Prof. Miyako Takaki, Department of Physiology II, Nara Medical University, 840 Shijo-cho, Kashihara, Nara 634-8521, Japan. Tel: +81-744-29-8829; fax: +81-744-23-4696; e-mail: firstname.lastname@example.org
The present study was aimed at elucidating how pacemaker activity (plateau potentials) (mean frequency: 15.9 ± 2.8 times min−1) from submucosal interstitial cells of Cajal (ICC-SM) control spontaneous contractions in the mouse proximal colon. Mechanical activities in the circular muscle direction showed high-frequency (mean frequency: 15.6 ± 2.7 times min−1) and low-amplitude (mean amplitude: 0.01 ± 0.005 g) (HFLA) rhythmic contractions. Simultaneous recordings of circular muscle mechanical activity and electrical activity from ICC-SM revealed that HFLA contractions were synchronized with plateau potentials (mean frequency: 15.9 ± 2.8 times min−1). Although low-frequency (3.5 ± 2.1 times min−1) and high-amplitude (0.12 ± 0.03 g) (LFHA) contractions in both longitudinal and circular muscle directions were synchronized with burst of action potentials in both muscle cells, these LFHA contractions were not synchronous with plateau potentials. Intracellular Ca2+ release from the internal stores through IP3 receptors is not a major factor to generate both action potentials differently from plateau potentials. Neither tetrodotoxin nor atropine affected the plateau potentials. The results reveal that the pacemaker activity from ICC-SM drives only the spontaneous HFLA (one-tenth amplitude of the LFHA circular and longitudinal muscle contractions) circular muscle contractions without control by enteric nerves.
Gastrointestinal tissues are spontaneously active and these smooth muscles generate electrical activities such as slow waves or spike potentials.1 Spontaneous pacemaker currents are generated in the interstitial cells of Cajal (ICC).2–5 A variety of driving mechanisms have been reported for the ICC networks in different tissues and species. (i) Submucosal interstitial cells of Cajal (ICC-SM) and circular smooth muscle cells form networks of gap junctions in the canine proximal colon and thus the ICC-SM drives circular muscle activity.6–9 (ii) The myenteric ICC (ICC-MY) drives smooth muscle activities in the colon, stomach or small intestine.2–4 (iii) In the rat colon, two independent and distinct pacemakers (ICC-SM and ICC-MY) drive the longitudinal and circular smooth muscle contractions: slow waves derived from the ICC-SM drive high-frequency and low-amplitude (HFLA) circular smooth muscle contractions and the cyclic oscillatory potentials derived from the ICC-MY drive both low-frequency and high-amplitude (LFHA) circular and longitudinal smooth muscle contractions.10 (iv) Summed signals from ICC-SM and ICC-MY excite circular smooth muscle electrical and mechanical activities in the canine proximal colon.5,6
In the mouse proximal colon, electrical activities recorded from ICC-SM are plateau potentials. The frequency of the plateau potentials is completely different from that of electrical activities in circular and longitudinal muscles. The electrical activities of circular and longitudinal muscle cells are generated at almost the same rhythms.11 Therefore, the role of plateau potentials in the ICC-SM remains unknown. The mouse proximal colon may differ from other parts of the gastrointestinal tract, but may be similar to the rat proximal colon.10
The present study was aimed at elucidating how the pacemaker activity (plateau potentials) derived from ICC-SM controls the spontaneous contractions in the mouse proximal colon. The results indicate that the pacemaker activity from ICC-SM controls the spontaneous HFLA circular muscle rhythmic contractions and the different pacemaker activity from that in ICC-SM controls the spontaneous LFHA synchronous circular and longitudinal contractions. A part of these experimental data was reported briefly in the 79th Annual Meeting of The Japanese Physiological Society.12
Materials and methods
Male mice (BALB/C strain), weighing 20–25 g, were killed by cervical dislocation following anaesthesia with sodium pentobarbital (0.05 mg g−1 i.p.). The animals were treated ethically according to the guidelines for the Care and Use of Animals approved by the Physiological Society of Japan. Proximal colon was isolated, opened along the mesenteric border, the mucosal layer removed and kept in modified Tyrode solution with the following ionic composition (in mmol L−1): NaCl, 136; KCl, 5.4; MgCl2, 1.0; NaH2PO4, 0.33; CaCl2, 1.8; d-glucose, 15.0; and HEPES, 5.0 (pH = 7.4 throughout the experiment).
