Altered electrical activity in colonic smooth muscle cells from dystrophic (mdx) mice


R. Serio Dipartimento di Biologia cellulare e dello Sviluppo, Laboratorio di Fisiologia generale, Università di Palermo, Viale delle Scienze, 90128 Palermo, Italy Tel.: 39 091 6577509; fax: 39 091 6577501 e-mail:


Because the colon from dystrophic (mdx) mice shows an altered motor pattern, probably due to neural disorders, our aim was to examine the electrophysiological properties of muscle cells and the functionality of nitrergic transmission in circular muscle from normal and mdx colon. Normal colonic cells (resting membrane potential [RMP] about −50 mV) showed spontaneous hyperpolarizations (inhibitory junction potentials; IJPs) and cyclic slow depolarizations were sometimes recorded. Mdx colon had a depolarized RMP (about –36 mV) and spontaneous IJPs, but the cyclic activity was never observed. In the normal colon, Nω-nitro- L-arginine methyl ester ( L-NAME) induced depolarization and abolished the cyclic activity. In the mdx colon, L-NAME caused a slight depolarization. Both preparations displayed the same value of RMP in the presence of L-NAME. In normals, neural stimulation induced nonadrenergic, noncholinergic IJPs composed of fast hyperpolarizations followed by a nitrergic slow hyperpolarization, selectively abolished by L-NAME. In the mdx colon the evoked IJPs were composed only of the initial fast hyperpolarization, the nitrergic component being absent. The hyperpolarization to sodium nitroprusside was not significantly different in both preparations. We conclude that the colon from animals lacking in dystrophin displays different electrophysiological features because of an impairment of nitric oxide function.


Dystrophin, the protein product of the Duchenne muscular dystrophy (DMD) gene,1 is a membrane-associated protein present in normal skeletal, cardiac and smooth muscle cells and in some neurones.2 It is absent in Duchenne patients and in mdx mice,1 the animal models of the disease.3 The relationship between the protein’s absence and muscle pathology is not yet understood and different hypotheses have been proposed. One of these speculates a role for nitric oxide in the pathogenesis of muscular dystrophy. It has been reported that in normal animal skeletal muscle, neuronal nitric oxide synthase (nNOS) is selectively present at the sarcolemmal membrane complexed with dystrophin-associated proteins, while in human DMD and in mdx mice, which lack dystrophin, nNOS is absent from the sarcolemma and accumulates in the cytosol, with a decrease in nNOS activity.4 Physiological actions for nNOS have been well characterized in the peripheral nervous system, where nitric oxide (NO) functions as a nonadrenergic, noncholinergic (NANC) transmitter in numerous pathways, including the gastrointestinal tract. Recent studies have shown some alterations in the motor pattern of proximal and distal colon from dystrophic mice5,6 possibly owing to an impairment of nitric oxide function.5,7

In light of these previous findings, our aim was to evaluate how the dystrophic process in mdx mice leads to an altered motility, and to investigate the possible differences in the electrophysiological properties of gastrointestinal smooth muscle between normal and mdx mouse mutants. In addition, the effects of drugs that interfere with pathways associated with NO signalling were studied in an attempt to clarify how the alteration in the functionality of nitrergic transmission contributes to the disease processes. Preliminary accounts of some of these results have appeared in abstract form.8



Experiments, authorized by the Ministero della Sanità (Rome, Italy) were performed on adult normal (C57BL/10SnJ) and dystrophic (mdx mutants; C57BL/10Sn-Dmd/J) male mice, killed by cervical dislocation. The abdomen was immediately opened and the entire colon was rapidly removed and placed in a dissecting dish filled with oxygenated Krebs solution. The colon was opened via a longitudinal incision made along the mesenteric border and washed of remaining faecal material. The mucosal layer was then gently removed. Longitudinally cut strips, 15 mm long, were dissected from the proximal or distal colon. Each tissue was dissected to a size of 1.5 mm wide over a length of 10 mm and then gradually became wider, to 3 mm. These strips were mounted in a partitioned stimulating chamber9 with the 1.5 mm wide section in the stimulation compartment. The preparation was continuously perfused at a constant rate (3.5 mL min–1) with oxygenated (95% O2 and 5% CO2) and prewarmed (37 °C) Krebs solution. Muscle strips were allowed to equilibrate for 90–120 min before experiments were begun. Nifedipine (1 μmol L–1) was added to the perfusate to reduce muscular contractions, and later, atropine and guanethidine (1 μmol L–1 each) were added to assess NANC conditions.

