Inhibitory Neural Regulation of the Ca2+ Transients in Intramuscular Interstitial Cells of Cajal in the Small Intestine

Gastrointestinal motility is coordinated by enteric neurons. Both inhibitory and excitatory motor neurons innervate the syncytium consisting of smooth muscle cells (SMCs) interstitial cells of Cajal (ICC) and PDGFRα+ cells (SIP syncytium). Confocal imaging of mouse small intestines from animals expressing GCaMP3 in ICC were used to investigate inhibitory neural regulation of ICC in the deep muscular plexus (ICC-DMP). We hypothesized that Ca2+ signaling in ICC-DMP can be modulated by inhibitory enteric neural input. ICC-DMP lie in close proximity to the varicosities of motor neurons and generate ongoing Ca2+ transients that underlie activation of Ca2+-dependent Cl− channels and regulate the excitability of SMCs in the SIP syncytium. Electrical field stimulation (EFS) caused inhibition of Ca2+ for the first 2–3 s of stimulation, and then Ca2+ transients escaped from inhibition. The NO donor (DEA-NONOate) inhibited Ca2+ transients and Nω-Nitro-L-arginine (L-NNA) or a guanylate cyclase inhibitor (ODQ) blocked inhibition induced by EFS. Purinergic neurotransmission did not affect Ca2+ transients in ICC-DMP. Purinergic neurotransmission elicits hyperpolarization of the SIP syncytium by activation of K+ channels in PDGFRα+ cells. Generalized hyperpolarization of SIP cells by pinacidil (KATP agonist) or MRS2365 (P2Y1 agonist) also had no effect on Ca2+ transients in ICC-DMP. Peptidergic transmitter receptors (VIP and PACAP) are expressed in ICC and can modulate ICC-DMP Ca2+ transients. In summary Ca2+ transients in ICC-DMP are blocked by enteric inhibitory neurotransmission. ICC-DMP lack a voltage-dependent mechanism for regulating Ca2+ release, and this protects Ca2+ handling in ICC-DMP from membrane potential changes in other SIP cells.


INTRODUCTION
In the gastrointestinal tract, muscle bundles making up the tunica muscularis have intrinsic mechanisms of excitability, and this has been described as myogenic activity. In fact this level of motor control is due not only to the functions of smooth muscle cells (SMCs), because the behavior of SMCs is modulated by interstitial cells [e.g., interstitial cells of Cajal (ICC) and cells labeled with antibodies to plateletderived growth factor receptor alpha (aka PDGFRα + cells)]. ICC and PDGFRα + cells are electrically coupled to SMCs (Zhou and Komuro, 1992a;Torihashi et al., 1993;Seki and Komuro, 1998;Horiguchi and Komuro, 2000), and the resulting cellular network has been referred to as the SIP syncytium . Conductance changes in one type of SIP cell causes changes in the membrane potentials and excitability of coupled cells. The SIP syncytium is innervated by enteric motor neurons, and each cell type expresses receptors that can bind to and transduce inputs from neurotransmitters released from motor neurons (Chen et al., 2007;Lee et al., 2017). Neural inputs are integrated by the SIP syncytium and the output sets the moment-to-moment excitability of the SMCs, generating the underlying basis for motility patterns such as phasic contractions, summation of phasic contractions to generate tone, peristalsis and segmentation.
ICC are present in all smooth muscle portions of the GI tract, and in the small intestine there are at least 2 populations of these cells. ICC in the myenteric plexus region (ICC-MY) generate pacemaker activity that develops into electrical slow waves (Langton et al., 1989;Ward et al., 1994;Huizinga et al., 1995;Ordog et al., 1999;Sanders et al., 2014b;Drumm et al., 2017). ICC within the deep muscular plexus region (ICC-DMP) are in close contact with varicosities of excitatory and inhibitory enteric motor neurons (Rumessen et al., 1992;Zhou and Komuro, 1992b;Blair et al., 2012), express receptors that can bind to major enteric motor neurotransmitters (Chen et al., 2007), and, as above, are electrically coupled to SMCs via gap junctions (Daniel et al., 1998;Seki and Komuro, 2001). These properties of ICC-DMP led to the suggestion that they may be innervated and involved in generating post-junctional responses to motor neurotransmission. In other regions of the GI tract loss of intramuscular ICC caused changes or disruption in normal motor neurotransmission (Burns et al., 1996;Ward et al., 2000Ward et al., , 2006Wang et al., 2003a;Iino et al., 2004;Klein et al., 2013;Sanders et al., 2014a). Mounting evidence also suggests that ICC-DMP are innervated and provide at least part of the receptive field for motor neurotransmission: (i) Due to the close, synaptic-like associations between ICC-DMP and nerve varicosities, neurotransmitter concentrations could be quite high near neurotransmitter receptors (Sanders et al., 2010;Bhetwal et al., 2013); (ii) functional immunohistochemistry has shown translocation of signaling molecules in ICC-DMP consistent with binding of muscarinic and NK1 receptors (Wang et al., 2003b;Iino et al., 2004); and (iii) a conductance unique to ICC-DMP (Ano1) is activated by motor neurotransmission (Zhu et al., 2011).
The precise mechanisms through which ICC transduce inputs from motor neurons are poorly understood, largely because past studies have relied upon in vitro experiments on isolated cells (in many cases on cells studied after several days in culture (Koh et al., 2000;D'antonio et al., 2009;So et al., 2009;Kim et al., 2012), studies on intact muscles utilizing techniques requiring fixation of tissues (Wang et al., 2003b;Iino et al., 2004), or studies using membrane permeable Ca 2+ sensors that load to varying degrees into all cells in tissues and provide, as a result, confusing and possibly misleading information about the Ca 2+ signaling in ICC (Huizinga et al., 2014;Zhu et al., 2016). Ca 2+ signaling, however, is important because a major conductance in ICC-DMP that is affected by neurotransmission is a Ca 2+ -activated Cl − conductance (Ano1; Zhu et al., 2011). We hypothesize that modulation of Ca 2+ transients in ICC constitutes a major mechanism regulated by enteric neurotransmission. Therefore, we have used optogenetics and mice expressing a geneticallyencoded Ca 2+ sensor (GCaMP3) expressed specifically in murine ICC to investigate the responses of ICC-DMP to enteric inhibitory neurotransmission. Our results show that enteric inhibitory neurotransmitters exert powerful inhibitory effects on Ca 2+ release, which would be expected to reduce activation of Ano1 and development of spontaneous transient inward currents (STICs) and reduce the excitatory drive exerted upon the SIP syncytium by ICC-DMP.