A small piece of rectangular tissue (2 mm × 10 mm) was isolated and mounted on a silicone rubber plate. Preparations were made to be longer in either circular or longitudinal muscle direction so that either circular (Fig. 1A) or longitudinal muscle contractions are recorded (Fig. 1B). The area (2 mm × 3 mm) was immobilized and fixed at the bottom of the recording chamber by using tiny pins (diameter: 0.1 mm) to record intracellular membrane potentials, either with the submucosal layer side uppermost (for cells in ICC-SM, Fig. 1-a), circular muscle side uppermost by partially (Fig. 1-b) or completely removing submucosal layer (Fig. 1-c) (for circular smooth muscle cells) or longitudinal muscle side uppermost (for longitudinal smooth muscle cells) with intact submucosal layer (Fig. 1-d). The recording chamber (8.0 mm wide × 20.0 mm long × 5.0 mm depth, with a capacity of about 1.0 mL) was made from Lucite plate, and the tissue segment was superfused with oxygenated warmed (35 °C) Tyrode solution at a constant flow rate of about 3 mL min−1.
Conventional microelectrode techniques were used to record electrical activity of a single cell from each tissue. Glass capillary microelectrodes (borosilicate glass tube with 1.2 mm OD) filled with 3 mol L−1 KCl had the tip resistance ranged between 50 and 80 MΩ. The opposite side of the fixed preparation was attached to the isometric force transducer (SS-1930; Nihon Kohden, Tokyo, Japan) under appropriate tension loading, to record force development in the circular or longitudinal muscle direction. The intracellular membrane potentials and muscle contractions thus recorded were displayed on a cathode ray oscilloscope (SS-7602; Iwatsu, Tokyo, Japan). The data were also acquired into a personal computer through an A/D converter (Axon Instruments Inc., Foster City, CA, USA) at 500 Hz, filtered at 100 Hz, and analysed with Axoscope 8 (Axon Instruments Inc.).
Drugs used were as follows: atropine sulphate, bis-(aminophenoxy) ethane-N, N, N′, N′-tetraacetic acid acetoxymethyl ester (BAPTA-AM), tetrodotoxin were purchased from Sigma Chemicals (St Louis, MO, USA), and 2-aminoethoxy-diphenyl borate (2-APB) was generously donated by Dr Mikoshiba of the University of Tokyo. 2-APB and BAPTA-AM were dissolved in dimethyl sulphoxide at concentrations of 5–10 mmol L−1. Other chemicals were dissolved in distilled water as a stock solution, and diluted further with Tyrode solution to desired concentrations (the ratios of the dilution were over 1 : 1000). The dilution procedures did not alter the pH of the Tyrode solution.
Values measured are expressed as the mean ± standard deviation (SD), with n representing the number of preparations taken from different animals. Differences between values were tested using paired Student's t-test, and P < 0.05 was considered significant.
Properties of electrical activities recorded from the proximal colon
With intracellular recordings, square-shaped potentials changes (plateau potentials) were obtained from submucosal layer uppermost tissues (Fig. 1-a) prepared from the proximal colon of mice (Fig. 2A). The cells generating these plateau potentials correspond to submucosal c-Kit-positive cells, i.e. ICC-SM.11 The plateau potentials were generated at the frequency ranging between 13 and 20 times min−1, with mean frequency of 15.9 ± 2.8 times min−1 (n = 10). The primary component had peak amplitude ranged between 12 and 36 mV (mean: 20.5 ± 9.3 mV; n = 10) and the rate of rise ranging between 70 and 330 mV s−1 (mean: 150 ± 60 mV s−1; n = 10), and the duration of plateau components measured at the foot ranged between 2.0 and 3.1 s (mean: 2.4 ± 0.4 s; n = 10). The resting membrane potential of cells generating plateau potentials ranged between −42 and −60 mV (mean: −50.3 ± 6 mV; n = 10). These properties of plateau potentials were similar to those reported previously11. Plateau potentials were not affected by 1 μmol L−1 tetrodotoxin or 1 μmol L−1 atropine (data not shown, n = 3 for each), suggesting that their generations were not causally related to periodical excitation of cholinergic enteric nerves.
