytes. T-type channel isoforms have been cloned from mouse brain (NM 009783), rat brain (AF027984), rat pancreas (AF125161), human brain (AF134986 and AF227750), human thalamus (AF227749), and human fetal kidney (AF227748 and AF227747). Sequences of these genes were aligned for a multiple sequence homology search, and the most homologous region was found to be in S5 in domain III, and in S4 in domain IV (see Lee et al. 1999). Degenerate primers were designed based on this homology to amplify T-type channel mRNA by RT-PCR. Products were readily generated from mouse brain extracts with 40 cycles of PCR (i.e. a 1009 bp fragment predicted from the cloned sequence, NM 009783, and a 750 bp fragment). No products were amplified from mouse colonic myocytes with the same techniques (Fig. 13). A 904 bp fragment predicted from the sequence was generated from mouse brain using nested primers and an additional 30 cycles of PCR. No products were generated from colonic myocytes (n = 4). A 497 bp product amplified from β-actin primers was used for an internal control for these experiments. These data provided no evidence for expression of T-type channels in the colonic smooth muscle cells, supporting the idea that ILVA is not due to T-type Ca2+ channels.

Figure 13.

RT-PCR: a representative ethidium bromide-stained agarose gel of RT-PCR products

PCR products were generated by using degenerate primers, forward primer TCaF1 and reverse primer TCaR2 for the low-voltage-activated T-type Ca2+ channel gene from the cDNAs of the brain tissues and colonic myocytes of mice. The second PCR was performed through the use of the nested primers, forward primer TCaF2 and reverse primer TCaR1. Products, 1009 bp and 750 bp, were readily detected at the first PCR in brain. Product of a 904 bp fragment was detected at the second PCR in brain, but not in colonic myocytes. A 497 bp product amplified from the β-actin gene was used for endogenous control and NTC was a non-template control. A 100 bp marker on each side indicated the size of the fragments.

DISCUSSION

Both low-voltage-activated (LVA) and high-voltage-activated (HVA) Ca2+ channels are expressed by gastrointestinal smooth muscle cells (Katzka & Morad, 1989; Lee & Sanders, 1993; Rich et al. 1993; Farrugia et al. 1995; Xiong et al. 1995). Under experimental conditions designed to minimize K+ currents we observed two voltage-dependent inward currents in murine colonic myocytes. One current was resolved by test depolarizations positive to -40 mV, and could be elicited from a holding potential of -50 mV. This current ran down over time and was inhibited by 0.5-1 μm nicardipine. We refer to this current as IHVA because it had properties commonly associated with L-type Ca2+ currents. A second component of the inward current was observed in the presence of nicardipine. We referred to the nicardipine-resistant current as ILVA. The maximal amplitude of ILVA was small (35 pA) and peak amplitude was achieved at a test potential of -40 mV. Ni2+ and mibefradil, common inhibitors of ‘T-type’ Ca2+ channels, blocked ILVA and shifted the voltage dependence of the conductance towards more positive potentials. Others have shown similar effects on T-type Ca2+ currents during Ni2+ block (Ertel & Ertel, 1997). The kinetic properties of ILVA in murine colonic myocytes were also similar to ‘T-type’ Ca2+ currents (i.e. rates of activation, inactivation, and recovery from inactivation), and ILVA showed no appreciable run-down during recordings lasting up to 60 min. Xiong et al. (1995) reported a T-type Ca2+ current in human colonic myocytes. Some of the properties of this current were similar to those observed for ILVA in the present study. For example, the T-type Ca2+ current in human colonic myocytes was insensitive to nifedipine and inhibited by Ni2+. However, the T-type Ca2+ current in human colonic cells reversed at about +60 mV, and this differed from the ILVA observed in murine colonic mycytes.

Several properties of ILVA suggested that it was not a ‘T-type’ Ca2+ current. T-type Ca2+ channels are equally permeable to Ca2+ and Ba2+ (Ertel & Ertel, 1997). When Ba2+ was used to replace external Ca2+, we found that ILVA could not be evoked. Even in the presence of external Ca2+, addition of Ba2+ inhibited ILVA. These data suggest that Ba2+ is not a permeant ion of the channels responsible for ILVA in colonic myocytes and that ILVA is not a ‘T-type’ Ca2+ current. Molecular studies also could not detect expression of known T-type channel isoforms.

