Spatial organization and coordination of slow waves in the mouse anorectum

Abstract

The internal anal sphincter (IAS) develops tone and is important for maintaining a high anal pressure while tone in the rectum is less. The mechanisms responsible for tone generation in the IAS are still uncertain. The present study addressed this question by comparing the electrical properties and morphology of the mouse IAS and distal rectum. The amplitude of tone and the frequency of phasic contractions was greater in the IAS than in rectum while membrane potential (E_m) was less negative in the IAS than in rectum. Slow waves (SWs) were of greatest amplitude and frequency at the distal end of the IAS, declining in the oral direction. Dual microelectrode recordings revealed that SWs were coordinated over a much greater distance in the circumferential direction than in the oral direction. The circular muscle layer of the IAS was divided into five to eight ‘minibundles’ separated by connective tissue septa whereas few septa were present in the rectum. The limited coordination of SWs in the oral direction suggests that the activity in adjacent minibundles is not coordinated. Intramuscular interstitial cells of Cajal and platelet-derived growth factor receptor alpha-positive cells were present in each minibundle suggesting a role for one or both of these cells in SW generation. In summary, three important properties distinguish the IAS from the distal rectum: (1) a more depolarized E_m; (2) larger and higher frequency SWs; and (3) the multiunit configuration of the muscle. All of these characteristics may contribute to greater tone generation in the IAS than in the distal rectum.

Key points

• The internal anal sphincter (IAS) develops tone important for maintaining high anal pressure and continence whereas tone in the rectum is less.

• To investigate tone generation, the electrical properties [membrane potential (E_m) and slow waves (SWs)] and morphology of the mouse IAS and distal rectum were compared.

• SWs were greatest in amplitude and frequency at the distal end of the IAS and declined toward the rectum. SWs were also coordinated to a greater degree in the circumferential than the oral direction.

• The circular muscle was divided into ‘minibundles’ in the IAS but not rectum. Intramuscular interstitial cells of Cajal and platelet-derived growth factor receptor alpha-positive cells were present in each minibundle making each a possible candidate for SW generation.

• The features that distinguish the IAS from rectum (i.e. depolarized E_m, larger and higher frequency SWs and multiunit configuration) are all properties that are predicted to result in greater tone generation.

Introduction

The internal anal sphincter (IAS) is a specialized region of muscle at the distal end (DE) of the gastrointestinal (GI) tract. It is responsible for generating approximately 70% of resting anal pressure (Bharucha, 2004; Rao & Meduri, 2011). To maintain a high anal pressure, the IAS remains contracted most of the time. In spite of the fundamental role played by the IAS, there is still uncertainty with regard to the basic mechanisms underlying tone generation in this muscle.

Although resting anal pressure is tonic, spontaneous contractile activity in isolated strips from the IAS consist of rapid frequency phasic contractions superimposed upon tone (Cobine et al. 2007; Duffy et al. 2012). In contrast, the rectum is predominantly a phasic muscle (Mutafova-Yambolieva et al. 2003; Patel & Rattan, 2006). Tone and phasic activity have been considered to be due to markedly different mechanisms in the IAS (Patel & Rattan, 2006), but both are nearly abolished by the L-type Ca²⁺ channel (Cav_L) antagonist nifedipine (Cobine et al. 2007) indicating a critical dependence upon the entry of Ca²⁺ through Cav_L channels.

The fundamental electrical event underlying phasic contractions in GI muscles is the slow wave (SW) (Szurszewski, 1987; Sanders et al. 2006; Huizinga et al. 2009). Each SW transiently increases the open probability of Cav_L channels, leading to a rise in intracellular Ca²⁺ and subsequent contraction. Although SWs in GI muscles are associated with phasic contractions, these events may also result in tone generation. For example, at higher SW frequencies, there may be insufficient time between SWs for the removal of all of the Ca²⁺ that entered during the SW depolarization ultimately resulting in [Ca²⁺]_i remaining above threshold for contraction. This is analogous to the partial tetanus that occurs in skeletal muscles when stimulus frequency is increased to the point where [Ca²⁺]_i does not fall below the mechanical threshold between stimuli (Watras, 2005). Furthermore, if the level of membrane potential (E_m) between SWs is sufficiently depolarized, some Ca²⁺ may enter continuously through Cav_L due to ‘window current’ (Imaizumi et al. 1989; Langton et al. 1989; Fleischmann et al. 1994). Finally, tone may arise from the summation of asynchronous phasic activity occurring in multiple independent motor units. This is analogous to the asynchronous firing of motor units in postural muscles that results in a relatively smooth continuous contraction of the skeletal muscle (Robinson, 2008). In calcium imaging studies of the urethra and gallbladder [two other visceral organs that contain spontaneously active ‘ICC-like’ pacemaker cells (Sergeant et al. 2000; Lavoie et al. 2007)], tone generation has also been proposed to arise from the summation of asynchronous activity in adjacent motor units (Balemba et al. 2006; Thornbury et al. 2011). In summary, several different electromechanical coupling mechanisms exist that may potentially contribute to tone generation in the IAS.

SWs in other GI muscles are known to arise from a specialized population of cells called ‘interstitial cells of Cajal' (ICC) and we have previously identified SWs and ICC in both the mouse and monkey IAS (McDonnell et al. 2008; Harvey et al. 2008; Cobine et al. 2010, 2011). However, the role of ICC in the generation of SWs in the IAS is still unknown. Recently another cellular candidate has been identified in the GI muscularis, i.e. platelet-derived growth factor receptor alpha-positive (PDGFRα⁺) cells. These cells have been shown to form gap junctions with GI smooth muscle (Horiguchi & Komuro, 2000) and to undergo fluctuations in E_m (Kurahashi et al. 2011). Thus, PDGFRα⁺ cells represent an additional cellular candidate for SW generation in the IAS.

The present study examined the electrical properties of the IAS and rectum to determine whether there are differences between these regions that may contribute to the greater tone generating capacity of the IAS. Toward this end, the spatial organization and coordination of electrical activity in the mouse anorectum was examined using single and dual microelectrode recording techniques. The specific questions addressed in this study included: (1) Are the electrical events that lead to Cav_L activation (i.e. SWs and overall depolarization of ‘resting’ membrane potential) greater in the region where tone generation is largest? (2) Is the muscle of the IAS organized into small bundles (i.e. ‘minibundles’) and if so, to what extent are SWs synchronized within and between minibundles? (3) How does the distribution of ICC and PDGFRα⁺ cell subtypes change from rectum to IAS and which cell subtypes are present in the region of greatest SW activity? Our results reveal that electrical events which activate Cav_L are greatest in the tone generating IAS. Minibundles were identified in the IAS and dual microelectrode recordings indicated that SWs within a minibundle were coordinated whereas SWs in adjacent minibundles were not. Finally, immunohistochemical studies revealed that while the distribution of PDGFRα⁺ cells was relatively constant throughout the anorectum the subtypes of ICC present changed. At the site of greatest SW activity (i.e. the distal IAS), only intramuscular ICC (ICC-IM) were present. Thus, ICC-IM and PDGFRα⁺ cells are both viable candidates for SW generation but functional studies are required to establish whether this is the case.

