Ca²⁺ signalling in mouse urethral smooth muscle in situ: role of Ca²⁺ stores and Ca²⁺ influx mechanisms

Abstract

Key points

• Contraction of urethral smooth muscle cells (USMCs) contributes to urinary continence. Ca²⁺ signalling in USMCs was investigated in intact urethral muscles using a genetically encoded Ca²⁺ sensor, GCaMP3, expressed selectively in USMCs.

• USMCs were spontaneously active in situ, firing intracellular Ca²⁺ waves that were asynchronous at different sites within cells and between adjacent cells.

• Spontaneous Ca²⁺ waves in USMCs were myogenic but enhanced by adrenergic or purinergic agonists and decreased by nitric oxide.

• Ca²⁺ waves arose from inositol trisphosphate type 1 receptors and ryanodine receptors, and Ca²⁺ influx by store‐operated calcium entry was required to maintain Ca²⁺ release events.

• Ca²⁺ release and development of Ca²⁺ waves appear to be the primary source of Ca²⁺ for excitation–contraction coupling in the mouse urethra, and no evidence was found that voltage‐dependent Ca²⁺ entry via L‐type or T‐type channels was required for responses to α adrenergic responses.

Abstract

Urethral smooth muscle cells (USMCs) generate myogenic tone and contribute to urinary continence. Currently, little is known about Ca²⁺ signalling in USMCs in situ, and therefore little is known about the source(s) of Ca²⁺ required for excitation–contraction coupling. We characterized Ca²⁺ signalling in USMCs within intact urethral muscles using a genetically encoded Ca²⁺ sensor, GCaMP3, expressed selectively in USMCs. USMCs fired spontaneous intracellular Ca²⁺ waves that did not propagate cell‐to‐cell across muscle bundles. Ca²⁺ waves increased dramatically in response to the α1 adrenoceptor agonist phenylephrine (10 μm) and to ATP (10 μm). Ca²⁺ waves were inhibited by the nitric oxide donor DEA NONOate (10 μm). Ca²⁺ influx and release from sarcoplasmic reticulum stores contributed to Ca²⁺ waves, as Ca²⁺ free bathing solution and blocking the sarcoplasmic Ca²⁺‐ATPase abolished activity. Intracellular Ca²⁺ release involved cooperation between ryanadine receptors and inositol trisphosphate receptors, as tetracaine and ryanodine (100 μm) and xestospongin C (1 μm) reduced Ca²⁺ waves. Ca²⁺ waves were insensitive to L‐type Ca²⁺ channel modulators nifedipine (1 μm), nicardipine (1 μm), isradipine (1 μm) and FPL 64176 (1 μm), and were unaffected by the T‐type Ca²⁺ channel antagonists NNC‐550396 (1 μm) and TTA‐A2 (1 μm). Ca²⁺ waves were reduced by the store operated Ca²⁺ entry blocker SKF 96365 (10 μm) and by an Orai antagonist, GSK‐7975A (1 μm). The latter also reduced urethral contractions induced by phenylephrine, suggesting that Orai can function effectively as a receptor‐operated channel. In conclusion, Ca²⁺ waves in mouse USMCs are a source of Ca²⁺ for excitation–contraction coupling in urethral muscles.

Key points

• Contraction of urethral smooth muscle cells (USMCs) contributes to urinary continence. Ca²⁺ signalling in USMCs was investigated in intact urethral muscles using a genetically encoded Ca²⁺ sensor, GCaMP3, expressed selectively in USMCs.

• USMCs were spontaneously active in situ, firing intracellular Ca²⁺ waves that were asynchronous at different sites within cells and between adjacent cells.

• Spontaneous Ca²⁺ waves in USMCs were myogenic but enhanced by adrenergic or purinergic agonists and decreased by nitric oxide.

• Ca²⁺ waves arose from inositol trisphosphate type 1 receptors and ryanodine receptors, and Ca²⁺ influx by store‐operated calcium entry was required to maintain Ca²⁺ release events.

• Ca²⁺ release and development of Ca²⁺ waves appear to be the primary source of Ca²⁺ for excitation–contraction coupling in the mouse urethra, and no evidence was found that voltage‐dependent Ca²⁺ entry via L‐type or T‐type channels was required for responses to α adrenergic responses.

Introduction

In mammals there is a reciprocal relationship between the contractile state of the bladder and urethra. The bladder suppresses muscle excitability as it fills with urine to accommodate the increase in volume, and during this period the urethra remains contracted to prevent leakage of urine and to maintain urinary continence (Brading, 1999). At the onset of micturition, the urethra relaxes and the bladder contracts to expel the stored urine (Brading, 1999; Hill, 2015).

A major factor closing the urethra during bladder filling is the summed contractions of urethral smooth muscle cells (USMCs), generating tone in a variety of species (Thornbury et al. 1992; Bridgewater et al. 1993; Bradley et al. 2004). Despite abundant evidence that USMCs, which make up the internal urethral sphincter, contribute to urethral closure in mammals (Tanagho & Smith, 1966; Donker et al. 1972; Sogbein et al. 1984; Conte et al. 1991; Bridgewater et al. 1993; Greenland et al. 1996; Janokowski et al. 2006) and the need for new pharmacological targets for the internal urethral sphincter to treat incontinence disorders (Lemack et al. 2007; Delancey et al. 2008; DeLancey, 2010; Staskin et al. 2011; Kedia et al. 2013; Alexandre et al. 2016; Kirschner‐Hermanns et al. 2016; Hokanson et al. 2017), there have been few studies on the control of USMC contractions performed on cells in situ. Isolated USMCs generate current responses and Ca²⁺ signals to depolarization or agonists, but these isolated cells are rarely rhythmically active (Sergeant et al. 2000; Hollywood et al. 2003; Bradley et al. 2004; Brading, 2006; Kyle, 2014; Drumm et al. 2014c).

In contrast to isolated USMCs, intact urethral tissues display spontaneous electrical activity. Studies on rabbits (Hashitani et al. 1996; Bradley et al. 2004) and guinea‐pigs (Hashitani & Edwards, 1999) showed that intact urethral muscles exhibit spontaneous electrical activity in the form of small amplitude spontaneous transient depolarizations and large amplitude slow waves. Imaging of rabbit urethral smooth muscles (USMs) has confirmed the spontaneous activity of these muscles. Studies by Hashitani et al. (2006) and Hashitani & Suzuki (2007) noted spontaneous Ca²⁺ oscillations in USMCs in situ, but the characteristics of these Ca²⁺ signals and their intracellular and extracellular sources were not examined in detail.

While the spontaneous Ca²⁺ activity of urethral interstitial cells has been examined (Drumm et al. 2014a), the nature of Ca²⁺ signalling in USMCs is mostly unknown. However, due to the importance of USMCs in generating urethral closure and the lack of effective, non‐surgical treatments for incontinence disorders (Miller et al. 1998; Thaker & Sharma, 2013), there is a need for a better understanding of the mechanisms regulating the contractile activity of these cells and to develop new targets for treatments of incontinence. In this study, we investigated Ca²⁺ signalling in the mouse urethra using Ca²⁺ imaging techniques and mice with USMC‐specific expression of the genetically encoded Ca²⁺ sensor GCaMP3. Imaging with optogenetic approaches provides important benefits over the use of membrane‐permeable Ca²⁺ dyes (Hashitani et al. 2006; Hashitani & Suzuki, 2007) including reduced photobleaching, use of high power magnification to record the details of subcellular Ca²⁺ signals, improved signal‐to‐noise characteristics, and loss of contaminating signals from cells other than the cells of interest in complex tissues. In contrast to isolated cells, USMCs in intact muscles were spontaneously active and exhibited intracellular Ca²⁺ waves. We examined how these spontaneous Ca²⁺ signals were modulated by neurotransmitters that affect urethral contraction and explored the intracellular and extracellular Ca²⁺ sources responsible for their generation in situ.

Methods

Ethical approval

All animals used and the protocols carried out in this study were in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. All procedures were approved by the Institutional Animal Use and Care Committee at the University of Nevada, Reno. We confirm that our experiments conform to the principles and regulations as described by Grundy (2015).

Animals

B6;129S‐Gt(ROSA)26Sor^{tm38(CAG‐GCaMP3)Hze}/J (GCaMP3 mice) and male wild type C57BL/6 mice were purchased from the Jackson Laboratory (Bar Harbor, ME, USA). Male mice expressing an inducible Cre Recombinase driven by the smooth muscle heavy chain promoter (SmMHC‐Cre mice) were generated by Dr Stefan Offermann (Max‐ Planck‐ Institute, Germany) and gifted to our lab from Dr James Stull of Southwestern University (TX, USA). SmMHC/Cre/eGFP (SMC‐eGFP), mice expressing a fluorescent eGFP molecule in cells expressing smooth muscle heavy chain mice were donated by Dr Michael Kotlikoff of Cornell University (NY, USA).

Tamoxifen preparation and administration

GCaMP3 mice were crossbred with SmMHC‐Cre mice and we refer to the resultant offspring mice as SmMHC‐Cre‐GCaMP3 mice throughout the manuscript. The mice were injected with tamoxifen at ages 6–8 weeks to induce Cre Recombinase and subsequent GCaMP3 expression in USMCs. Tamoxifen (Sigma, St. Louis, MO, USA, T5648; 80 mg) was dissolved in 800 μl of ethanol (Pharmco‐Aaper, 200 proof, absolute, anhydrous) by vortexing for 20 min. Then, 3.2 ml of safflower (generic) was added to create solutions of 20 mg ml⁻¹, which were sonicated for 30 min prior to injection. Mice were injected i.p. with 0.1 ml of tamoxifen solution (2 mg tamoxifen) for three consecutive days. Mice were used 10 days after the first injection and the subsequent expression of GCaMP3 was confirmed by genotyping.

Tissue preparation

Mice were anaesthetized by inhalation of isoflurane (Baxter, Deerfield, IL, USA) and then killed by cervical dislocation before being used for experimentation. The abdomen of the mice was opened, and the bladder and proximal urethra were removed and placed in Krebs–Ringer–bicarbonate solution (KRB). The urethra was opened along a longitudinal incision and the urothelium was carefully removed by sharp dissection.

