Abstract
Type 2 ryanodine receptor (RyR2) is a Ca2+ release channel on the endoplasmic (ER)/sarcoplasmic reticulum that plays a central role in the excitation-contraction coupling in the heart. Hyperactivity of RyR2 has been linked to ventricular arrhythmias in patients with catecholaminergic polymorphic ventricular tachycardia and heart failure, where spontaneous Ca2+ release via hyperactivated RyR2 depolarizes diastolic membrane potential to induce triggered activity. In such cases, drugs that suppress RyR2 activity are expected to prevent the arrhythmias, but there is no clinically available RyR2 inhibitors at present. In this study, we searched for RyR2 inhibitors from a well-characterized compound library using a recently developed ER Ca2+-based assay, where the inhibition of RyR2 activity was detected by the increase in ER Ca2+ signals from R-CEPIA1er, a genetically encoded ER Ca2+ indicator, in RyR2-expressing HEK293 cells. By screening 1535 compounds in the library, we identified three compounds (chloroxylenol, methyl orsellinate, and riluzole) that greatly increased the ER Ca2+ signal. All of the three compounds suppressed spontaneous Ca2+ oscillations in RyR2-expressing HEK293 cells and correspondingly reduced the Ca2+-dependent [3H]ryanodine binding activity. In cardiomyocytes from RyR2-mutant mice, the three compounds effectively suppressed abnormal Ca2+ waves without substantial effects on the action-potential-induced Ca2+ transients. These results confirm that ER Ca2+-based screening is useful for identifying modulators of ER Ca2+ release channels and suggest that RyR2 inhibitors have potential to be developed as a new category of antiarrhythmic drugs.
SIGNIFICANCE STATEMENT We successfully identified three compounds having RyR2 inhibitory action from a well-characterized compound library using an endoplasmic reticulum Ca2+-based assay, and demonstrated that these compounds suppressed arrhythmogenic Ca2+ wave generation without substantially affecting physiological action-potential induced Ca2+ transients in cardiomyocytes. This study will facilitate the development of RyR2-specific inhibitors as a potential new class of drugs for life-threatening arrhythmias induced by hyperactivation of RyR2.
Introduction
Ryanodine receptors (RyRs) are Ca2+ release channels in the endoplasmic (ER)/sarcoplasmic reticulum and three genetically distinct isoforms of RyR (RyR1–3) with 65%–70% amino acid homology have been identified in mammalian tissues. Among them, RyR2 is indispensable in cardiac excitation-contraction coupling (Bers, 2001; Woll and Van Petegem, 2022; Keefe et al., 2023), in which RyR2 is activated by Ca2+ entry through L-type Ca2+ channel via a Ca2+ induced Ca2+ release mechanism. On the other hand, RyR1 plays a key role in skeletal muscle contraction and opens by depolarization-induced Ca2+ release mechanism via physical interactions with dihydropyridine receptor (Rios and Pizarro, 1991). RyR3 is present in low abundance in various tissues, and several physiologic roles have been reported.
Genetic mutations in RyR2 are associated with various arrhythmogenic heart diseases, including catecholaminergic polymorphic ventricular tachycardia (CPVT), idiopathic ventricular tachycardia, and long QT syndrome (Tester et al., 2004; Medeiros-Domingo et al., 2009; Priori and Chen, 2011; Nozaki et al., 2020; Sun et al., 2021; Hirose et al., 2022; Woll and Van Petegem, 2022). Of these, CPVT mutations are of the gain-of-function type and account for approximately 90% of diseases linked to RyR2 mutations. When gain-of-function-type mutant RyR2 is further activated under strong sympathetic stimulation, spontaneous Ca2+ release occurs without Ca2+ entry from L-type Ca2+ channel. This spontaneous Ca2+ release, in turn, activates inward sodium-calcium exchanger currents to cause delayed afterdepolarization and triggered activity, leading to arrhythmias (Tsien et al., 1979; Lakatta, 1992; Keefe et al., 2023). In addition to the arrhythmic disorders linked to genetic mutations, it has been reported that excessive activation of RyR2 in chronic heart failure (HF) also causes arrhythmias (Dridi et al., 2020; Benitah et al., 2021; Szentandrássy et al., 2022).