Methods

Ethical approval

The maintenance of all animals was in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. The experiments and procedures were under the approval of the Institutional Animal Use and Care Committee at the University of Nevada.

Tissue preparation

Adult (3–4-month-old) C57BL/6 wild-type (WT) mice and smMHC^Cre-eGFP mice of both sexes were obtained from Jackson Laboratories (Bar Harbor, ME, USA; n = 56 mice). The animals were killed with isoflurane (Baxter, Deerfield, IL, USA) followed by cervical dislocation.

The rectoanal region was removed and loosely pinned in a dissecting dish containing cold Krebs–Ringer buffer solution (KRBS) of the following composition (in mm): NaCl 118.5, KCl 4.7, CaCl₂ 2.5, MgCl₂ 1.2, NaHCO₃ 23.8, KH₂PO₄ 1.2 and dextrose 11.0. KRBS was maintained at pH 7.4 at 37°C by bubbling to equilibrium with 95%O₂–5%CO₂. The rectoanal region was cleared of all adhering skeletal muscle, fat and glands, opened longitudinally and the mucosa removed. The final length of muscle prepared was dependent upon the experiment performed as described below. Guanethidine (1 μm) and atropine (1 μm) were included in all bathing solutions to eliminate the influence of adrenergic and cholinergic neural inputs respectively.

Immunohistochemistry

Whole mount preparations

Whole mount preparations were created from the final 10 mm of the GI tract and were prepared and labelled with anti-Kit antibody as previously described (Cobine et al. 2011).

Cryostat sections

To examine the relationship of interstitial cells to smooth muscle cells (SMC) the distal GI tract of the smMHC^Cre-eGFP mouse was retained as a tube. The mucosa was removed after inverting the tube and then returning it to the original configuration. Muscular tubes were prepared for cryostat sectioning as previously described (Cobine et al. 2011). Cryostat sections of fixed tubes were cut perpendicular to the circular muscle (CM) layer in 10–20 μm thin sections using a Leica CM 3050 cryostat (Leica Microsystems, Wetzlar, Germany). Those sections containing both sides of the gut wall were retained and labelled with either anti-Kit or anti-PDGFRα antibodies as previously described (Cobine et al. 2011).

Imaging

Rectoanal preparations labelled using immunohistochemical techniques were examined with a Zeiss LSM 510 Meta confocal microscope (Carl Zeiss, Thornwood, NY, USA). The images generated are digital composites of Z-series of scans 0.25–1 μm optical sections. The construction of final images was completed using Zeiss LSM 5 Image Examiner Software, Adobe Photoshop CS5 Software and CorelDRAW X4 Software.

Masson's trichrome staining

For Masson's trichrome staining, a flat sheet of the final 5 mm of the GI tract was used. Skeletal muscle and mucosa were retained as well as the skin and hair at the DE of the anal canal. Tissues were fixed in formalin overnight and then embedded in paraffin wax before cutting into 3–4 μm thick sections. Paraffin wax was removed with xylene and sections were rehydrated with solutions with decreasing ethanol content followed by water. Bouin's fluid was added for 1 h at 56°C to fix the dye before washing with water. Slides were placed in Weigert's haematoxylin for 5 min then washed with water before placing in Biebrich scarlet-acid fuchsin for 15 min. Slides were then washed in distilled water and placed in phosphomolybdic/phosphotungstic acid for 15 min then aniline blue stain for 20 min and rinsing with distilled water. Preparations were then placed in 1% acetic acid for 4 min then dehydrated by rinsing with 95% ethanol once and 100% ethanol twice, followed by xylene before covering with a coverslip. All solutions were received from American MasterTech (Lodi, CA, USA). Images were scanned and digitized with Bioimagene, Inc. iScan^© developed by Dr S.H. Barsky, Department of Pathology, University of Nevada (Reno, NV, USA).

Contraction experiments

For contractile experiments the final 2 mm of the GI tract was cut into two pieces, i.e. a ‘0–1 mm segment’ (IAS) and an adjacent ‘1–2 mm segment’ (distal rectum) as shown in Fig. 1A. Contraction was recorded by attaching the muscle strips to a strain gauge (Gould, Cleveland, OH, USA) and a stable mount and subsequently immersing them in tissue baths containing 3 ml of oxygenated KRBS maintained at 37°C. Muscles were given an initial stretch of 0.25 g followed by a 60 min equilibration period.

Figure 1. Diagrams of microelectrode placement during single and dual recordings in the mouse anorectum.

Shown are diagrams of anorectal strips placed submucosal side up with the DE of the gastrointestinal tract located at the bottom of each diagram. The blue bars indicate placement of electrodes to stimulate nerves. A, configuration used to record activity in the IAS (0–1 mm) and the distal rectum (1–2 mm). Recordings were made by sequentially placing the microelectrode in the centre of each muscle strip (see Fig. 3). A 3 mm wide strip of anorectum was used for the remaining recordings. B, configuration used to map the changes in electrical activity with distance in the oral and circumferential directions. A reference point was selected that was 0.5 mm oral to the DE and halfway across the muscle strip. This point was designated ‘0.5, 0 mm’, i.e. 0.5 mm from the DE and 0 mm away from the centre of the muscle strip (red dot). Recordings made in the oral direction were all along the centre of the muscle strip (0 mm) and ranged between 0.25 and 2.5 mm from the DE. Recordings made in the circumferential direction were all 0.5 mm from the DE and ranged between 0 and 2.25 mm from the centre of the muscle strip (see Fig. 4). C and D, configuration used for dual microelectrode recordings. The first microelectrode was placed at the reference position (0.5, 0 mm) and a second microelectrode was placed at various distances in the oral (C) or the circumferential (D) direction (see Fig. 5). DE, distal end; IAS, internal anal sphincter.

Phasic contractile frequency was calculated by counting the number of peaks occurring during 1 min. Maximum relaxation was determined by the addition of the nitric oxide donor, sodium nitroprusside (SNP, 10 μm) and the L-type calcium channel blocker nifedipine (1 μm) while maximum contraction was determined by the addition of KCl (60 mm). Tone was defined as the amount of contraction between the troughs of the phasic contractions and maximum relaxation. Total spontaneous contraction was defined as the amount of contraction between average peak phasic contractions and maximum relaxation. To compare activity in the IAS and rectum spontaneous contractions were normalized to the maximum contraction occurring with 60 mm KCl (for more detail see Duffy et al. 2012).

