Excitatory Neuronal Responses of Ca2+ Transients in Interstitial Cells of Cajal in the Small Intestine

Abstract Interstitial cells of Cajal (ICC) regulate smooth muscle excitability and motility in the gastrointestinal (GI) tract. ICC in the deep muscular plexus (ICC-DMP) of the small intestine are aligned closely with varicosities of enteric motor neurons and thought to transduce neural responses. ICC-DMP generate Ca2+ transients that activate Ca2+ activated Cl- channels and generate electrophysiological responses. We tested the hypothesis that excitatory neurotransmitters regulate Ca2+ transients in ICC-DMP as a means of regulating intestinal muscles. High-resolution confocal microscopy was used to image Ca2+ transients in ICC-DMP within murine small intestinal muscles with cell-specific expression of GCaMP3. Intrinsic nerves were stimulated by electrical field stimulation (EFS). ICC-DMP exhibited ongoing Ca2+ transients before stimuli were applied. EFS caused initial suppression of Ca2+ transients, followed by escape during sustained stimulation, and large increases in Ca2+ transients after cessation of stimulation. Basal Ca2+ activity and the excitatory phases of Ca2+ responses to EFS were inhibited by atropine and neurokinin 1 receptor (NK1) antagonists, but not by NK2 receptor antagonists. Exogenous ACh and substance P (SP) increased Ca2+ transients, atropine and NK1 antagonists decreased Ca2+ transients. Neurokinins appear to be released spontaneously (tonic excitation) in small intestinal muscles and are the dominant excitatory neurotransmitters. Subcellular regulation of Ca2+ release events in ICC-DMP may be a means by which excitatory neurotransmission organizes intestinal motility patterns.


Introduction
Muscles of the gastrointestinal (GI) tract are innervated by both excitatory and inhibitory enteric motor neurons (Furness, 2012), and motility patterns of the gut depend on the outputs of the enteric nervous system. Neural inputs are overlaid on the basal excitability of the smooth muscle cells (SMCs) that line the walls of GI organs. SMC excitability is determined by ionic conductances and Ca 2ϩ sensitization mechanisms intrinsic to these cells but also by interstitial cells that are electrically coupled to SMCs.
Together SMCs and interstitial cells, i.e., interstitial cells of Cajal (ICC) and platelet-derived growth factor receptor ␣-immunopositive (PDGFR␣ ϩ ) cells (Komuro, 2006;Iino et al., 2009;Blair et al., 2012;Baker et al., 2013), form an electrical syncytium, known as the SIP syncytium . It is the integrated output of these cells that determines the basal excitability of GI smooth muscle tissues and ultimately the responses to enteric motor neurons and other higher order regulatory pathways (Sanders et al., 2014a).
In the small intestine, a network of ICC in the myenteric region (ICC-MY) serves as the pacemaker cells that generate and actively propagate electrical slow waves and organize contractile activity into a phasic pattern that underlies segmental contractions (Langton et al., 1989;Ward et al., 1994;Ordög et al., 1999;Sanders et al., 2014a). Another class of ICC are distributed within smooth muscle bundles in the deep muscular plexus (ICC-DMP; Rumessen et al., 1992;Zhou and Komuro, 1992) throughout the smooth muscle organs of the GI tract. ICC-DMP are innervated by motor neurons and transduce part of the input from enteric motor neurons (Wang et al., 2003b;Iino et al., 2004;Ward et al., 2006). This conclusion is based on the fact that ICC-DMP are: (1) closely apposed to varicosities of enteric motor neurons, forming synaptic-like contacts, i.e., Ͻ20 nM (Rumessen et al., 1992;Zhou and Komuro, 1992); (2) express major receptors for enteric motor neurotransmitters (Sternini et al., 1995;Vannucchi et al., 1997;Chen et al., 2007); (3) display evidence of receptor binding, receptor internalization, and translocation of signaling molecules on nerve stimulation (Wang et al., 2003b;Iino et al., 2004); and (4) electrically coupled to SMCs via gap junctions (Daniel et al., 1998;Daniel and Wang, 1999;Seki and Komuro, 2001). Experiments in other regions of the GI tract, where ICC-IM are lost in mutant animals have shown distinct changes in motor neurotransmission in the absence of ICC (Daniel and Posey-Daniel, 1984;Burns et al., 1996;Komuro et al., 1999;Ward et al., 2000;Klein et al., 2013). Nevertheless, there is controversy about the importance of ICC in neurotransmission, and some investigators have argued that ICC are not important elements in enteric nerve responses (Goyal and Chaudhury, 2010;Goyal, 2016).
