Interstitial cells of Cajal associated with the myenteric plexus (ICC-MP) are pacemaker cells of the small intestine, producing the characteristic omnipresent electrical slow waves, which orchestrate peristaltic motor activity and are associated with rhythmic intracellular calcium oscillations. Our objective was to elucidate the origins of the calcium transients. We hypothesized that calcium oscillations in the ICC-MP are primarily regulated by the sarcoplasmic reticulum (SR) calcium release system. With the use of calcium imaging, study of the effect of T-type calcium channel blocker mibefradil revealed that T-type channels did not play a major role in generating the calcium transients. 2-Aminoethoxydiphenyl borate, an inositol 1,4,5 trisphosphate receptor (IP3R) inhibitor, and U73122, a phospholipase C inhibitor, both drastically decreased the frequency of calcium oscillations, suggesting a major role of IP3 and IP3-induced calcium release from the SR. Immunohistochemistry proved the expression of IP3R type I (IP3R-I), but not type II (IP3R-II) and type III (IP3R-III) in ICC-MP, indicating the involvement of the IP3R-I subtype in calcium release from the SR. Cyclopiazonic acid, a SR/endoplasmic reticulum calcium ATPase pump inhibitor, strongly reduced or abolished calcium oscillations. The Na-Ca exchanger (NCX) in reverse mode is likely involved in refilling the SR because the NCX inhibitor KB-R7943 markedly reduced the frequency of calcium oscillations. Immunohistochemistry revealed 100% colocalization of NCX and c-Kit in ICC-MP. Testing a mitochondrial NCX inhibitor, we were unable to show an essential role for mitochondria in regulating calcium oscillations in the ICC-MP. In summary, ongoing IP3 synthesis and IP3-induced calcium release from the SR, via the IP3R-I, are the major drivers of the calcium transients associated with ICC pacemaker activity. This suggests that a biochemical clock intrinsic to ICC determines the pacemaker frequency, which is likely directly linked to kinetics of the IP3-activated SR calcium channel and IP3 metabolism.
- interstitial cells of Cajal
- myenteric plexus
- intestinal pacemaker activity
- sarcoplasmic reticulum
- inositol 1,4,5 trisphosphate receptor
- phospholipase c
interstitial cells of cajal (ICC) are pacemaker cells (33, 56) responsible for initiating and propagating electrical slow waves (13) essential for several motor patterns of the gastrointestinal tract (23, 51). The ICC associated with the myenteric plexus of the small intestine (ICC-MP) spontaneously oscillate between high and low intracellular calcium (57, 59) at the same frequency as their electrical pacemaker activity (45). Before calcium oscillations were demonstrated, intracellular calcium was already thought to be a major factor in initiating pacemaker activity; the frequency of pacemaker activities generated by the ICC associated with the submuscular plexus in the dog colon (37) and the ICC-MP from the mouse small intestine (38) were dose dependently reduced by cyclopiazonic acid (CPA), consistent with involvement of calcium release from the sarcoplasmic reticulum (SR). No effect was observed with ryanodine, suggesting no involvement of ryanodine-sensitive calcium release channels. Rhythmic inositol 1,4,5 trisphosphate (IP3)-induced calcium release may be part of the pacemaker-generating mechanism because blockers of IP3-induced calcium release abolished slow-wave activity in the mouse small intestine (38). Importantly, mice without IP3 receptors (IP3R) do not exhibit slow-wave activity (54). At the single cell level, pacemaker currents, recorded in ICC from the mouse small intestine, were abolished by inhibition of IP3-induced calcium release (58).
The role of cell membrane calcium channels and exchangers in pacemaking has not been fully elucidated and will be different in various subtypes of ICC and different in different organs (40). Mouse intestine ICC, chemically isolated and put into short-term culture, displayed both nifedipine- (or nicardipine-) sensitive and nifedipine-insensitive, mibefradil-sensitive inward currents, suggesting the presence of both L- and T-type calcium channels (27). L-type calcium currents were also identified in ICC from the guinea pig stomach (28). In mouse intestinal ICC, the α1H Ca2+ channel subunit was identified (14), and both T-type and L-type calcium channels appear upregulated compared with smooth muscle using differential gene expression (9).