The conventional antiarrhythmic drugs for ventricular arrhythmias in CPVT and HF include Na+ channel blockers, β-blockers and Ca2+ channel blockers. However, they sometimes fail to prevent sudden cardiac death. In arrhythmias induced by hyperactivation of RyR2, drugs that inhibit RyR channels are expected to effectively suppress the arrhythmias, but currently there are no clinically available RyR2 specific inhibitors. To date, several drugs, such as carvedilol (Zhou et al., 2011), flecainide (Watanabe et al., 2009), dantrolene (Kobayashi et al., 2010), EL20 (Klipp et al., 2018), and S107 (Lehnart et al., 2008), have been proposed to prevent the hyperactivity of mutated RyR2 (Szentandrássy et al., 2022). However, these drugs also interact with molecules other than RyR2. At present, it is difficult to predict how efficiently RyR2 inhibition itself can suppress arrhythmia. Thus, efforts to find RyR2-specific inhibitors are strongly required.
High-throughput screening is a powerful method for rapidly evaluating many chemical compounds, which greatly accelerates drug discovery. We have reported that the ER Ca2+ concentration ([Ca2+]ER) in HEK293 cells expressing wild-type (WT) and mutant RyRs was inversely correlated with the Ca2+ release activity of RyRs at resting cytosolic Ca2+ concentrations ([Ca2+]cyt) (Murayama et al., 2015, 2016; Kurebayashi et al., 2022), and developed an efficient high-throughput screening platform to find RyR1 inhibitors using HEK293 cells expressing mutant RyR1 and R-CEPIA1er (Murayama et al., 2018; Murayama and Kurebayashi, 2019).
In this study, we applied this method to identify RyR2 inhibitors by screening a chemical library of well-characterized drugs (1535 compounds). We successfully identified three compounds (chloroxylenol, methyl orsellinate, and riluzole) that prevent Ca2+ leakage from the ER to increase [Ca2+]ER in HEK293 cells expressing RyR2. The three hit compounds exhibited RyR isoform specificities and dose-dependent RyR2 inhibition. These compounds suppressed Ca2+ waves in isolated cardiomyocytes without substantial effects on action-potential-induced Ca2+ transients. Our results indicate that RyR2 inhibitors are promising candidates for novel anti-arrhythmic drugs.
Materials and Methods
Generation of Stable and Inducible HEK293 Cell Lines Expressing RyRs
HEK293 cells stably expressing ER Ca2+ sensor protein R-CEPIA1er (Suzuki et al., 2014) together with inducibly expressing wild-type RyR2 were generated as described previously (Murayama et al., 2018; Murayama and Kurebayashi, 2019; Kurebayashi et al., 2022). Briefly, a full-length mouse RyR2 cDNA was cloned in a tetracycline-induced expression vector (pcDNA5/FRT/TO; Life Technologies, Carlsbad, CA, USA) (Tong et al., 1997). Flp-In T-REx HEK293 cells (Thermo Fisher Scientific) were co-transfected with this expression vector and pOG44 Flp-recombinase expression vector in accordance with the manufacturer’s instructions. Clones with suitable doxycycline-induced expression of RyR2 were selected and used for experiments. The cells were then transfected with cDNA of R-CEPIA1er for its stable expression. Cells were cultured in Dulbecco’s Modified Eagle’s Medium supplemented with 10% fetal calf serum, 2 mM L-glutamine, 15 μg/ml blasticidin, 100 μg/ml hygromycin, and 400 μg/ml G418. When testing the effects of compounds on other RyR isoforms, we used HEK293 cells co-expressing RyR1 and R-CEPIA1er, and RyR3 and R-CEPIA1er that were generated by methods similar to those described above (Murayama et al., 2018; Murayama and Kurebayashi, 2019; Kurebayashi et al., 2022). To test the effects of compounds on mutant RyR2s, RyR2-R2474S, -R4497C, -R176Q, -N2386I, phospho-null triple mutant S2807A/S2813A/S2030A (RyR2-S3A), and phospho-mimetic triple mutant (RyR2-S3D) cells (Uehara et al., 2017; Iyer et al., 2020; Kurebayashi et al., 2022) these cells were infected with baculovirus vector to express R-CEPIA1er.
