RyR1-specific requirement for depolarization-induced Ca²⁺ sparks in urinary bladder smooth muscle

Ryanodine receptors (RyRs) are intracellular Ca²⁺-release channels located in the membrane of intracellular stores in numerous cell types. In smooth muscle cells, spontaneous or triggered RyR-mediated Ca²⁺ releases generate localized events called Ca²⁺ sparks (Arnaudeau et al., 1996). These Ca²⁺ signals contribute to the control of the cell excitability and smooth muscle relaxation (Nelson et al., 1995). Summation of sparks results in the generation of Ca²⁺ waves that control contraction of smooth muscle (Macrez and Mironneau, 2004).

Smooth muscle cells have multiple and different expression patterns of the three RyR subtypes 1, 2 and 3 (RyR1, RyR2 and RyR3) (Coussin et al., 2000; Fill and Copello, 2002). The regulation of these RyR subtypes has been studied in tissues that express only one subtype, or in single-channel studies in lipid bilayers. If the major activator of Ca²⁺ release through RyRs is Ca²⁺ itself, numerous proteins are associated with and regulate RyRs (Fill and Copello, 2002), the most important proteins among which are FK506-binding proteins (FKBPs) of 12 and 12.6 kDa (FKBP12 and FKBP12.6). First described as immunophilins, these FKBPs have been demonstrated to bind to Ca²⁺ and regulate Ca²⁺ flux through RyRs. Both FKBP12 and FKPB12.6 are expressed in almost all muscle cells but display selectivity binding to certain RyR subtypes. The high affinity of FKBP12.6 towards RyR2 makes it the main partner of that isoform, whereas FKBP12 can potentially regulate all three RyR subtypes, displaying a selectivity of RyR1>RyR3>RyR2 (Chelu et al., 2004).

Activation of the voltage-dependent Ca²⁺ current evokes Ca²⁺-induced Ca²⁺ release (CICR) in the form of Ca²⁺ sparks or Ca²⁺ waves in urinary bladder myocytes (Imaizumi et al., 1998; Collier et al., 2000). RyR2 has been suggested to be the primary contributor of Ca²⁺ sparks in urinary bladder myocytes based on the increase in the number of spontaneous and triggered Ca²⁺ sparks in mice that do not express FKBP12.6 (FKBP12.6^–/– mice) (Ji et al., 2004). However, all three isoforms of RyR are expressed in mouse detrusor muscle (Ji et al., 2004) and several studies suggested a role for RyR1 in smooth muscle Ca²⁺ waves or Ca²⁺ sparks (Coussin et al., 2000; Yang et al., 2005; Du et al., 2005; Du et al., 2006).

Here, we used Ryr1-null mice (hereafter referred to as RyR1^–/– mice) to determine the specific role of RyR1 in Ca²⁺ signaling in urinary bladder myocytes. As RyR1^–/– mice die immediately after birth, we used the detrusor muscle of mice at embryonic days 20-21 (E20-E21) for our studies. Like adult muscle, the embryonic detrusor muscle expresses all three RyR isoforms. We studied spontaneous, caffeine-induced and depolarization-induced Ca²⁺ sparks in both wild-type (WT) and RyR1^–/– myocytes and our results suggest a major role for RyR1 in controlling depolarization-induced Ca²⁺ sparks in urinary bladder. Relevance of these results in urinary bladder function and discrimination between different spark sites are discussed.

In detrusor muscle of adult mice, all three RyR subtypes mRNA have been detected by RT-PCR (Ji et al., 2004). Here, we show by western blot that detrusor smooth muscle of WT embryonic mice also expresses the three subtypes RyR1, RyR2 and RyR3 (Fig. 1). As expected, RyR1^–/– mouse urinary bladder lacks the expression of RyR1 but expresses RyR2 and RyR3 (Fig. 1). Absence of cross-reactions of the three anti-RyR antibodies have been described previously (Jeyakumar et al., 1998; Jeyakumar et al., 2001; Jeyakumar et al., 2002) and their use in immunostaining, showing specific inhibition of RyR subtypes by microinjection of antisense oligonucleotides (Fritz et al., 2005), further supports their specificity.