Animals
GCaMP3-floxed mice (B6.129S-Gt(ROSA) 26Sor tm38(CAG−GCaMP3)Hze /J) and their corresponding wild-type siblings (C57BL/6) were purchased from Jackson Laboratories (Bar Harbor, MN, USA) and subsequently crossed with Kit-Cre mice (c-Kit +/Cre−ERT2 ) provided by Dr. Dieter Saur (Technical University Munich, Munich, Germany). Kit-Cre-GCaMP3 mice underwent treatment with tamoxifen at 6-8 weeks of age (2 mg for 3 consecutive days), as previously described (Baker et al., 2016), to induce activation of Cre recombinase in ICC and activate expression of GCaMP3. After tamoxifen (15 days); Kit-Cre-GCaMP3 mice were anesthetized by isoflurane inhalation (Baxter, Deerfield, IL, USA) and killed by cervical dislocation. All animals used for these experiments were handled in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and approved by the Institutional Animal Use and Care Committee at the University of Nevada, Reno [Animal assurance # D16-00311 (A3500-01)].

Tissue Preparation
Following an abdominal incision, 2 cm segments of jejunum were removed and bathed in Krebs-Ringer bicarbonate solution (KRB). The jejunal segments opened along the mesenteric border, and intra-luminal contents were removed by washing with KRB. Mucosal and sub-mucosal layers were removed by sharp dissection, and the remaining tunica muscularis was pinned out in a Sylgard coated dish.

Drugs and Solutions
Tissues were maintained by perfusing with KRB containing (mmol/L): NaCl, 120.35; KCl, 5.9; NaHCO 3 , 15.5; NaH 2 PO 4 , 1.2; MgCl 2 , 1.2; CaCl 2 , 2.5; and glucose, 11.5. The KRB was bubbled with a mixture of 97% O 2 -3% CO 2 and warmed to 37 ± 0.2 • C. All drugs were purchased from Tocris Bioscience (Ellisville, Missouri, USA) and dissolved in solvents recommended by the manufacturer to create appropriate stock solutions. Final concentration used for experiments were obtained by diluting with KRB. All work was performed according to biosafety level II regulations.