A fundamental mechanism involved in the activation of ICC (as pacemakers and in regulating the excitability of GI muscles) is Ca 2ϩ release from intracellular stores (van Helden and Imtiaz, 2003;Lee et al., 2007;Baker et al., 2016;Drumm et al., 2017). Ca 2ϩ release is important because it activates Ca 2ϩ -activated Clchannels (CaCCs), encoded by Ano1, that are strongly expressed in ICC (Chen et al., 2007;Gomez-Pinilla et al., 2009;Zhu et al., 2015). We have used mice expressing Ca 2ϩ sensors specifically in ICC to investigate the Ca 2ϩ transients generated by ICC in intact intestinal muscles Drumm et al., 2017).
Excitatory neurotransmission in the gut is mediated predominantly via cholinergic neurotransmitters and neurokinins. The tachykinin (TKs) family of peptides [substance P (SP), neurokinin A (NKA) and NKB] is expressed throughout the GI tract (Holzer and Holzer-Petsche, 2001;Cipriani et al., 2011;Mitsui, 2011;Steinhoff et al., 2014). SP, NKA, and NKB are preferentially mediated through the stimulation of neurokinin 1 receptor (NK1), NK2, and NK3 G protein-coupled receptors. Both NK1 and NK2 receptors mediate contractile effects in the gut. Smooth muscle electrical, and motor events induced by electrical field stimulation (EFS) can involve both NK1 and NK2 receptors. But functional evidence supports the involvement of the NK1 subtype in mediating nonadrenergic noncholinergic (NANC) contractions to EFS in the mouse small intestine (Iino et al., 2004;De Schepper et al., 2005).
In the present study, we investigated the hypothesis that a major mechanism by which enteric motor neurotransmitters regulate ICC is through modulation of Ca 2ϩ release events. To test this hypothesis, we explored whether excitatory neural inputs to ICC-DMP are coupled to Ca 2ϩ release and characterized the nature of the Ca 2ϩ responses that constitute this transduction pathway for postjunctional excitatory transmission.

Animals
GCaMP3-floxed mice (B6.129S-Gt(ROSA)26 Sor tm38(CAG-GCaMP3)Hze /J) and their wild-type siblings (C57BL/6) were acquired from The Jackson Laboratory and crossed with Kit-Cre mice (c-Kit ϩ/Cre-ERT2 ), provided by Dr. Dieter Saur (Technical University Munich, Munich, Germany). Kit-Cre-GCaMP3 mice (both sexes) were injected with tamoxifen at six to eight weeks of age (2 mg for three consecutive days), as previously described  to activate Cre recombinase and GCaMP3 expression. 15 days after tamoxifen injection, Kit-Cre-GCaMP3 mice were anaesthetized by isoflurane inhalation (Baxter) and killed by cervical dislocation. All animals used for these experiments were handled in compliance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals, and the protocols were approved by the Institutional Animal Use and Care Committee at the University of Nevada, Reno.

Tissue preparation
Segments of jejunum (2 cm in length) were removed from mice and bathed in Krebs-Ringer bicarbonate solu-National Institutes of Health National Institute of Diabetes and Digestive and Kidney Diseases Grants P01 DK41315 and P30-GM110767. *S.A.B. and B.T.D. contributed equally to this work. tion (KRB). Jejunal segments were opened along the mesenteric border and luminal contents were washed away with KRB. The mucosa and sub-mucosa layers were removed by sharp dissection, and the remaining tunica muscularis was pinned flat within a Sylgard coated dish.
All drugs were purchased from Tocris Bioscience and dissolved in the solvents recommended by the manufacturer to obtain stock solutions. Final concentrations used in experiments were obtained by dilution into KRB.

Fluorescence-activated cell sorting (FACS), RNA extraction, and quantitative PCR (qPCR)
Jejunal ICC were dispersed from Kit ϩ/copGFP mice as previously described Zhu et al., 2011). ICC were sorted and purified by FACS (FACSAria II; Becton-Dickinson) using an excitation laser (488 nm) and emission filter (530/30 nm). Sorting was performed using a 130-m nozzle and a sheath pressure of 12 psi.