Embryos of RyR1^–/– mice were then used to investigate the role of RyR1 in detrusor muscle. In single embryonic myocytes from both WT or RyR1^–/– mice, localized spontaneous Ca²⁺ events were frequently observed (Fig. 2A). These spontaneous events were also observed in patch-clamped cells and they were RyR-dependent, since pre-treatment of the cells with ryanodine (10 μM) or intracellular diffusion of an anti-RyRcom antibody (targeted against RyR1, RyR2 and RyR3) through the patch pipette of whole-cell patch-clamped WT myocytes fully inhibited them (data not shown). These localized spontaneous Ca²⁺ events display the characteristics of the previously described spontaneous Ca²⁺ sparks in smooth muscle (Table 1) (Arnaudeau et al., 1996; Jaggar et al., 2000). For these two main reasons, spontaneous localized Ca²⁺ signals observed in embryonic urinary bladder myocytes were considered to be Ca²⁺ sparks.

Only a few WT cells (five out of 30 cells) displayed spontaneous Ca²⁺ waves, whereas these propagating spontaneous Ca²⁺ signals were frequently observed in RyR1^–/– myocytes (20 out of 43 cells). The functionality of Ca²⁺ signaling in RyR1^–/– myocytes was attested by comparing Ca²⁺ signals activated by both membrane depolarization and the direct RyR activator caffeine. Caffeine (10 mM) induced both Ca²⁺ sparks and global Ca²⁺ waves in WT as well as RyR1^–/– myocytes (Fig. 2B). In RyR1^–/– myocytes, the amplitude of Ca²⁺ waves was reduced compared with WT (ΔF/F₀= 1.25±0.35, n=10 in RyR1^–/– myocytes vs 2.85±0.56, n=13 in WT myocytes) whereas the properties of the Ca²⁺ sparks were not different between the two cell types (Table 1). These results show that the absence of RyR1 expression selectively alters caffeine-induced RyR-dependent global Ca²⁺ signals. By contrast, Ca²⁺ sparks were not affected in RyR1^–/– myocytes, suggesting that RyR2 and RyR3 are sufficient for both spontaneous and caffeine-induced Ca²⁺ sparks (Fig. 3).

The most physiological activation of RyR is achieved by membrane depolarization producing a Ca²⁺ influx through voltage-gated Ca²⁺ channel, which in turn activates RyRs. We investigated the specific roles of the RyR subtypes in depolarization-induced Ca²⁺ sparks in WT and RyR1^–/– myocytes. In WT myocytes, two types of Ca²⁺ signals could be activated by short depolarizing pulses (100 mseconds, 50 mV, holding potential –40 mV): Ca²⁺ sparks were seen in 28 out of 52 tested cells and small Ca²⁺ waves of low amplitude were seen in 17 out of the 52 tested cells (Fig. 2C and Table 1). Seven cells out of 52 did not respond to depolarization, although caffeine did produce a Ca²⁺ signal in all cells (data not shown). Increasing the duration of depolarizing pulses to 500 mseconds induced Ca²⁺ waves in almost all cells (ten out of 13 cells; data not shown). The depolarization-induced small Ca²⁺ waves had amplitudes similar to the Ca²⁺ sparks (ΔF/F₀=0.43±0.09, n=28 and 0.30±0.07, n=17 for sparks and waves, respectively) but they decreased much more slowly (half decay time = 42.1±7.9 mseconds, n=28 and 1247±29 mseconds, n=17 for sparks and waves, respectively) and spread much more into the cytosol (full width at half maximal amplitude = 2.03±0.05 μm, n=28 and 10.1±4.6 μm, n=17 for sparks and waves, respectively). Importantly, we never observed depolarization-induced Ca²⁺ sparks in RyR1^–/– myocytes however, depolarizing pulses induced Ca²⁺ waves in 25 out of 49 tested cells. The Ca²⁺ waves activated in RyR1^–/– myocytes displayed spatio-temporal properties similar to those activated in WT myocytes (Table 1).