Intracellular electrical recording

Cells within the circular muscle layer were impaled with glass microelectrodes filled with 3 mol L–1 KCl and having resistances of 50–90 MΩ. Successful impalement of circular muscle cells was indicated by a sharp negative deflection in membrane potential followed by the recording of spontaneously occurring inhibitory junction potentials (IJPs) and of responses to electrical field stimulation (EFS). Membrane potential was measured with an high input impedance electrometer (WPI-intra 767) and outputs were displayed on an oscilloscope (Tektronix 5113) and reproduced on chart paper (Grass Physiograph 79D). The muscle strips were stimulated by a Grass S88 electrical stimulator through a Grass stimulus isolation unit (SIU5).

For experiments in which electrotonic potentials were recorded intracellularly from the circular muscle, an Abe and Tomita bath modified according to Bywater et al.10 was employed. In particular, a ‘T’-shaped preparation was used. The preparations were trimmed such that the cross-bar of the ‘T’ was cut parallel to the longitudinal muscle cells and the vertical portion was cut parallel to the circular muscle cells. The portion of the tissue cut parallel to the circular muscle cells was placed between large external plate electrodes that were used to produce conditioning hyperpolarization in the cells of circular muscle layer. In order to monitor the membrane potential changes evoked by the passage of polarizing current, two recording microelectrodes were used. Prior to microelectrode impalement of a muscle cell, the position of an extracellularly placed electrode was adjusted relative to the position of the intracellular microelectrode (while still extracellular) so that the applied current did not generate a potential difference between the two microelectrodes. The intracellular recording microelectrode was then advanced and impaled into an adjacent smooth muscle cell and the potential difference between the two microelectrodes was taken as the actual membrane potential.

Statistical analysis

All data are given as means ± SE. n refers to the number of animal preparations on which observations were made. Statistical analysis was performed by means by Student’s t-test or analysis of variance, where appropriate. A probability value of less than 0.05 was regarded as significant.


The Krebs solution used in this study contained (in mmol L–1) NaCl 119; KCl 4.5; MgSO4 2.5; NaHCO3 25; KH2PO4 1.2, CaCl2 2.5 and glucose 11.1. The following drugs were used: L-arginine hydrochloride, atropine sulphate, guanethidine monosulphate, tetrodotoxin (TTX), nifedipidine, Nω-nitro- L-arginine methyl ester ( L-NAME) and sodium nitroprusside (SNP) (all from Sigma Chemical Corporation, St. Louis, MO, USA). A stock solution of nifedipine was prepared in ethanol and all the other drugs were dissolved in distilled water. The working solutions were prepared fresh on the day of the experiment by diluting the stock solutions in Krebs and were added to the perfusing solution. Lastly, experiments were conducted in a darkened laboratory because of the photosensitive nature of nifedipine and SNP.


Spontaneous electrical activity

Normal animals.

In the presence of nifedipine (1 μmol L–1), the circular smooth muscle cells of the proximal and distal colon from normal mice had an average membrane potential of –50.17 ± 2.21 mV (n=20) and of –47.20 ± 1.47 mV (n=15), respectively. The electrical activity consisted of tetrodotoxin-sensitive spontaneous hyperpolarizations (IJPs) of variable amplitude and frequency (Fig. 1). In 35% of the preparations, irregular fluctuations of membrane potentials consisting of spontaneous slow depolarizations (about 16 mV high) with superimposed rapid oscillations were recorded (Fig. 2). The duration of this oscillation, estimated at 50% of peak amplitude, was about 60 s and the interval between cyclic depolarization varied from 30 to 140 s in different preparations. The NO synthase inhibitor, L-NAME (100 μmol L–1) induced, within 20 min after its addition, a sustained depolarization of smooth muscle cells (RMP –52 ± 3.4 mV in control and –28.7 ± 2.4 mV in L-NAME; P < 0.05; n=15). No effect was observed on the spontaneously occurring IJPs. L-NAME abolished cyclic depolarizations in the preparations showing them and they were not restored by repolarization of the membrane back to the value before L-NAME treatment (Fig. 2). The NO donor, SNP (1 μmol L–1) produced a membrane hyperpolarization of 12.2 ± 1.4 mV (n=4) that slowly reversed after washout.