Responses of ICC to Intrinsic Nerve Stimulation
Neural responses were elicited by electrical field stimulation (EFS; 1-20 Hz, 0.5 ms pulse duration; 10-15 v; 5 s trains) generated by a Grass stimulator Grass S48 stimulator (Quincy, MA, USA) and delivered via two platinum electrodes placed on either side of muscle strips. Responses evoked by EFS were completely abolished by tetrodotoxin (TTX: 1 µM, data not shown).
Fluorescence Activated Cell Sorting (FACS), RNA Extraction, and Quantitative PCR Jejunal ICC were dispersed from Kit +/copGFP mice as previously described (Zhu et al., 2009). Enriched populations of ICC were sorted by FACS (FACSAria II; Becton-Dickinson) using an excitation laser (488 nm) and emission filter (530/30 nm). Sorting was performed using a 130-µm nozzle and a sheath pressure of 12 psi. RNA was prepared from sorted ICC and dispersed unsorted 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 listed in Table 1. Quantitative PCR (qPCR) was performed using SYBR green chemistry on the 7500 HT Real-time PCR System (Applied Biosystems) and analyzed, as previously described (Baker et al., 2016).

Calcium Imaging
For imaging studies, the muscles were equilibrated with continuous perfusion of warmed KRB solution at 37 • C for 1 h. 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 60x 1.0 NA CFI Fluor lens (Nikon instruments INC, NY, USA). GCaMP3, expressed in ICC within the jejunal muscles, was excited at 488 nm using a laser coupled to a Borealis system (ANDOR Technology, Belfast, UK). Emitted

Gene
Sequence GenBank accession number fluorescence (>515 nm) was captured using a high-speed EMCCD Camera (Andor iXon Ultra; ANDOR Technology, Belfast, UK). Image sequences were acquired at 33 fps using MetaMorph software (Molecular Devices Inc., CA, USA). In some experiments images were acquired with an Eclipse E600FN microscope (Nikon Inc., Melville, NY, USA) equipped with a 60x 1.0 CFI Fluor lens (Nikon instruments Inc., NY, USA). In this system, GCaMP3 was excited at 488 nm (T.I.L.L. Polychrome IV, Grafelfing, Germany), as previously described (Baker et al., 2013). All Ca 2+ imaging experiments were performed in the presence of nicardipine (100 nM) to minimize movement artifacts resulting from contractions.

Calcium Event Analysis
Analysis of Ca 2+ activity in ICC-DMP was performed, as described previously (Baker et al., 2016). Briefly, movies of Ca 2+ 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 initial pre-processing analysis.

Statistics
Ca 2+ event frequency in ICC-DMP was expressed as the number of events fired per cell per second (s −1 ). Ca 2+ event amplitude was expressed as F/F 0 , the duration of Ca 2+ events was expressed as full duration at half maximum amplitude (FDHM), and Ca 2+ event spatial spread was expressed as µm of cell propagated per Ca 2+ event. Unless otherwise stated, data is represented as mean ± standard error (S.E.M.). Statistical analysis was performed using either a student's t-test or with an ANOVA test where appropriate (data was tested for normality using a D'Agostino-Pearson omnibus normality test). In all statistical analyses, P < 0.05 was taken as significant. P < 0.05 are represented by a single asterisk ( * ), P < 0.01 are represented by two asterisks ( * * ), P < 0.001 are represented by three asterisks ( * * * ) and P < 0.0001 are represented by four asterisks ( * * * * ). 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 data set.

Enteric Nerve Stimulation Produces Inhibition of Ca 2+ Transients in ICC-DMP
Ca 2+ transients in ICC-DMP were ongoing and stochastic in nature, as previously reported (Baker et al., 2016). In the absence of stimulation there was no evidence of coordination between the events occurring within single cells or in other cells within a field-of-vision (FOV), suggesting there was no voltage-dependent regulation of Ca 2+ transients in ICC-DMP. Electrical field stimulation (EFS; 10 Hz, 0.5 ms for 5 s trains) resulted in multiphasic responses (inhibition of Ca 2+ transients followed by enhancement of these events), as shown in a representative spatiotemporal map (ST map) and traces of Ca 2+ transients (Figures 1A,B). Prior to EFS, Ca 2+ transients fired from multiple sites (representative sites marked with white arrows along the vertical axis of the ST map). During EFS (5 s), an initial inhibitory period was observed in which Ca 2+ transients ceased for ∼2 s. The firing of Ca 2+ transients escaped from inhibition during the final 3 s of EFS, and a robust increase in Ca 2+ transient firing was apparent during this period and after cessation of EFS ( Figures 1A,B).