Measurement of electrical activity

Muscle strips were pinned submucosal (SM) side up to the base of a recording chamber and superfused with KRBS at 36°C. Cells were impaled with glass microelectrodes filled with 3 m KCl (tip resistances 60–150 MΩ) (Fig. 1A). To maintain impalements, tissues were initially bathed for 20 min in 20 μm wortmannin (a MLCK inhibitor) followed by a 45 washout period in regular KRBS before beginning recordings. Wortmannin eliminates contraction of the tissue while retaining SW activity (Duffy et al. 2012).

To compare the electrical properties of the IAS (0–1 mm) and distal rectum (1–2 mm), strips were mounted next to one another in the recording chamber between two platinum electrodes to stimulate nerves. Cells in the centre of each muscle strip were impaled with a microelectrode (Fig. 1A). ‘Resting’ membrane potential (E_m) was determined as the average values between SWs. SW amplitude was determined by averaging SW peaks.

The methods used to determine the gradient of SW amplitude and frequency in the circumferential and oral directions are described in Fig. 1B. To determine if SWs were ‘coordinated’ the tissue configuration and electrode placement shown in Fig. 1C and D were used. SWs at two locations (i.e. reference location designated ‘1st’ and a second location designated ‘2nd’) were considered coordinated if they maintained a constant temporal relationship to one another over the duration of the dual microelectrode recording. This relationship was evaluated by determining the time at which peak depolarization occurred with successive SWs at both the 1st and 2nd location. The time at which peak SW depolarization occurred at the first location was then subtracted from that at the second location (i.e. 2nd₁ – 1st₁, 2nd₂ – 1st₂…2nd_n – 1st_n) to generate an ongoing tally of the time differences (dt) between corresponding SW peaks at the 1st and 2nd position. Values for dt were then plotted for 30 successive SW peaks and linear regression performed on these data (see Fig. 5B and C). If the slope of the line was not significantly different from zero (P > 0.05), the SWs were considered coordinated (see Fig. 5B) whereas if the slope was different from zero (P < 0.05) the SWs were considered non-coordinated (see Fig. 5C). The percentage of dual impalements with coordinated activity was tabulated for all recordings to establish the final relationship between the percentage coordinated SWs and distance in the oral and circumferential directions (see Fig. 5D). Coordinated SWs included those with varying degrees of phase shift between the SW peaks occurring at the 1st and 2nd positions. To quantitate the percentage phase shift of coordinated SWs (see Fig. 5E) mean dt for each dual recording was divided by the mean duration of the SW cycle in that recording (e.g. at 60 cpm, the mean SW duration is 1 s) and expressed as a percentage. Table 1 includes the frequency of SWs at both positions for all dual recordings as well as the mean ± s.d. of dt for coordinated SWs and the percentage phase shift of coordinated SWs.

Figure 5. Coordination of SW activity in the oral and circumferential directions (dual microelectrode recordings).

Sample traces of 25 SWs recorded simultaneously from two positions in the circumferential direction (A). Each successive SW is numbered above. Aa, example of coordinated SW activity for an electrode separation of 0.5 mm. SW frequency at both positions was 79.0 cpm. Ab, example of non-coordinated SW activity obtained with an electrode separation of 1 mm. The point at which SWs were synchronized has been assigned as ‘0’. Before and after ‘0’ SW peaks become increasingly separated from one another as the frequency of SWs at the first position (black) is significantly greater than at the second position (red) (i.e. 77.0 vs. 75.8 cpm respectively). B, plot of the difference in time (dt) between SW peaks at the first and second positions (y-axis) with successive SWs (x-axis) for the recording shown in (Aa). The slope of the line through these data obtained by linear regression was not different from 0 (P > 0.05) thus SWs were considered coordinated. C, plot of dt between SW peaks for the recording shown in (Ab), SWs in this recording were considered non-coordinated as the slope of the line was significantly greater than 0 (P < 0.05). D, summary plot of the percentage of dual impalements exhibiting coordinated SW activity with different electrode separations in the oral and circumferential directions. SWs were coordinated over a greater distance in the circumferential than in the oral direction. A complete listing of SW frequencies and the mean dt ± s.d. for coordinated SWs is included in Table 1. n values for dual impalements (recordings/animals) oral direction: 0.125 mm (6, 5), 0.25 mm (5, 5), 0.5 mm (7, 5); circumferential direction: 0.125 (5, 4), 0.5 mm (13, 5), 1 mm (13, 6) and 2 mm (18, 5). SW, slow wave.

Table 1.