Calcium imaging

The muscular coat of the urethra was pinned to the base of a 5 ml volume, 60 mm diameter Sylgard‐coated dish. The urethra preparation was continuously perfused with warmed KRB solution at 37°C for an equilibration period of 1 h before experiments were initiated. Following the equilibration period, Ca²⁺ imaging of circularly orientated USMCs was performed with an Eclipse E600FN microscope (Nikon Inc., Melville, NY, USA) equipped with a 60× 1.0 CFI Fluor lens (Nikon). GCaMP3 was excited at 488 nm. Fluorescence was captured using a FITC HQ series fluorescence filter cube (Nikon). This cube had a 510 nm dichroic with a barrier filter of 535 nm, as previously described (Drumm et al. 2017). The pixel size using this acquisition configuration was 0.225 μm. Image sequences were collected at 33 frames s⁻¹ using a high‐speed EMCCD Camera (Andor iXon X₃; ANDOR Technology) and TILLvisION software (T.I.L.L. Photonics GmbH, Grafelfing, Germany). For pharmacological tests, a control period of 30 s was recorded and then KRB containing the appropriate concentration of drug was perfused into the bath for 12–15 min to allow full tissue penetration before another 30 s recording was acquired to ascertain the effects of the drug.

Analysis of Ca²⁺ transients

Analysis of Ca²⁺ transients in urethral cells was performed as previously described (Drumm et al. 2014b; Baker et al. 2016). Movies of in situ USMC Ca²⁺ activity were converted to a stack of TIFF (tagged image file format) images and imported into ImageJ (version1.40, National Institutes of Health, MD, USA, http://rsbweb.nih.gov/ij). Spatio‐temporal (ST) maps of Ca²⁺ activity in USMCs were generated by drawing a single pixel line along the mid axis of the cell and invoking the ‘reslice’ function in ImageJ; this generated an ST map which was then calibrated for distance and time. The basal Ca²⁺ fluorescence was acquired from regions of the cell that displayed the most uniform and least intense fluorescence (F ₀). The fluorescence value throughout the cell was then divided by the calculated F ₀ value to calibrate the ST map for the amplitude of Ca²⁺ events as F/F ₀. Ca²⁺ transient frequency in USMCs was expressed as the number of events fired per cell per minute (min⁻¹). The amplitude of Ca²⁺ transients was expressed as ΔF/F ₀, the duration of Ca²⁺ transients was expressed as full duration at half‐maximum amplitude (FDHM) and the spatial spread of Ca²⁺ transients was expressed as micrometres of cell propagated per Ca²⁺ transient. Ca²⁺ wave velocity was calculated by drawing a line along a propagating wave front on an ST map and invoking the ‘Wave Speed’ plugin in ImageJ to produce read‐outs of Ca²⁺ wave velocity. In experiments where drugs were applied to USMCs, quantification of Ca²⁺ transient parameters in control and in the presence of a drug would be analysed in three to five representative cells in a field of view (FOV) per animal used in experimentation. A series of time control experiments was performed in which a control period of 20 s of USMC Ca²⁺ activity was recorded and 15 min later the same FOV was recorded in the absence of any changes in experimental conditions or pharmacological agents for another 20 s. No significant differences were observed in the frequency (P = 0.51), amplitude (P = 0.99), duration (P = 0.96) or spatial spread (P = 0.54) of USMC Ca²⁺ waves recorded from the same cells in consecutive recordings (data not shown, no. of cells (c) = 12, no. of animals (n) = 5).

Cell isolation and fluorescence‐activated cell sorting (FACS)

USMCs were isolated from SMC‐eGFP⁺ mice for analysis of gene transcripts. Urethral tissues were obtained as described above in the ‘Tissue preparation’ section. The urothelium was removed and then the remaining muscle segments were cut into fine pieces and incubated in dispersion medium containing per millilitre of Ca²⁺‐free KRB solution (see ‘Drugs and solutions’ below): 3 mg collagenase (Sigma type 1A), 0.2 mg protease (Sigma type XXIV), 2 mg bovine serum albumin (Sigma) and 2 mg trypsin inhibitor (Sigma) for 10–15 min at 37°C. Tissues were then washed in Ca²⁺‐free KRB for 10 min to obtain isolated USMCs. USMCs were sorted by FACS (FACSAria II; Becton‐Dickinson) with an excitation laser (488 nm) and emission filter (530/30 nm). USMC sorting was carried out with a 130 μm nozzle at a sheath pressure of 12 p.s.i. and sort rate of 1000–3000 events s⁻¹. Live cells, gated on exclusion of Hoechst 33258 viability indicator (data not shown), were subsequently gated on GFP fluorescence intensity.

RNA extraction and quantitative PCR

Total RNA was isolated from unsorted urethral cells and sorted (purified) USMCs using an illustra RNAspin Mini RNA Isolation Kit (GE Healthcare), and First‐Strand cDNA was synthesized using SuperScript III (Invitrogen, Carlsbad, CA, USA), according to the manufacturer's instructions. The PCR primers used and their GenBank accession numbers are listed in Table 1. PCR was performed using GoTaq DNA Polymerase (Promega, Madison, WI, USA) and PCR products were analysed on 2% agarose gels and visualized by ethidium bromide. Quantitative PCR (qPCR) was performed with the same primers as PCR using SYBR green chemistry on the 7500 HT Real‐time PCR System (Applied Biosystems) and analysed as described previously (Baker et al. 2016; Drumm et al. 2017). Gene expression levels are described in relation to the expression of the housekeeping gene Gapdh ± standard deviation.

Table 1.

Gene primers used in qPCR analysis

Gene name	Sequence (sense primer on top)	GenBank accession no.
Gapdh	GCCGATGCCCCCATGTTTGTGA GGGTGGCAGTGATGGCATGGAC	NM_008084
Pdgfra	ATGACAGCAGGCAGGGCTTCAACG CGGCACAGGTCACCACGATCGTTT	NM_011058
Kit	CGCCTGCCGAAATGTATGACG GGTTCTCTGGGTTGGGGTTGC	NM_021099
Myh11	CTACACTGCGCAATACCACG TGCCAGGATCTCATAGCGTT	NM_013607
Uchl1	CGATGGAGATTAACCCCGAGATG TTTTCATGCTGGGCCGTGAG	NM_011670
Trpc1	CTGTGGATTATTGGGATGATTTGGTCA CACTTTGAGGGCAAAGGTTGCCAA	NM_011643
Trpc2	CTCCTCCTGGCTGGGCTTG GGTGAAGCTGAGCATGCTGGTG	NM_011644
Trpc3	CAGCCAACACGATATCAGATAAT AAGTTCATAACGAAGGCTGGAGATATC	NM_019510
Trpc4	CCATGACTGGTGGAATCTAAT GGTGCCACATGTCCCATGATT	NM_016984
Trpc5	CATCTCCCTGGTAGTGCTGCTG CACCTTCATCAAAGTAACTCATCCAGAG	NM_009428
Trpc6	GTGGTGCTTGCTGTTGTCATTG CTTCAAATCTGTCAGCTGCATTCATGAC	NM_013838
Tpsab1	TGTGGGACCGGATGTTGCTGAC AGTCGGGGTGTGTGATGATCTGG	NM_031187
Chrm1	GCAACGCCTCTGTCATGAAT AACGAAGGAAACCAACCACG	NM_001112697
Chrm2	GGTGTCTCCCAGTCTAGTGCAAGG ATGTCTGCCTAGAGTTGTCATCTTTGGA	NM_203491
Chrm3	TGTGGCCAGCAATGCTTCTGTCATGA CCACAGGACAAAGGAGATGACCCAAG	NM_033269
Chrm4	ATGTGTGACTGCCATCGAGA GGATGAAGGCCAGCAGAATG	NM_007699
Chrm5	GTCAAAGAGAGGAAAGCGGC TAGCACAGCCAGTAACCCAA	NM_205783
Adra1a	GCCCATTGGGTCCTTCTTCCCG ACTCCTGGCTGGAGCATGGG	NM_013461
Adra1b	AGTGCTCGGCTACTGGGTGC TGACCAGCGTGGGGTACTGC	NM_007416
Adra1d	CGTCCTGCCTCTGGGTTCTCT AGCAGGGGTAGATGAGCGGGT	NM_013460
Npr1	ACATGGTGGTATCAGGGCTC GTGTGAATTCCAATGCGCAAG	NM_008727
Npr2	GAAGGTGCGGAATGGTCAGA CCTTCCTTGTTAAACCGGCG	NM_173788
Prkg1a	AACTCCACAAATGCCAGTCG GGTCGTGGAAGGACCTGTAG	NM_001013833
Prkg1b	GATTTACAGTATGCGCTCCAGG AGCTTCTGGATCAGTTCGTC	NM_011160
P2ry1	ACCGAGGTGCCTTGGTCGGT CCGGTCTTGGTCAGGGCACA	NM_008772
P2ry2	AGCACCATCAATGGCACCTGGGA CACGACGTTCAGGCACAACCC	NM_008773
P2ry6	ACCGCACTGTGTGCTACGAC CGGCGGGCCATGCGACAATA	NM_183168
P2ry12	GGCTGACGTCACTGAACGCCTG TCTCTTCGCTTGGTTCGCCACC	NM_027571
P2ry13	CAACACCACTGGGATGCAGGGCT GCTGGGGATGTGGACGAACACC	NM_028808
P2ry14	ACCCTCCAAACCAGCCCTGC AGGCCCATGAGAAAGTCAGCCA	NM_133200
P2rx1	CCACGCTTCAAGGTCAACAGGC GATACCAACCACCCCGCCCTTC	NM_008771
P2rx2	TTGTGCATGGACAGGCAGGGAAA ACTTGAGGGGTGCCTTGGGGT	NM_153400
P2rx3	CTTCCGTGGGAGTGGGGACTGTT GCCCCCGAGTCTGTGGACTGC	NM_145526
P2rx4	TGACGCTGGTGTGCCAACGC GGCAGAAGGGATCCGTCCGC	NM_011026
P2rx5	TCCGTCTCAGGGGGAGAACGT TCCCCAGCGTGACAGTCGGT	NM_033321
P2rx6	GAGAGTGGTGCTGTGCCCAGGAAG CGGAACACTGGGCAGTACGGA	NM_011028
Itpr1	TTATCAGCACCTTAGGCTTGGTTGA ATCTGTAGTGCTGTTGGCCCCG	NM_010585
Itpr2	CAACCCAGGCTGCAAAGAGGTGA AGGTCGTCCGAAGGAAAATGTGCT	NM_019923
Itpr3	CTTTGGGGCTGGTGGATGACCGTTG TGCAGCTTCTGCAGCAATACCACA	NM_080553
Ryr1	GTCAGTTCGAGCCCTGCAGGAG GCAACTCAGGTACATACGACTGTGT	NM_009109
Ryr2	TCCCCCGGACCTGTCTATCTGC GGCCTCCACCTTGAGCAGTCTTC	NM_023868
Ryr3	TCCTCGTCAGTGTGTCCTCTGAAA CATGGCCACCGAGTAAGTATCCTTC	NM_177652
Orai1	GTTCACTTCTACCGCTCCCT GTGCCCGGTGTTAGAGAATG	NM_175423
Orai2	CACAAGGGCATGGATTACCG CCCTGCTCAGGTAGAGCTTC	NM_178751
Orai3	GGCTGAAGTTGTTCTGGTGG TGGAAGGCTGTTGTGATGTG	NM_198424

Contraction experiments

Ring preparations of mouse urethra (urothelium intact) were attached with tied sutures to a Gould strain gauge and a stable mount and immersed in tissue baths containing 15 ml of oxygenated KRB solution maintained at 37°C. Urethra preparations were initially stretched to 0.25 g of tension followed by a 60 min equilibration period. Viability of tissues was evaluated by their ability to generate reproducible contractile responses to a bolus application of phenylephrine (PE). Contractile data were recorded using AcqKnowledge software and urethral contractions were normalized as a percentage of a bolus PE contraction.