Figure 1.

 Intracellular microelectrode recordings from the colonic circular muscle of normal and mdx mice. Electrical activity consisted in both preparations of irregularly occurring spontaneous hyperpolarizations. Note that RMP in mdx colon is lower than that in the normal colon.

Figure 2.

 Effects of L-NAME in the colonic circular muscle of normal and mdx mice. Traces show the electrical activity before and after 20-min perfusion with L-NAME as indicated. Normal animals: the spontaneous cyclic depolarizations are shown. L-NAME induced depolarization and abolished spontaneous cyclic activity, which was not restored when the membrane was repolarized by electrical current back to the value prior to treatment. Mdx mice: spontaneous cyclic depolarizations were not observed even when membrane was hyperpolarized by current. L-NAME induced a slight depolarization. Note that in the presence of L-NAME there is no longer any difference in the value of membrane potentials between normal and mdx colon.

Mdx animals.

In the mdx mice, circular muscle cells from the colon had a more depolarized membrane potential (proximal colon –37.83 ± 0.87 mV, n=18; distal colon –37.44 ± 1.47 mV, n=16; P < 0.05 compared to the values in normal animals) and the electrical activity consisted only of the irregularly occurring TTX-sensitive IJPs (Fig. 1). Spontaneous cyclic depolarization was not observed even when the membrane was hyperpolarized with electrical current to the values observed in the normal animals (Fig. 2). In the mdx colon also, L-NAME caused membrane depolarization (RMP –38.3 ± 1.7 mV in control and –25.0 ± 2.9 mV in L-NAME; P < 0.05; n=10) (Fig. 2). Once again, spontaneous IJPs were not affected by L-NAME. The membrane potential values attained in the presence of L-NAME in normal and in mdx colon were not statistically different (P > 0.05; n=10). SNP (1 μmol L–1) was able to induce a membrane hyperpolarization of 11.7 ± 1.8 mV (n=4; P > 0.05 when compared to the values in normal animals).

Responses to transmural electrical stimulation

Normal animals.

The amplitude of the evoked IJPs has been reported to vary during the cycling of myoelectric complexes.11 Therefore, the responses to inhibitory NANC nerve-stimulation experiments in normal animals were carried out only in the preparations lacking the cyclic depolarizations. Under NANC conditions, EFS with single pulses, 0.6 ms duration, at variable stimulus strength induced an inhibitory response, an IJP, characterized by two distinct components: an initial rapidly developing hyperpolarization followed by a slower and smaller amplitude hyperpolarization (Fig. 3). According to the terminology of Shuttleworth et al.12 we have termed the initial hyperpolarization the ‘fast component’ and the second one the ‘slow component’ of IJPs. The amplitude of both components of the IJPs was dependent upon the strength of the applied stimulus reaching a maximum at 80 V (Fig. 4). The duration of the slow component, measured as the time between maximum hyperpolarization and return to the membrane potential recorded before stimulation, was only slightly dependent on the stimulus strength (Fig. 4). Both components of the IJP were blocked by TTX (1 μmol L–1; n=3). L-NAME (100 μmol L–1) did not affect the fast IJP component directly, but slightly increased its amplitude, probably as a result of induced depolarization. However, the slow component of the evoked responses was abolished at all stimulus strengths tested (Fig. 3 and Fig. 4). Long duration exposure to L-arginine (0.1 mmol L–1, 1 h) had no effect on membrane potential or on IJPs. No difference in IJP features between proximal and distal colon were found.

Figure 3.