RNA was prepared from sorted ICC and dispersed jejunal cells of the tunica muscularis before sorting using an illustra RNAspin Mini RNA Isolation kit (GE Healthcare). The PCR primers used and their GenBank accession numbers are provided in Table 1. qPCR was performed using SYBR green chemistry on the 7500 HT Real-time PCR System (Applied Biosystems) and analyzed, as previously described . All datasets were normalized to the housekeeping gene Gapdh.

Calcium imaging
Jejunal muscle sheets (5.0 ϫ 10.0 mm) were pinned to the base of a 5-ml, 60-mm diameter Sylgard-coated dish. The muscles were perfused with warmed KRB solution at 37°C for an equilibration period of 1 h. Fluorescence imaging was performed with a spinning-disk confocal microscope (CSU-W1 spinning disk; Yokogawa Electric Corporation) mounted to an upright Nikon Eclipse FN1 microscope equipped with a 60ϫ 1.0 NA CFI Fluor lens (Nikon Instruments Inc). GCaMP3, expressed solely in ICC, was excited at 488 nm using a laser coupled to a Borealis system (ANDOR Technology). Emitted fluorescence (Ͼ515 nm) was captured using a high-speed EM-CCD Camera (Andor iXon Ultra; ANDOR Technology). Image sequences were collected at 33 fps using Meta-Morph software (Molecular Devices Inc). Additional Ca 2ϩ imaging data were acquired with an Eclipse E600FN microscope (Nikon Inc.) equipped with a 60ϫ 1.0 CFI Fluor lens (Nikon instruments Inc). In this system, GCaMP3 was excited at 488 nm (T.I.L.L. Polychrome IV), as previously described (Baker et al., 2013). All Ca 2ϩ imaging experiments were performed in the presence of nicardipine (100 nM) to minimize contractile movements.

Calcium event analysis
Analysis of Ca 2ϩ activity in ICC-DMP was performed, as described previously . Briefly, movies of Ca 2ϩ activity in ICC-DMP were converted to a stack of TIFF (tagged image file format) images and imported into custom software (Volumetry G8c, GW Hennig) for analysis. Tissue movement was stabilized to ensure accurate measurement of Ca 2ϩ transients from ICC-DMP. Whole cell ROIs were used to generate spatio-temporal (ST) maps of Ca 2ϩ activity in individual ICC-DMP. ST maps were then imported as TIFF files into ImageJ (version 1.40, National Institutes of Health; http://rsbweb.nih.gov/ij) for post hoc quantification analysis of Ca 2ϩ events.

Experimental design and statistical analysis
Ca 2ϩ event frequency in ICC-DMP was expressed as the number of events fired per cell per second (s Ϫ1 ). Ca 2ϩ event amplitude was expressed as ⌬F/F 0 , the duration of Ca 2ϩ events was expressed as full duration at half maximum amplitude (FDHM), and Ca 2ϩ event spatial spread was expressed as m of cell propagated per Ca 2ϩ event. Unless otherwise stated, data are represented as mean Ϯ SEM. Statistical analysis was performed using either a Student's t test or with an ANOVA with a Dunnett post hoc test where appropriate. In all statistical analyses, p Ͻ 0.05 was taken as significant; p Ͻ 0.05 are represented by a single asterisk ‫,)ء(‬ p Ͻ 0.01 are represented by two asterisks ‫,)ءء(‬ and p Ͻ 0.001 are represented by three asterisks ‫.)ءءء(‬ When describing data throughout the text,  n refers to the number of animals used in that dataset while c refers to the numbers of cells used in that same dataset.

Results
Postjunctional modulation of Ca 2؉ signaling in ICC-DMP by enteric nerve stimulation ICC-DMP displayed intracellular Ca 2ϩ transients that fired in a stochastic manner (Fig. 1), as reported previously . Ca 2ϩ transients were generated at multiple sites along the length of individual ICC-DMP and were typically localized, demonstrating no mechanism for active or regenerative propagation of these events within individual cells or between cells and no extrinsic mechanism of entrainment, as has been previously suggested (Huizinga et al., 2014). Ca 2ϩ transients in ICC-DMP exhibit a range of frequencies, amplitudes, durations and spatial spread . ICC are thought to be intermediaries in enteric neurotransmission, relaying signals from enteric neurons to smooth muscle cells, that are electrically coupled to ICC (Daniel et al., 1998;Daniel and Wang, 1999). Therefore, we investigated how Ca 2ϩ transients are modulated by enteric neurons activated by EFS.