Drugs and solutions

Tissues were perfused with KRB containing (mmol l⁻¹): NaCl, 120.35; KCl, 5.9; NaHCO₃, 15.5; NaH₂PO₄, 1.2; MgCl₂, 1.2; CaCl₂, 2.5; and glucose, 11.5. KRB was bubbled with a mixture of 97% O₂–3% CO₂ and warmed to 37 ± 0.2°C. For experiments with 0 mm [Ca²⁺]_o, no CaCl₂ was added to the KRB and 0.5 mm EGTA was added. Note that in cell isolation protocols, EGTA was not added to Ca²⁺ free solutions for cell dispersal. ATP, caffeine, carbachol (CCh), Diethylamine NONOate (DEA NONOate), tetracaine, phenylephrine, cyclopiazonic acid (CPA), nifedipine and nicardipine; Sigma‐Aldrich (St Louis, MO, USA). FPL 64176, isradipine, ryanodine, tetrodotoxin (TTX), thapsigargin and SKF 96365; Tocris Bioscience (Ellisville, MO, USA). NNC‐550396 and TTA‐A2; Alomone Labs (Jerusalem, Israel). GSK‐7975A; Aobious (Gloucester, MA, USA). Xestospongin C; Cayman Chemical (Ann Arbor, MI, USA). All drugs were dissolved in the manufacturer‐recommended solvent and then diluted into KRB solution for use in experiments.

Statistics

Unless otherwise stated in the text, data are represented as mean ± standard error of the mean (SEM). Statistical analyses were performed using Student's t tests or an ANOVA with a multiple comparison Tukey post hoc test as appropriate. In all statistical analyses, P < 0.05 was taken as significant. When describing data, n refers to the number of animals used in that dataset while c refers to the number of cells analysed in that same data set.

Results

Nature of USMC Ca²⁺ activity

In situ Ca²⁺ imaging of the urethras of SmMHC‐Cre‐GCaMP3 mice revealed that USMCs within smooth muscle bundles fired spontaneous Ca²⁺ events that manifested as intracellular Ca²⁺ waves that arose from discrete initiation sites (Fig. 1 A–C) and propagated in either a uni‐ or bi‐directional fashion (Fig. 1 D–F). Ca²⁺ waves in USMCs were almost exclusively intracellular, with little to no intercellular propagation within USMC bundles (n = 31). Analysis of USMC intracellular Ca²⁺ waves was performed by spatio‐temporal (ST) mapping of Ca²⁺ activity as described in the Methods. An ST map of the Ca²⁺ waves from the cell highlighted in Fig. 1 is shown in Fig. 2 A, with the corresponding fluorescence trace and 3‐D plots from the ST map in Fig. 2 B–D. Using this analysis (Drumm et al. 2015; Baker et al. 2016) we tabulated USMC Ca²⁺ waves in terms of their firing frequency, amplitude, duration, spatial spread, velocity and number of sites of initiation for Ca²⁺ events (i.e. firing sites) per USMC. Across 301 cells from 31 SmMHC‐Cre‐GCaMP3 mice, 5500 Ca²⁺ events were tabulated, and these data were plotted as frequency histograms in Fig. 2 E (and see also Table 2). In many instances Ca²⁺ wave velocity could not be accurately measured due to the lack of a uniform, uni‐directional propagating wave front. Thus, the histogram in Fig. 2 Ee represents Ca²⁺ wave velocities of 726 Ca²⁺ waves.

Figure 1. Spontaneous intracellular Ca²⁺ waves in USMCs.

A, representative image of a USMC bundle imaged from a SmMHC‐Cre‐GCaMP3 mouse urethra with a 60× objective. The white arrow highlights a cell of interest. B–F, time‐lapse montage of spontaneous intracellular Ca²⁺ activity in USMCs in situ with a bi‐directional Ca²⁺ wave in the highlighted cell emphasized throughout. The initiation site of this Ca²⁺ wave is indicated in F by the white asterisks. Panels are colour coded as F/F ₀ as shown in B.

Figure 2. Characterization of USMC intracellular Ca²⁺ events.

A, representative spatio‐temporal (ST) map of spontaneous intracellular Ca²⁺ waves recorded in situ. The ST map is calibrated for amplitude as F/F ₀ and colour coded accordingly. B, x–y plot of the Ca²⁺ activity represented in A. C, 3‐D plot representation of the ST map shown in A. D, 3‐D plot profile of 5 representative Ca²⁺ waves from the ST map in A. E, summary histograms showing the distribution of values for USMC Ca²⁺ event frequency (a), amplitude (b), duration (c), spatial spread (d), velocity (e) and number of firing sites (f) (c = 301, n = 31).

Table 2.

Quantification of USMC Ca²⁺ wave parameters

Ca2+ wave parameter	Min	Max	Mean	Median	SEM	SD
Frequency (min−1)	2	224	51	36	2.3	39.6
Amplitude Δ(F/F 0)	0.01	1.4	0.2	0.15	0.002	0.1
FDHM (ms)	60	930	226	210	0.9	67.7
Spatial spread (μm)	0.4	117.3	25.2	18.9	0.3	19.6
Velocity (μm s−1)	31	601.3	108.5	88.8	2.4	65
No. of firing sites per USMC	1	8	3.7	3	0.1	1.7

As illustrated by the data in Fig. 2 Ef, USMCs often contained multiple firing sites, ranging from a single site to as many as eight (mean 3.7 ± 0.1; Fig. 2 Ef, c = 301, n = 31). Figure 3 A shows an ST map of a USMC firing Ca²⁺ transients from multiple sites of origin, four of which are indicated by coloured arrows to the left of the ST map and the activity that occurred in these regions is plotted as colour‐coded traces in Fig. 3 B. These traces illustrate that while Ca²⁺ waves initiated at one firing site can propagate to another region of initiation, there was no apparent relationship between the timing of Ca²⁺ transient firing at different sites. Instead, Ca²⁺ transients in USMCs appeared to fire in a stochastic manner with little to no temporal coordination.

Figure 3. Asynchronous firing of USMC intracellular Ca²⁺ events.

A, representative ST map of USMC Ca²⁺ activity showing multiple Ca²⁺ firing sites along the length of the cell, 4 of which are indicated by the black, blue, red and green arrows. B, colour coded plot profiles of the Ca²⁺ activity at each of the corresponding colour coded Ca²⁺ firing sites indicated in A. C, image from a FOV of USMCs in situ recorded with a 60× objective showing 2 adjacent cells illustrated by the green and red ROIs. The activity of these adjacent cells is plotted below the image. D, ST maps of cell 1 and cell 2 represented in C which have been uniformly coloured to show all Ca²⁺ activity in the cell as either red (cell 1) or green (cell 2). These ST maps are then merged at the bottom of the panel.

The stochastic pattern of firing was also apparent between adjacent cells. Figure 3 C shows a FOV with two adjacent highlighted USMCs (as indicated by the green and red regions of interest (ROIs)). The activity of cells 1 and 2 over 30 s is plotted as fluorescence traces below the FOV and shows that there is little or no correlation between Ca²⁺ transients in the two cells. This was further examined by comparing ST maps from the two cells. Figure 3 D shows ST maps from the highlighted cells where all Ca²⁺ signals were thresholded to be a uniform red (cell 1) or green colour (cell 2). When these two ST maps were merged, it was apparent that there was little overlap, and using the same analysis in cells from five different animals there was only 14 ± 1.2% overlap of Ca²⁺ signals in adjacent USMCs. These data indicate that the intracellular activity of one USMC was largely independent of the activity occurring in adjacent USMCs. This asynchronous activity was also observed in relation to urethral contractions. As shown in Movie S1 in the online Supporting information, the firing of intracellular Ca²⁺ waves in USMCs was associated with small contractions (60× objectives were required to resolve these movements) of individual USMCs in muscle bundles. These contractions, like the intracellular Ca²⁺ waves occurring asynchronously among cells, did not spread cell‐to‐cell across the muscle bundles. These observations suggest that cellular contractions, occurring independently and asynchronously in many cells within bundles and across many bundles within the tissue, summate to generate tonic contractions of the urethra.

Neuronal control of USMC Ca²⁺ activity

We next sought to evaluate whether spontaneous Ca²⁺ waves observed in USMCs are modulated by neural inputs. As shown in Fig. 4 A, tetrodotoxin (TTX; 1 μm) did not alter Ca²⁺ events in USMCs. TTX did not affect frequency (Fig. 4 Ba, P = 0.76, c = 18, n = 6), amplitude (Fig. 4 Bb, P = 0.06, c = 18, n = 6), duration (Fig. 4 Bc, P = 0.63, c = 18, n = 6), or spatial spread (Fig. 4 Bd, P = 0.36, c = 18, n = 6) of USMC Ca²⁺ waves, suggesting that the basal level Ca²⁺ events are myogenic in origin and are not under either basal inhibitory or excitatory neural regulation in urethral muscles in situ.

Figure 4. The effects of TTX and carbachol on USMC Ca²⁺ transients.

Aa and b, representative ST maps of Ca²⁺ transient firing within a single USMC recorded in situ during control conditions (a) and after incubation with 1 μm TTX (b). Ba–d, summary data showing the effect of 1 μm TTX on USMC Ca²⁺ transient frequency (a), amplitude (b), duration (c) and spatial spread (d) (c = 18, n = 6). C, relative expression of cellular identification genes in FACS sorted and unsorted urethral cells from SMC‐eGFP mice determined by qPCR analysis, normalized to Gapdh expression (n = 4). Genes examined are Kit (tyrosine kinase receptor, found in interstitial cells of Cajal), Pdgfra found in PDGFRα⁺ interstitial cells, Myh11 (smooth muscle myosin), Uch11 (neural marker encoding PGP 9.5) and Tpsab1 (mast cell tryptase). D, relative expression of muscarinic receptor genes in FACS sorted and unsorted urethral cells from SMC‐eGFP mice determined by qPCR analysis, normalized to Gapdh expression (n = 4). Ea and b, representative ST maps of Ca²⁺ transient firing within a single USMC recorded in situ during control conditions (a) and after incubation with 10 μm CCh (b). Fa–d, summary data showing the effect of 10 μm CCh on USMC Ca²⁺ transient frequency (a), amplitude (b), duration (c) and spatial spread (d) (c = 13, n = 4). ns, not significant.