Time-Lapse [Ca2+]ER Measurement Using Multi-Well Plate Fluorometer
Time-lapse [Ca2+]ER measurements were performed using the FlexStation3 fluorometer (Molecular Devices, San Jose, CA, USA), as described previously (Murayama et al., 2018; Murayama and Kurebayashi, 2019). Briefly, HEK293 cells were seeded on 96-well, flat, clear-bottomed black microplates (#3603; Corning, New York, NY, USA) at day 1, expression of RyR2 was induced by adding doxycycline (2 μg/ml) to the culture medium on day 2, and time-lapse ER Ca2+ measurements were carried out on day 3. Before the measurements, the culture medium in individual wells was replaced with 90 μl of HEPES-Krebs solution (140 mM NaCl, 5 mM KCl, 2 mM CaCl2, 1 mM MgCl2, 11 mM glucose, and 5 mM HEPES, pH 7.4). R-CEPIA1er was excited at 560 nm and fluorescence emitted at 610 nm was captured every 10 seconds for 300 seconds. At 100 seconds after the start of measurement, 60 μl of compound solution, containing 25 μM test compound in HEPES-Krebs solution, was applied to the wells of the reading plate at a final concentration of 10 μM. The change in fluorescence induced by the compounds was expressed as F/F0, in which average fluorescence intensity of the last 100 seconds (F) was normalized to that of the initial 100 seconds (F0). We screened 1535 well-characterized drugs in the Tokyo Medical and Dental University (TMDU) chemical compound library at a concentration of 10 μM. Measurements were performed at 37°C.
Analysis of High-Throughput Screening Data
Assay quality was determined based on positive (1 mM tetracaine) and negative [0.1% DMSO] controls, as indexed by Z’ factor: where σP and σN are the S.D.s of positive and negative controls, and μP and μN are the means of positive and negative controls, respectively.
Reagents
The following reagents were purchased after finding hit compounds from the initial TMDU library screening. Chloroxylenol and methyl orserllinate (supply name: methyl 2,4-dihydroxy-6-methylbenzoate) were purchased from Toronto Research Chemicals (Toronto, Canada), riluzole (supply name: 2-amino-6-(trifluoromethyl)benzothiazole) and 4-chloro-3-ethylphenol (4-CEP) from Tokyo Chemical Industry (Tokyo, Japan), and 4-chloro-m-cresol (4-CMC, supply name: 4-chloro-3-methylphenol) from Fujifilm Wako Chemicals (Kyoto, Japan). All the above reagents were dissolved in dimethyl sulfoxide (DMSO) and used at a final concentration of 0.1% DMSO.
Simultaneous Measurement of [Ca2+]ER and [Ca2+]cyt in Single HEK293 Cells Expressing RyR2s
For measurements of Ca2+ signals in individual HEK293 cells, [Ca2+]cyt and [Ca2+]ER signals were simultaneously monitored using genetically encoded Ca2+ indicators, G-GECO1.1 (Zhao et al., 2011) and R-CEPIA1er (Suzuki et al., 2014), respectively (Kurebayashi et al., 2022). Cells were transfected with G-GECO1.1 and R-CEPIA1er cDNA 26–28 hours before measurements. Doxycycline was added to the medium at the same time as transfection. G-GECO1.1 and R-CEPIA1er were excited at 488 nm and 561 nm, respectively, through a 20× objective lens, and light emitted at 525 nm and 620 nm, respectively, was simultaneously captured with an EM-CCD camera at 700 millisecond intervals (Model 8509; Hamamatsu Photonics, Hamamatsu, Japan). [Ca2+]cyt and [Ca2+]ER signals were monitored for 4 minutes in HEPES-Krebs solution, 4 minutes in the presence of a test compound, 3 minutes in HEPES-Krebs solution, and then 1.5 minutes in 10 mM caffeine-containing Krebs solution. At the end of the measurement, the cells were perfused with the following solutions to obtain [Ca2+]cyt and [Ca2+]ER calibrations: 0Ca-Krebs solution (140 mM NaCl, 5 mM KCl, 1 mM MgCl2, 11 mM glucose, 10 mM HEPES, pH 7.4), BAPTA-0Ca-Krebs solution containing 5 mM 1,2-bis(o-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid (BAPTA) and 20 μM ionomycin, 0Ca-Krebs solution, and then 20Ca-Krebs solution containing 20 mM CaCl2 and 20 μM ionomycin (Kurebayashi et al., 2022). Fmin and Fmax values were obtained with the BAPTA-0Ca-Krebs solution and 20Ca-Krebs solution, respectively.