EFS (10 Hz, 0.5 ms for 5-s trains) caused two distinct Ca 2ϩ responses: (1) an initial inhibitory phase; (2) an excitatory response that occurred largely after cessation of EFS (Movie 1). The initial inhibitory response at the onset of EFS lasted about ϳ2 s. During this phase, Ca 2ϩ transients in ICC-DMP ceased ( Fig. 1A-C). In the final 3 s of EFS, Ca 2ϩ transients escaped from inhibition leading to an excitatory response that persisted into the period after cessation of the stimulus (Fig. 1A). These effects are illustrated by an ST map and Ca 2ϩ activity traces in Figure  1B,C. This example demonstrates that in the final 3 s of EFS and particularly in the 5 s after cessation of EFS, Ca 2ϩ transients were increased relative to the control period, and firing sites within ICC-DMP increased their firing frequency. We also found that the initiation sites for Ca 2ϩ transients varied temporally in response to EFS (Fig.  1B). These responses were mediated by neuronal inputs, as they were blocked by tetrodotoxin (TTX, 1 M, data not shown). As above, after the onset of EFS, an inhibitory response phase occurred, but in subsequent experiments we concentrated on the excitatory aspects of the neural responses.
The excitatory Ca 2ϩ response to EFS was quantified during the final 3 s of EFS ( Fig. 2A, blue dashed box) and in the 5 s immediately following EFS (post-EFS; Fig. 2A, green dashed box). In the pre-EFS period, the control frequency of Ca 2ϩ transients was 1.04 Ϯ 0.08 events s Ϫ1 , and this was increased significantly during the final 3-s period of EFS to 1.8 Ϯ 0.15 events s Ϫ1 (Fig. 2B, p Ͻ 0.0001, n ϭ 23, c ϭ 56). The frequency of Ca 2ϩ transients in the post-EFS period was also significantly increased from control, firing on average at 2.1 Ϯ 0.1 events s Ϫ1 (Fig. 2B, p Ͻ 0.0001, n ϭ 23, c ϭ 56). There was a significant increase in Ca 2ϩ transient amplitude in the final 3 s of EFS from 0.8 Ϯ 0.06-1.1 Ϯ 0.05 ⌬F/F 0 (Fig. 2C, p Ͻ 0.05, n ϭ 23, c ϭ 56), although there was no significant increase in amplitude in the post-EFS period compared to control (Fig. 2C, p Ͼ 0.05, n ϭ 23, c ϭ 56). Ca 2ϩ transient duration increased in the final 3 s of EFS from 193 Ϯ 3.7 to 219.6 Ϯ 7.9 ms (Fig. 2D, p Ͻ 0.01, n ϭ 23, c ϭ 56) and was also significantly increased in the post-EFS period, increasing to 222 Ϯ 6.5 ms (Fig. 2D, p Ͻ 0.001, n ϭ 23, c ϭ 56). Ca 2ϩ transient propagation spread was also increased in the final 3 s of EFS from 11 Ϯ 0.6 to 15.4 Ϯ 0.9 m (Fig. 2E, p Ͻ 0.001, n ϭ 23, c ϭ 56) and was also significantly increased, as compared to control, in the post-EFS period, with Ca 2ϩ transients propagating an average of 12.9 Ϯ 0.6 m (Fig. 2E, p Ͻ 0.05, n ϭ 23, c ϭ 56). The number of Ca 2ϩ firing sites in ICC-DMP was decreased significantly during the final 3 s of EFS (p Ͻ 0.001) and during the post-EFS period (p Ͻ 0.001; Fig. 2F, n ϭ 23, c ϭ 56). This is likely a result of the increased propagation spread of Ca 2ϩ transients during these periods, as shown in Figure 2E. As the frequency of Ca 2ϩ transients increased and they propagated over longer distances, individual firing sites may summate to create the increase in propagation distances observed during the final seconds of EFS and post-EFS. This could lead to an apparent reduction in firing sites, as the underlying sites were masked by propagating Ca 2ϩ waves. A small increase in Ca 2ϩ transient propagation velocity, that did not reach significance, was also observed during the final 3 s of EFS and during the post-EFS period (p Ͻ 0.05; Fig.  2G, n ϭ 23, c ϭ 56).