Slow-wave activity is shown to be persistent in the presence of L-type calcium channel blockers in almost all studies (31) (21) and so are spontaneous calcium oscillations (2, 59). The role of T-type calcium channels in slow-wave generation is less well defined. Mibefradil, 10 μM, hardly affected slow-wave activity in one study (3), whereas 2 μM reduced the frequency from 40 to 35/min in another study (14), suggesting a minor role in slow-wave generation. However, it was demonstrated that mibefradil abolished calcium transients in the ICC-MP of the mouse ileum at concentrations between 0.1 μM and 1.0 μM. Calcium may also enter through nonselective cation channels, which likely contribute significantly to pacemaker currents (12, 26, 32). Particular attention has been given to transient receptor potential channels (20, 25, 26, 40, 57). In the urethra, the sodium calcium exchanger (NCX) in reverse mode has been shown to be involved in calcium influx in ICC (7).
Despite progress in understanding the role of calcium in pacemaker generation, the exact means of intracellular calcium regulation in ICC is still unknown. The objective of the present study was to increase our understanding of the mechanisms by which the intracellular calcium oscillations are initiated in the ICC-MP, specifically to assess the role of voltage-sensitive calcium channels vs. IP3-induced calcium release in the generation of the calcium transients.
MATERIALS AND METHODS
Procedures to obtain mouse tissue were approved by the McMaster University Animal Research Ethics Board. Adult CD1 mice, 6–8 wk old, were euthanized by cervical dislocation. The entire length of the small intestine was immediately removed from the animal and placed in oxygenated Krebs solution containing 1 μM nicardipine.
A longitudinal muscle myenteric plexus (LMMP) tissue preparation was prepared as follows. A small segment, about 1 cm in length, of jejunum was cut and placed into a Sylgard dish. The segment was then cut along the mesenteric border and pinned out flat with the mucosa facing upward. The mucosa and submucosa were carefully peeled away from the muscle layers and discarded. The circular muscle (CM) layer was then carefully peeled away form the LM layer, exposing the myenteric plexus. Some areas were left with CM intact to help maintain the integrity of the tissue.
Fluo-4 AM calcium imaging was used to visualize calcium oscillations in the ICC-MP. Tissue preparations were incubated with 5 μM Fluo-4 AM, containing 0.02% pluronic F-127, for 20 min at room temperature. A washout with Krebs solution at 37°C was performed for a period of 20 min to allow for full deesterification of AM esters. Tissue preparations were continuously perfused with fresh oxygenated Krebs solution containing 1 μM nicardipine. Nicardipine was added to avoid strong contractile activity, which makes analysis difficult or impossible.
Live tissue c-kit staining and double labeling with Fluo-4 loading.
To determine whether the cells exhibiting Fluo-4 AM calcium oscillations were in fact ICC, c-Kit staining was done. The technique for live tissue ICC staining was previously described (15); ACK2 (rat anti-mouse c-Kit antibody; eBioscience, San Diego, CA) was conjugated to Alexa Fluor 594 (Alexa Fluor 594 Monoclonal Antibody Labeling Kit; Molecular Probes, Burlington, ON, Canada) before staining. Tissue preparations were incubated with 5 g/ml ACK2-Alexa Fluor 594 in 1 ml Krebs solution for 1.5 h at 40°C.
Drugs and solutions.
Physiological Krebs solution (pH 7.35–7.40) containing (in mM) 118.1 NaCl, 1.0 NaH2PO4, 1.2 MgSO4, 2.5 CaCl2, 4.8 KCl, 11.1 glucose, and 25 NaHCO3 was used and continuously oxygenated with 95% O2 and 5% CO2. Mibefradil, CPA, and 2-aminoethoxydiphenyl borate (2-APB) were acquired from Sigma Aldrich (St. Louis, MO), and KB-R7943 and U73122 were purchased from Tocris Bioscience (Ellisville, MO). All pharmacological stock solutions were suspended with Krebs solution to the desired concentration and administered to the tissue via a constant perfusion system.