To examine the effects of test compounds on [Ca2+]cyt oscillations, cells were loaded with 4 µM fluo-4 AM in culture medium for 30 minutes at 37°C and then incubated with HEPES Krebs solution. Fluo-4 was excited at 488 nm through a 20× objective lens and light emitted at 525 nm was captured with the EM-CCD camera at 700-millisecond intervals. All measurements were performed at 26°C by perfusing solutions using an in-line solution heater/cooler (Warner Instruments, Holliston, MA, USA).
[3H]Ryanodine Binding
The assay was carried out as described previously (Kurebayashi et al., 2022). In brief, microsomes prepared from HEK293 cells stably expressing the RyR2s were incubated with 5 nM [3H]ryanodine for 1 hour at 25°C in medium containing 0.17 M NaCl, 20 mM 3-morpholino-2-hydroxypropanesulfonic acid (pH 7.0), 2 mM dithiothreitol, 1 mM adenosine monophosphate, 1 mM MgCl2, and various concentrations of free Ca2+ buffered with 10 mM EGTA. Free Ca2+ concentrations were calculated using WEBMAXC STANDARD (https://somapp.ucdmc.ucdavis.edu/pharmacology/bers/maxchelator/webmaxc/webmaxcS.htm). Protein-bound [3H]ryanodine was separated by filtration through polyethyleneimine-treated glass filters (Filtermat B; PerkinElmer, Waltham, MA) using a Micro 96 Cell Harvester (Skatron Instruments, Lier, Norway). Nonspecific binding was determined in the presence of 20 μM unlabeled ryanodine.
Ca2+ Imaging in Isolated Mouse Single Cardiac Myocytes
All animal-handling procedures were in accordance with the guidelines and approved by the ethics committees of Juntendo University School of Medicine. For Ca2+ imaging in isolated cardiomyocytes, 2-month-old WT and homozygous RyR2-R420W (Okudaira et al., 2014) and 3-month-old homozygous Tnnt2-ΔK210 mice (Du et al., 2007; Odagiri et al., 2014) were used.
Mice were deeply anesthetized with pentobarbital sodium (100 mg/kg i.p.), and their hearts were excised and rinsed in Krebs solution. Single cardiac myocytes were isolated from the ventricles of mice using an established enzymatic method (Shioya, 2007) and loaded with Cal520-AM (AAT Bioquest, Inc., Pleasanton, CA, USA). Cells were field-stimulated at 0.5 Hz in HEPES-buffered Tyrode’s solution (140 mM NaCl, 5 mM KCl, 1.5 mM CaCl2, 1 mM MgCl2, 0.3 mM NaH2PO4, 11 mM glucose, 10 mM HEPES, pH 7.4) with or without test compounds. Cal520 was excited at 488 nm through a 20× objective lens and light emitted at 525 nm was captured with an EM-CCD camera at 8–14 millisecond intervals (Model 8509; Hamamatsu Photonics). Fluorescence signals (F) of Cal520 in individual cells were determined using region of interest (ROI) analysis, from which the cell-free background fluorescence was subtracted. To confirm intracellular Ca2+ signals with high resolution, video images were acquired at 6-millisecond intervals using a 60x objective lens. All measurements were carried out at 30°C by perfusing solutions through an in-line solution heater/cooler (Warner Instruments, Holliston, MA, USA).
Statistics
Data are presented as the mean ± S.D. Statistical analysis was performed using Prism 9 (GraphPad Software, Inc., La Jolla, CA). Unpaired Student’s t test was used for comparisons between two groups. One-way analysis of variance (ANOVA) followed by Dunnett’s test was performed to compare multiple groups with one factor, while two-way ANOVA followed by Tukey’s test was used to compare multiple groups in experiments with two factors. Statistical significance was defined as P < 0.05 when compared with the negative control. These analyses were performed as planned and recorded in the protocols.
Results
Validation of Screening Platform for RyR2 Inhibitors
In this study, we used HEK293 cells stably expressing R-CEPIA1er (Suzuki et al., 2014) together with inducibly expressing WT RyR2 as a screening platform for RyR2 inhibitors. Fig. 1A illustrates the concept of the ER Ca2+-based screening for RyR2 inhibitors. [Ca2+]ER is generally determined by the balance between Ca2+ release via the Ca2+ release channels and Ca2+ uptake by sarcoplasmic/endoplasmic reticulum Ca2+-ATPase Ca2+ pumps (Murayama et al., 2018; Kurebayashi et al., 2022). We have demonstrated that expression of WT and CPVT mutant RyR2s reduce [Ca2+]ER by spontaneous Ca2+ release and that [Ca2+]ER is inversely correlated with the channel activity: channels with greater activity cause more reduced [Ca2+]ER (Fujii et al., 2017; Kurebayashi et al., 2022). Because induction of RyR2 expression causes spontaneous Ca2+ release to moderately lower [Ca2+]ER in HEK293 cells (Figs. 1, A-a and A-b) (Fujii et al., 2017; Kurebayashi et al., 2022), the inhibition of RyR2 by drugs will increase [Ca2+]ER by means of Ca2+ uptake (Fig. 1A-c). In contrast, further activation of RyR2 by drugs will enhance Ca2+ release, leading to a further decrease in [Ca2+]ER (Fig. 1A-d).