We next tested the effects of neurotransmitter agonists on urethral Ca²⁺ waves. These experiments were performed in conjunction with PCR experiments to detect relative expression of genes encoding receptors and molecules pertinent to USM neurotransmission. Individual USMCs were isolated from SMC‐eGFP⁺ mice and purified using FACS, as described in the Methods. The sorted cells were evaluated for purity by quantifying the expression of genes encoding cell‐specific markers, such as Kit (interstitial cells of Cajal found in the urethra; McHale et al. 2006), Pdgfra encoding platelet‐derived growth factor receptor α (marker of fibroblast‐like cells in visceral smooth muscles, Kurahashi et al. 2011; Koh et al. 2012; Peri et al. 2015), Myh11 (myosin heavy chain; smooth muscle cells), Uch11 encoding PGP 9.5 (nerve cells) and Tpsab1 (mast cell tryptase). Enriched USMCs isolated from SMC‐eGFP⁺ mice showed minimal expression of these genes with the exception of the smooth muscle marker Myh11 (Fig. 4 C), and the expression of this gene was enriched in the sorted USMCs, as compared to unsorted cells (1.72 ± 0.1 in sorted USMCs (normalized to Gapdh expression) vs. 0.4 ± 0.01 in unsorted cells, n = 4). Thus, FACS provided a means of determining which neurotransmitter receptor subtypes are specific to USMCs. qPCR showed that USMCs expressed genes encoding muscarinic receptor subtypes for Chrm2 (M₂), Chrm3 (M₃) and Chrm5 (M₅) (Fig. 4 D). However, in spite of muscarinic receptor expression, effects of exogenous carbachol (CCh; 10 μm; Fig. 4 E) on USMC Ca²⁺ wave frequency (Fig. 4 Fa, P > 0.99, c = 13, n = 4), amplitude (Fig. 4 Fb, P = 0.39, c = 13, n = 4), duration (Fig. 4 Fc, P = 0.91, c = 13, n = 4), or spatial spread (Fig. 4 Fd, P = 0.36, c = 13, n = 4) were not resolved.

The role of the sympathetic, adrenergic pathway in modulating USMC Ca²⁺ waves was also characterized. This is reported to be the primary excitatory neural pathway mediating contractions of USMs, and sympathetic effects have been reported to be due to expression and binding of α1 adrenoceptors (Ek et al. 1977; Andersson et al. 1984; Ito & Kimoto, 1985; Kimoto et al. 1987; Mattiasson et al. 1989; Chen & Brading, 1992; Andersson, 2001; Kedia et al. 2013). Genes encoding all three α1 adrenoceptors were expressed in sorted USMCs (Fig. 5 A). The α1 adrenoceptor agonist phenylephrine (PE; 10 μm) caused dramatic increases in the firing of USMC Ca²⁺ waves (Fig. 5 B and C). PE nearly doubled the frequency of Ca²⁺ waves from 46.9 ± 5.7 events min⁻¹ to 88.8 ± 9.6 events min⁻¹ (Fig. 5 Da, P < 0.0001, c = 34, n = 10). PE also increased the duration of Ca²⁺ events from 225 ± 7.6 ms to 263 ± 11.5 (Fig. 5 Dc, P = 0.0005, c = 34, n = 10) and the spatial spread of Ca²⁺ events, increasing propagation distance from 21.7 ± 2 μm in control to 36.5 ± 4.9 μm (Fig. 5 Dd, P = 0.0025, c = 34, n = 10). However, PE did not affect Ca²⁺ wave amplitude significantly (Fig. 5 Db, P = 0.5, c = 34, n = 10). As shown in Movies S2 and S3 in the online Supporting information, PE did not synchronize firing of USMC Ca²⁺ waves or generate what might be referred to as a global rise in Ca²⁺. Instead, PE increased the intracellular activity of all the cells in the FOV by increasing the frequency, duration and spatial spread of intracellular Ca²⁺ waves, and the asynchronous nature of firing of Ca²⁺ waves was retained in all cells in the FOV.

Figure 5. Phenylephrine increases USMC Ca²⁺ transient firing.

A, relative expression of adrenergic α1 receptor genes in FACS sorted and unsorted urethral cells from SMC‐eGFP mice determined by qPCR analysis, normalized to Gapdh expression (n = 4). Ba and b, representative ST maps of Ca²⁺ transient firing within a single USMC recorded in situ during control conditions (a) and after incubation with 10 μm phenylephrine (b). C, 3‐D plots of the ST maps shown in A. Da–d, summary data showing the effect of 10 μm phenylephrine on USMC Ca²⁺ transient frequency (a), amplitude (b), duration (c) and spatial spread (d) (c = 34, n = 10). ns, not significant, ^** p < 0.01, ^*** p < 0.001, ^**** p < 0.0001.

We also investigated inhibitory neural regulation of USMs, which has been reported to be mediated by the nitric oxide (NO)–guanylate cyclase (GC)–protein kinase G (PKG) pathway (Thornbury et al. 1992; Bridgewater et al. 1993; Dokita et al. 1994; Zygmunt et al. 1995; Waldeck et al. 1998; Costa et al. 2001; Kedia et al. 2013). USMCs expressed genes required for the NO–GC–PKG pathway (Fig. 6 A), and sorted USMCs showed enrichment in transcripts of genes encoding GCα (Npr1; 0.1 ± 0.02 in sorted USMCs compared to 0.01 ± 0.001 in unsorted cells) and GCβ (Npr2; 0.08 ± 0.01 in sorted USMCs compared to 0.04 ± 0.005 in unsorted cells). Application of the NO donor (DEA NONOate; 10 μm) nearly abolished Ca²⁺ transients in USMCs (Fig. 6 B and C), decreasing USMC firing frequency by 93% (from 55.4 ± 8.1 min⁻¹ in control to 3.7 ± 1.8 min⁻¹ in the presence of DEA NONOate; Fig. 6 Da, P < 0.0001, c = 19, n = 6). DEA NONOate decreased Ca²⁺ wave amplitude from 0.27 ± 0.06 ΔF/F ₀ in control to 0.03 ± 0.01 ΔF/F ₀ in DEA NONOate (Fig. 6 Db, P = 0.0012, c = 19, n = 6), Ca²⁺ event duration was decreased from 213 ± 11.5 ms in control to 58 ± 23.4 ms in the presence of DEA NONOate (Fig. 6 Dc, P < 0.0001, c = 19, n = 6), and the propagation spread of Ca²⁺ transients was decreased from 34.1 ± 3.9 μm to 5.3 ± 2.3 μm (Fig. 6 Dd, P < 0.0001, c = 19, n = 6).

Figure 6. Nitric oxide decreases USMC Ca²⁺ transient firing.

A, relative expression of guanylate cyclase and protein kinase G genes in FACS sorted and unsorted urethral cells from SMC‐eGFP mice determined by qPCR analysis, normalized to Gapdh expression (n = 4). Ba and b, representative ST maps of Ca²⁺ transient firing within a single USMC recorded in situ during control conditions (a) and after incubation with 10 μm DEA NONOate (b). C, 3‐D plots of the ST maps shown in A. Da–d, summary data showing the effect of 10 μm DEA NONOate on USMC Ca²⁺ transient frequency (a), amplitude (b), duration (c) and spatial spread (d) (c = 19, n = 6). ^** p < 0.01, ^**** p < 0.0001.

We also explored the actions of purinergic substances on USMC Ca²⁺ waves. Sorted USMCs showed expression of P2X receptor genes P2rx1–4 and P2rx6 (Fig. 7 A). There was enriched expression in USMCs of P2rx1 (0.01 ± 0.002 sorted USMCs vs. 0.0015 ± 0.0002 unsorted cells), P2rx2 (0.003 ± 0.0004 sorted USMCs vs. 0.0009 ± 0.0004 unsorted cells) and P2rx6 (0.007 ± 0.001 sorted USMCs vs. 0.002 ± 0.0004 unsorted cells). The only P2Y receptor gene resolved in USMCs was P2ry14 (Fig. 7 B; 0.28 ± 0.05 sorted USMCs vs. 0.16 ± 0.018 unsorted cells), while unsorted cells showed expression of genes encoding P2ry1, 2, 6, 12, 13 and 14 (Fig. 7 B).

Figure 7. The effects of purines on USMC Ca²⁺ transients.

A, relative expression of P2X receptor genes in FACS sorted and unsorted urethral cells from SMC‐eGFP mice determined by qPCR analysis, normalized to Gapdh expression (n = 4). B, relative expression of P2Y receptor genes in FACS sorted and unsorted urethral cells from SMC‐eGFP mice determined by qPCR analysis, normalized to Gapdh expression (n = 4). C, representative ST map showing the effect of 10 μm ATP on Ca²⁺ transient firing in USMCs. Da–d, summary data showing the effect of 10 μm ATP on USMC Ca²⁺ transient frequency (a), amplitude (b), duration (c) and spatial spread (d) (c = 5, n = 3). ns, not significant, ^** p < 0.01, ^**** p < 0.0001.

At resting tone, purines such as adenosine triphosphate (ATP) exert excitatory effects on urethral muscles (Callahan & Creed, 1981; Bradley et al. 2010), and we hypothesized, due to expression of several P2X receptor paralogues, that purines would increase Ca²⁺ events in USMCs similar to PE. We found that ATP exerted excitatory effects on USMC Ca²⁺ waves (Fig. 7 C). ATP significantly increased USMC Ca²⁺ wave frequency from 36 ± 10.4 min⁻¹ to 91.2 ± 11.8 min⁻¹ (Fig. 7 Da, P < 0.0001, c = 5, n = 3). ATP also increased the amplitude of Ca²⁺ waves from 0.06 ± 0.002 ΔF/F ₀ to 0.16 ± 0.01 ΔF/F ₀ (Fig. 7 Db, P = 0.0012, c = 5, n = 3); however, the duration and spatial spread of Ca²⁺ transients were not changed significantly by ATP (Fig. 7 Dc and d, P = 0.25, P = 0.24, c = 5, n = 3).