I LVA reversed close to 0 mV. We considered the possibility that ILVA could be contaminated by Cl currents, although the reversal potential was not affected by Cl replacement. It should be noted that the peak amplitude of ILVA was decreased by less than 10 % by Cl replacement with isethionate and methanesulfonic acid. The mechanism of block of ILVA by Cl replacement was not investigated. Reversal of the current at 0 mV suggested the possibility that ILVA could be a non-selective cation conductance (NSCC). Therefore, we examined the effects of changing external Na+ concentration on ILVA. We found that reduction of external Na+ decreased the amplitude of ILVA and shifted the reversal potential to more negative values. These data suggest that Na+ is a charge carrier of ILVA. It should also be noted that ILVA was blocked in solutions containing Na+ in which Ca2+ was removed and reduced in solutions containing Ca2+ in which Na+ was reduced. These findings suggest a possible regulatory role of external Na+ and Ca2+ on ILVA.

In human colonic myocytes current attributed to ‘T-like’ Ca2+ channels was not affected by removal of Na+ but was abolished by removal of Ca2+ (Xiong et al. 1995). In murine myocytes decreased external Ca2+ blocked ILVA, and increased Ca2+ enhanced the amplitude of the current and shifted the reversal potential to more positive potentials. These data suggest that Ca2+ is also a charge carrier for the conductance responsible for ILVA. Additional experiments showed that low concentrations of Gd3+ (2 μm) completely inhibited ILVA. Taken together, these observations suggest that ILVA is a novel voltage-gated non-selective current expressed by murine colonic myocytes.

It should also be noted that ILVA was also blocked by external Cs+. Application of Cs+ has been used to identify the presence of hyperpolarization-activated currents (If; see Denyer & Brown, 1990); however, the activation and inactivation properties of ILVA clearly distinguish it from an If. These observations suggest that inhibition of inward current by Cs+ in some gastrointestinal muscles may not be a reliable indicator of the presence of an If current.

Morel et al. (1997) reported a novel Ca2+ conductance in murine duodenal myocytes that was not blocked by N-, L-, T-, P- or Q-type Ca2+ channel blockers. However, the current, isolated using holding potential changes and oxodipine (10 μm), was similar in many properties to L-type Ca2+ currents (e.g. resolution threshold with depolarization to -45 mV, reached maximum current at 0 mV, and rapidly deactivated upon repolarization). These authors referred to this current as a non-L-type Ca2+ current (dihydropyridine-insensitive), although experiments to determine the ionic selectivity of this conductance were not performed. In the present study ILVA was similar to the current noted by Morel and coworkers (1997) in terms of dihydropyridine insensitivity and lack of ‘run-down’, but the threshold potential for ILVA was more negative (-60 mV) than the non-L-type Ca2+ current, and peak ILVA occurred at -40 mV.

At present the physiological significance of ILVA is unknown. At resting membrane potentials in colonic myocytes (i.e. approximately -60 mV; Koh et al. 1999) little of ILVA would be available, and therefore inward current through this conductance would not be likely to participate significantly in action potentials unless availability was altered by agonists. Hyperpolarization of membrane potential is a characteristic response when colonic muscles are stimulated with enteric inhibitory transmitters (e.g. ATP, nitric oxide, vasoactive intestinal polypeptide; Keef et al. 1993, 1994). The hyper-polarization response (lasting up to several seconds) might increase the availability of ILVA channels and this conductance may contribute to the active repolarization of membrane potential (i.e. the restoration of resting membrane potential at the end of inhibitory stimuli) and rebound excitation (i.e. depolarizations that follow inhibitory stimuli). In addition, the availability of ILVA may be affected by excitatory agonists, and future experiments will test whether this conductance is regulated by enteric neurotransmitters and hormones.

In summary, ILVA is activated at negative potentials (half-activated at -60 mV) and cannot be elicited from a holding potential more positive than -60 mV. ILVA is insensitive to block by dihydropyridines, does not run down upon cell dialysis, and is carried by Ca2+. These are properties generally associated with T-type Ca2+ currents (Ertel & Ertel, 1997), although several additional properties of ILVA suggest that it is not a typical T-type Ca2+ current: (i) Ba2+ is not a charge carrier for ILVA and in solutions containing Ca2+, Ba2+ was found to be an inhibitor of ILVA; (ii) T-type Ca2+ channels have low Na+ permeability, but Na+ is a charge carrier of ILVA; (iii) T-type Ca2+ channels are unaffected by Cs+, but ILVA was blocked by external Cs+; (iv) Gd3+, although not a specific inhibitor or NSCC, inhibited ILVA. We could find no report that low concentrations of Gd3+ block T-type Ca2+ channels. Therefore, ILVA does not appear to be a T-type Ca2+ current. Our observations suggest that ILVA is mediated by a novel voltage-dependent, non-selective cation conductance.

Acknowledgements

These experiments were supported by research grants to S. D. Koh (DK57169) and K. M. Sanders (DK41315).

Ancillary