Fig. 1B shows representative actual effects of tetracaine and caffeine, a known inhibitor and activator of ryanodine receptors (RyRs), respectively, on simultaneously measured [Ca2+]cyt and [Ca2+]ER signals in single HEK293 cells expressing WT RyR2. As previously reported, cells expressing RyR2 show spontaneous [Ca2+]cyt oscillations and corresponding periodic [Ca2+]ER decrease (Jiang et al., 2007; Uehara et al., 2017; Kurebayashi et al., 2022). The [Ca2+]ER level was increased by an application of tetracaine (1 mM) with termination of spontaneous [Ca2+]cyt oscillations (left), whereas the [Ca2+]ER was substantially decreased by the application of caffeine (3 mM) (right).
Fig. 1C shows representative effects of tetracaine and caffeine on the R-CEPIA1er fluorescence signals in WT RyR2 cells measured with a FlexStation3 microplate reader. The cells were seeded in 96-well plates on the first day and RyR2 expression was induced by doxycycline on the next day. Twenty-four hours after the induction, R-CEPIA1er fluorescence signals were measured using the FlexStation3 fluorometer, which obtains the summation of fluorescence signals from many cells in the optical field (1 mm diameter). The application of 3 mM caffeine rapidly decreased the R-CEPIA1er signal, whereas tetracaine (1 mM) gradually increased the fluorescence signal, which reached a plateau at around 200 seconds. The fluorescence ratio (F/F0) before and after the application of compounds was determined by normalizing the average fluorescence intensity for the last 100 seconds (F) to that for the initial 100 seconds (F0) (Fig. 1C). These results are consistent with the above idea that [Ca2+]ER monitors RyR2 activity.
To quantitatively validate the assay system, we determined the coefficient of variation (CV), the ratio of the standard deviation to the mean and Z’ factor using 96 wells in each condition. Histograms of F/F0 showed that fluorescence signals changed by tetracaine (μ = 1.675 and σ = 0.086) and caffeine (μ = 0.517, σ = 0.032), where μ and σ are the mean and S.D., respectively. were perfectly separated from that changed by DMSO (μ = 0.975 and σ = 0.023) (Fig. 1D). Coefficient of variation values for tetracaine, caffeine, and DMSO were 0.05, 0.06, and 0.02, respectively, which satisfied the criterion for high-throughput screening (coefficient of variation < 0.1). The Z’ factors were calculated by eq. 1 as 0.533 and 0.640 for tetracaine and caffeine, respectively (see Materials and Methods), which were within the acceptable range (Z’ > 0.5).
Screening of a Chemical Compound Library for RyR2 Inhibitors by [Ca2+]ER Measurements
Using this screening platform based on [Ca2+]ER measurements, we screened a TMDU chemical compound library of well-characterized drugs (1535 compounds) at 10 μM, which was the same as that used for the search for RyR1 inhibitors (Murayama et al., 2018). The overall results with WT RyR2 cells in duplicate assays are shown in Fig. 1E. We identified three compounds (#1–#3) with F/F0 values greater than 1.5 (Fig. 1E). They were chloroxylenol (#1), methyl orsellinate (#2), and riluzole (#3) (Fig. 1F).