Intracellular sources of Ca²⁺ contributing to USMC Ca²⁺ waves

Ca²⁺ waves in USMCs rely on functional sarcoplasmic reticulum (SR) stores, as sarco(endo)plasmic reticulum Ca²⁺‐ATPase (SERCA) antagonists thapsigargin (10 μm; Fig. 8 A and B, c = 4, n = 3) or cyclopiazonic acid (CPA; 10 μm; Fig. 8 C and D, c = 15, n = 5) abolished Ca²⁺ waves in these cells. Ca²⁺ release from the SR involves the opening of SR Ca²⁺ channels, such as inositol trisphosphate receptors (IP₃Rs), ryanodine receptors (RyRs) or a combination of the two (Berridge et al. 2000). Functional RyRs are tetrameric proteins of which there are three subtypes, RyR1, RyR2 and RyR3 (Amundson & Clapham, 1993), and there are also three IP₃R subtypes, IP₃R1, IP₃R2 and IP₃R3 (Berridge et al. 2000). Analysis of transcripts from sorted USMCs and unsorted urethral cells by qPCR showed that the dominant gene encoding SR Ca²⁺ release channels in urethral cells was Itpr1 (encodes IP₃R1), but expression of Itpr2 was also observed (Fig. 8 E). USMCs expressed all three gene paralogues encoding RyRs (Ryr1–3; Fig. 8 F). The data also showed enrichment of Itpr1, Ryr2 and Ryr3 in USMCs in comparison to the unsorted cell population.

Figure 8. Ca²⁺ transients in USMCs rely on functional SR stores.

Aa and b, representative ST maps showing the effect of 10 μm thapsigargin on Ca²⁺ transient firing in USMCs. Ba–d, summary data showing the effect of 10 μm thapsigargin on USMC Ca²⁺ transient frequency (a), amplitude (b), duration (c) and spatial spread (d) (c = 4, n = 3). Ca and b, representative ST maps showing the effect of 10 μm CPA on Ca²⁺ transient firing in USMCs. Da–d, summary data showing the effect of 10 μm CPA on USMC Ca²⁺ transient frequency (a), amplitude (b), duration (c) and spatial spread (d) (c = 13, n = 5). E, relative expression of IP₃R genes in FACS sorted and unsorted urethral cells from SMC‐eGFP mice determined by qPCR analysis, normalized to Gapdh expression (n = 4). F, relative expression of RyR genes in FACS sorted and unsorted urethral cells from SMC‐eGFP mice determined by qPCR analysis, normalized to Gapdh expression (n = 4). ^* p < 0.05, ^** p < 0.01, ^**** p < 0.0001.

Figure 9 A shows that blocking IP₃Rs with 100 μm 2‐aminoethyl‐diphenylborinate (2‐APB) reduced Ca²⁺ wave activity in USMCs and this was quantified in Fig. 9 B. 2‐APB reduced Ca²⁺ wave frequency from 45.2 ± 8.4 min⁻¹ in control to 2.9 ± 1.3 min⁻¹ in the presence of the IP₃R antagonist (Fig. 9 Ba, P < 0.0001, c = 19, n = 6). The amplitude of Ca²⁺ waves was reduced by 2‐APB from 0.16 ± 0.03 ΔF/F ₀ to 0.04 ± 0.015 ΔF/F ₀ (Fig. 9 Bb, P = 0.003, c = 19, n = 6). Ca²⁺ wave duration was reduced from 230 ± 6.2 ms to 66.1 ± 23.5 ms (Fig. 9 Bc, P < 0.0001, c = 19, n = 6) and the spatial spread of Ca²⁺ waves was reduced from 18.75 ± 1.8 μm to 7 ± 2.8 μm (Fig. 9 Bd, P = 0.0008, c = 19, n = 6). However, it should be noted that non‐specific effects on Ca²⁺ handling other than blocking IP₃Rs have been attributed to 2‐APB (Peppiatt et al. 2003), and therefore we also tested another IP₃R antagonist. Xestospongin C (XeC; 1 μm) reduced Ca²⁺ waves (Fig. 9 Ca and b); Ca²⁺ wave frequency was reduced from 66.9 ± 10.1 min⁻¹ in control to 37.8 ± 8.8 min⁻¹ (Fig. 9 Da, P < 0.0001, c = 20, n = 5), amplitude was reduced from 0.15 ± 0.02 ΔF/F ₀ to 0.08 ± 0.014 ΔF/F ₀ (Fig. 9 Db, P = 0.008, c = 20, n = 5), and spatial spread was reduced from 27.8 ± 3.3 μm to 17.6 ± 4 μm (Fig. 9 Dd, P = 0.013, c = 20, n = 5) in the duration of Ca²⁺ waves in response to XeC were not observed (Fig. 9 Dc, P = 0.75, c = 20, n = 5).

Figure 9. The effect of blocking IP₃Rs on USMC Ca²⁺ waves.

Aa and b, representative ST maps showing the effect of 100 μm 2‐APB on Ca²⁺ transient firing in USMCs. Ba–d, summary data showing the effect of 100 μm 2‐APB on USMC Ca²⁺ transient frequency (a), amplitude (b), duration (c) and spatial spread (d) (c = 19, n = 6). Ca and b, representative ST maps showing the effect of 1 μm xestospongin C (XeC) on Ca²⁺ transient firing in USMCs. Da–d, summary data showing the effect of XeC on USMC Ca²⁺ transient frequency (a), amplitude (b), duration (c) and spatial spread (d) (c = 20, n = 5). ns, not significant, ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, ^**** p < 0.0001.

We also studied the effects of blocking RyRs with the compounds tetracaine and ryanodine. The RyR antagonist tetracaine (100 μm) had inhibitory effects on USMC Ca²⁺ waves similar to those of XeC and 2‐APB (Fig. 10 A). Tetracaine reduced the frequency of Ca²⁺ waves from 37 ± 6.2 min⁻¹ to 9.5 min⁻¹ (Fig. 10 Ba, P < 0.0001, c = 18, n = 6). Tetracaine reduced the amplitude of Ca²⁺ waves from 0.14 ± 0.025 ΔF/F ₀ to 0.07 ± 0.02 ΔF/F ₀ (Fig. 10 Bb, P = 0.008, c = 18, n = 6), the duration from 234 ± 6.5 ms to 130 ± 29.6 ms (Fig. 10 Bc, P = 0.004, c = 18, n = 6), and spatial spread from 19.9 ± 1.6 μm to 9.1 ± 2.4 μm (Fig. 10 Bd, P = 0.005, c = 18, n = 6). Application of ryanodine (100 μm) had similar effects to tetracaine (Fig. 10 C). Ryanodine reduced Ca²⁺ wave frequency from 40.1 ± 6.6 min⁻¹ to 4.8 ± 2.8 min⁻¹ (Fig. 10 Da, P < 0.0001, c = 14, n = 5). Ryanodine reduced the Ca²⁺ wave amplitude from 0.09 ± 0.006 ΔF/F ₀ to 0.02 ± 0.007 ΔF/F ₀ (Fig. 10 Db, P < 0.0001, c = 14, n = 5), the duration from 215.3 ± 7.7 ms to 90.5 ± 34.7 ms (Fig. 10 Dc, P = 0.0025, c = 14, n = 5), and spatial spread from 20.6 ± 2.4 μm to 3.6 ± 1.6 μm (Fig. 10 Dd, P < 0.0001, c = 14, n = 5).

Figure 10. The effect of blocking RyRs on USMC Ca²⁺ waves.

Aa and b, representative ST maps showing the effect of 100 μm tetracaine on Ca²⁺ transient firing in USMCs. Ba–d, summary data showing the effect of 100 μm tetracaine on USMC Ca²⁺ transient frequency (a), amplitude (b), duration (c) and spatial spread (d) (c = 18, n = 6). Ca and b, representative ST maps showing the effect of 100 μm ryanodine on Ca²⁺ transient firing in USMCs. Da–d, summary data showing the effect of 100 μm ryanodine on USMC Ca²⁺ transient frequency (a), amplitude (b), duration (c) and spatial spread (d) (c = 5, n = 14). ^* p < 0.01, ^*** p < 0.001, ^**** p < 0.0001.

Extracellular sources of Ca²⁺ contributing to USMC Ca²⁺ waves

Our findings suggest that Ca²⁺ waves in USMCs depend upon Ca²⁺ release from the SR via IP₃R1 and RyRs. Nevertheless, Ca²⁺ influx mechanisms are typically required for maintenance of Ca²⁺ stores (Berridge & Galione, 1988; Yao & Parker, 1994; Sneyd et al. 1995, 2004). Therefore, we investigated the impact of Ca²⁺ influx on Ca²⁺ waves. Removal of [Ca²⁺]_o (Ca²⁺ free solution also contained 0.5 mm EGTA) caused all Ca²⁺ transient activity in USMCs to cease within 10 min (Fig. 11 A and B, c = 19, n = 5); however, reducing [Ca²⁺]_o by 50% to 1.25 mm or doubling [Ca²⁺]_o to 5 mm had little effect on Ca²⁺ waves (Fig. 11 C). The effects of zero [Ca²⁺]_o may have been due to depletion of SR Ca²⁺ stores resulting from a lack of store refilling after release events. Application of caffeine (10 mm) was used to test this idea. Under control conditions caffeine (10 mm) induced large amplitude Ca²⁺ transients in USMCs throughout the muscles (0.7 ± 0.1 ΔF/F ₀; P = 0.002, c = 7, n = 4; Fig. 11 Da). After incubation in Ca²⁺ free solution, caffeine (10 mm) induced only ∼5% of maximal Ca²⁺ released during control conditions (0.03 ± 0.01 ΔF/F ₀; Fig. 11 Db and c). Thus, in Ca²⁺ free conditions, SR luminal Ca²⁺ appears to decrease to a point where spontaneous Ca²⁺ release events cease.

Figure 11. Generation of USMC Ca²⁺ transients requires extracellular Ca²⁺ influx to replenish SR stores.

Aa and b, representative ST maps showing the effect of Ca²⁺ free solution on Ca²⁺ transient firing in USMCs. Ba–d, summary data showing the effect of Ca²⁺ free solution on USMC Ca²⁺ transient frequency (a), amplitude (b), duration (c) and spatial spread (d) (c = 19, n = 5). Ca–d, dose–response graphs showing the effect of lowering or increasing extracellular Ca²⁺ on USMC Ca²⁺ transients (c = 7, n = 4). Da and b, representative ST maps showing the effect of Ca²⁺ free solution on 10 mm caffeine‐induced Ca²⁺ transients in USMCs. Summary data are shown in Dc (c = 7, n = 4). ^* p < 0.01, ^**** p < 0.0001.