In tissues with circular muscle layer exposed upside by removing the submucosal layer (Fig. 1-b), rhythmic electrical activity was recorded in circular muscle cells (Fig. 2B). The resting membrane potential ranged between −40 and −56 mV (mean: −45.1 ± 5.9 mV; n = 19) and the frequency ranged between 2 and 7 times min−1 (mean: 3.9 ± 1.5 times min−1; n = 19). The properties of electrical activity of circular muscle cells were similar to those reported previously.11
In tissues with the longitudinal muscle layer uppermost (Fig. 1-d), rhythmic electrical activity was recorded in longitudinal muscle cells (Fig. 2C). Although the form of electrical activity in longitudinal muscle cells somewhat differed from those of the plateau potential and the circular muscle electrical activity, the resting membrane potential ranging between −42 and −57 mV (mean: −48.1 ± 5.4 mV) and the frequency ranging between 2 and 7 times min−1 (mean: 3.4 ± 1.2 times min−1; n = 21) mimicked those of the circular muscle electrical activity. The properties of electrical activity of longitudinal muscle cells were similar to those reported previously.11
Properties of longitudinal and circular mechanical activities recorded from the proximal colon
In tissues with submucosal layer attached (Fig. 1A-a), LFHA periodical rhythmic contractions were recorded in the circular muscle direction (Fig. 3A). The frequency of these contractions ranged between 2 and 6 times min−1 (mean: 3.9 ± 1.5 times min−1; n = 15) and the amplitude of these contractions ranged between 0.03 and 0.2 g (mean: 0.12 ± 0.07 g; n = 15). Between these contractions, HFLA rhythmic contractions in the circular direction were also recorded (Figs 3A and 4A). The frequency of these contractions ranged between 12 and 21 times min−1 (mean: 15.6 ± 2.7 times min−1; n = 15) and the amplitude ranged between 0.003 and 0.015 g (mean: 0.01 ± 0.005 g; n = 15). The mean frequency was nearly equal to that of the plateau potentials, 15.9 ± 2.8 times min−1 (n = 10). LFHA periodical rhythmic contractions in the same type of tissue (Fig. 1B-a) were also recorded in the longitudinal muscle direction (Fig. 3B). The frequency of these contractions ranged between 2 and 4.5 times min−1 (mean: 3.4 ± 1 times min−1; n = 15) and the amplitude of these contractions ranged between 0.03 and 0.19 g (mean: 0.1 ± 0.04 g; n = 15). HFLA longitudinal contractions were not recorded between LFHA longitudinal contractions in all cases. There are no significant differences in the frequency and amplitude of LFHA contractions recorded in the preparations with submucosal layer (Fig. 1-a,d) between circular (Fig. 1A) and longitudinal muscle directions (Fig. 1B).
Simultaneous recordings of mechanical and electrical activities in the proximal colon
Simultaneous recordings of mechanical and electrical activity were made to elucidate how electrical activity generated from each cell regulates the two patterns of mechanical activity in 37 tissues. Almost all the preparations used in this series of experiments (n = 30) were submucosal layer attached.
Simultaneous recordings of plateau potentials and circular muscle or longitudinal muscle mechanical activities
In 12 tissues (Fig. 1-a), the LFHA contractions in circular (Fig. 1A) (CMM in Fig. 4A; n = 6) and longitudinal directions (Fig. 1B) (LMM in Fig. 4B; n = 6) were not generated synchronously with the plateau potentials (ICC-SM in Fig. 4A,B). Although the rhythm of the LFHA contractions was quite different from that of the plateau potentials generated from ICC-SM (Fig. 4A,B), the HFLA contractions were generated synchronously with the plateau potentials from ICC-SM (Fig. 4A).
Circular muscle mechanical activities and longitudinal and circular electrical activities with and without submucosal layer
In five tissues (Fig. 1A-d), the mechanical activities were recorded in the circular muscle direction but the electrical activities were recorded in longitudinal muscle cells. Between the LFHA contractions synchronous with burst of action potentials recorded in longitudinal muscle cells, the HFLA rhythmic circular muscle contractions were observed (CMM in Fig. 5A).
In seven tissues without submucosal layer (Fig. 1A-c), the HFLA rhythmic circular muscle contractions were not observed (CMM in Fig. 5B), although the LFHA periodic rhythmic contractions were still recorded in the circular muscle direction. Mean frequency (3.5 ± 2.1 times min−1; n = 7) and amplitude (0.12 ± 0.03 g; n = 7) of the LFHA circular muscle contractions synchronous with burst of action potentials in circular muscle cells were unchanged after removal of the submucosal layer (Fig. 5B).
Role of intracellular Ca2+ on electrical activities recorded in circular and longitudinal muscle cells
Intracellular Ca2+ plays an important role to generate plateau potentials in ICC-SM11. Experiments were carried out to investigate the cellular mechanisms underlying for the generation of the LFHA contractions synchronous with longitudinal and circular muscle electrical activities. Chemicals tested were BAPTA-AM, an intracellular Ca2+ chelator,13 and 2-APB, an inhibitor of IP3 receptor-mediated release of Ca2+ from internal stores.14, 15 Each concentration of BAPTA-AM and 2-APB is sufficient to abolish plateau potentials of cells in ICC-SM.11
Application of 20 μmol L−1 BAPTA for 15 min did not change the frequency of burst of action potentials in circular muscle cells (control: 3.7 ± 1.1 times min−1; in BAPTA: 3.7 ± 1.1 times min−1; n = 6; P > 0.05) or in longitudinal muscle cells (control: 3.7 ± 1.1 times min−1; in BAPTA: 3.7 ± 1.1 times min−1; n = 6; P > 0.05) without changing resting membrane potentials (LME in Fig. 6A).