 Inhibitory responses to electrical field stimulation in normal and mdx mice and effects of L-NAME. NANC IJPs were evoked by electrical field stimulation (single pulse, 0.6 ms, 100 V) in circular muscle of colon from normal and mdx mice, in the absence and in the presence of L-NAME. Note that L-NAME selectively blocked the slow component of IJPs observed only in normal colon.

Figure 4.

 Summary of the effects of L-NAME on the inhibitory responses to electrical field stimulation in normal animals. Histograms show the effects of 100 μmol L–1L-NAME on the amplitude of the ‘fast’ and the ‘slow’ components and on the repolarization time (interval between maximum hyperpolarization and return to the membrane potential recorded before stimulation) of IJPs evoked by single pulses at different stimulus strength. Note that L-NAME affects only the slow component of IJPs. Data are expressed as means ±SE (n=10). *P < 0.05 when compared to the respective value obtained in the normal animals.

Mdx animals.

In the mdx proximal colon the responses to single pulse EFS consisted of a fast, brief hyperpolarization followed by a rapid repolarization and return to the membrane potential observed before stimulation within 1.5 s (Fig. 3). The amplitude of the IJP was dependent upon the stimulus strength reaching a maximum at 100 V (Fig. 5). The slower, longer-lasting hyperpolarization following the fast component, observed in the recordings from normal animals, was not observed in mdx mice (Figs 3,5). TTX (1 μmol L–1; n=3) abolished these responses. L-NAME did not affect the amplitude or the repolarization time of the evoked IJPs (Figs 3,5). Long duration exposure to L-arginine (0.1 mmol L–1, 1 h) had no effect on membrane potential or on IJPs, and did not restore the slow component (at 100 V, IJP duration was 1.4 ± 0.2 s and 1.6 ± 0.4 s before and after L-arginine pretreatment, respectively; n=4). Once again, the responses to EFS in the proximal colon were similar to those recorded in the distal colon.

Figure 5.

 Summary of the effects of L-NAME on the inhibitory responses to electrical field stimulation in mdx animals. Histograms show the effects of 100 μmol L–1L-NAME on the amplitude and on the repolarization time (interval between maximum hyperpolarization and return to the membrane potential recorded before stimulation) of IJPs evoked by single pulses at different stimulus strength. Note that L-NAME fails to affect the IJPs. Data are expressed as means ±SE (n=10).


Enteric inhibitory neurotransmission by NANC transmitters regulates gastrointestinal muscles. Among the transmitters coexpressed and probably coreleased by the inhibitory nerve terminals, an important role is played by NO. Circular muscle of murine colon has been shown to have a high density of neurones containing NOS-like immunoreactivity13,14 and different NO functions have been established. In fact, in this tissue tonic release of NO seems to be responsible for the regulation of migrating myoelectric complexes,11,15 and a component of the inhibitory response to nerve stimulation is due to NO release.12,16 Our working hypothesis was that defective nitrergic neurotransmission was responsible for the altered motor pattern observed in dystrophic mice. In fact, our previous investigation has shown that the proximal colon from dystrophic mice has a different motor pattern that appears to be the consequence of a disruption of the motor co-ordination by neural inputs, probably due to an inadequate ongoing release of NO.5 Moreover, Azzena and Mancinelli7 reported that changes in nitrergic transmission are responsible for the reversed polarity of propulsive peristaltic activity observed in the distal colon of dystrophic animals.

A role for NO in the pathogenesis of muscular dystrophy has been already put forward. In fact, in normal skeletal muscle, nNOS is selectively present at the sarcolemmal membrane, interacting with dystrophin-associated proteins. Lack of dystrophin in human DMD and in mdx mice causes translocation of nNOS from the sarcolemma to cytosol, with a decrease in nNOS activity.4 As well as skeletal muscle, splice variants of nNOS with a specific domain for membrane association have been determined in mouse brain17,18 and in rat small intestine19 from normal animals. Thus, it could be possible that in the gastrointestinal tract of mdx mice, lack of dystrophin causes a different cellular localization of nNOS.