Currents attributable to L‐type Ca²⁺ channels have been reported in human and rabbit USMCs (Hollywood et al. 2003; Bradley et al. 2004), and antagonists of these channels reduced tone in pig urethra (Bridgewater et al. 1993; Brading, 2006) and Ca²⁺ oscillations in rabbit USMCs (Hashitani & Suzuki, 2007). In the present experiments, nifedipine (1 μm; Fig. 12 A) failed to inhibit the frequency (P = 0.19), amplitude (P = 0.08), duration (P = 0.34) or spatial spread (P = 0.79) of USMC Ca²⁺ waves (Fig. 12 B, c = 22, n = 6; Fig. 12 A and B). Nicardipine (1 μm) also had no resolvable effect on Ca²⁺ wave frequency (P = 0.83), amplitude (P = 0.81), duration (P = 0.4) or spatial spread (P = 0.83) (c = 15, n = 5; Fig. 12 C and D). Lack of L‐type Ca²⁺ channel contribution to USMC Ca²⁺ waves was also supported by showing that an L‐type Ca²⁺ agonist, FPL 64176 (1 μm; Fig. 13 A) had no effect on the frequency (P = 0.97), amplitude (P = 0.2), duration (P = 0.19) or spatial spread (P = 0.42) of Ca²⁺ waves (Fig. 13 B, c = 18, n = 6). In some experiments, instead of perfusing the drug into the bath, FPL 64176 (1 μm) was applied directly via a single bolus. This method of application induced a large Ca²⁺ transient, lasting 700 ± 98.4 ms with an amplitude of 0.7 ± 0.1 ΔF/F ₀ (n = 4; Fig. 13 C). However, the effect was transient, and regular Ca²⁺ wave activity, similar to control events, developed during the remainder of the recording period (Fig. 13 C, n = 4). We also tested another L‐type channel antagonist that blocks voltage dependent calcium 1.3 (Cav 1.3) as well as Cav1.2 channels, isradipine (Lipscombe et al. 2004). Isradipine (1 μm; Fig. 13 D) also had no significant effects on USMC Ca²⁺ wave frequency (P = 0.4), amplitude (P = 0.33), duration (P = 0.68) or spatial spread (P = 0.28; Fig. 13 E, c = 14, n = 4).

Figure 12. Spontaneous Ca²⁺ transients in USMCs are insensitive to L‐type Ca²⁺ channel blockers.

Aa and b, representative ST maps showing the effect of 1 μm nifedipine on Ca²⁺ transient firing in USMCs. Ba–d, summary data showing the effect of 1 μm nifedipine on USMC Ca²⁺ transient frequency (a), amplitude (b), duration (c) and spatial spread (d) (c = 22, n = 6). Ca and b, representative ST maps showing the effect of 1 μm nicardipine on Ca²⁺ transient firing in USMCs. Da–d, summary data showing the effect of 1 μm nicardipine on USMC Ca²⁺ transient frequency (a), amplitude (b), duration (c) and spatial spread (d) (c = 15, n = 5). ns, not significant.

Figure 13. The effect of FPL 64176 and isradipine on USMC Ca²⁺ transients.

Aa and b, representative ST maps showing the effect of 1 μm FPL 64176 on Ca²⁺ transient firing in USMCs. Ba–d, summary data showing the effect of 1 μm FPL 64176 on USMC Ca²⁺ transient frequency (a), amplitude (b), duration (c) and spatial spread (d) (c = 18, n = 6). C, representative ST map showing the effect of rapidly applying 1 μm FPL 64176 to USMC Ca²⁺ transients. Da and b, representative ST maps showing the effect of 1 μm isradipine on Ca²⁺ transient firing in USMCs. Ea–d, summary data showing the effect of 1 μm isradipine on USMC Ca²⁺ transient frequency (a), amplitude (b), duration (c) and spatial spread (d) (c = 14, n = 4). ns, not significant.

We next tested the effects of L‐type Ca²⁺ channel antagonists on the excitatory responses to PE. As described previously, PE (10 μm) increased USMC Ca²⁺ waves dramatically (Fig. 14 A and B). In this series of experiments PE increased firing frequency from 38.8 ± 6.9 min⁻¹ to 101 ± 14.8 min⁻¹ (Fig. 14 Ba, P = 0.001, c = 12, n = 4), duration from 251.2 ± 15.8 ms to 322.7 ± 20.2 ms (Fig. 14 Bc, P = 0.01, c = 12, n = 4) and spatial spread from 19.8 ± 3.4 μm to 31.3 ± 4.3 μm (Fig. 14 Bd, P = 0.01, c = 12, n = 4). Nifedipine (1 μm) had no effect on Ca²⁺ wave frequency (Fig. 14 Ba, P > 0.05, c = 12, n = 4), amplitude (Fig. 14 Bb, P > 0.05, c = 12, n = 4), duration (Fig. 14 Bc, P > 0.05, c = 12, n = 4) or spatial spread (Fig. 14 Bd, P > 0.05, c = 12, n = 4) in response to PE.

Figure 14. PE excitatory response is insensitive to L‐type Ca²⁺ channel inhibition.

Aa–c, representative ST maps showing Ca²⁺ transient firing in USMCs during control conditions (a), in the presence of 10 μm phenylephrine (b) and in the presence of 10 μm phenylephrine + 1 μm nifedipine (c). Ba–d, summary data showing the effect of 1 μm nifedipine on the phenylephrine response on USMC Ca²⁺ transient frequency (a), amplitude (b), duration (c) and spatial spread (d) (c = 12, n = 4). C, representative contractile data showing the effect of 1 μm nifedipine on urethral contractions induced by phenylephrine and 90 mm extracellular K⁺. D, summary data showing the effect of 1 μm nifedipine on urethral tone (a) and contractions induced by phenylephrine (b) and 90 mm extracellular K⁺ (c) (n = 6). ns, not significant, ^** p < 0.01, ^*** p < 0.001.

As described above, PE (10 μm) caused large amplitude contractions of USM rings (Fig. 14 C). These effects were reproducible, after a 15 min washout period, with no significant rundown in consecutive PE responses (P = 0.44, n = 6). Contractile responses to PE were not blocked by nifedipine (1 μm). Summary data show that nifedipine had no effect on basal tone (Fig. 14 Da, P = 0.58, n = 6) or contractile responses to PE (Fig. 14 Db, P = 0.59, n = 6). Contractions of USM rings were also enhanced by 90 mm KCl (data not shown), and after 10 min washout periods, these responses were reproducible to subsequent additions of 90 mm KCl (data not shown, P = 0.099, n = 6). Responses to 90 mm KCl, however, were reduced by 84.4% by nifedipine (Fig. 14 Dc, P = 0.002, n = 6). Application of 90 mm KCl to USMCs caused responses similar to FPL 64176 in terms of Ca²⁺ wave activity (data not shown). After addition of 90 mm KCl, a single global Ca²⁺ transient occurred and then Ca²⁺ waves resumed, as before 90 mm KCl was added (n = 5).

Isolated USMCs have also been reported to exhibit T‐type Ca²⁺ currents (Hollywood et al. 2003; Bradley et al. 2004), and we investigated the possibility that Ca²⁺ influx occurs through a T‐type Ca²⁺ conductance. The T‐type Ca²⁺ channel antagonist, TTA‐A2 (1 μm), did not have inhibitory effects on USMC Ca²⁺ wave frequency (P = 0.39), amplitude (P = 0.15), duration (P = 0.1) or spatial spread (P = 0.51) (Fig. 15 A and B, c = 16, n = 5). Similarly, NNC‐550396 (1 μm) had no effect on Ca²⁺ wave frequency (P = 0.4), amplitude (P = 0.23) or spatial spread (P = 0.52) (Fig. 15 C and D, c = 13, n = 4). However, we observed a small but significant increase in Ca²⁺ wave duration in the presence of NNC‐550396 (Fig. 15 Dc, P = 0.01, c = 13, n = 4).

Figure 15. USMC Ca²⁺ waves are insensitive to T‐type Ca²⁺ channel blockers.

Aa and b , representative ST maps showing the effect of 1 μm TTA‐A2 on Ca²⁺ transient firing in USMCs. Ba–d, summary data showing the effect of 1 μm TTA‐A2 on USMC Ca²⁺ transient frequency (a), amplitude (b), duration (c) and spatial spread (d) (c = 16, n = 5). Ca and b, representative ST maps showing the effect of 1 μm NNC‐550396 on Ca²⁺ transient firing in USMCs. Da–d, summary data showing the effect of 1 μm NNC‐550396 on USMC Ca²⁺ transient frequency (a), amplitude (b), duration (c) and spatial spread (d) (c = 13, n = 4). ns, not significant, ^* p < 0.05.

Although L‐type and T‐type Ca²⁺ channel antagonists did not interfere with spontaneous and PE‐evoked Ca²⁺ signalling in USMCs, extracellular Ca²⁺ depletion inhibited Ca²⁺ waves, suggesting that Ca²⁺ influx is necessary for maintenance of Ca²⁺ store function. Therefore, we investigated whether store operated Ca²⁺ entry (SOCE) might be necessary for the maintenance of Ca²⁺ waves in these cells. SOCE has been best characterized in non‐excitable cells (Prakriya & Lewis, 2015); however, studies have also shown that SOCE is essential for store refilling and maintaining Ca²⁺ release in a number of excitable cells (Trebak et al. 2013).

SR Ca²⁺ store refilling resulting from SOCE occurs either through transient receptor potential cation (TRPC) channels or Orai Ca²⁺ channels (Trebak & Putney, 2017). USMCs expressed genes for Trpc1, 3, 4 and Trpc6 (Fig. 16 A, n = 4), and higher relative expression of Trpc3 (i.e. 0.014 ± 0.0006 sorted USMCs vs. 0.008 ± 0.00015 unsorted cells), Trpc4 (0.009 ± 0.00007 sorted USMCs vs. 0.0018 ± 0.00005 unsorted urethral cells) and Trpc6 (0.019 ± 0.0006 sorted USMCs vs. 0.0045 ± 0.00016 unsorted urethral cells) was noted (Fig. 16 A, n = 4). USMCs also express all three Orai genes (Fig. 16 B, n = 4).

Figure 16. Store operated Ca²⁺ entry (SOCE) in USM.

A, relative expression of TRPC genes in FACS sorted and unsorted urethral cells from SMC‐eGFP mice determined by qPCR analysis, normalized to Gapdh expression (n = 4). B, relative expression of Orai channel genes in FACS sorted and unsorted urethral cells from SMC‐eGFP mice determined by qPCR analysis, normalized to Gapdh expression (n = 4). C, representative contractile data showing a protocol to induce SOCE in USM. Basal USM contraction is shown at the start of the trace when intracellular Ca²⁺ stores were passively depleted by incubating the tissue in 0 mm Ca²⁺ solution and thapsigargin (10 μm). Reintroduction of [Ca²⁺]_o (2.5 mm) in the continued presence of thapsigargin led to development of a sustained tonic contraction that was insensitive to nifedipine. D, contractile trace showing the effect of GSK 7975A (1–10 μm) on SOCE induced tonic contraction of USM. E, summary data showing the effects of nifedipine and GSK 7975A on the sustained tonic contraction induced in USM using the SOCE protocol described in C and D. Contractions are expressed as fold changes normalized to contractile level in 0 mm Ca²⁺ + thapsigargin (10 μm; n = 8). ns, not significant, ^* p < 0.05, ^** p < 0.01.