EFS-evoked frequency-dependent excitatory Ca 2؉ responses in ICC-DMP
We examined whether the Ca 2ϩ responses in ICC-DMP were dependent on the frequency of EFS. EFS was applied to muscles from 1 to 20 Hz (0.5 ms, 5-s trains). No changes in Ca 2ϩ transient parameters were resolved during 1-Hz stimulation (Fig. 3A, n ϭ 5, c ϭ 16), although a significant increase in the frequency of Ca 2ϩ transients occurred in the post-EFS period (Fig. 3A, p Ͻ 0.05, n ϭ 5, c ϭ 16). Higher EFS frequencies (5, 10, and 20 Hz) increased Ca 2ϩ transients significantly during EFS (final 3 s) and during the post stimulus period (Fig. 3B,C). For example, 5 Hz EFS increased the firing frequency (final 3 s) to 2.5 Ϯ 0.6 events s Ϫ1 , which was significantly greater than control values of 1.6 Ϯ 0.4 events s Ϫ1 (p Ͻ 0.05, n ϭ 5, c ϭ 16). EFS 5 Hz also increased Ca 2ϩ transient frequency during the post-EFS period to 2.5 Ϯ 0.4 events s Ϫ1 , as compared to 1.6 Ϯ 0.4 events s Ϫ1 in control (p Ͻ 0.01, n ϭ 5, c ϭ 16). During EFS, the amplitude and duration of Ca 2ϩ transients were not significantly changed at all frequencies tested (p Ͼ 0.05). However, Ca 2ϩ transient duration increased during the post-EFS period at 5 Hz from 198 Ϯ 10 to 228 Ϯ 13.1 ms (p Ͻ 0.05, n ϭ 5, c ϭ 16). The spread of Ca 2ϩ transients was not significantly affected by 1-Hz EFS but increased significantly at 5 Hz during EFS (final 3 s; increased from 10.1 Ϯ 0.9 to 18.6 Ϯ 2.6 m (p Ͻ 0.05, n ϭ 5, c ϭ 16). At 20 Hz, the spatial spread increased from 8.5 Ϯ 0.6 to 11.9 Ϯ 1.2 m during EFS (p Ͻ 0.05, n ϭ 5, c ϭ 14). The change in firing frequency (% change) for each stimulus 1, 5 10, and 20 Hz was calculated and plotted in Figure 3D,E during EFS (final 3 s; Fig. 3D) and after the stimulus period (Fig. 3E). The firing of Ca 2ϩ transients was dependent on the stimulus frequency during both periods (Fig. 3D,E).

Expression of excitatory cholinergic and neurokinin receptors in ICC
Excitatory neurotransmitters mediate responses by binding to specific post junctional receptors. In the case of excitatory enteric neurotransmission, responses have been attributed to muscarinic (M2 and M3) receptors and neurokinin (NK1 and NK2) receptors expressed in small intestinal muscles (Lavin et al., 1998;Stadelmann et al., 1998;Iino et al., 2004;Faussone-Pellegrini and Vannucchi, 2006). In this study, we sorted ICC (CopGFP-Kit ϩ cells) from small intestinal muscles of Kit ϩ/copGFP mice by FACS from, as previously described , and characterized the expression of Chrm2 and Chrm3 transcripts and Tacr1 and Tacr2 transcripts. We noted higher expression of Chrm3 in ICC in comparison to Chrm2 normalized to the housekeeping gene Gapdh (Chrm3: 0.043 Ϯ 0.001; Chrm2: 0.029 Ϯ 0.002, P ϭ 0.001, n ϭ 4). Chrm3 transcripts were also higher in ICC relative to unsorted cells (total cell population). Tacr1 was also highly expressed in ICC (Tacr1: 0.06 Ϯ 0.01, n ϭ 4), and expression of Tacr2 was not resolved in ICC. Thus, the dominant receptor transcripts in ICC were Chrm3 and Tacr1.
When cholinergic and NK1 receptors were both antagonized by adding both atropine (1 M) and RP 67580 (1 M), pronounced inhibition of Ca 2ϩ transients persisted during the final 3 s of EFS and during the post stimulus period, as shown in Figure 10A-F (n ϭ 4, c ϭ 20).
Next, we inhibited cholinergic and neurokinin transmission with atropine and RP 67580 in the presence of L-NNA and MRS 2500. Under these conditions all Ca 2ϩ transients were significantly diminished across all parameters tabulated, as shown in Figure 13A-F (n ϭ 5, c ϭ 32).