Imaging and data analysis.
Imaging was performed on an upright Nikon eclipse FN1 microscope using ×10, ×20, and ×40 objectives with a GFP filter to excite the Fluo-4 AM dye at a wavelength of 488 nm. Image recordings were obtained using a Quantem 512SC camera at a frame rate of 5–10 frames per second and Nikon-NIS Elements, advanced research, imaging software. Calcium oscillation traces were created in the Nikon-NIS elements program using a minimum of three regions of interest (ROIs) per experiment. ROIs were selected free hand and encompassed the cell body of the ICC to be analyzed. The ROIs measured the average pixel intensity in that region for the duration of the recording. These measurements were then exported to Excel and Clampfit for further analysis of frequency and amplitude. Frequency was calculated as the average number of oscillations per minute. The potential decline of the amplitude of calcium transients in response to drugs was assessed as follows: amplitude was calculated as the average peak intensity, in arbitrary units, and presented in text and bar graphs as the percent decline of the signal and was compared with the percent decline of control values. The t-test was used on two samples with equal variances to determine statistically significant differences at P = 0.05. Control values will be different dependent on exposure protocols. In each set of experiments, the exposure protocols for control and test conditions were identical. Separate control experiments were completed for each experimental condition.
Double-immunohistochemical labeling of c-Kit and NCX and c-Kit, IP3R-I, IP3R-II, and IP3R-III.
Proximal jejunums from four CD1 mice were removed. Both wholemount and frozen sections were prepared. The same LMMP wholemount preparations as calcium image study were made under the dissection microscope. Other tissues were embedded in Tissue-Tek and frozen with liquid nitrogen. Frozen sections of 10 μm thick were cut with a Cryostat and mounted on coated slides. Both wholemounts and sections were fixed in ice-cold acetone at 4°C for 10 min for anti-c-Kit staining. After c-Kit staining, tissues were either fixed again in cold 4% paraformaldehyde for 10 min for anti-NCX staining or continuously processed for IP3Rs staining. Tissues were incubated with 2% albumin bovine serum for 1 h to block the nonspecific staining. The incubation time with primary antibodies [rat anti-c-Kit (1:200), Cedarlane Laboratory, Burlington, ON Canada; rabbit anti-NCX (1:100), rabbit anti-IP3R type I (IP3R-I) (1:100), goat anti-IP3R-II (1:100), and goat anti-IP3R-III (1:100); Santa Cruz Biotechnology, Santa Cruz, CA] were 18–24 h at room temperature. After being rinsed in PBS, tissues were incubated with secondary antibodies. For double staining of c-Kit/NCX and c-Kit/IP3R-I, Cy2-conjugated anti-rat IgG (1:50, for c-Kit staining) and Cy3-conjugated anti-rabbit IgG (1:500, for NCX and IP3R-I staining) were used. For double staining of c-Kit/IP3R-II and c-Kit/IP3R-III, Cy2-conjugated anti-rat IgG (1:50, for c-Kit staining) and Cy3-conjugated anti-goat IgG (1:500, for IP3R-II and IP3R-III staining) were used. All secondary antibodies were from Jackson ImmunoResearch (West Grove, PA). The buffer for rinsing and antibody dilution was 0.05 M PBS (pH 7.4) with 0.3% Triton X-100. Negative control staining was performed by omitting primary antibodies from the incubation solutions. Reactions were examined with a confocal microscope (LSM 510; Zeiss, Jena, Germany) with excitation wavelength appropriate for the Cy2 (492 nm) and Cy3 (550 nm).
c-Kit-positive ICC exhibit calcium oscillations.
Spontaneous rhythmic oscillations in intracellular calcium, visualized with Fluo-4 AM (Fig. 1B), occurred in c-Kit-positive cells (Fig. 1A) that constitute the network of ICC-MP (Fig. 1, Supplemental Movie S1; supplemental material for this article is available online at the American Journal of Physiology Gastrointestinal and Liver Physiology website).