We used contractile experiments to determine whether USM possesses a significant SOCE apparatus, as the expression studies would suggest. In USM ring preparations, SOCE was activated by passive depletion of SR Ca²⁺ stores by blocking the SERCA pump with thapsigargin (10 μm) while incubating the muscle in 0 mm Ca²⁺ bathing solution (Fig. 16 C). Upon reintroduction of [Ca²⁺]_o (2.5 mm), in the continued presence of 10 μm thapsigargin, a large sustained tonic contraction developed in USM ((2.7 ± 0.3)‐fold increase, normalized to 0 mm Ca²⁺ + thapsigargin conditions, P < 0.01, Fig. 16 E, n = 8). This sustained, tonic increase in contraction after reintroduction of Ca²⁺ is indicative of SOCE (Putney, 1986). The sustained contractile response was not due to activation of L‐type Ca²⁺ channels, as it was insensitive to nifedipine (1 μm; Fig. 16 C and E, P > 0.05, n = 8). However, the Orai antagonist GSK 7975A (1–10 μm) inhibited the SOCE response (Fig. 16 D). GSK 7975A (1 μm) decreased the contraction activated by SOCE by 32 ± 7% (P < 0.05, n = 8), and it was further reduced by 63 ± 5% and 65 ± 5% by 3 μm (P < 0.01, n = 8) and 10 μm GSK 7975A (P < 0.01, n = 8), respectively (Fig. 16 E).

We then tested the effect of SOCE blockers on Ca²⁺ waves in USMCs. SKF 96365 (10 μm), a relatively non‐selective SOCE antagonist (Harteneck & Gollasch, 2011) inhibited Ca²⁺ waves in USMCs (Fig. 17 A), reducing Ca²⁺ wave frequency from 74.6 ± 12.2 min⁻¹ in control to 14.3 ± 3.9 min⁻¹ (Fig. 17 Ba, P = 0.002, c = 20, n = 5). SKF 96365 also reduced Ca²⁺ wave amplitude from 0.08 ± 0.006 ΔF/F ₀ to 0.04 ± 0.007 ΔF/F ₀ (Fig. 17 Bb, P < 0.0001, c = 20, n = 5), duration from 270 ± 14.4 ms to 185 ± 37.2 ms (Fig. 17 Bc, P = 0.018, c = 20, n = 5), and spread from 17.9 ± 1.9 μm in control to 7.2 ± 1.6 μm (Fig. 17 Bd, P < 0.0001, c = 20, n = 5). The Orai antagonist, GSK 7975A (1 μm), also inhibited USMC Ca²⁺ waves (Fig. 17 C), reducing frequency from 52.5 ± 9.2 min⁻¹ to 23.4 ± 5.2 min⁻¹ (Fig. 17 Da, P < 0.0001, c = 17, n = 6), amplitude from 0.1 ± 0.01 ΔF/F ₀ to 0.06 ± 0.01 ΔF/F ₀ (Fig. 17 Db, P = 0.011, c = 17, n = 6), duration from 264 ± 12.4 ms in control to 194 ± 22.8 ms in the presence of GSK 7975A (Fig. 17 Dc, P = 0.01, c = 17, n = 6) and spatial spread of Ca²⁺ waves from 25.2 ± 2.4 to 15.9 ± 2.25 μm (Fig. 17 Dd, P = 0.003, c = 17, n = 6).

Figure 17. SOCE blockers inhibit Ca²⁺ transients in USMCs.

Aa and b , representative ST maps showing the effect of 10 μm SKF 96365 on Ca²⁺ transient firing in USMCs. Ba–d, summary data showing the effect of 10 μm SKF 96365 on USMC Ca²⁺ transient frequency (a), amplitude (b), duration (c) and spatial spread (d) (c = 20, n = 5). Ca and b, representative ST maps showing the effect of 1 μm GSK 7975A on Ca²⁺ transient firing in USMCs. Da–d, summary data showing the effect of 1 μm GSK 7975A on USMC Ca²⁺ transient frequency (a), amplitude (b), duration (c) and spatial spread (d) (c = 17, n = 6). ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, ^**** p < 0.0001.

Inhibition of Ca²⁺ waves in USMCs would be predicted to affect responses to PE, so we tested the effects of GSK 7975A on the contractile effects of PE. PE (0.1–30 μm) increased urethral tone (Fig. 18 A); however, when the muscles were treated with GSK 7975A (1 μm), contractions elicited by 10 and 30 μm PE were reduced by 23.3% (P < 0.05) and 25.8% (P < 0.05), respectively (Fig. 18 A and B, n = 6). Increasing the concentration of GSK 7975A caused dose‐dependent inhibition of PE contractions (e.g. 3 μm GSK 7975A reduced the response to 10 μm PE by 32.5%, Fig. 18 B, P < 0.05, n = 6) and 10 μm GSK 7975A reduced the response to 10 μm PE by 62.23% (Fig. 18 B, P < 0.001, n = 6).

Figure 18. SOCE can influence PE induced contractions in USM.

A, representative contractile data showing the effect of 1–10 μm GSK 7975A on urethral contractions induced by increasing concentration of PE (0.1–30 μm). B, summary data showing the effect of 1–10 μm GSK 7975A on urethral contractions induced by increasing concentration of PE (0.1–30 μm, n = 6). ^* p < 0.05, ^** p < 0.01, ^**** p < 0.0001.

Discussion

In the current study, we characterized spontaneous Ca²⁺ signalling in mouse USMCs in situ, utilizing a smooth muscle specific, optogenetic Ca²⁺ sensor, GCaMP3. Using traditional cell permeable Ca²⁺ indicator dyes, Hashitani and Suzuki (2007) reported the occurrence of spontaneous Ca²⁺ oscillations in rabbit USMCs in situ, and consistent with our observations, these authors reported that Ca²⁺ oscillations rarely propagated cell‐to‐cell. Use of GCaMP3 allowed monitoring of subcellular Ca²⁺ events in USMCs in great detail. We found that Ca²⁺ waves in murine USMCs were largely confined to single cells and were associated with small asynchronous contractions of USMCs. Ca²⁺ waves arose from multiple firing sites along the lengths of USMCs, and firing of events was stochastic in nature, with little coordination between Ca²⁺ firing sites. The activity of adjacent cells within the same bundle was also stochastic, and little or no coordination between USMCs in the same bundle was observed. Robust contraction of USM was elicited by PE, which also increased the frequency, duration and spread of intracellular USMC Ca²⁺ waves without causing a global rise in [Ca²⁺]_i. Thus, USM contractility can be modulated by influencing the occurrence and characteristics of intracellular Ca²⁺ waves. Summation of this asynchronous activity appears to be the source of sustained tone in the urethra.

The spontaneous activity of USMCs in situ observed in the current study contrasts with the lack of spontaneous activity observed in isolated USMCs (Sergeant et al. 2000; Hollywood et al. 2003; Bradley et al. 2004; Brading, 2006; Kyle, 2014). Isolated rabbit USMCs display no spontaneous Ca²⁺ signals, but large amplitude Ca²⁺ responses were evoked by 60 mm [K⁺]_o (Drumm et al. 2014c). Spontaneous Ca²⁺ events were ubiquitous in USMCs in situ, and most events were manifest as propagating Ca²⁺ waves. These contrasting findings between isolated USMCs and cells in situ highlight the need for studies of USMCs in situ to better understand the physiological regulation of urethral contraction.

USMC contraction is thought to have an important myogenic component because urethral tone is resistant to nerve blockade in most mammals (Thornbury et al. 1992; Bridgewater et al. 1993; Brading, 1999). In alignment with this concept, we found that basal Ca²⁺ wave activity in USMCs was resistant to TTX. While basal Ca²⁺ waves were myogenic in nature they were modulated by neurotransmitters that are known to influence urethral contraction. We found that isolated murine USMCs express genes encoding muscarinic receptors (M₂, M₃ and M₅). M₂ and M₃ receptor expression was previously reported in whole muscle preparations of rabbit (Mutoh et al. 1997) and pig urethras (Yamanishi et al. 2002). However, the role of parasympathetic innervation in the urethra is unclear. Some investigators have reported that cholinergic agonists CCh or ACh elicit contractions of urethral smooth muscles (Ek et al. 1977; Sogbein et al. 1984; Mutoh et al. 1997; Yamanishi et al. 2002), but others have argued that blocking cholinergic nerves has no effect on urethral closure pressure (Ulmsten & Andersson, 1977). Urethral smooth muscle contains a sparse population of cholinergic nerves compared to the bladder (Gosling & Dixon, 1975). Our findings with CCh suggested that cholinergic input has little or no direct effect on USMCs and the main excitatory influence on murine USMCs is adrenergic.

We found that mouse USMCs express all three α1 adrenoceptor subtypes (Kimoto et al. 1987; Chen & Brading, 1992; Andersson, 2001) and spontaneous Ca²⁺ waves were dramatically increased by application of PE, which also led to robust contractions of murine urethral smooth muscles. Conversely, murine USMCs expressed high levels of genes encoding GC and PKG, and Ca²⁺ waves were decreased by a nitric oxide donor. The inhibitory effects on Ca²⁺ waves is likely to occur through the NO→GC→PKG pathway, which is known as the main inhibitory pathway exerting relaxation on the internal urethral sphincter during micturition (Thornbury et al. 1992; Bridgewater et al. 1993; Dokita et al. 1994; Zygmunt et al. 1995; Waldeck et al. 1998; Costa et al. 2001).

Purinergic neurotransmission in the urethra is controversial. On pre‐contracted muscles, purines, such as ATP, have been reported to induce relaxation of hamster and pig urethral smooth muscle (Pinna et al. 1998, 2005; Hernandez et al. 2009). However, when applied to basal activity, ATP induces contraction in rabbit urethral smooth muscle (Bradley et al. 2010, 2011). In accordance with the latter findings, our results show that ATP enhanced Ca²⁺ waves in USMCs and notable contractions of the muscles while imaging. Bradley et al. (2011) suggested the excitatory effects of purines were due to P2X receptors in USMCs, and we found P2rx1–4 and P2rx6 receptor genes expressed in murine USMCs. We did not resolve P2Y receptor genes in USMCs, with the exception of P2ry14. However, P2Y receptor gene expression was observed in the unsorted cell populations, so P2Y expression is likely to occur in cell types other than USMCs. For example, others have reported that non‐contractile interstitial cells from the urethra respond to P2Y agonists (Bradley et al. 2010). ATP and other purinergic mediators released from the urothelium are thought to affect contractions in the bladder (Birder & Andersson, 2013; Merrill et al. 2016); however, at present, it is unknown whether urothelial mediators affect urethral contractility. In the current study, we found that ATP exerted significant excitatory effects on USMC Ca²⁺ waves, thus it is possible that purines released from urothelium could have paracrine effects on USMCs. Mice with optogenetic sensors expressed in USMCs, as utilized in this study, will serve as valuable tools for assessing the role of urothelium‐derived mediators on urethral excitation–contraction coupling in future studies.