Since the same depolarizing protocol was able to induce either Ca²⁺ sparks (WT) or Ca²⁺ waves (WT and RYR1^–/–), we hypothesized that different voltage-gated Ca²⁺ current densities could be associated with each type of Ca²⁺ response. However, current densities measured in WT cells displaying Ca²⁺ sparks or Ca²⁺ waves (4.1±0.8 pA/pF, n=13) were not significantly different from those measured in RyR1^–/– cells displaying only Ca²⁺ waves (4.9±0.6 pA/pF, n=13), suggesting that the pattern of Ca²⁺ response probably depends on regulation of the RyR channels rather than on the amount of Ca²⁺ ions flowing through the voltage-gated channels. Taken together, these results show that, although RyR1 is not required for spontaneous or caffeine-activated Ca²⁺ sparks, it is needed to activate depolarization-induced Ca²⁺ sparks.

Different expression patterns of RyR isoforms might be responsible for the variability of Ca²⁺ responses triggered by the same stimulus (Macrez and Mironneau, 2004). We compared RyR2 and RyR3 expression in embryonic WT and RyR1^–/– detrusor muscle. A direct comparison of the western blot luminescence is not appropriate because luminescence intensity highly depends on the coupling efficiency between primary and secondary antibodies. Therefore, we calculated the ratios of luminescence associated with RyR subtypes and inositol (1,4,5)-trisphosphate receptor subtype 3 [Ins(1,4,5)PR3] on the same PVDF membranes. The expression of Ins(1,4,5)PR3 does not vary in the bladder during development (Rosemblit et al., 1999) and its large molecular mass (260 kDa) provides a partial loading control.

The protein samples were analyzed for RyR expression using the specific anti-RyR2 and anti-RyR3 antibodies (Fig. 4A, upper panel). Lack of protein recognition was verified by chemiluminescence to assess stripping efficiency. PVDF membranes were then incubated with the specific anti-Ins(1,4,5)PR3 antibody (Fig. 4A, lower panel). This protocol allowed us to quantify the luminescence associated with proteins arising from the same samples, allowing comparison. Specificity of the anti-Ins(1,4,5)PR3 antibody has been verified in previous western blots and immunostaining, revealing expression of Ins(1,4,5)PR3 in rat ureteric myocytes versus no expression in portal vein myocytes (Morel et al., 2003). Fig. 4B clearly shows that neither RyR2 nor RyR3 expression differs between WT and RyR1^–/– detrusor smooth muscle. Similar experiments were carried out with dystrophin (molecular mass is 427 kDa for the complete isoform) as a loading control and conclusions were the same (data not shown).

Since RyR1 seems to be specifically required for depolarization-induced Ca²⁺ sparks, we investigated whether RyR isoforms had specific subcellular localizations. RyR1-associated staining was seen in WT myocytes but not in RyR1^–/– myocytes (Fig. 5A), whereas RyR2 and RyR3 were observed in both types of myocytes (Fig. 5B,C), confirming western blot analysis. We compared subcellular localization of each RyR isoform by analyzing fluorescence profiles of stained myocytes. RyR1 is exclusively distributed near the plasma membrane (profile a), whereas RyR2 expression seems more homogeneous and is associated with the membrane and deeper parts of the cytosol (profiles c and d). Finally, RyR3 is most highly expressed near the nucleus (profiles e and f). No difference in the intracellular location of RyR2 and RyR3 was seen between WT and RyR1^–/– myocytes.

Loss of perimembraneous RyR1 isoform in KO myocytes could explain why depolarization-induced Ca²⁺ sparks were not detected in RyR1^–/– myocytes. In this hypothesis, RyR1 is necessary for depolarization-induced Ca²⁺ sparks, whereas RyR2 and RyR3 are sufficient to produce spontaneous and caffeine-induced Ca²⁺ sparks and waves. However this cannot explain the increased number of cells responding to depolarization with Ca²⁺ waves in RyR1^–/– myocytes (35% versus 75% in WT versus RyR1^–/– myocytes).

Buck et al. (Buck et al., 1997) have demonstrated that the RyR-associated protein FKBP12 appears to be overexpressed in RyR1^–/– mouse skeletal muscle. This protein is a well-known regulator of RyR-dependent Ca²⁺ signaling. Thus, we investigated whether alterations in FKBP protein expression can be detected in RyR1^–/– detrusor smooth muscle.