SW characteristics from dual microelectrode recordings

1st Position (mm)	2nd Position (mm)	Electrode separation (mm)	Frequency 1st (cpm)	Frequency 2nd (cpm)	Slope = 0	dt Mean ± s.d.	Mean % phase shift
Circumferential
0.5, 0	0.5, 0.125	0.125	47.7	47.7	Y	−0.095 ± 0.025	7.5
0.5, 0	0.5, 0.125	0.125	51.8	51.8	Y	−0.053 ± 0.023	4.6
0.5, 0	0.5, 0.125	0.125	80.1	80.1	Y	−0.005 ± 0.046	0.7
0.5, 0	0.5, 0.125	0.125	67.5	67.5	Y	−0.007 ± 0.028	0.8
0.5, 0	0.5, 0.125	0.125	69.1	69.1	Y	0.043 ± 0.038	5.0
0.5, 0	0.5, 0.5	0.5	43.4	43.4	Y	−0.058 ± 0.042	4.2
0.5, 0	0.5, 0.5	0.5	55.2	55.2	Y	0.007 ± 0.05	0.6
0.5, 0	0.5, 0.5	0.5	65.8	65.8	Y	0.019 ± 0.036	2.1
0.5, 0	0.5, 0.5	0.5	67.0	67.0	Y	0.021 ± 0.04	2.3
0.5, 0	0.5, 0.5	0.5	77.7	77.7	Y	−0.047 ± 0.077	6.1
0.5, 0	0.5, 0.5	0.5	77.2	77.2	Y	0.001 ± 0.051	0.1
0.5, 0	0.5, 0.5	0.5	79.0	79.0	Y	0.009 ± 0.032	1.2
0.5, 0	0.5, 0.5	0.5	84.1	84.1	Y	0.025 ± 0.045	3.5
0.5, 0	0.5, 0.5	0.5	83.1	83.1	Y	−0.028 ± 0.032	3.9
0.5, 0	0.5, 0.5	0.5	57.0	54.0	N	—	—
0.5, 0	0.5, 0.5	0.5	49.4	58.9	N	—	—
0.5, 0	0.5, 0.5	0.5	72.0	42.0	N	—	—
0.5, 0	0.5, 0.5	0.5	72.0	56.0	N	—	—
0.5, 0	0.5, 1.0	1.0	47.5	47.5	Y	−0.001 ± 0.04	0.1
0.5, 0	0.5, 1.0	1.0	75.9	75.9	Y	0.471 ± 0.044	40.4
0.5, 0	0.5, 1.0	1.0	77.7	77.7	Y	−0.096 ± 0.035	12.4
0.5, 0	0.5, 1.0	1.0	65.6	65.6	Y	0.144 ± 0.052	15.7
0.5, 0	0.5, 1.0	1.0	66.2	66.2	Y	0.299 ± 0.099	33.0
0.5, 0	0.5, 1.0	1.0	77.0	76.5	N	—	—
0.5, 0	0.5, 1.0	1.0	77.0	75.8	N	—	—
0.5, 0	0.5, 1.0	1.0	75.5	69.5	N	—	—
0.5, 0	0.5, 1.0	1.0	81.0	72.0	N	—	—
0.5, 0	0.5, 1.0	1.0	81.0	72.0	N	—	—
0.5, 0	0.5, 1.0	1.0	69.3	65.0	N	—	—
0.5, 0	0.5, 1.0	1.0	78.5	74.6	N	—	—
0.5, 0	0.5, 1.0	1.0	69.5	68.0	N	—	—
0.5, 0	0.5, 2.0	2.0	79.0	73.5	N	—	—
0.5, 0	0.5, 2.0	2.0	78.9	73.7	N	—	—
0.5, 0	0.5, 2.0	2.0	77.7	73.1	N	—	—
0.5, 0	0.5, 2.0	2.0	78.0	76.9	N	—	—
0.5, 0	0.5, 2.0	2.0	83.3	82.9	N	—	—
0.5, 0	0.5, 2.0	2.0	79.5	79.8	N	—	—
0.5, 0	0.5, 2.0	2.0	78.4	77.1	N	—	—
0.5, 0	0.5, 2.0	2.0	79.1	78.0	N	—	—
0.5, 0	0.5, 2.0	2.0	78.2	77.1	N	—	—
0.5, 0	0.5, 2.0	2.0	77.1	76.7	N	—	—
0.5, 0	0.5, 2.0	2.0	78.2	76.6	N	—	—
0.5, 0	0.5, 2.0	2.0	63.7	60.2	N	—	—
0.5, 0	0.5, 2.0	2.0	67.0	66.8	N	—	—
0.5, 0	0.5, 2.0	2.0	50.2	48.9	N	—	—
0.5, 0	0.5, 2.0	2.0	77.5	73.7	N	—	—
0.5, 0	0.5, 2.0	2.0	72.6	71.5	N	—	—
0.5, 0	0.5, 2.0	2.0	72.0	70.5	N	—	—
0.5, 0	0.5, 2.0	2.0	69.2	78.9	N	—	—
Oral
0.5, 0	0.625, 0	0.125	57.5	57.5	Y	0.031 ± 0.04	3.0
0.5, 0	0.625, 0	0.125	68.7	68.7	Y	0.037 ± 0.053	4.2
0.5, 0	0.625, 0	0.125	67.1	67.1	Y	0.098 ± 0.041	11.0
0.5, 0	0.625, 0	0.125	53.6	53.6	Y	0.029 ± 0.023	2.6
0.5, 0	0.625, 0	0.125	62.3	62.3	Y	−0.015 ± 0.047	1.6
0.5, 0	0.625, 0	0.125	57.0	66.0	N	—	—
0.5, 0	0.75, 0	0.25	63.6	63.6	Y	−0.116 ± 0.062	12.3
0.5, 0	0.75, 0	0.25	64.9	64.9	Y	−0.256 ± 0.075	27.7
0.5, 0	0.75, 0	0.25	60.5	59.4	N	—	—
0.5, 0	0.75, 0	0.25	78.0	75.0	N	—	—
0.5, 0	0.75, 0	0.25	59.1	60.0	N	—	—
0.5, 0	1.0, 0	0.5	57.5	55.1	N	—	—
0.5, 0	1.0, 0	0.5	66.9	57.4	N	—	—
0.5, 0	1.0, 0	0.5	72.0	66.0	N	—	—
0.5, 0	1.0, 0	0.5	64.1	63.2	N	—	—
0.5, 0	1.0, 0	0.5	75.5	72.8	N	—	—
0.5, 0	1.0, 0	0.5	62.2	63.1	N	—	—
0.5, 0	1.0, 0	0.5	71.6	66.0	N	—	—

Abbreviations: 1st position, reference electrode position; 2nd position, second electrode position; cpm, cycles per minute; dt, time difference between successive peaks at the 1st and 2nd positions; N, no; SWs, slow waves; Y, yes. The first microelectrode (1st) was positioned in the centre of the muscle strip 0.5 mm from the distal end of the internal anal sphincter (column 1) whereas the second electrode (2nd; column 2) was positioned at various distances away from 1st in the circumferential or oral direction (see Methods and Fig. 1). Electrode separation is listed in column 3. The dual recordings, which were determined to have ‘coordinated’ SWs, are listed in column 6 (Y), i.e. the relationship between SW peaks over time was constant resulting in a line with a slope of 0 (see Methods and Fig. 5). Note that the SW frequencies (columns 4 and 5) in coordinated dual recordings (Y) are the same whereas those in non-coordinated recordings differ (N). The mean difference in time (dt) between the paired SW peaks in coordinated recordings (±s.d.) is listed in column 7 and the mean percentage phase shift of these paired SWs is listed in column 8.

Data acquisition and statistics

Data were collected, stored and analysed by computer using a data acquisition program (AcqKnowledge 3.9.1; Biopac systems, Inc., Goleta, CA, USA). Individual data points are expressed as means ± s.e.m. and n values represent the number of mice. Data sets were compared with a Student's t test and considered significantly different when P < 0.05. Values for SW amplitude and frequency in the oral and circumferential directions with distance were fit by linear regression using GraphPad Prism 3.02 software (San Diego, CA, USA), which also determined whether the slope was significantly different from zero.

Bundle area and width

Muscle bundle organization in the IAS and rectum was analysed by first identifying and outlining bundles in CorelDRAW X4. The identified bundles were analysed using Volumetry G8a (Dr G.W. Hennig, Department of Physiology and Cell Biology, University of Nevada, Reno, NV, USA). Bundles were density sliced and particle flood-filled to quantify the number of pixels per bundle. A calibration factor was applied to convert pixels into mm². To determine the average width per bundle the length of muscle containing bundles was divided by the number of bundles.

Drugs

Atropine sulphate, guanethidine, SNP and nifedipine were purchased from Sigma-Aldrich (St Louis, MO, USA). Wortmannin was purchased from Santa Cruz Biotechnology Inc. (Dallas, TX, USA). Atropine, guanethidine and SNP were dissolved in deionized water. Nifedipine was dissolved in ethanol. Wortmannin was dissolved in DMSO.