General characteristics of calcium oscillations.
The ICC-MP displayed rhythmic oscillations in calcium at a frequency of 22.8 ± 4.8 cycles/min (cpm) (n = 9). These oscillations were spontaneous in nature and were synchronized within the ICC-MP network (Fig. 2A). Under control conditions there was a reduction in the amplitude of calcium transients over time attributable to bleaching of the photo-sensitive Fluo 4 AM dye and potentially attributable to loss of dye from cells (Fig. 2A). To minimize this, recordings were made during short periods of light exposure. In all figures, recordings of 30-s duration are shown at various time points.
Voltage-gated calcium channels.
L-type calcium channels were not essential in the generation of the calcium transients because all experiments were performed in the presence of nicardipine (1 μM). T-type calcium channels have been demonstrated to play a role in calcium oscillations in the ICC-MP (45); we attempted to confirm this using the T-type calcium channel blocker, mibefradil. However, mibefradil (0.5 μM) had no effect on the frequency of calcium transients (Fig. 2B). The average frequency of calcium oscillations in the ICC-MP under control conditions for the first 30 s of recording was 22.3 ± 3.6 cpm. Twenty-one minutes after addition of mibefradil, the frequency of oscillations was 19.3 ± 4.8 cpm (n = 8, P = 0.27). There was also no effect of 0.5 μM mibefradil on the amplitude of calcium transients (compared to controls) for the duration of the experiments. After 21 min, the average amplitudes were 37.8 ± 9.1% of the intial value at t = 0 min (see materials and methods) in the presence of mibefradil vs. 34.9 ± 8.1% for control, not significantly different (P = 0.57).
The addition of 5 μM KB-R7943, a reverse-mode NCX inhibitor, to the ICC-MP caused a decrease in calcium oscillation frequency (Fig. 2C; Supplemental Movie S2). The average control frequency of 26.8 ± 8.5 cpm declined to 5.8 ± 3.4 cpm (n = 6, P = 0.001) 17 min after the addition of KB-R7943. The addition of KB-R7943 reduced the amplitude of calcium transients in the ICC-MP network after 17 min to 10.2 ± 7.3% compared with control values of 24.4 ± 11.3% (n = 6, P = 0.03).
IP3-mediated calcium release from the SR.
The addition of the intracellular IP3-induced calcium release inhibitor, 2-APB (30 μM), reduced the frequency of calcium oscillations in the ICC-MP (Fig. 3A) from 24.6 ± 7.0 cpm to 9.9 ± 5.9 cpm (n = 5, P = 0.048) at 16 min. In addition to a decrease in frequency, the calcium transients in two of the five experiments became unsynchronized (Fig. 3A). Furthermore, the time needed to complete one oscillation of calcium became longer after modulating the IP3 receptor (2.8 ± 0.3 s vs. 6.1 ± 2.1 s after 2-APB). The duration of the calcium transient increased probably because the rate of calcium reuptake decreased. The average amplitude of transients after the addition of 2-APB was 36.9 ± 16.5% compared with 24.4 ± 5.5% for controls; not significantly different (P = 0.20). In fact, 2-APB increased the amplitude of transients in some individual cells (Fig. 3A).
A role for IP3 was also investigated by blocking phospholipase C using U73122 (5 μM), inhibiting the hydrolysis of phosphatidylinositol 4,5-bisphosphate to IP3. This reduced the frequency of calcium oscillations in the ICC-MP from 26.4 ± 8.9 cpm to 7.2 ± 13.0 cpm at 27 min (n = 5, P < 0.05) (Fig. 3B). The amplitude of calcium transients was also significantly reduced by U73122; transients had an average amplitude of 5.8 ± 6.6%, 27 min after exposure to U73122, compared with control values of 18.5 ± 5.5% (P = 0.008). In two out of five experiments, the oscillations were abolished. To assure specificity of action of U73122, the inactive analog, U73343 (5 μM), was also investigated. U73343 had no effect on the frequency of calcium transients where the initial frequency, before the addition of U73343, was 22.5 ± 6.6 cpm and the frequency of oscillations 27 min after the addition of U73343 was 23.0 ± 5.8 cpm (Fig. 3C, n = 4, P = 0.6). The amplitude of calcium transients was also unaffected by U73343; transients had an average amplitude of 16.8 ± 5.7% 27 min after exposure to U73343, compared with control values of 18.5 ± 5.5%.