Ratios of FKBP12- or FKBP12.6-associated and glyceraldehyde-3-phosphate dehydrogenase (GAPDH)-associated fluorescence were calculated and compared in WT and RyR1^–/– mice (Fig. 6). GAPDH is an enzyme classically used as an internal standard for RT-PCR because of its ubiquitous and stable expression in mouse tissues. Interestingly, RT-PCR data suggest that FKBP12 mRNA levels are higher than FKBP12.6 mRNA levels in detrusor muscle (Fig. 6A,B) as previously described in cardiac and skeletal myocytes (Timerman et al., 1996). More interestingly, a significant difference was found between WT and RyR1^–/– mice, where FKBP12:GAPDH ratios were found to be 1.48±0.22 (n=6) versus 0.82±0.07 (n=6) for WT versus RyR1^–/– myocytes, respectively. These results suggest a decrease in FKBP12 expression in RyR1^–/– detrusor muscle (Fig. 6B). The already low FKBP12.6:GAPDH ratios did not allow detection of specific changes between WT and RyR1^–/– detrusor muscle (Fig. 6B).

Western blot analysis of RyR1^–/– and WT mouse detrusor muscle further supports the inhibition of FKBP expression in RyR1^–/– myocytes. A common anti-FKBP12 and anti-FKBP12.6 antibody used in western blots (Fig. 6C,D) did not allow discrimination between the two proteins but revealed a significant decrease in total FKBP protein expression in RyR1^–/– myocytes compared with a smooth muscle α-actin control (Fig. 6C,D).

FKBP12 and FKBP12.6 are thought to regulate RyR-mediated Ca²⁺ release by stabilizing the channel in its closed state (Wehrens et al., 2004). To test the hypothesis that this decrease in FKBP expression could provide an explanation for the increased number of waves activated by membrane depolarization in RyR1^–/– versus WT detrusor muscle, we used rapamycin to pharmacologically uncouple FKBP from RyR, and studied functional changes in Ca²⁺ signaling. First, addition of rapamycin (1 μM) to the extracellular medium caused a significant increase in the caffeine-induced Ca²⁺-response amplitude in WT but not in RyR1^–/– urinary bladder myocytes (Fig. 7A). Second, higher concentrations of rapamycin induced either Ca²⁺ sparks or waves (Fig. 7B,C, n=9) in WT myocytes, revealing a tonic inhibitory role for FKBP on RyR-mediated Ca²⁺ release. These effects are not observed in RyR1^–/– myocytes, in which acute application of a high concentration of rapamycin only induced a very slow and weak increase in Ca²⁺ concentration (Fig. 7D, n=11).

Taken together, these results indicate that RyR-mediated Ca²⁺ release is negatively regulated by FKBP in WT embryonic urinary bladder myocytes. RyR1^–/– myocytes, which express less FKBP, lack such negative regulation. This could explain why `non-regulated' RyR2 and RyR3 cannot produce Ca²⁺ sparks in RyR1^–/–myocytes, and thus respond to short depolarizations with Ca²⁺ waves.