Results

Comparison of spontaneous contractile activity in the mouse internal anal sphincter and rectum

Strips of the IAS (0–1 mm from the DE of the GI tract) and distal rectum (1–2 mm from the DE) were mounted in tissue baths to record contractile activity. Both muscles developed spontaneous phasic contractions and tone (Fig. 2A and B) although tone and phasic contractile frequency were significantly greater in the IAS than in the distal rectum (Fig. 2C and D). Phasic contractions in the distal rectum also differed in that they occurred in bursts with periods of quiescence between (Fig. 2Ba). In contrast, phasic contractions in the IAS were continuous (Fig. 2Aa), although they waxed and waned as previously described (Duffy et al. 2012). In both muscles, tone and phasic contractions were abolished by the nitric oxide donor SNP (10 μm; Fig. 2A and B). Total spontaneous contractile activity (normalized to peak KCl-induced contraction; Fig. 2Ab and Bb) was significantly greater in the IAS than in the distal rectum (Fig. 2D).

Figure 2. Comparison of spontaneous contractile activity in the mouse IAS and rectum.

A, the IAS (0–1 mm) displayed both phasic contractions and tone (Aa, Ab), which were abolished by addition of the nitric oxide donor, SNP. Maximum contraction was elicited with KCl (Ab). B, the distal rectum (1–2 mm) developed phasic contractions but much less tone. As in the IAS, SNP abolished tone and phasic contractions (Ba, Bb) and KCl was used to elicit maximum contraction (Bb). C, frequency of phasic contractions was significantly greater (*) in the IAS than in the distal rectum, P < 0.05. D, total spontaneous contractile activity and tone (normalized to the maximum KCl contraction) were significantly greater (*) in the IAS than in the distal rectum, P < 0.05. n = 5 animals. IAS, internal anal sphincter; SNP, sodium nitroprusside.

Comparison of spontaneous electrical activity in the mouse internal anal sphincter and rectum

The electrical activity of the IAS (0–1 mm) and distal rectum (1–2 mm) were compared by impaling cells in the centre of each muscle strip. SWs (20–100 mV s⁻¹ rise time), were present in both muscles (Fig. 3A) but the amplitude and frequency of these events was significantly greater in the IAS than in the distal rectum (Fig. 3C and D). ‘Resting’ E_m (determined as the average value of E_m between SW) was also significantly less negative in the IAS than in the distal rectum (i.e. −44.3 ± 0.8 mV, n = 27 vs. −48 ± 0.6 mV, n = 24; respectively, Fig. 3B). Spikes (250–900 mV s⁻¹) were also observed in 25 ± 8% of cells in the IAS and 46 ± 11% of cells in the distal rectum. These events often followed nerve-evoked inhibitory junction potentials but some were also superimposed upon SWs.

Figure 3. Comparison of electrical activity in the mouse IAS and rectum.

A, sample traces of electrical activity in the IAS (0–1 mm) and distal rectum (1–2 mm) recorded with single microelectrodes (see Methods, Fig. 1A). SWs were smaller in amplitude and of lower frequency in the rectum (upper trace) than in the IAS (lower trace). A single shock of electrical field stimulation was applied to stimulate nerves (NS) giving rise to an inhibitory junction potential. B, membrane potential (between SWs) in the IAS (0–1 mm) was significantly (*) less negative than the distal rectum (1–2 mm). C, frequency of SWs was significantly greater (*) in the IAS (0–1 mm) than in the distal rectum (1–2 mm). D, amplitude of SWs in the IAS (0–1 mm) was significantly greater (*) than in the distal rectum (1–2 mm); P < 0.05. n = 27 animals. IAS, internal anal sphincter; NS, nerve stimulation; SW, slow wave.

To map the characteristics of electrical activity throughout the IAS and distal rectum additional recordings were made from larger muscle strips that encompassed the final 3 mm of the GI tract. The changes in electrical activity in the oral direction were determined by recording from cells at nine different positions located 0.25–2.5 mm from the DE (see Figs 1B and 4B). SWs were of greatest amplitude at the most distal position (i.e. 0.25 mm) averaging 14.1 ± 2.5 mV (n = 7) and declined in a linear manner in the oral direction to 3.9 ± 0.8 mV (n = 9) at 2.5 mm from the DE (Fig. 4Ba). The frequency of SWs also declined from 69.7 ± 3.2 cpm (n = 6) to 50 ± 2.6 cpm (n = 3) in the oral direction (Fig. 4Bb). To determine whether there were also changes in SW properties in the circumferential direction seven additional recordings were made from 0 to 2.25 mm distance from the centre of the muscle strip (see Figs 1B and 4C). In contrast to the oral direction, there were no significant differences in either the frequency or amplitude of SWs in the circumferential direction (Fig. 4C; P > 0.05).

Figure 4. Relationship between SW amplitude and frequency and distance in the oral and circumferential directions (single microelectrode recordings).

A, sample traces of electrical activity recorded from various positions in a strip of anorectum including: reference position (0.5, 0 mm, Aa), a point 2 mm away from centre in the circumferential direction (0.5, 2 mm, Ab) and a point 2 mm oral to the distal end of the gastrointestinal tract (2, 0 mm, Ac). B and C, plots of the relationship between SW amplitude (Ba, Ca) and frequency (cpm; Bb, Cb) with distance obtained by recording activity at various distances in either the oral (B) or circumferential (C) directions. Zero distance in (B) represents the distal end, whereas zero distance in (C) represents the centre of the muscle strip. Data fit by linear regression. The slope of the line was significantly different (P < 0.05) from zero in the oral (B) but not in the circumferential (C) direction. n = 24 animals. See Fig. 1B for further explanation of recording positions. SW, slow wave.

Degree of coordination of slow waves in the mouse anorectum

The coordination of SW activity in the oral and circumferential directions was evaluated with dual microelectrode recordings as shown in Fig. 1C and D. The first microelectrode was placed 0.5 mm from the DE in the centre of the muscle strip (0.5, 0 mm) and the second microelectrode was placed at distances between 0.125 and 0.5 mm from the first recording site in the oral direction or between 0.125 and 2 mm in the circumferential direction. SW activity was then simultaneously recorded from both positions. SWs were considered coordinated if they maintained a constant temporal relationship to one another over the duration of the dual microelectrode recording (see Methods). An example of coordinated SWs is shown in Fig. 5Aa. When the time difference (dt) between SW peaks at the two positions was plotted for successive SWs linear regression generated a line with a slope not significantly different from zero (P > 0.05) confirming the coordinated nature of these SWs (Fig. 5B). In contrast, other dual recordings identified SWs that were not coordinated (e.g. Fig. 5Ab). In this case, the difference in SW frequency between the two positions resulted in a continuous change in the relationship between SWs at the two sites. Plotting the dt between SW peaks at the 1st and 2nd positions resulted in a line with a slope significantly different from zero (P < 0.05), confirming the lack of coordination between the sites (Fig. 5C).