Ca 2+ Firing Sites in ICC-DMP Have Variable Escape From Inhibition Characteristics
In this and a previous study we noted a range in the number of Ca 2+ firing sites in ICC-DMP, from a single site to 13 sites (Baker et al., 2016). In the present study ICC-DMP averaged 5.2 ± 0.4 firing sites per cell ( Figure 2D, n = 19, c = 48). The nature of Ca 2+ transients during sustained EFS (5 s) was examined in more detail by the analysis described in Figure 2. Similar to the previous example, Ca 2+ transients ceased during the initial 2 s of EFS, as shown in the ST map (Figure 2A). During the final 3 s of EFS, Ca 2+ firing sites escaped from inhibition, but each site escaped at different times, as illustrated by 3-D plots (Figure 2Bii). Ca 2+ transients in the cell shown in Figure 2 originated from 2 distinct firing sites, as indicated by the white arrows in the 3-D plots (Figure 2Bii). The activities of these sites are also plotted as line traces in Figure 2C. Site 1 was the first site to escape from inhibition, with a ∼2 s delay from the onset of EFS to the first Ca 2+ transient that occurred at this site ( Figure 2C). A greater period of inhibition was observed at site 2; a delay of ∼2.6 s occurred at this site ( Figure 2C). The inhibitory period from the onset of EFS to the first appearance of a Ca 2+ transient at all firing sites in ICC-DMP averaged 2.4 ± 0.1 s (range 0.2-4.5 s; Figure 2E, n = 19, c = 48). The delay periods describing the escape from inhibition are plotted as a summary histogram in Figure 2F. The average inhibitory period at all sites was 3.8 ± 1.8 s, but ranged from 0.2 to 9.9 s ( Figure 2F, n = 19, c = 48).
It was also apparent that different ICC-DMP within a FOV escaped inhibition at variable times and did not show a coordinated escape response. For example, a FOV containing several ICC-DMP is shown in Figure 3A. Two cells are highlighted by the red and green ROIs, and the Ca 2+ transients in these cells are displayed in ST maps in Figures 3B,C. When these ST maps were merged, it can be seen that cell 1 and 2 escaped the inhibitory effects of EFS at different points in time, with a single Ca 2+ firing site active in cell 2 before anything occurred in cell 1 ( Figure 3D). Another example is provided in which Ca 2+ transients in 3 ICC-DMP in a FOV were plotted in Figure 3E. Here again, the cells did not escape inhibition at the same time points. This example also illustrates the point that Ca 2+ transients in all ICC-DMP ceased at the onset of EFS (Figures 3D,E).

Effects of SIP Syncytium Hyperpolarization on Ca 2+ Transients
The purine neurotransmitter(s) in GI muscles cause significant hyperpolarization due to opening of apamin-sensitive, small conductance Ca 2+ -activated K + channels (Banks et al., 1979;Matsuda et al., 2004;Gallego et al., 2008), but recent studies have shown these responses are generated by another celltype in the SIP syncytium (Kurahashi et al., 2011. Therefore, purinergic responses conveyed to ICC-DMP would be in the form of membrane hyperpolarization. To simulate this type of response, we tested whether hyperpolarization of the SIP syncytium by MRS 2365 (acting by hyperpolarization of PDGFRα + cells; Kurahashi et al., 2014) or pinacidil (acting by hyperpolarization of SMCs; Kito et al., 2005) affected Ca 2+ transients in ICC-DMP.
We also tested the PKG inhibitors seeking a positive control for their poor performance against nitrergic effects in ICC. Contractile experiments were performed on muscles of the colon, aorta and corpus cavernosum. Neither KT 5823 nor Rp-8-pCPT-cGMPS reduced the inhibitory effects of sodium nitroprusside (SNP; 100 nM) or DEA NONOate (10 µM) on muscles pre-contracted with norepinephrine (NE; 100 nM) or carbachol (CCh; 10 µM) (data not shown). Taken together, the lack of effects of commercial PKG inhibitors on smooth muscles, in general, might be due to poor penetration of the drugs into cells in intact tissues or targets besides PKG contributing to inhibitory responses. This is a question in need of further evaluation. FIGURE 10 | Expression of genes encoding nitrergic and peptidergic signaling molecules in ICC. (A) Quantitative PCR (qPCR) data showing the relative expression of transcripts for Gucy1a1 and Gucy1b1, protein kinase cGMP-dependent type 1: Prkg1, and inositol-1,4,5 triphosphate receptor I-associated G kinase substrate (IRAG; Mrvi1) in sorted small intestinal ICC by FACS and unsorted cells (total cell population). qPCR data is expressed as relative expression, normalized to Gapdh, n = 4. (B) qPCR data showing the relative expression of transcripts for Vipr1 and Vipr2 (VIP receptors) and Adcyplr1 (PACAP receptor) in FACS sorted small intestinal ICC and unsorted cells (total cell population). qPCR data is expressed as relative expression, normalized to Gapdh, n = 4.