We report here the specific contribution of RyR1 in some but not all Ca²⁺ release signals in urinary bladder myocytes using RyR1^–/– mice. Both caffeine-induced global Ca²⁺ increases and depolarization-induced Ca²⁺ sparks were affected by the lack of expression of RyR1, indicating a role for this isoform in the normal Ca²⁺ signaling of embryonic urinary bladder myocytes. Apart from its well-characterized function in excitation-contraction coupling in skeletal muscle, RyR1 has also been shown – by specific isoform expression inhibition by antisense oligonucleotides – to be crucial for CICR and Ca²⁺ sparks in vascular smooth muscles (Coussin et al., 2000). Moreover, several studies report expression of RyR1 and suggest RyR1-mediated Ca²⁺ release in airway smooth muscle and neurons (Du et al., 2006; De Crescenzo et al., 2006). However, our conclusion that RYR1 is required for depolarization-induced Ca²⁺ sparks in urinary bladder does not oppose the proposal that RYR2 contributes to the formation of Ca²⁺ sparks in urinary bladder smooth muscle based on alteration of spontaneous and triggered Ca²⁺ sparks in FKBP12.6^–/– mice (Ji et al., 2004). These two conclusions might be reconcilable if both RyR1 and RyR2 are required to form functional clusters producing Ca²⁺ sparks activated by depolarization. We previously proposed the idea of a dual requirement for both RyR1 and RyR2 based on antisense oligonucleotides in vascular myocytes (Coussin et al., 2000). Conversely, spontaneous Ca²⁺ sparks are present in both WT and RyR1^–/– urinary bladder myocytes, indicating that RyR2 and RyR3 are sufficient for activation of spontaneous Ca²⁺ sparks in these myocytes. These results are in agreement with previous data suggesting the contribution of RyR2 to the formation of Ca²⁺ sparks in urinary bladder myocytes (Ji et al., 2004).

Together, these results suggest that Ca²⁺ sparks activated by different means result from activation of different RyR subtypes. This seducing scheme is also supported by Yang et al. showing that caffeine and endothelin may activate different Ca²⁺ sparks in pulmonary artery myocytes (Yang et al., 2005), and by ZhuGe et al. who reported multiple Ca²⁺ sparks sites in smooth muscle cells (ZhuGe et al., 2004).

To ensure that the differences seen between the two types of myocytes were due to the loss of RyR1, we analyzed the expression of other RyR isoforms in WT versus RyR1^–/– urinary bladder myocytes. We observed no differences in the expression of RyR2 and RyR3 between the two strains of mice. Similarly, in the RyR3 null mouse, RyR1 and RyR2 expression is also not affected (Löhn et al., 2001).

Since we recorded spontaneous Ca²⁺ sparks in both WT and in RyR1^–/– myocytes, whereas depolarization-induced Ca²⁺ sparks were never activated in RyR1^–/– myocytes, we reasoned that RyR1 has a specific subcellular localization that supports this Ca²⁺ signaling specificity. This is supported by our immunolocalization experiments, showing RyR1 near the surface membrane whereas RyR2 and RYR3 were found more scattered throughout the cytosol or in the perinuclear endoplasmic reticulum in WT myocytes. Immunostaining of RyR2 and RYR3 was unaffected in RyR1^–/– myocytes. Peri-membranous expression of RyR1 has also been reported by other groups in different tissues (Du et al., 2006; De Crescenzo et al., 2006), this specific localization might support a specific action mechanism for RyR1. Indeed, RyR1 is known to physically interact with Cav1.1 during the excitation-contraction coupling of skeletal muscle. Physical interaction of RyR with neuronal L-type Ca²⁺ channels has also been observed (Chavis and Westbrook, 1996) and voltage-gated activation of RyR1 without Ca²⁺ influx through L-type channels seems to be possible in nerve terminals (De Crescenzo et al., 2006). In addition to this voltage-gated direct activation, RyR1 might also be localized specifically at the near-plasma-membrane endoplasmic reticulum to bridge the Ca²⁺ entry through L-type Ca²⁺ channels to RyR clusters to yield efficient CICR. This mechanism seems to occur in portal vein and urinary bladder myocytes, where Ca²⁺ influx is required for depolarization-induced Ca²⁺ sparks and waves (Coussin et al., 2000; Kotlikoff, 2003).

In urinary bladder, RyR1 localized near the plasma membrane would rapidly sense Ca²⁺ influx during depolarization and allow Ca²⁺ sparks in WT myocytes, whereas in RyR1^–/– myocytes Ca²⁺ entering through L-type channel has to spread before reaching the deeper localization of RyR2 and RyR3 inducing the obligatory propagating Ca²⁺ signal.

Our results point out a decreased expression of FKBP12 and FKBP12.6 proteins with a marked decrease of FKBP12 mRNA in RyR1^–/– myocytes. Such decreased FKBP expression observed in RyR1^–/– detrusor muscle contrasts with the conserved expression of FKBP revealed in skeletal muscle membrane preparations of RyR1^–/– mice by western blot (Buck et al., 1997).