Using the above criteria we found that a greater percentage of dual recordings were coordinated in the circumferential direction than in the oral direction (Fig. 5D). With an electrode, separation of 0.125 mm most cells in the oral and circumferential directions were coordinated (i.e. 83% and 100% respectively). However, when electrode separation was increased to 0.5 mm no recordings in the oral direction were coordinated as compared to 69% in the circumferential direction. With greater electrode separation in the circumferential direction, the percentage of coordinated recordings declined further. Thus, with 1 mm separation 38.5% of recordings were coordinated whereas at 2 mm none of the 18 dual recordings had coordinated activity (Fig. 5D) although the differences in SW frequency between the two positions was usually quite small (e.g. 79.5 vs. 79.8 cpm; see Table 1).

SWs that maintained a constant temporal relationship to one another (i.e. those that were coordinated) did not always have peaks that occurred simultaneously. Rather, in some cases there was a substantial offset between SW peaks (i.e. a phase shift). To describe the temporal relationship of SW peaks to one another further we plotted the mean phase shift between coordinated SWs peaks as a function of electrode separation. As seen in Fig. 5E the phase shift between coordinated SW peaks at 0.125 mm in the oral direction and 0.125–0.5 mm in the circumferential direction was less than 5%. However, with greater electrode separations in either direction, a greater phase shift was seen. Thus, true ‘synchronous’ electrical activity in the terminal rectoanal region is rather limited.

Anatomical characterization of the mouse anorectum

The morphology of the mouse rectoanal region was examined in sections stained with Masson's trichrome (Fig. 6A). The luminal surface of the rectum consisted of a well-developed mucosa with epithelia and prominent glandular cells, lamina propria and an underlying muscularis mucosa. At a distance of 0.5–1 mm from the DE, the muscularis mucosa ended and the mucosa transitioned into simple anoderm. The rectum consisted of both CM and longitudinal muscle (LM) layers. There was a slight thickening of the CM in the IAS whereas the LM became more diffuse and usually terminated before the end of the CM layer. Distal to the IAS was the external anal sphincter, which overlapped the IAS by ∼650 μm. The CM layer of the IAS was divided into minibundles separated by connective tissue septa. In contrast, few septa were observed in the distal rectum. The area per bundle was not significantly different between preparations (n = 6; Fig. 6B and C) with an overall average of 0.014 ± 0.001 mm²/bundle. The mean number of bundles per preparation was 6.7 ± 0.5 with an average bundle width of ∼134 μm (see Methods).

Figure 6. Anatomy of the mouse anorectum.

A, sagittal section of the final 2 mm of the mouse gastrointestinal tract, including the IAS and distal rectum. The preparation was stained with Masson's trichrome to identify smooth muscle (pink) and connective tissue (blue). Additional structures staining pink include EAS, mucosa, anoderm, MM, MG, glands, hair follicles and blood vessels. The IAS CM layer is divided into distinct bundles separated by connective tissue septa (outlined in white for further clarity). Bundles in this preparation were present over a distance of ∼1.2 mm (range for six muscles: 0.6–1.2 mm) followed by a continuous CM layer in the rectum. The LM layer in the rectum is apparent as a compact structure that becomes diffuse in the anal direction, ending (in this preparation) ∼0.25 mm before the DE of the gastrointestinal tract. B, tracings of muscle bundles from the distal most 1 mm section from two additional mouse rectoanal regions. C, plot of the average muscle bundle area (mm²) in six rectoanal preparations (see Methods). There was no significant difference in bundle area between preparations (P > 0.05). CM, circular muscle; DE, distal end; EAS, external anal sphincter; IAS, internal anal sphincter; LM, longitudinal muscle; MG, myenteric ganglia; MM, muscularis mucosa; My, myenteric; SM, submucosa.

Distribution of interstitial cells of Cajal in the anorectum

Kit⁺ ICC were identified by immunohistochemical techniques in whole mount preparations of WT mouse anorectum and in cryostat sections of smMHC^Cre-eGFP mouse anorectum. Examination of the SM surface of whole mount preparations revealed clusters of stellate-shaped ICC (ICC-SM) in the rectum. The density of ICC-SM declined distally with ICC-SM ending before the DE (Fig. 7A). Spindle-shaped ICC-IM were present throughout the CM layer of the IAS and rectum at a uniform density (Fig. 7B). In contrast, ICC-IM density was lower in the LM than in the CM layer and these cells ended before the DE (Fig. 7C). A second class of stellate-shaped ICC was present at the myenteric surface of the rectum (ICC-My). ICC-My formed a loose network in the rectum but were absent from the IAS. The differences in morphology of ICC in the rectum and distal IAS are apparent in Fig. 7D–M, which shows Kit⁺ labelling at various levels within the muscularis in two confocal stacks. Whereas three cell populations are apparent in the rectum, i.e. ICC-SM (Fig. 7J), ICC-IM (Fig. 7K) and ICC-My (Fig. 7L), only one cell population, i.e. ICC-IM persists until the end of the IAS (Fig. 7E–H).

Figure 7. Distribution of ICC in the mouse anorectum.

Images from whole mount preparations of wild-type mouse anorectum are shown. Kit⁺ ICC populations were identified using immunohistochemical techniques. A–C, distribution and morphology of ICC at various depths within the muscularis visualized with confocal microscopy in a 2.3 mm long segment of the anorectum (each figure is composed of seven overlapping 20× images). The DE of the gastrointestinal tract is located on the right. A, morphology of Kit⁺ cells at the submucosal surface of the preparation. A diffuse population of stellate-shaped ICC (examples highlighted with arrows) can be seen at the rectal SM surface. The density of these cells declines in the anal direction ending ∼0.5 mm before the DE. B, morphology of Kit⁺ cells imaged at a depth ∼30% through the muscularis (i.e. within the CM layer). Spindle-shaped ICC-IM can be seen throughout the 2.3 mm section at this level with a relatively uniform density. C, morphology of Kit⁺ cells imaged at a depth ∼90% through the muscularis (i.e. at the level of the LM layer). A sparse population of spindle-shaped ICC-IM (examples highlighted with arrows) can be seen in the rectal LM layer running perpendicular to those in the CM layer. The density of these cells declines in the anal direction ending ∼0.2 mm before the DE. D, confocal stack of ICC in the final 230 μm of the gastrointestinal tract. E–H, subsections of (D) showing the submucosal surface (E), and regions ∼30% (F), ∼60% (G) and ∼90% (H) through the muscularis. All sections contain CM ICC-IM while a single LM ICC-IM can also be seen in (H). I, confocal stack of ICC in the distal rectum (∼2.5 mm from the DE). J–M, subsections of (I) containing the submucosal surface (J), and regions ∼30% (K), ∼60% (L) and ∼90% (M) through the muscularis. Four different populations of ICC are apparent in these images, including ICC-SM (J), CM ICC-IM (K), ICC- myenteric (L) and LM ICC-IM (M). CM, circular muscle; DE, distal end; ICC, interstitial cells of Cajal; IM, intramuscular; LM, longitudinal muscle; SM, submucosa.