Vasoactive Intestinal Peptide (VIP) Modulation of ICC-DMP Ca 2+ Transients
Nitrergic modulation of ICC-DMP transients is an important aspect of inhibitory neurotransmission, but neuropeptides (VIP and PACAP: pituitary adenylate cyclase-activating peptide) are also released from nerve terminals and might modulate ICC-DMP activity. Therefore, we examined the expression profile of peptide receptors in sorted ICC from small intestinal muscles and characterized expression of VIP receptors (Vipr1 and Vipr2) and PACAP receptor (Adcyap1r1). We noted elevated expression in all of peptidergic receptors in ICC relative to unsorted cells (total cell population). Vipr1 transcripts were higher in ICC in comparison to unsorted cells (Vipr1 in ICC: 0.006 ± 0.0005 vs. unsorted cells: 0.004 ± 0.0004, P = 0.001, n = 4; Figure 10B). Also Vipr2 in ICC: 0.006 ± 0.0006 vs. unsorted cells 0.0005 ± 0.00001 (P = 0.0001, n = 4; Figure 10B) and Adcyap1r1 in ICC: 0.01 ± 0.001; unsorted cells: 0.0008 ± 0.00001 (P = 0.0001, n = 4; Figure 10B). Peptidergic receptors are abundant in ICC suggesting that ICC has the machinery to mediate inhibitory peptidergic transmission.
While VIP 6-28 increased the basal level of Ca 2+ transient firing in a similar manner to L-NNA (Figure 14), VIP 6-28 did not significantly relieve EFS-evoked inhibitory responses (Figures 15A-C). Neither the frequency (P = 0.11), amplitude (P = 0.16), or spatial spread (P = 0.58) of Ca 2+ transients was significantly affected during the initial 2 s period of EFS VIP 6-28 (10 µM) (Figure 15C, n = 5, c = 16). However, the duration of Ca 2+ transients in the initial 2 s period of EFS was significantly increased from 59 ± 26.9 ms in control to 110 ± 29.5 ms in the presence of VIP 6-28 (Figure 15Ciii, P = 0.023, n = 5, c = 16). It is possible that multiple receptors mediate responses to inhibitory peptides in ICC.