These FKBP downregulations in detrusor muscle are accompanied by a defect of RyR-mediated Ca²⁺ release regulation by rapamycin in RyR1^–/– myocytes. Caffeine-induced Ca²⁺ release is not significantly affected by rapamycin in RyR1^–/– myocytes, whereas it is largely increased in WT myocytes. Moreover, rapamycin rapidly triggers Ca²⁺ sparks and waves in WT myocytes but not in RyR1^–/– myocytes. Taking together that rapamycin uncouples both FKBP12 and FKBP12.6 from their associated RyR, and that FKBP proteins appear to be downregulated in RyR1^–/– myocytes, these results suggest that in RyR1^–/– myocytes the remaining pool of FKBP proteins is not sufficient to allow a rapid demonstration of the activation of RyR2 and RyR3 by rapamycin.

Since RyR2 and RyR3 expression in RyR1^–/– myocytes is unchanged, the lower activity of rapamycin in RyR1^–/– myocytes – which express much less FKBP than WT myocytes – is in agreement with studies showing that detrusor muscle Ca²⁺ signals are affected by FKBP12.6 knockout (Ji et al., 2004). In the RyR1 knockout model, loss of RyR1 appears to be compensated by a decreased content of FKBP to boost Ca²⁺ signals that do not require specific anchoring to the plasma membrane (i.e. spontaneous sparks, caffeine-induced signals).

In conclusion, although the duet FKBP12.6-RyR2 plays an essential physiological role in smooth muscle and cardiac cells (Xin et al., 2002; Ji et al., 2004), we show here that, in addition to signals from RyR2, RyR1 fine-tunes Ca²⁺ signals activated by membrane depolarization in urinary bladder smooth muscle. Our study also reveals that one of the side effects of the RyR1 knockout is a decrease in FKBP protein expression and that this combination of events significantly alters Ca²⁺ signaling in urinary bladder smooth muscle.

C57BL/6 (WT) and Ry1⁵³ (RyR1^+/–) (Buck et al., 1997) mouse strains were used. All animals were kept in a transgenic animal facility (authorization number A-33-063-003). RyR1^+/+, RyR1^–/+ and RyR1^–/– mice were obtained from heterozygous mating of Ry1⁵³ mice and selected by PCR for the neomycin-resistance gene that was at the KpnI site in exon 10 inserted in the Ryr1 gene to disrupt its transcription. All experiments presented are from homozygous RyR1^+/+ or RyR1^–/– mice.

The investigation conforms to the European community and French guiding principles in the care and use of animals. Authorization to perform animal experiments was obtained from the Préfecture de la Gironde (France). Pregnant mice and their embryos were killed by cervical dislocation. Urinary bladders were dissected to remove vessels and adventitia and cut into several pieces. Tissues were incubated (5 minutes, 37°C) in a physiological solution (Hank's balanced salt solution, HBSS) containing 1 mg/ml papain, 1 mM dithiothreitol, washed twice with fresh medium, incubated in HBSS containing 100 μM Ca²⁺ and 1 mg/ml collagenase (10 minutes, 37°C), washed again and placed in an enzyme-free HBSS and triturated with a fire-polished Pasteur pipette to release cells. Cells were seeded on glass plates, maintained in short-term primary culture in M199 medium containing 5% fetal calf serum (FCS), 20 U/ml penicillin and 20 U/ml streptomycin and kept in an incubator at 95% air and 5% CO₂ at 37°C. Myocytes were studied within 4-30 hours of isolation in a physiological solution containing 130 mM NaCl, 5.6 mM KCl, 1 mM MgCl₂, 11 mM glucose and 10 mM HEPES pH 7.4 with NaOH.