Our data suggest Ca²⁺ waves in USMCs arose from the release of Ca²⁺ from intracellular stores. Hashitani et al. (2006) reported previously that Ca²⁺ oscillations in rabbit USMCs were not consistently blocked by SERCA pump inhibitors, but we found that the SERCA pump inhibitors thapsigargin and CPA abolished Ca²⁺ transients in mouse USMCs in all experiments. The SR channels responsible for Ca²⁺ release appeared to involve both IP₃Rs and RyRs, as antagonists of both receptors blocked Ca²⁺ waves. Itpr1 was the dominant Ca²⁺ release channel gene expressed by USMCs, but the inhibitory effects of tetracaine and ryanodine in addition to the expression of Ryr1–3 genes suggest the possibility that clusters of RyRs may initiate or amplify Ca²⁺ release from IP₃R1. Interaction between IP₃Rs and RyRs to generate intracellular Ca²⁺ waves has been reported in other smooth muscle types. In rat portal vein myocytes, Ca²⁺ waves could be induced by both release of caged IP₃ or by applications of caffeine suggesting the involvement of IP₃Rs and RyRs in the generation of Ca²⁺ waves (Boittin et al. 1998, 1999). In rat gastric myocytes CCh induced whole cell Ca²⁺ events that were inhibited by IP₃R or RyR blockers (White & McGeown, 2002). One type of SR Ca²⁺ channel might act as an ‘initiator’ or an ‘amplifier’ of a primary Ca²⁺ transient. In pressurized arteries from murine cremaster muscles, Ca²⁺ waves were blocked by inhibiting IP₃Rs with 2‐APB, but initiating Ca²⁺ sparks remained. Tetracaine blocked both events suggesting that RyRs were the initiating spark and Ca²⁺ release from IP₃Rs amplified the sparks into propagating waves (Westcott et al. 2012). A similar situation occurs in rabbit urethral interstitial cells where RyRs also initiate waves in rabbit urethral interstitial cells, which is then amplified to a propagating wave by additional Ca²⁺ release from IP₃Rs (Drumm et al. 2015). Our data also suggest cooperation between RyRs and IP₃Rs in USMCs; further work will be needed to clarify if an ‘initiator’ and ‘amplifier’ relationship exists between the two receptor types.

Studies on isolated USMCs have shown that these cells possess Ca²⁺ conductances, such as L‐ and T‐type Ca²⁺ channels (Hollywood et al. 2003; Bradley et al. 2004). Using membrane permeable dyes, Hashitani and Suzuki (2007) reported that nicardipine reduced the amplitude of Ca²⁺ oscillations in rabbit USMCs. In murine USMCs, neither dihydropyridines nor the L‐type Ca²⁺ channel agonist FPL 64176 had any apparent effect on Ca²⁺ signalling. Our findings suggesting no role for L‐type Ca²⁺ channels in urethral Ca²⁺ waves are not without precedent. Larsson et al. (1984) found that nifedipine failed to abolish contractions of rabbit urethra induced by noradrenaline (NA), suggesting that other Ca²⁺ influx or release pathways are important in this tissue. Nifedipine also failed to decrease urethral closure pressure in cats (Mawby et al. 1991). Nifedipine also failed to inhibit contractions of sheep urethra induced by NA and was 1000 times less effective at blocking NA induced contractions than those elicited by 124 mm K⁺ (Garcia‐Pascual et al. 1991). Nifedipine had no effect on human urethral tone in vitro or in vivo (Forman et al. 1978). T‐type Ca²⁺ currents have also been reported in USMCs (Hollywood et al. 2003; Bradley et al. 2004; Fry & Jabr, 2014). However, specific T‐type Ca²⁺ channel blockers that potently block T‐type currents (Huang et al. 2004; Kraus et al. 2010; Drumm et al. 2017) had no effect on Ca²⁺ waves in USMCs.

The reliance on Ca²⁺ influx to sustain Ca²⁺ waves and the lack of effects of L‐type and T‐type Ca²⁺ conductances led us to hypothesize that Ca²⁺ stores in USMCs may be refilled by SOCE. The phenomenon of SOCE was first described in non‐excitable cells (Putney, 1986), but it also occurs in excitable muscle cells (Gibson et al. 1998; Quinn et al. 2004; Bradley et al. 2005; Trebak et al. 2013; Prakriya & Lewis, 2015). SOCE depends upon the Orai Ca²⁺ channels (Trebak & Putney, 2017), and these channels are gated by interactions with an SR luminal Ca²⁺ sensor (STIM) (Liou et al. 2005; Feske et al. 2006; Zhang et al. 2006; Soboloff et al. 2012). USMCs expressed transcripts for all three Orai channel paralogues, and SKF 96365, a poorly selective Orai antagonist, and GSK 7975A, a more selective antagonist, inhibited Ca²⁺ waves in USMCs. Taken together, these data suggest that SOCE is an important mechanism for sustaining spontaneous Ca²⁺ activity in USMCs, and this Ca²⁺ transport system also has an important role in sustaining urethral contractions. Our data suggest that Orai channels, possibly due to expression of Orai1 and Orai3 (Fig. 16 B; Mignen et al. 2008), fill the role of a receptor‐operated channel, supplying the Ca²⁺ required for responses to PE.

A similar situation has recently been reported in airway smooth muscle, where Ca²⁺ oscillations can be induced by agonists such as methacholine (MCh) (Boie et al. 2017). MCh‐induced Ca²⁺ oscillations were relatively insensitive to high concentrations of nifedipine, although nifedipine could reduce Ca²⁺ oscillations induced by high external K⁺. Despite the lack of effect of blocking L‐type Ca²⁺ channels, the Orai1 blocker GSK 7975A blocked MCh‐induced Ca²⁺ oscillations in these airway smooth muscle cells (Boie et al. 2017), similar to the effects of GSK 7975A on basal USMC Ca²⁺ waves and responses to PE in the current study. The observations led the authors of the study on airway muscles to conclude that Ca²⁺ entry due to SOCE, and not via L‐type Ca²⁺ channels, contributes to the control of airway smooth muscle tone.

In many smooth muscle cells Ca²⁺ transients are not coupled directly to contractions but have seemingly paradoxical effects by regulating membrane potential in such a manner as to restrain contractility. For example, Ca²⁺ sparks are linked to activation of large‐conductance Ca²⁺‐activated K⁺ (BK) channels and tend to cause membrane hyperpolarization or at least to moderate depolarization (Nelson et al. 1995; Jagger et al. 2000). Hyperpolarization tends to reduce activation of voltage‐dependent Ca²⁺ channels and restrains the development of contractions. In other smooth muscles Ca²⁺ release leads to activation of Ca²⁺‐activated Cl⁻ conductances, resulting in inward current and depolarization (ZhuGe et al. 1998; Bootman et al. 2001). The open probability of voltage‐dependent Ca²⁺ channels increases in response to depolarization, so Ca²⁺ transients in these cells are indirectly linked to contraction. Our results suggest that Ca²⁺ waves in USMCs are directly responsible for excitation–contraction coupling, constituting a pure example of pharmaco‐mechanical coupling (Somlyo et al. 1999).

In conclusion, we characterized spontaneous and evoked Ca²⁺ signalling in mouse USMCs in situ using cell specific expression of GCaMP3. Ca²⁺ waves in USMCs appear to be a source of Ca²⁺ for the activation of the contractile apparatus instead of an indirect mechanism for regulating contraction via activation of Ca²⁺‐dependent conductances in the plasma membrane. Ca²⁺ waves were greatly stimulated by excitatory neurotransmitters and inhibited by NO, the major inhibitory neurotransmitter. Responses to the major sympathetic neurotransmitter, working through α1‐adrenoceptors, did not depend upon depolarization‐induced Ca²⁺ influx through activation of L‐type Ca²⁺ channels, which are expressed in USMCs, because dihydropyridines did not affect contractile responses to PE. Generation of Ca²⁺ waves in USMCs depends upon release of Ca²⁺ from the SR mediated by both IP₃Rs and RyRs. A Ca²⁺ entry mechanism is apparent in the regulation of Ca²⁺ waves, however, because Ca²⁺‐free external solution and SOCE antagonists inhibited Ca²⁺ waves. Our results also provide evidence that Orai channels provide a receptor‐operated Ca²⁺ entry mechanism required to cause excitation–contraction coupling in response to PE. This study of Ca²⁺ signalling in USMCs in situ provides new insights into the physiological mechanisms responsible for contractions in urethral muscles.

Additional information

Competing interests

None declared.

Author contributions

B.T.D., G.P.S. and K.M.S. contributed to the conception and design of the experiments. B.T.D., B.E.R., C.A.C., S.A.B., G.P.S., M.A.H., K.D.T. and K.M.S. contributed to the collection, analysis and interpretation of data. All experiments were performed at University of Nevada, Reno School of Medicine. B.T.D., B.E.R., C.A.C., S.A.B., G.P.S., M.A.H., K.D.T. and K.M.S. contributed to drafting the article and/or revising it critically for important intellectual content. All authors have approved the final version of the manuscript and agree to be accountable for all aspects of the work. All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed.

Funding

This project was supported by exploratory funding from the University of Nevada and from P01 DK41315 from the NIDDK that supported Core laboratories providing molecular and immunohistochemical support. USMCs were sorted by the FACS/flow cytometry shared resource lab that is supported by a Phase III COBRE award (P30‐GM110767). B.T.D. and S.A.B. received salary support from R01 DK‐091336 and K.M.S. received salary support from P01 DK41315.

Supporting information

Disclaimer: Supporting information has been peer‐reviewed but not copyedited.

Movie S1. Spontaneous intracellular Ca²⁺ waves in USMC are associated with USMC contractions.

Movie S2. Spontaneous Ca²⁺ waves in USMC in situ.

Movie S3. Phenylephrine increases Ca²⁺ waves in USMC.

Biography

Bernard Drumm received his PhD in Physiology from Dundalk Institute of Technology, Ireland in 2013. His research investigates how Ca²⁺ signalling impacts the functions of visceral smooth muscle organs. His doctoral work, studying Ca²⁺ wave propagation in urethral cells was followed by a postdoctoral position at the University of Nevada, Reno (UNR), where he studied the regulation of pacemaker and neuroeffector mechanisms in gastrointestinal interstitial cells of Cajal. Dr Drumm is currently a Research Assistant Professor at UNR, where he uses optogenetic approaches to investigate intestinal motility and lower urinary tract smooth muscle contractility.