Discussion
Innervation of GI muscles by enteric motor nerves and the integrated firing of these neurons is essential for generating archetypal motility patterns (Spencer et al., 2016). ICC are innervated by enteric motor neurons, and their responses to neurotransmitters contribute to complex postjunctional responses of the SIP syncytium (Ward et al., 2000;Iino et al., 2004). In the case of the small intestine, ICC-DMP are an intramuscular type of ICC that are closely associated with and innervated by motor neurons (Rumessen et al., 1992;Zhou and Komuro, 1992;Wang et al., 2003b;Iino et al., 2004;Ward et al., 2006). We recently described the properties of spontaneous Ca 2ϩ transients that occur in the absence of extrinsic stimuli in these cells . In the present study we investigated the effects of excitatory enteric motor neurotransmission on Ca 2ϩ transients in ICC-DMP, because these events mediate activation of CaCC, the ion channels responsible for the electrophysiological postjunctional excitatory responses to nerve stimulation in small intestinal muscles (Zhu et al., 2011). EFS of intrinsic neurons resulted in three-component effects on Ca 2ϩ transients: a brief inhibitory period (ϳ2 s), a period of escape from inhibition during sustained EFS, and a period of strong excitation after cessation of the stimuli (poststimulus or "rebound" excitation). The complexity of these responses is likely due to the fact that the enteric nervous system contains both inhibitory and excitatory motor neurons (Furness, 2012), and EFS can be expected to activate both classes of neurons.
In the mouse small intestine, the neurokinin component of the excitatory neural inputs to ICC-DMP was dominant. Our experiments also suggest that tonic release of neurokinins and binding to NK1 receptors is responsible for significant drive in generating the Ca 2ϩ transients observed under basal conditions in ICC-DMP . Thus, the Ca 2ϩ transients observed in the absence of applied stimuli are not "spontaneous" and do not appear to be driven intrinsically within ICC-DMP. Excitatory neurotransmitters greatly increased Ca 2ϩ transients in ICC-DMP, and this mechanism likely underlies a portion of the postjunctional electrophysiological response to excitatory neural regulation (Zhu et al., 2011(Zhu et al., , 2015. ICC-DMP are plentiful and in close contact with varicosities of enteric motor neurons in the DMP region of the small intestine (Rumessen et al., 1992;Zhou and Komuro, 1992). We confirmed that ICC express receptors required for excitatory motor neurotransmission (e.g., muscarinic and neurokinin receptors), and transcripts for M3 (Chrm3) and NK1 (Tacr1) were enriched in ICC-DMP versus unsorted cells. However, transcripts of Chrm2 were also  present, suggesting these receptors and coupling to effectors via Gi/Go may also have a role in transduction or modulation of excitatory neurotransmission. Our findings are consistent with previous studies showing muscarinic receptors and NK1 receptor expression in ICC with immunohistochemical techniques (Sternini et al., 1995;Vannucchi et al., 1997;Stadelmann et al., 1998;Iino et al., 2004;Iino and Nojyo, 2006;Ward et al., 2006;Sanders et al., 2014b).
This study demonstrates that ICC-DMP receive and transduce excitatory neural inputs in the small bowel. Previous studies predicted this finding from morphologic observations (Rumessen et al., 1992;Zhou and Komuro, 1992;Wang et al., 2003a;Iino et al., 2004;Faussone-Pellegrini, 2006;Shimizu et al., 2008) and by showing that cholinergic excitatory neural responses develop in phase with the development of ICC-DMP and blocking Kit receptors causes parallel loss of ICC and cholinergic neural responses . Excitatory neurotransmission caused PKC translocation in ICC-DMP that was blocked by atropine (Wang et al., 2003b), demonstrating functional cholinergic innervation and muscarinic responses in these cells. ACh binding to M3 receptors can enhance Ca 2ϩ release in ICC-DMP via generation of inositol 1,4,5-trisphosphate (IP 3 ) which activates Ca 2ϩ release from the endoplasmic reticulum (ER). All of the molecular components of this pathway are expressed in ICC, as shown by transcriptome analyses (Chen et al., 2007;Lee et al., 2017). Previous direct observation of ICC-DMP in situ has shown that Ca 2ϩ transients are due to Ca 2ϩ release from intracellular stores (e.g., ER), mediated, in part, by IP 3 R . Increasing Ca 2ϩ release in ICC leads to activation of CaCC, and the inward current generated by thousands of ICC-DMP in whole muscles would provide a net depolarizing influence that would summate with slow wave depolarizations, increase the likelihood of action potentials being generated during the plateau phase of slow waves (i.e., period of peak depolarization), and enhance the amplitude of phasic contractions (Zhu et al., 2011).