Application of 10 μmol L−1 2-APB for 10 min depolarized the membrane potentials in circular (23.7 ± 3.8 mV; n = 4) and longitudinal muscle cells (29 ± 10.6 mV; n = 4), and increased the frequency of burst of action potentials and finally generated action potentials continuously (LMR in Fig. 6B).
In smooth muscle tissues of the mouse proximal colon, three types of spontaneously active cells were recorded, as reported previously.11 Visualization of the impaled cells using neurobiotin injection indicates that the cells generating plateau potentials are ICC-SM and the cells generating burst of action potentials are circular muscle cells and longitudinal muscle cells, respectively11. In recording the mechanical activity in the mouse proximal colon, we found the HFLA rhythmic contractions in the circular muscle direction and the LFHA periodical rhythmic contractions synchronously in the longitudinal and circular muscle directions. These mechanical activity patterns are similar to those of the rat colon.10 However, no clear evidence exists as to how the pacemaker activity (plateau potentials) from ICC-SM controls the spontaneous contractions in the mouse proximal colon due to the difficulty in simultaneous recordings of electrical and mechanical activities. We succeeded in making simultaneous recordings of mechanical and electrical activities in any combinations of circular and longitudinal muscles.
The results from the simultaneous recordings of mechanical and electrical activities indicate that HFLA rhythmic contractions in the circular muscle direction were synchronized with plateau potentials from ICC-SM and were abolished after removal of the submucosal layer where ICC-SM was distributed. Thus, the circular muscle HFLA contractions are driven by plateau potentials generated by ICC-SM. On the contrary, LFHA periodic rhythmic contractions with completely different frequencies from that of the plateau potentials were synchronous in circular and longitudinal muscle directions. The LFHA contractions were synchronized with a burst of action potentials in both circular and longitudinal muscle cells. These mechanical and electrical activities were not changed after removal of the submucosal layer. This result suggests that the LFHA contractions and the burst of action potentials are not generated by ICC-SM, but by the different pacemaker mechanisms.
We conclude that the HFLA circular contractions are driven by the pacemaker activity from ICC-SM and the LFHA circular and longitudinal contractions are driven independently by the different pacemaker activity in the mouse proximal colon. The activity recorded in the mouse colon is very similar to the dog, i.e. two discrete pacemaker sites. The main difference is that the ICC-MY appears to be dominant one in the mouse whereas ICC-SM may be dominant in the canine model.16–18
We could not obtain any evidence about the presence of the cells generating the different pacemaker activities from that in ICC-SM, but we speculated that a pacemaker different from the ICC-SM is involved. We accidentally (only three times) recorded the cyclic oscillatory potentials in some cells in different preparations (our unpublished observation). These cyclic oscillatory potentials were recorded from the longitudinal muscle layer side unintentionally. Although the morphological features of these cells could not be detailed, they may be ICC-MY or ICC distributed in the longitudinal muscle layer (ICC-LM) and may contribute in generating different pacemaker activities.
In the mouse colon, we have previously reported that plateau potentials from ICC-SM are abolished by CPA or 2-APB and greatly inhibited by BAPTA, indicating that IP3 is one of the key factors to the generation of plateau potentials.11 In contrast to the plateau potentials, bursts of action potentials are neither abolished nor inhibited by 2-APB or BAPTA in the present study. The results of 2-APB and BAPTA indicate that the release of Ca2+ from the internal stores through IP3 receptors does not play a major role for the different pacemaker activities from that in ICC-SM.
We conclude that in the mouse proximal colon, the pacemaker activity from ICC-SM regulates only the spontaneous HFLA (one-tenth of the LFHA contractions) rhythmic circular muscle contractions,19 the different pacemaker activity from that in ICC-SM regulates the spontaneous LFHA periodical rhythmic contractions synchronously in the circular and longitudinal muscle directions. In contrast to the pacemaker activity in ICC-SM, intracellular Ca2+ release from the internal stores through IP3 receptors does not play a major role for the generation of the different pacemaker activity from that in ICC-SM. Further studies are needed to evidence the presence of the different pacemaker cells.
This work was supported by Grants-in-aid for Scientific Research (14370189, 14657311) from the Ministry of Education, Science, Sports and Culture of Japan.