DISCUSSION
In this study we examined enteric inhibitory regulation of Ca 2+ transients in ICC-DMP, the intramuscular class of ICC in the small intestine. Ca 2+ handling mechanisms in ICC are of interest because Ca 2+ release from stores couples to activation of Ano1 channels and the generation of STICs (Zhu et al., 2011(Zhu et al., , 2015. In the case of ICC-DMP, Ca 2+ transients have only localized influence within cells, and no evidence for propagation of Ca 2+ transients between cells, even over long periods of observation, was obtained (Baker et al., 2016). Localized, stochastic events occurring in hundreds or thousands of ICC-DMP could have significant influence on the excitability of cells of the SIP syncytium to which ICC-DMP are electrically coupled by abundant gap junctions (Zhou and Komuro, 1992a;Torihashi et al., 1993;Seki and Komuro, 1998). We found that nitrergic mechanisms are the primary neural inhibitory regulators of Ca 2+ transients in ICC-DMP, and regulation by purines was not resolved. As in many other cells, nitrergic input was transduced by binding to its natural receptor, sGC, and downstream effects were mediated by cGMP (Bult et al., 1990;Moncada et al., 1991;Pfeifer et al., 1998;Somlyo and Somlyo, 2003). Our observation, linking cGMP to inhibition of Ca 2+ release events in ICC-DMP, is a novel aspect of nitrergic regulation, and as discussed below, this is likely to be one of the fundamental inhibitory mechanisms of nitrergic regulation in GI motility. We also provide novel evidence that peptidergic neurotransmission is superimposed on ICC-DMP and provides a portion of tonic inhibition in ICC-DMP.
Interstitial cells, ICC and PDGFRα + cells, are functional elements of the SIP syncytium that regulate the excitability of SMCs in all smooth muscle regions of the GI tract (Sanders et al., 2014b). Interstitial cells transduce different parts of the motor neural inputs that regulate GI motility (Burns et al., 1996;Ward et al., 2000;Iino et al., 2004;Kurahashi et al., 2011;Baker et al., 2015). A role for ICC in neurotransmission was suggested from morphological studies that described close associations between varicose nerve terminals and ICC (Rumessen et al., 1992;Zhou and Komuro, 1992b;Faussone-Pellegrini, 2006;Blair et al., 2012). Studies in animals lacking intramuscular ICC demonstrated that these cells contribute to post-junctional responses to both excitatory and inhibitory neurotransmission (Burns et al., 1996; , and spatial spread (iv) in ICC-DMP during control conditions (pre-EFS), and in the 1st 2 s of EFS (n = 5, c = 16). ns = P > 0.05, *P < 0.05, **P < 0.01. Wang et al., 2003a;Iino et al., 2004;Klein et al., 2013;Sanders et al., 2014a) and cell specific knock down of sGC in ICC  have also been consistent with a role for ICC in neurotransmission. In the small intestine, inhibitory and excitatory post-junctional neural responses follow age-dependent development of ICC-DMP (Ward et al., 2006), and blocking Kit with neutralizing c-Kit antibody caused reduction in ICC-DMP and loss of cholinergic and nitrergic neural responses. These previous findings are in agreement with the results of the present study: (i) ICC-DMP have the necessary molecular machinery to transduce signals arising from motor neurons; (ii) ICC-DMP are innervated by enteric inhibitory motor neurons; (iii) Inhibitory neurotransmission regulates the occurrence of Ca 2+ transients in ICC-DMP, thus controlling Ca 2+ transients necessary for activation of electrophysiological responses in the SIP syncytium (Zhu et al., 2011(Zhu et al., , 2015. We focused on neuromodulation of Ca 2+ transients in ICC-DMP because Ca 2+ transients are coupled to activation of Ano1 channels and STICs in these cells (Zhu et al., 2011(Zhu et al., , 2015. Thus, regulation of Ca 2+ transients, which are ongoing in these cells, provides a means of bi-directional regulation of excitability of SMCs. Turning STICs off reduces net inward current in the SIP syncytium, and such a signal would favor stabilization of excitability; increasing STICs increases net inward current in the SIP syncytium and adds a depolarizing influence that increases the excitability of SMCs. In a previous study, we showed that inhibition of basal nerve activity with TTX increases Ca 2+ transients (Baker et al., 2016). In the present study we demonstrated that inhibition of NO synthesis with L-NNA, or inhibition of cGMP synthesis with ODQ also increased Ca 2+ transients. This would lead to increased inward current and at least partial blockade of what has been termed "tonic inhibition" of GI muscles (Wood, 1972). Inhibitory peptides also appear to contribute to tonic inhibition because Ca 2+ transients were increased by VIP 6-28. We also found that stimulation of intrinsic neurons by EFS caused a brief period of inhibition (∼2 s) in which ongoing Ca 2+ transients in ICC-DMP were largely abolished. Shutting off of STICs during the initial phase of EFS may explain a portion of the hyperpolarization responses to nerve stimulation seen in electrophysiological recordings (Stark et al., 1991).
It is difficult to reconcile generalized tissue level responses to EFS (e.g., electrophysiological events and contractions) with events occurring in a single type of ICC. Neurotransmission may affect conductances and Ca 2+ sensitization mechanisms in multiple cells (SMCs, PDGFRα + cells, and in other types of ICC) leading to non-linear changes in voltage-dependent conductances, membrane potential and excitation-contraction coupling. At present there are no organ-specific, unified models of the many responses in GI muscles that might be initiated or repressed by stimulation of enteric neurons.