IP3R-I, IP3R-II, and IP3R-III immunoreactivity.
Section staining revealed that IP3R-I (Fig. 4) reactivity was much stronger in LM than in CM cells, whereas IP3R-II (Supplemental Fig. S1) and IP3R-III (Supplemental Fig. S2) immunoreactivities were very weak in both muscle layers. Many IP3R-I-positive cells were found around but separate from the myenteric ganglia (Fig. 4), whereas IP3R-II- (Supplemental Fig. S1) and IP3R-III-positive cells (Supplemental Fig. S2) were mainly located within the myenteric ganglia, occasionally scattered outside the ganglia. Double labeling with c-Kit revealed IP3R-I immunoreactivity in ICC-MP. Most ICC-MP expressed IP3R-I immunoreactivity, whereas only about 50% of IP3R-I-positive cells expressed c-Kit immunoreactivity (Fig. 4, C, F, and I). Double labeling of c-Kit/IP3R-II (Supplemental Fig. S1, C, F, and I) and c-Kit/IP3R-III (Supplemental Fig. S2, C, F, and 2I) displayed neither IP3R-II nor IP3R-III immunoreactivity in ICC-MP.
Calcium refilling of the SR.
CPA, at a concentration of 10 μM, reduced the frequency of calcium transients (Fig. 5A) and in three of five experiments completely abolished calcium oscillations at 35 min. CPA inhibits the SR calcium ATPase, which is responsible for pumping calcium into the SR. The frequency under control conditions was an average of 25.0 ± 13.5 cpm (n = 5) and was reduced to 4.8 ± 5.5 cpm (n = 5, P = 0.02) at 30 min. The amplitude of calcium oscillations was also reduced 30 min after the addition of CPA to 10.5 ± 9.4% compared with control amplitudes after 30 min of 18.5 ± 5.5% (P = 0.02).
NCX-positive cells in musculature.
Cross-section staining using antibodies to NCX and c-Kit showed that ICC-MP express NCX positivity (Fig. 6). They were frequently observed encompassing the NCX-negative myenteric ganglia. Most smooth muscle cells, especially the circular muscle cells, were also NCX positive, but the NCX immunoreactivity in ICC was stronger than in adjacent smooth muscle cells (Fig. 6A2). c-Kit-positive mast cells in the mucosa and submucosa were NCX negative (Fig. 6, A1–C1). Double labeling of wholemount preparations revealed 100% colocalization of c-Kit and NCX in ICC-MP (Fig. 6, D–F).
Modulators of mitochondrial function.
CGP-37157, a mitochondrial NCX inhibitor, was used to test the potential role of mitochondria in the generation and maintenance of calcium oscillations in the ICC-MP. Calcium oscillations under control conditions had a frequency of 21.3 ± 7.4 cpm and were not significantly affected by the addition of CGP-37157. Thirty minutes after the addition of 20 μM CGP-37157 the frequency of oscillations was 16.4 ± 9.1 cpm (Fig. 7, n = 9, P = 0.09). However, in two of nine experiments, a drastic decline in frequency was observed. The amplitude of oscillations was variable but on average unchanged in the presence of the mitochondrial NCX inhibitor where experiments with CGP-37157 had an amplitude intensity of 15.9 ± 7.5% compared with control values of 18.5 ± 5.5% (P = 0.81).