To examine further the relationship of ICC to SMC sagittal cryostat sections of the smMHC^Cre-eGFP anorectum were created and labelled with an anti-Kit antibody. SMC in these sections were clearly visible because they express GFP (green) (Fig. 8A). In the IAS, SMC in the CM layer were divided into sections (i.e. minibundles) whereas the rectal CM layer was a more continuous structure. ICC-IM were distributed throughout each minibundle of the IAS but were absent from septal structures. These cryostat sections further support the conclusions reached from whole mount preparations, i.e. that ICC-SM and ICC-My are present in the rectum but end before the distal IAS while ICC-IM are distributed throughout the anorectum. Sections from the same smMHC^Cre-eGFP mouse anorectum were labelled with an anti-PDGFRα antibody to examine the distribution of PDGFRα⁺ cells (Fig. 8B). These sections revealed that PDGFRα⁺ cells were densely distributed throughout the anorectum, including the SM, My, IM and Ser regions as well as within septal structures.

Figure 8. Distribution of ICC and PDGFRα⁺ cells in the smMHC^Cre-eGFP mouse anorectum.

Images from sagittal cryostat sections (2.5 mm long) of the smMHC^Cre-eGFP mouse anorectum are shown. Kit⁺ ICC and PDGFRα⁺ cells (both red) were identified using immunohistochemical techniques while SMC were identified through expression of green fluorescent protein (green). A, distribution of ICC (red) and SMC (green) in the anorectum. ICC-SM and ICC-My are apparent along the SM and My edges of the CM layer in rectum but disappear before reaching the distal IAS. In contrast, ICC-IM is present throughout the CM layer of the rectum and IAS. Spindle-shaped ICC-IM can also be seen in the LM layer ending before the DE of the IAS. Division of the CM layer into bundles (green) can clearly be seen within the IAS. B, distribution of PDGFRα⁺ cells (red) and SMC (green) in an adjacent section from the same anorectum as (A). PDGFRα⁺ cells are present throughout the anorectum, including the SM edge (PDGFRα⁺-SM), within the CM and LM layers (PDGFRα⁺-IM), at the My edge (PDGFRα⁺-My) and within the Ser (PDGFRα⁺-Ser). There was no obvious change in the density of these cells from the rectum to the IAS. Both images are composites of three images taken at 10×. CM, circular muscle; DE, distal end; ICC, interstitial cells of Cajal; IAS, internal anal sphincter; IM, intramuscular; LM, longitudinal muscle; My, myenteric; PDGFRα, platelet-derived growth factor receptor alpha; Ser, serosa; SM, submucosa; SMC, smooth muscle cells.

Discussion

The IAS and rectum are adjacent segments of the GI tract with differing functional roles. The ability of the IAS to generate tone is well documented and is important for the maintenance of high resting anal pressure. In contrast, pressure in the rectum is lower and the contractile activity of this muscle is predominantly phasic. In this study, we have compared various morphological and functional properties of the IAS to those of the distal rectum. Our results indicate that the IAS differs in a number of important ways from the rectum. These differences probably contribute to the greater tone generating capacity of the IAS.

The morphology of the mouse internal anal sphincter and distal rectum differ

Cross-sections of the mouse rectoanal region visualized with Masson's trichrome staining techniques revealed that the CM layer of the IAS (0–1 mm from the DE) was divided into five to eight ‘minibundles’ separated by connective tissue septa. Minibundles averaged ∼134 μm in width and formed muscular segments oriented in the circumferential direction. In contrast, few septal structures were observed in the distal rectum (1–2 mm from the DE). The organization of muscle into bundles in the mouse IAS is similar to that previously described for the monkey IAS except that, because of its larger size (i.e. ∼200× bigger) the monkey IAS was divided into hundreds of minibundles rather than just a few. Because of this organization we suggested that minibundles may function independently of one another (Cobine et al. 2010). The present study further explored this hypothesis by simultaneously recording electrical events from different sites with dual microelectrodes as discussed below.

The morphology and distribution of ICC also differed between the mouse IAS and rectum. ICC-My, ICC-SM and ICC-IM were all present in the rectum. However, ICC-My declined in the distal direction terminating before the start of the IAS whereas ICC-SM terminated before the end of the IAS. In contrast, spindle-shaped ICC-IM were present throughout the muscularis of the anorectum. This arrangement is similar to ICC in the monkey rectoanal region except that in monkeys, ICC-IM in the CM layer of the IAS are highly branched cells. The reason for this difference is unclear although, as suggested above, differences in scale may play a role. For example, the highly branched ICC-IM of the monkey IAS may provide better electrical coupling to SMC. If ICC-IM are indeed pacemaker cells, then this arrangement could provide better coordination of the activity of a minibundle located within the thick (∼2 mm) CM layer of the monkey IAS versus the thin (∼0.1 mm) CM layer of the mouse.

Other possible cellular candidates for slow wave generation in the internal anal sphincter

As the only type of ICC present in the distal IAS are ICC-IM we have speculated that these are the cells responsible for SW generation. However, in a previous study of the W/W^v mouse IAS, we found that SWs persisted (Duffy et al. 2012), while Kit-labelling of ICC-IM could not be detected (Cobine et al. 2011). It is possible that ICC-IM remain in the W/W^v mouse IAS but that Kit expression has fallen below detectable levels. Alternatively, if ICC-IM are indeed absent then SWs must be generated by some other cell type. PDGFRα⁺ cells are also present in the mouse IAS (Cobine et al. 2011) and previous studies have shown that these cells can mediate the electrical events associated with purinergic neuromuscular transmission (Kurahashi et al. 2011). In the present study, we found a dense population of PDGFRα⁺ cells throughout the IAS (Fig. 8B) making them an additional candidate for SW generation. Finally, SWs may be generated by SMC. Regardless of whether SWs are generated by ICC-IM, PDGFRα⁺ cells or SMC, the ionic conductances expressed by these cells in the distal IAS must be unique as all three cell types are present throughout the anorectum. Our studies of the monkey rectoanal region revealed marked changes in the morphology of ICC-IM in rectum versus IAS (i.e. from spindle-shaped to stellate-shaped cells). Thus, a possible additional shift in the ionic conductances expressed by these cells is not unprecedented. Indeed, there is evidence that ICC-IM in the guinea-pig gastric corpus and pylorus can serve as pacemaker cells (Van Helden & Imtiaz, 2003; Hirst & Edwards, 2006).