During 5 s periods of EFS, Ca 2+ firing sites within ICC-DMP escaped from inhibition. In our experiments this occurred within about 2 s from the onset of stimulation. However, the Ca 2+ release sites escaped inhibition with different temporal characteristics. The heterogeneity in the periods before escape from inhibition occurred is further demonstration of the independence of Ca 2+ release sites, even within individual cells, and supports the conclusion that there is no cellular or multi-cellular correlation between Ca 2+ release sites within the ICC-DMP network (Baker et al., 2016). The variability in the period before escape from inhibition might be attributed to several factors. There may be variability in the size or molecular composition of different release sites (e.g., relative balance between RyRs and IP 3 Rs or relative density or distribution of SERCA pumps). We showed that Ca 2+ release events were virtually blocked if neurokinin receptor antagonists were present (Baker et al., 2018). Thus, it was not possible to investigate inhibitory responses in isolation of excitatory neural inputs. Therefore, another factor affecting the escape from inhibition might be the relative density of excitatory varicosities and distribution of post-junctional receptors and/or effectors along the lengths of ICC-DMP. Greater excitatory or inhibitory neural inputs at a given site might accelerate or delay the escape from inhibition. Likewise, a greater concentration of post-junctional inhibitory or excitatory pathway components may impact the rate of escape.
Soluble guanylyl cyclase (sGC) is expressed in ICC and is the main receptor/transducer of the inhibitory effects of NO in the GI tract (Shuttleworth et al., 1993;Salmhofer et al., 2001;Iino et al., 2008Iino et al., , 2009Cobine et al., 2014;Lies et al., 2014Lies et al., , 2015Sanders, 2016). Previous immunohistochemical studies have shown that sGC-α and sGC-β are both expressed in ICC of lower esophageal sphincter, stomach, small intestine, caecum, colon, and internal anal sphincter (Salmhofer et al., 2001;Iino et al., 2009;Cobine et al., 2014;Lies et al., 2014Lies et al., , 2015, and at least from immunohistochemical analyses, sGC is more abundant in ICC than in SMCs. We confirmed these findings in the small intestine and found high expression of Gucy1a1 and Gucy1b1 in sorted ICC relative to the unsorted cell population (which would contain SMCs). Thus, ICC have the receptor and the molecular apparatus to transduce nitrergic signals and produce cGMP. Others have found that knock down of Gucy1b1 in ICC, using Cre-LoxP technology, abolished nitrergic inhibitory junction potentials (IJP) in gastric fundus and reduced IJPs in the colon, but these authors also reported that knockdown of Gucy1b1 in SMCs also either reduced the amplitude or shortened IJPs (Lies et al., 2014). Another study concluded that nitrergic relaxation of fundus muscles depends upon sGC in both ICC and SMCs (Groneberg et al., 2013). We found that nitrergic inhibition of Ca 2+ transients in ICC-DMP depends upon the sGC, as inhibitors and activators of sGC effectively modulated Ca 2+ release.
How cGMP regulates Ca 2+ release in ICC is complicated. The traditional view is that cGMP-dependent protein kinase-1 (PKG1; encoded by Prkg1) is the principal downstream signaling molecule mediating nitrergic responses, and Prkg1 is expressed in ICC in the small intestine (Salmhofer et al., 2001), as also confirmed by the present study. cGMP is thought to activate PKG1 and cause phosphorylation of downstream targets (Xue et al., 2000;Hofmann, 2005). These targets in ICC have not been defined precisely. However, one study showed that cell-specific knockdown of Prkg1 in ICC reduced NO-dependent inhibitory junction potentials in colonic smooth muscles . A signaling molecule downstream of PKG1 appears to be inositol triphosphate receptor (IP 3 R)-associated cGMP-kinase substrate (IRAG; encoded by Mrvi1), and this gene is also expressed in ICC of the small intestine. IRAG co-precipitates with IP 3 R and was found to be indispensable for cGMP regulation of Ca 2+ release in model cells or cultured human colonic SMCs (Schlossmann et al., 2000;Fritsch et al., 2004). IRAG is phosphorylated by PKG1β at Ser696 and suppresses Ca 2+ release from IP 3 R1 (Masuda et al., 2010;Schlossmann and Desch, 2011), and transcripts of the PKG1β splice variant (NM_ 011160) were 23-fold more abundant than transcripts of the PKG1α variant (NM_001013833) from RNA-seq of small intestinal ICC (Lee et al., 2017). As shown in the present study by real-time PCR, all of these signaling molecules are present in small intestinal ICC and more strongly expressed in ICC than in the general population of cells dispersed from the tunica muscularis of the jejunum. Therefore, this pathway might represent the primary mechanism for nitrergic suppression of Ca 2+ transients and waves in ICC. However, a recent paper showed that nitrergic relaxation was only slightly reduced in murine internal anal sphincter muscles of PKG −/− mice, and L-NNA abolished relaxations to nitrergic nerve stimulation in both wildtype and PKG −/− mice (Cobine et al., 2014). These findings suggest that significant cGMP-dependent, but PKG independent, pathways may contribute to nitrergic responses, and pathways specific to ICC will require additional investigation. In the present study PKG inhibitors had no effect on nitrergic responses, but these drugs appear to have penetration problems in whole muscles because they also failed to block nitrergic responses in several smooth muscle preparations (colon, aorta, and corpus cavernosum). PKG inhibitors also failed to have any significant effects on ICC pacemaker potentials (Koh et al., 2000;Shahi et al., 2014). Taken together, PKG inhibitors do not appear to be suitable for in situ studies, and genetic models with combinations of deleted genes and expression of optogenetic sensors appear to be needed for future studies to address downstream signaling mechanisms responsible for neural regulation of ICC.