The ICC-MP of the small intestine show spontaneous calcium oscillations that are linked to the generation of the electrical slow waves (45, 57, 59). The present study contributes to our knowledge about the origin of the calcium transients. Block of IP3 synthesis reduced calcium transients immediately, suggesting constitutive IP3 synthesis in ICC. Depending on the kinetics of synthesis and breakdown and possible feedback mechanisms, this may result in an oscillating IP3 concentration that may contribute to the rhythmicity of the calcium transients. This may work in concert with rhythmic IP3-induced calcium release, where the rate of refilling may determine the frequency of release (8). Rhythmic release may also relate to calcium-induced inhibition and activation of the IP3 receptor (55). Our data are consistent with the hypothesis that the NCX, working in reverse mode, is critical for refilling of the calcium store. Calcium release from the SR involves IP3R-I, not IP3R-II and IP3R-III that are not present in ICC-MP but are prominent in the enteric nervous system.
Influx of calcium through voltage-activated ion channels probably does not play a significant role in the initiation of the transients. First, L-type calcium channels do not play a role because the experiments reported here were performed in the presence of nicardipine (1 μM), consistent with a study on calcium transients in ICC in cultured tissue from the murine stomach (36), and numerous studies have shown that L-type calcium channel blockade does not affect slow-wave activity (22, 31, 39, 50). Second, T-type calcium channels do not appear to play a major role either because 0.5 μM mibefradil had no effect on the frequency or amplitude of calcium transients in the ICC-MP. It has been previously reported that T-type calcium channels are essential for the influx of calcium necessary for proper pacemaker function in the ICC-MP (45). It was demonstrated that mibefradil abolished calcium transients in the ICC-MP of the mouse ileum at concentrations between 0.1 μM and 1.0 μM within 10 min (45). This is in contrast to our observations. Our data are consistent with studies from Boddy and colleagues (6), who showed only a small decrease in frequency of the rhythmic slow-wave-driven contractions in the mouse intestine by mibefradil (0.5 μM). Our data are also consistent with the observation that 10 μM mibefradil hardly affects slow-wave activity (3).
IP3-sensitive calcium release from the SR is thought to play a critical role in the regulation of slow-wave frequency in the intestine (37, 38, 58) and, among others, in the stomach (29), renal pelvis (35), and portal vein (16). In addition, IP3 knockout mice do not generate slow waves in the stomach (54). In the present study, 2-APB reduced the calcium oscillation frequency or abolished the activity of the ICC-MP network, consistent with a study on cultured tissues from the mouse stomach in which 2-APB and xestospongin completely and rapidly abolished calcium transients in ICC (36).
Calcium oscillations occur in synchrony across the network studied. 2-APB disrupted this synchrony. If IP3-induced calcium release is the key event underlying pacemaker generation, how is synchrony achieved? Possibly through synchronization of calcium and/or another gap junction-permeable substance such as IP3. Block of gap junction permeability causes loss of synchrony of calcium activity (45). 2-APB might disrupt calcium or IP3 signaling. It is also possible that 2-APB directly affects cell-to-cell coupling as was found in the urinary bladder (18). Another possibility is that 2-APB caused asynchrony because the drug did not act within exactly the same time span or in exactly the same manner in each cell.
The role of IP3 was also investigated by assessing the effect of phospholipase C activity. U73122, a blocker of phospholipase C, reduced the calcium oscillation frequency and amplitude, whereas the inactive analog U73343 had no effect. Consistent with our data, U73122 reduced both the amplitude and frequency of the electrical slow wave (38). This suggests that ongoing synthesis of IP3 is required, hence the presence of constitutively active phospholipase C, possibly associated with continuous dephosphorylation of IP3 via a 5-phosphatase and/or phosphorylation via a 3-kinase (46).
The present study reveals that the IP3R responsible for calcium release in ICC-MP is IP3R-I, also observed in cultured tissue from mouse stomach ICC (36), and not IP3R-II or IP3R-III, which are present in enteric nerves.
It is still a mystery in most cell types with calcium oscillatory activity what is causing the rhythmicity. Two possibilities are generally brought forward, rhythmically activated ion channels and/or ion transporters or the presence of an intracellular clock (5). This paper does not provide the final answer, but the data are consistent with the presence of the clock within the signal transduction machinery (47). Activation and delayed inhibition by intracellular calcium of the IP3R is a possibility, but our data are also consistent with oscillating levels of IP3 (19) as the main clock. Frequency regulation could be achieved by calcium feedback on IP3 synthesis and/or breakdown by calcium (47).