Total RNA was extracted from urinary bladder detrusor smooth muscle cells using a RNA preparation kit (Epicentre, Madison, WI), following the instructions of the supplier. The RNA concentration was determined by OD₂₆₀ with an Eppendorf biophotometer (Eppendorf, Le Pecq, France). The reverse transcription reaction was performed on 50 ng RNA using Sensiscript-RT kit (Qiagen, Hilden, Germany). Total RNA was incubated with oligodT₍₁₅₎ primers (Promega, Lyon, France) at 65°C for 5 minutes. After a 5-minute cooling time at 4°C, the RT mix was added and the mixture was incubated for 60 minutes at 37°C. PCR was performed with 0.25 μg cDNA, 1.25 units of HotStartTaq DNA polymerase (Qiagen), 1 mM of each primer and 200 μM of each deoxynucleotide triphosphate, in a final volume of 25 μl. PCR conditions were 95°C for 15 minutes, 35 cycles at 94°C for 1 minute, 60°C for 1 minute and 72°C for 1 minute. Thereafter, samples were kept at 72°C for 10 minutes for final extension and then stored at 4°C. PCRs were performed with a thermal cycler (Eppendorf). Amplification products were separated by electrophoresis (2% agarose gel) and visualized by ethidium bromide staining. Gels were photographed with EDAS120 and analyzed with KDS1D 2.0 software (Kodak Digital Science, Paris, France). Sense (s) and antisense (as) oligonucleotides were designed with lasergene software (DNASTAR, Madison, WI) according to the sequences deposited with GenBank (accession numbers NM008019, NM016863 and XM111014 for FKBP12, FKBP12.6 and GAPDH respectively). The nucleotide sequences and the length of the expected PCR products for each primer pairs were: FKBP12(s) 5′-AGCGATGGGTTAACTTAGAATAGA-3′ and FKBP12(as) 5′-TGCCAACCCAGGGACAGATA-3′ (387 bp); FKBP12.6(s) 5′-CCCCAGGAGACGGAAGGACA-3′ and FKBP12.6(as) 5′-GTGGGGATGATTAAATGGCTGACA-3′ (487 bp); GAPDH(s) 5′-CCCTTATTGACCTCAACTACATGGT-3′ and GAPDH(as) 5′-GAGGGGCCATCCACAGTCTTCTG-3′ (463 bp).

Detrusor muscle from urinary bladder, cardiac and skeletal muscles and brain were homogenized in 10% SDS. After centrifugation, supernatants were collected and the protein content was measured according to the method by Bradford (Bradford, 1976). Total proteins (50 μg or 100 μg of each tissue sample) were treated with Laemmli sample buffer containing 5% β-mercaptoethanol, heated for 5 minutes at 95°C and separated by SDS-polyacrylamide gel electrophoresis (6% separating gel with 3% stacking gel for RyR subtypes and Ins(1,4,5)PR3 and 11% separating gel and 3% stacking gel for FKBP) together with a high-molecular weight standard (Invitrogen, Cergy-Pontoise, France) that enables the identification of proteins up to 500 kDa. The resolved proteins were transferred to a polyvinylidene difluoride (PVDF, Bio-Rad) membrane (1 hour at 100 V at 4°C). Non-specific binding was blocked by incubating PVDF membrane for 1 hour with 50 g/l dry milk in PBS complemented with 0.1% Tween 20 (PBS-T). Blots were incubated overnight at 4°C with the primary antibody (1:500 for anti-RyR1, anti-RyR2 and anti-RyR3; 1:1000 for anti-Ins(1,4,5)PR3; 1:3000 for anti-FKBP). After extensive washes in PBS-T, membranes were incubated 2 hours with a 1:3000 to 1:5000 dilution of horseradish peroxidase (HRP)-coupled anti-rabbit or anti-mouse IgG in PBS-T complemented with 5 g/l dry milk. Specific antigen detection was performed using the chemoluminescent ECLplus western blotting detection system (Amersham Biosciences, Orsay, France). Gels were analyzed with KDS1D 2.0 software. The specific anti-RyR1, anti-RyR2, anti-RyR3 antibodies were supplied by the Sydney Fleischer lab (Vanderbilt University, Nashville, TN) and were produced using the following peptides as epitopes: residues 4476-4486 (11 amino acids) of rabbit RyR1 (Jeyakumar et al., 2002); residues 1344-1365 (22 amino acids) of rabbit RyR2 (Jeyakumar et al., 2001) and residues 4326-4336 (11 amino acids) of rabbit RyR3 (Jeyakumar et al., 1998). The mouse Ins(1,4,5)PR3-specific antibody was obtained from Transduction Laboratory (Lexington, KY) and the rabbit anti-FKBP12-FKBP12.6 antibody was obtained from Affinity Bioreagents (Ozyme, Saint-Quentin, France).