The lack of purinergic effects on Ca 2+ transients in ICC-DMP might seem surprising since it is well-known that purines contribute significantly to enteric inhibitory regulation of GI muscles (Burnstock et al., 1970;Gallego et al., 2014;Jimenez et al., 2014;Sanders, 2016). P2Y1 receptors mediate purinergic enteric neural inhibition in GI muscles, as shown by pharmacological and gene deactivation studies (Gallego et al., 2012Hwang et al., 2012;Gil et al., 2013). However, dominant expression of P2ry1 is found in PDGFRα + interstitial cells, not SMCs or ICC, and purinergic inhibitory effects are mediated through PDGFRα + cells (Kurahashi et al., 2011Baker et al., 2013Baker et al., , 2015. P2Y1 receptor agonists hyperpolarize PDGFRα + cells by activation of small conductance, Ca 2+ activated K + channels, and hyperpolarization responses are conducted to other SIP cells (Kito et al., 2014). Thus, the lack of purinergic effects on ICC is compensated by effects of purines on PDGFRα + cells.
Pinacidil, through activation of K ATP in GI SMCs, also causes hyperpolarization of the SIP syncytium (Koh et al., 1998;Kito et al., 2005), but this agonist had no effect on Ca 2+ transients in ICC-DMP. Likewise, hyperpolarization of PDGFRα + cells by the P2Y1 specific agonist MRS2365 (which has no effect on GI muscles lacking P2Y1 receptors; Hwang et al., 2012); also had no effect on Ca 2+ transients in ICC-DMP. These data demonstrate that substances that cause openings of K + channels and exert hyperpolarizing effects on the SIP syncytium, do not interfere with the Ca 2+ release events occurring in the ICC-DMP component of the syncytium. We also know from previous studies that depolarization does not affect Ca 2+ transients in ICC-DMP. Our imaging studies were conducted on full thickness jejunal muscles that undergo periodic depolarizations from slow wave activity; yet there is no periodic behavior in the firing of Ca 2+ transients that might indicate regulation of Ca 2+ release by a voltage-dependent mechanism (Baker et al., 2016). In fact our data illustrate an important design feature of the SIP syncytium: By lacking a voltage-dependent mechanism that coordinates Ca 2+ release events, ICC-DMP are protected from the effects of compounds having membrane potential effects in other SIP cells. This allows neural regulation of ICC-DMP without having this mechanism pre-activated or deactivated by events occurring in other cells.
Peptidergic inhibitory neurotransmission also contributes to regulation of motility in the small intestine (Ekblad et al., 2000;Lazar et al., 2001;Matsuyama et al., 2002;Sanders, 2016). In the present study we found that Ca 2+ transients in ICC-DMP are also regulated by inhibitory peptides. VIP 6-28 enhanced basal Ca 2+ transient activity, suggesting ongoing release of peptidergic neurotransmitters and contributions from peptides to tonic inhibition. Neurotransmission involving inhibitory peptides during EFS is more complicated and may involve binding of transmitters to multiple post-junctional receptors, as several are expressed and VIP6-28 failed to block neural responses.
In summary, Ca 2+ transients in ICC-DMP are suppressed under basal conditions by TTX and this appears to occur by blocking release of NO and inhibitory peptides from intrinsic neurons. EFS caused inhibition of Ca 2+ transients, but ICC-DMP escaped from inhibition during 5 s trains of stimuli. The inhibitory period was due maingly to nitrergic effects mediated by cGMP. Purinergic inputs, that occur in parallel to release of NO in the GI muscles, did not affect Ca 2+ transients in ICC-DMP and agonists that hyperpolarize other cells in the SIP syncytium also were ineffective in modulating Ca 2+ transients in ICC-DMP. These data demonstrate a lack of voltage-dependent regulation of Ca 2+ transients in ICC-DMP. Peptidergic neurotransmission can also modulate Ca 2+ ICC-DMP, but the receptor(s) responsible for these effects are complex. Ca 2+ transients initiate inward currents in ICC-DMP that are conducted to other cells in the SIP syncytium. Thus, suppression of Ca 2+ transients in ICC-DMP by inhibitory neural inputs would tend to reduce SMC excitability and reduce contractile force in the small intestine.

AUTHOR CONTRIBUTIONS
SB, BD, and KS: Conception and design of the experiments; SB, BD, CC, KK, and KS: Collection, analysis, and interpretation of data; SB, BD, and KS: Drafting the article and revising it critically for intellectual content. All authors read and approved the manuscript before submission.

FUNDING
This study was funded by a Program Project Grant from the NIDDK, P01 DK41315-29. Cells were sorted by the Fluorescence activated cell sorting cell sorting and flow cytometry Core lab supported by P30 GM110767.