Our results strengthen the hypothesis of an impairment of NO function in mdx colon which results in changes in the electrical activity. However, the defect in nitrergic transmission is not in the mechanisms necessary to transduce NO signals (i.e, second messengers or ion channels) nor in L-arginine deficiency.

In particular, the present study shows that circular muscle cells of mdx colon display differences in some electrophysiological features, which may be related to the possible impairment of nitrergic pathways. In particular, the main characteristic of the mdx colon, as compared with the normal, was a more depolarized average resting membrane potential. Moreover, the cyclic depolarizations present in normal colon, although irregular, were never observed in mdx colon. Spencer et al.20 reported that, in mouse colon, membrane potential between myoelectric complexes was maintained under tonic inhibition via spontaneous release of NO and that suppression of inhibitory inputs was involved in the depolarization phase of the myoelectric complexes. The observation that our preparations depolarized in the presence of L-NAME confirms that NO contributes to determine the level of membrane potential. Interestingly, after blockade of NO synthesis, normal and mdx colon display the same resting membrane potential, suggesting that a reduced nitrergic inhibitory influence could account for the depolarized resting membrane potentials observed in mdx colon. Moreover, L-NAME abolished cyclic activity, when present, in the normal animals. The associated depolarization was not the cause of this, because repolarization back to the original membrane potentials did not bring back the cyclic activity. These observations confirm the conclusion of Spencer et al.20 regarding an involvement of NO in the genesis of the membrane oscillations. Therefore, the lack of cyclic depolarization in mdx colon can be ascribed to an impaired NO function. Moreover, it does not seem to be a consequence of the level of membrane potential, as membrane repolarization did not reveal cyclic oscillations.

Evidence of a role for NO in the murine colonic function and its impairment in dystrophic conditions is strengthened by the results for NANC electrically evoked responses. In normal murine colon, as in other gastrointestinal muscles,21[22][23][24][25]–26 IJPs show two phases, a fast and a slow hyperpolarization. The initial fast component of IJPs was similar in both preparations and unaffected by L-NAME. The neurotransmitter(s) responsible for this initial phase of IJP is currently unknown, but it would activate an apamin-sensitive conductance.16,27 Moreover, the same conductance would be involved in the spontaneously occurring IJPs.16 Although the identification of this neurotransmitter was not the aim of our study, we can conclude that the neural pathway involved in the fast phase of IJPs and in the spontaneously occurring IJPs is not altered in mdx colon. Moreover, NO is the mediator responsible for the small-amplitude, longer-lasting component of IJPs observed in normal mouse colon, as it was abolished by L-NAME. In contrast, in the mdx colon the evoked IJP lacks the nitrergic component, at least at the EFS parameters we used. It has to be noted that, in our previous experiments, the NO-dependent relaxations observed in the colon of mdx mice were not significantly different from those of normal animals.5 However, in mechanical experiments, a period of stimulation longer than we used in the present study was applied. A longer period of stimulation may result in the production of enough NO to overflow onto smooth muscle cells and directly elicit mechanical responses in these cells.

In the gastrointestinal tract, inputs from the enteric nervous system can be mediated or transduced by specific classes of interstitial cells of Cajal (ICC).28,29 An essential role for intramuscular ICC has been shown in the murine stomach30 and in mouse lower oesophageal and pyloric sphincters31 where they are supposed to be the effectors that transduce NO signals into hyperpolarization response. However, the observation that mdx colon is able to respond to sodium nitroprusside in the same way as the normal colon suggests that the transduction and effector mechanisms that convert NO to hyperpolarization response, wherever they are expressed, are intact in the mdx colon. Lastly, the failure of L-arginine to restore the nitrergic component of IJPs in mdx colon suggests that L-arginine deficiency is not responsible for inefficient inhibitory neurotrasmission.

In conclusion, the present studies show that colon from animals lacking dystrophin displays different electrophysiological features. An impairment of NO-mediated neurotransmission, but not NO response, is likely to be involved.


This work was supported by grant no. 1134 from Comitato Telethon Fondazione ONLUS-Italy. The authors are grateful to Professor Emeritus E. E. Daniel (McMaster University, Hamilton, Canada) for helpful advice and suggestions.