The inhibitor of calcium uptake from the SR, CPA, has been shown to reduce slow-wave frequency in the ICC-MP (38). This is consistent with our observation that CPA reduced the frequency and amplitude of the calcium oscillations and in several experiments abolished the activity after 30 min. Hence, block of calcium refilling of internal stores quickly disrupts calcium oscillations and pacemaker activity.
The inhibitor of the reverse-mode NCX KB-R7943 quickly reduced the calcium-transient frequency and amplitude, suggesting that it plays a role in the maintenance of calcium oscillations. This is consistent with our observation of a high density of NCX protein in the ICC; NCX immunoreactivity was stronger in ICC than in smooth muscle cells. This confirmed a previous study (10) although we found the density in ICC to be much stronger compared with smooth muscle cells and we did not observe NCX in ganglia. The colocalization of NCX and Caveolin 1 in ICC suggested that NCX might be localized in the caveolae (10).
The influx of calcium through the NCX may be a possible source of calcium for the refilling of the SR. A role for NCX, working in reverse mode, has been implicated in modulating pacemaker frequency in rabbit urethral ICC (17, 53). KB-R7943 completely and reversibly abolished calcium transients in urethral ICC and also substantially reduced basal calcium immediately after its addition to the cells (7). Other candidates for calcium influx are nonselective cation channels (26, 32, 36).
Mitochondria play a critical role in all metabolic processes of the cell including pacemaking. Pacemaking has been shown to be highly sensitive to block of metabolic function (1, 11). Pacemaking is also highly sensitive to temperature, seen as evidence for an association with metabolic processes (30, 44). Mitochondria play a role in cell calcium homeostasis by releasing and taking up calcium; because mitochondria show a variety of oscillatory behaviors, they may play a role in any type of calcium oscillation (4, 42). Carbonyl cyanide m-chlorophenyl-hydrazone is a protonophore that inhibits oxidative phosphorylation; it depolarizes smooth muscle of the guinea pig stomach and inhibits slow-wave activity (24), which was interpreted as evidence that mitochondrial calcium release was essential for slow-wave activity. Studies on the urethra of the rabbit concluded that spontaneous calcium waves in ICC were regulated by the calcium buffering capacity of the mitochondria (52). Our present results suggest that the mitochondria are not essentially involved in calcium handling related to the generation and maintenance of calcium oscillation in mouse small intestine ICC-MP. However, we report that, in some experiments, CGP-37157 affected the calcium transient, suggesting that mitochondria can affect intracellular calcium also under the conditions of the present study.
Limitations of this study are related to the nonspecific nature of some of the pharmacological agents used. 2-APB blocks the IP3 receptor, but it can also affect calcium release-activated channels (43, 48); clearly, the role of the calcium release-activated calcium channel protein 1 (Orai1/CRAC) channel in intestinal pacing needs further investigation. CGP-37157 inhibits SR/endoplasmic reticulum calcium ATPase and also activates ryanodine receptors (41). We did not find any effect of ryanodine on intestinal slow-wave activity (38). On the other hand, ryanodine was shown to affect calcium oscillations in cell cluster preparations from the mouse ileum (2) although not in ICC of the dog colon (49); this needs further study.
In summary, ongoing IP3 synthesis and IP3-induced calcium release from the SR, via the IP3R-I, are the major drivers of the calcium transients associated with ICC pacemaker activity. This suggests that a biochemical clock intrinsic to ICC determines the pacemaker frequency, which is likely directly linked to kinetics of the IP3-activated calcium channel and IP3 metabolism. This bears similarity to the recently studied calcium clock in cardiac pacemaker cells (34).
This study was supported by CIHR operating grant no. MOP12874.
No conflicts of interest, financial or otherwise, are declared by the authors.
We appreciated support from Dr. Sean Parsons.
- Copyright © 2011 the American Physiological Society