Dose-Dependent Effects of Hit Compounds on WT and Mutant RyR2s
To further characterize the hit compounds, we examined dose-dependent effects on [Ca2+]ER in RyR2 WT cells using the same assay system (Fig. 2A). Chloroxylenol showed the highest potency (EC50 = 0.3 μM), followed by methyl orsellinate (EC50 = 1.1 μM) and riluzole (EC50 = 9.9 μM). Since RyR2 inhibitors are expected to be used for patients carrying mutant RyR2, we also tested these compounds on CPVT-linked mutant RyR2s, R2474S and R4497C, which strongly and moderately activate the channel activity, respectively (Kurebayashi et al., 2022). The three compounds also similarly suppressed the CPVT-linked mutant RyR2s, R2474S and R4497C (Fig. 2, B and C), as well as WT RyR2. The inhibitory effects of these compounds were also confirmed with arrhythmogenic right ventricular dysplasia/cardiomyopathy-associated mutant RyR2s such as R176Q (Kannankeril et al., 2006) and N2386I (Tiso et al., 2001), as well as RyR2 mutants at the three important phosphorylation sites (Huke and Bers, 2008; Lanner et al., 2010; Potenza et al., 2019), a phospho-null triple mutant S2807A/S2813A/S2030A (RYR2 S3A), and a phospho-mimetic triple mutant of RyR2 S2807D/S2813D/S2030D (RyR2 S3D) (Supplemental Fig. 1).
To examine the RyR isoform specificity, the effects of these compounds on WT RyR1-WT and RyR3-WT cells were also examined (Fig. 2, D and E). Because [Ca2+]ER of RyR1-WT and RyR3-WT cells was almost fully loaded due to least spontaneous Ca2+ release activity (Murayama et al., 2018), measurements were performed in the presence of 4 mM (for RyR1-WT) or 3 mM (for RyR3-WT) caffeine, which enhances RyR channel activity to moderately reduce [Ca2+]ER. The addition of tetracaine increased [Ca2+]ER (see dotted lines in Fig. 2, D and E). Riluzole slightly increased [Ca2+]ER in the RyR1-WT cells at 30 μM, whereas chloroxylenol and methyl orsellinate clearly reduced [Ca2+]ER in the RyR1-WT at 3 μM or higher (Fig. 2D). These compounds had little effect on RyR3 (Fig. 2E). Therefore, the three compounds all suppress RyR2, although chloroxylenol and methyl orsellinate have some activating effects on RyR1.
The chemical structure of chloroxylenol is similar to those of 4-CMC and 4-CEP, which are well-known potent RyR1 activators (Zorzato et al., 1993; Westerblad et al., 1998) (Fig. 2F). We therefore examined the effects of these drugs on RyR1- and RyR2-expressing cells. The 4-CEP at 0.3–10 μM increased F/F0 in RyR2 cells, indicating that 4-CEP also has some RyR2-inhibiting effect (Fig. 2G), whereas 4-CMC did not exhibit RyR2 inhibition. All of these compounds showed trends of decreasing F/F0 of RyR2 cells from the peak at 30 μM, suggesting that they have a Ca2+-releasing effect on RyR2 at higher concentrations, i.e., 100–300 μM or more. On RyR1-expressing cells, these compounds all similarly reduced F/F0, at 3 µM or more, indicating their Ca2+-releasing effects on RyR1 (Fig. 2H). These results indicate that the activating effect of chloroxylenol on RyR1 is similar to those of 4-CMC and 4-CEP (Zorzato et al., 1993; Westerblad et al., 1998).
Effects of Hit Compounds on [3H]ryanodine Binding
Next, we examined the effects of the three hit compounds on RyR2 activity via a Ca2+-dependent [3H]ryanodine binding assay (Fig. 3A). Since ryanodine specifically binds open RyR channels, [3H]ryanodine binding reflects the RyR2 channel activity (Fujii et al., 2017; Kurebayashi et al., 2022). Chloroxylenol (3 μM) and methyl orsellinate (10 μM) suppressed maximal activity and shifted the Ca2+ dependence to the right (Fig. 3A, left). Riluzole (10 and 100 μM) showed similar effects on [3H]ryanodine binding (Fig. 3A, right). Fig. 3B shows the activity measured at pCa 5, a Ca2+ concentration close to that under physiologic conditions. All three compounds suppressed [3H]ryanodine binding with IC50s similar to those obtained with the [Ca2+]ER-based assay (chloroxylenol, 0.13 μM; methyl orsellinate, 1.6 μM: riluzole, 5.3 μM).