The contractile and electrical activity of the mouse internal anal sphincter and distal rectum differ

Contractile measurements revealed significantly greater tone in the 0–1 mm muscle strip than in the adjacent 1–2 mm muscle strip. A well-developed mucosa was also present in the 1–2 mm muscle strip whereas the 0–1 mm muscle strip contained anoderm. For these reasons, we have referred to the 0–1 mm muscle strip as the IAS. On the other hand, the rectum is a much longer structure [i.e. ∼8× longer than the IAS in humans (Netter, 2006)], thus the 1–2 mm muscle strip contains only the distal-most section of the rectum. Further differences in functional properties are therefore likely to be present in portions of the rectum that are more proximal. In spite of their close proximity, the IAS differed from the distal rectum in a number of ways, including greater tone, higher frequency phasic contractions, a more depolarized ‘resting’ E_m and larger, higher frequency SWs. The electrical properties that distinguish the IAS from the distal rectum are properties that will increase Cav_L open probability, calcium entry and tone. This link between Cav_L and tone is further supported by the fact that tone is nearly abolished by Cav_L inhibitors (Cook et al. 1999; Cobine et al. 2007).

To examine the origin and spread of SWs further, electrical activity was recorded at various positions in the oral and circumferential direction in intact strips of anorectum. These measurements revealed that SW amplitude and frequency were greatest at the DE of the IAS and gradually declined in the oral but not the circumferential direction. The decline in SW frequency indicates that SWs are not simply generated at the DE of the IAS and conducted in the oral direction. For example, in the canine colon, SWs do arise from one surface (i.e. the SM edge of the CM layer) and conduct into the interior where their amplitude and rate of rise are reduced but not their frequency (Smith et al. 1987). A more likely explanation for our results is that the cells generating pacemaker currents are distributed throughout the IAS and distal rectum but the ionic conductances expressed by these cells changes in the oral direction. In a similar manner, although ICC are present throughout the mouse and guinea-pig gastric corpus and antrum, SW frequency declines in the distal direction (Ordog et al. 2002; Domae et al. 2008).

Relative degree of slow wave coordination in the anorectum

To determine the extent to which the electrical activity in minibundles was coordinated dual microelectrode recordings were undertaken. These measurements revealed a high degree of coordination between cells separated by 0.125 mm in either the circumferential or the oral direction (i.e. 100% and 83% of impalements respectively). However, coordination declined markedly in the oral direction with greater distances between electrodes (i.e. 40% of impalements at 0.25 mm and 0% at 0.5 mm). The average width of a minibundle was estimated to be ∼134 μm. Thus, some limited coordination appears to be present in adjacent minibundles but beyond this, coordination was absent. These data provide evidence that functionally, minibundles are largely independent of one another.

In contrast to the oral direction, coordination of SWs in the circumferential direction was greater, i.e. with 0.5 mm electrode separation no SWs were coordinated in the oral direction whereas in the circumferential direction 69% were coordinated. Furthermore, there was very little phase shift between coordinated SWs at 0.5 mm distance in the circumferential direction (i.e. 2.3 ± 0.6%). However, at 1 mm electrode separation the percentage of coordinated impalements declined to 38.5% (Fig. 5D) and the phase shift increased to 20.3 ± 7.3% (Fig. 5E). As the circumference of the IAS is ∼10 mm, even in the circumferential direction, the degree to which cells are coordinated and in phase with one another is relatively limited (i.e. <10% of the circumference).

The overall consistency of SW amplitude and frequency in the circumferential direction (Fig. 4) suggests that SWs are generated from multiple sites within minibundles and are coordinated because of electrical coupling between cells. Even the lack of coordination with greater distances between electrodes is compatible with this concept. From location to location, different cells will be responsible for SW generation introducing more potential for phase shifts between regions and finally non-coordinated activity. The distribution of ICC-IM (and PDGFRα⁺ cells) throughout each minibundle is in keeping with this ‘multiple site’ proposal.

The degree of coordination observed in the oral and circumferential directions suggests that the ‘unit’ of muscular activity in the IAS is approximately the width of a minibundle (134 μm) and a length <1 mm. Thus, ample opportunity exists in both the oral and the circumferential directions for summation of asynchronous activity to contribute to tone generation in the IAS. Tone generation in the urethra (Thornbury et al. 2011) and gallbladder (Balemba et al. 2006) have also been attributed to summation of asynchronous phasic activity of motor units.

Differences in myofilament sensitivity versus electrical activity

Studies by others have provided evidence that the myofilament sensitivity to calcium in the IAS is greater than that of the rectum (Patel & Rattan, 2006). The present study does not discount a possible difference in the sensitivity of the myofilaments to calcium in the mouse IAS versus rectum. However, as noted above, both tone and phasic contractions in the IAS are nearly abolished by Cav_L inhibitors making the status of Cav_L a key feature determining the degree of tone in this muscle. Furthermore, motor neurons give rise to substantial changes in E_m that will further modulate Cav_L activity (Duffy et al. 2012). Thus, differences in the electrical properties of the IAS versus the rectum and their modulation by enteric motor neurons are key features regulating motor activity in these muscles.

In summary, this study has identified three important features of the IAS that distinguish it from the distal rectum, namely: (1) larger and higher frequency SWs; (2) a more depolarized E_m; and (3) the multiunit configuration of the muscle. It is possible that all of these properties contribute to the greater tone generating capacity of the IAS than rectum.

Glossary

BSA

bovine serum albumin

Cav_L

L-type calcium channel

CM

circular muscle

CPM

cycles per minute

DE

distal end

E_m

membrane potential

GI

gastrointestinal

IAS

internal anal sphincter

ICC

interstitial cells of Cajal

KRBS

Krebs–Ringer bicarbonate solution

LM

longitudinal muscle

MG

myenteric ganglia

MLCK

myosin light chain kinase

My

myenteric

PBS

phosphate buffer solution

PDGFRα

platelet-derived growth factor receptor alpha

Ser

serosa

SMC

smooth muscle cells

SM

submucosa

SNP

sodium nitroprusside

SW

slow wave

WT

wild-type

Additional information

Competing interests

The authors of this study have no competing interests.

Author contributions

All experiments were carried out in the laboratory of K.D.K. K.A.H., C.A.C. and KDK were involved in the collection, analysis and interpretation of data. S.M.W. critically revised the manuscript. C.A.C. and KDK were responsible for the concept and design of experiments and drafting of the article. All authors have read and approved the final submission.

Funding

Grant funding DK078736 to K.D.K. and S.M.W. Support for the Zeiss LSM510 confocal microscope was provided by 1 S10 RR16871.