While our observations suggest innervation and contributions from cholinergic nerves to postjunctional excitatory responses, our data also suggest that neurokinins are the dominant excitatory neurotransmitters affecting Ca 2ϩ transients in ICC-DMP in the mouse small intestine. ICC-DMP are closely associated with SP containing nerve fibers, and ICC-DMP express NK1 receptors (Iino et al., 2004;Faussone-Pellegrini, 2006;Shimizu et al., 2008) which is consistent with our observation that excitatory transmission to ICC-DMP was mediated through NK1 receptors. Previous studies have shown that exposure of small intestinal muscles to SP or stimulation of motor neurons causes internalization of NK1 receptors in ICC (Lavin et al., 1998;Iino et al., 2004). Our experiments showed that two NK1 receptor antagonists greatly attenuated basal Ca 2ϩ transients and suppressed responses of ICC-DMP to EFS. The strong inhibitory effects of NK1 antagonists on Ca 2ϩ transients could possibly be due to off-target effects on Ca 2ϩ stores or Ca 2ϩ release mechanisms; however, nonspecific effects do not appear to be significant because responses to CCh on Ca 2ϩ transients were intact in the presence of the NK1 antagonist, RP 67580. Taken together these findings support the importance of neurokinin signaling in shaping motility patterns in the small intestine.
The degree to which basal Ca 2ϩ transients were affected by NK1 antagonists in the present study was somewhat surprising. These results suggest ongoing release of neurokinins (i.e., tonic excitation), similar in concept to the tonic inhibition phenomena observed in many GI muscles (Wood, 1972;Lyster et al., 1995). Although this phenomenon has not been described previously in the small intestine, tonic activation of NK1 receptors has been proposed in other systems (Henry et al., 1999;Jasmin et al., 2002). In the present study attenuation of Ca 2ϩ transients by the NK1 receptor antagonists may be caused by continuous release of neurokinins or persistence of the ligand in the spaces between motor nerve varicosities and ICC-DMP.
The enhanced relative reliance on neurokinins for excitatory effects may be due, in part, to the high expression of NK1 receptors by ICC-DMP which does not appear to be true for intramuscular ICC in the colon (Lee et al., 2017). NK1 receptors also couple to cellular responses through activation of phospholipase C and generation of IP 3 (Steinhoff et al., 2014). Thus, there is a signaling pathway available for the enhancement of Ca 2ϩ transients in ICC-DMP. However, it should also be noted that transfection of neurokinin receptors in model cells has also been associated with activation of adenylate cyclase and production of cAMP (Steinhoff et al., 2014), a pathway not typically linked to enhanced release of Ca 2ϩ . Generation of cAMP and stimulation of cAMP-dependent protein kinase is known to enhance Ca 2ϩ sequestration into stores by phosphorylation of phospholamban (highly expressed in ICC; Lee et al., 2017) and stimulation of SERCA (Stammers et al., 2015). Perhaps increased loading of Ca 2ϩ stores contributes to augmentation of Ca 2ϩ transient amplitude and spatial spread by neurokinins, and enhancing the rate of recovery of Ca 2ϩ into stores after a release event, reducing the time required for a given store to generate another Ca 2ϩ transient.
In summary, this study supports the idea that significant neural regulation occurs in the intramuscular class of ICC in the small intestine (ICC-DMP). As discussed above, much of the excitatory response was mediated through NK1 receptors that are expressed largely by ICC-DMP (Sternini et al., 1995;Vannucchi et al., 1997;Iino et al., 2004). Responses to EFS were attenuated by NK1 antagonists. Previous studies have shown that electrophysiological responses in ICC-DMP are linked to Ca 2ϩ release events (Zhu et al., 2011;Zhu et al., 2015), suggesting that Ca 2ϩ transients in ICC-DMP couple to generation of inward currents and depolarizing effects on the SIP syncytium. NK2 receptors, expressed largely by SMCs (Cipriani et al., 2011), were apparently not involved in responses of ICC-DMP to neurokinins released from nerve terminals, because an NK2 antagonist had no effect on responses. The effectiveness of neurokinins as neurotransmitters in the tunica muscularis of the small intestine may be spa-tially confined by concentrations achieved in postjunctional spaces to a subset of neurokinin receptors expressed by ICC-DMP.