Cells were incubated in physiological solution containing 2 μM fluo-4 acetoxymethylester (Fluo4-AM) for 25 minutes at 37°C. They were then washed and allowed to cleave the dye to the active fluo-4 compound for 10 minutes. Images were acquired using the line-scan mode (6 mseconds per scan) of a Bio-Rad MRC 1024ES (Bio-Rad, Paris, France) confocal attachment connected to a Nikon Diaphot microscope. Excitation light was delivered by a 25 mW argon ion laser (Ion Laser Technology, Salt Lake City, UT) through a Nikon Plan Apo 60×, oil immersion, 1.4 NA objective lens. Fluo-4 was excited at 488 nm and emitted fluorescence was filtered and measured at 540±30 nm. The fluorescence value of each pixel in the line (F) was divided by the fluorescence value of the same pixel at rest levels (F₀), fluorescence signals are thus expressed as pixel per pixel fluorescence ratios (F/F₀). Amplitudes of the responses are expressed as ΔF/F₀, which represents the difference between maximal ratio and ratio at rest level. Image processing and analysis were performed using Lasersharp 2000 software (Bio-Rad) and IDL software (RSI, Boulder, CO), respectively.

Caffeine was applied by pressure ejection from a glass pipette for the period indicated in the figures. All experiments were carried out at 26±1°C. Voltage clamp experiments were carried out with a standard patch-clamp technique using a List EPC-7 patch-clamp amplifier (Darmstadt-Eberstadt). Patch pipettes had resistance of 3-4 MΩ. Intracellular medium contained 130 mM CsCl, 10 mM HEPES pH 7.3, 50 μM Fluo-4.

We investigated the intracellular localization of the three RyR subtypes in WT and RyR1^–/– smooth muscle urinary bladder myocytes using the specific anti-RyR subtype antibodies described above. The immunostaining procedure has been previously described (Fritz et al., 2005). Briefly, cells were fixed in paraformaldehyde 4% vol/vol and glutaraldehyde 0.05%, permeabilized in saponin (PBS, 1 mg/ml saponin, 2% FCS) and incubated overnight with one of the three anti-RyR subtype antibodies (1:100) at 4°C. After extensive washes with an antibody-free permeabilization solution, cells were incubated with the goat anti-rabbit or goat anti-mouse secondary antibodies coupled with Alexa Fluor 488 (1:250, 1 hour, room temperature). Cells were then washed to remove any unfixed secondary antibody, post-fixed and mounted in Vectashield (AbCys, Paris, France) before fluorescence measurements by confocal microscopy in 0.5-μm sections. Localization and intensity of fluorescence were analyzed with IDL software (RSI). Comparison of cells was possible by keeping acquisition parameters constant.

Fluo-4 and Fluo4-AM were from Interchim (Montluçon, France), caffeine was from Merck and ryanodine from Calbiochem (Nottingham, UK). Medium M199, streptomycin, penicillin, papain and collagenase were from Invitrogen (Cergy Pontoise, France). All primers were synthesized by and purchased from Eurogentec (Seraing, Belgium). The secondary HRP-coupled and Alexa Fluor 488-coupled goat anti-mouse- and anti-rabbit-antibodies were obtained from Santa Cruz Biotechnology (Santa-Cruz, CA) and anti-RyRcom antibody directed against the C-terminal region of RyR1, RyR2 and RyR3 was from (Calbiochem, Darmstadt, Germany; catalog number 559279). All other chemicals were from Sigma.

Data are expressed as the mean ± s.e.m.; n represents the number of events. Significance was calculated using Student's t-test. Values of P<0.05 were considered significant.

This work was supported by the Centre National de la Recherche Scientifique, by the Agence Nationale de la Recherche (05-PCOD-029-03 to NM) and NIH (1P01 AR17605 to P.D.A.).