Effects of Hit Compounds on Single-Cell Ca2+ Homeostasis in HEK293 Cells Expressing RyR2
The above screening results indicate that RyR2 activity is suppressed by these hit compounds. Because HEK293 cells expressing RyR2 show spontaneous Ca2+ oscillations (Jiang et al., 2007; Uehara et al., 2017; Kurebayashi et al., 2022) as shown in Fig. 1B, we investigated how these compounds affect Ca2+ oscillations caused by RyR2. Fig. 4A shows simultaneous measurements of [Ca2+]cyt and [Ca2+]ER signals from individual HEK293 cells expressing WT RyR2. Chloroxylenol (1 μM), methyl orsellinate (10 μM), and riluzole (10 μM) completely suppressed the spontaneous Ca2+ oscillations associated with the increased [Ca2+]ER (Fig. 4B). The washout of the drugs immediately restored the Ca2+ oscillation, indicating that the effects of these compounds are reversible.
We have previously reported that the enhancement of RyR2 activity by CPVT-linked mutations will result in an increase in Ca2+ oscillation frequency and a decrease in the amplitude of Ca2+ transients due to the decrease in [Ca2+]ER (Kurebayashi et al., 2022). We investigated whether these RyR2 inhibitors could make the enhanced Ca2+ oscillations in CPVT mutant RyR2s close to that of WT RyR2. Fig. 4C shows the effects of various concentrations of chloroxylenol, which has the highest affinity for RyR2 among the three compounds, on cytoplasmic Ca2+ oscillations in cells expressing WT and CPVT-linked mutant RyR2s. In RyR2-WT cells, Ca2+ oscillation was substantially reduced at 0.3 μM and almost disappeared at 1 μM. Compared with the WT cells, mutant RyR2 cells, R2474S and R4497C cells exhibited more frequent with smaller-amplitude Ca2+ oscillations in normal Krebs solution (Fig. 4, C–E). The addition of chloroxylenol to R2474S and R4497C cells increased the amplitudes of Ca2+ transients and decreased the frequencies of Ca2+ oscillations in a dose-dependent manner (Fig. 4, C–E). Notably, oscillation frequency and peak amplitude of mutant cells in the presence of 0.3 μM chloroxylenol were similar to those of WT cells in the absence of the drug, suggesting that RyR2 inhibitors can ameliorate abnormal Ca2+ homeostasis in RyR2-expressing cells at around their IC50s.
Effects of Hit Compounds on Ca2+ Signals in Isolated Cardiomyocytes from Disease Model Mice
The above results demonstrate that the hit compounds suppress RyR2 in HEK293 cells. Next, we investigated whether the hit compounds can suppress abnormal Ca2+ signals, such as Ca2+ waves in cardiomyocytes. For this purpose, we used cardiomyocytes from mice harboring CPVT-linked RyR2 mutation, R420W (Nishio et al., 2006; Okudaira et al., 2014). The R420W cardiomyocytes stimulated at 0.5 Hz showed few Ca2+ waves in normal Krebs solution but showed frequent Ca2+ waves in the presence of 10−7 M isoproterenol (ISO) (Fig. 5, A and B, Supplemental Video 1). We tested the effects of the RyR2 inhibitors at a concentration near their IC50s (i.e., 0.3 μM for chloroxylenol, 3 μM for methyl orsellinate, and 10 μM for riluzole) on ISO-induced Ca2+ waves. Fig. 5C shows typical Ca2+ transients and Ca2+ waves in RyR2-R420W cardiomyocytes stimulated at 0.5 Hz. As shown in Fig. 5, C-a and D-a, Ca2+ waves were observed in normal Krebs and after 5 minute of incubation with vehicle (0.1% DMSO). In the presence of 0.3 μM chloroxylenol, cardiomyocytes showed normal Ca2+ transient in response to field stimulation; however, the frequency of Ca2+ waves dramatically decreased (average ninefold reduction compared with ISO) (Fig. 5, C-b and 5D-b, also see Supplemental Videos 1 and 2). High resolution videos in the presence of isoproterenol alone and isoproterenol plus chloroxylenol are shown in Supplemental Video 3 and Supplemental Video 4 respectively. Similar reduction in Ca2+ frequencies were also seen in the presence of 3 μM methyl orsellinate (average 10-fold) (Fig. 5, C-c and D-c) and 10 μM riluzole (average 14-fold) (Fig. 5, C-d and D-d).
We also examined the effects of the three hit compounds on Ca2+ waves in cardiomyocytes from dilated cardiomyopathy model mice, tnnt2 ΔK210, which develop progressive severe heart failure and ventricular arrhythmia with aging (Suzuki et al., 2012; Odagiri et al., 2014). These mice exhibit increased phosphorylation levels of RyR2 in the heart, suggesting enhanced RyR2 activity in the tnnt2 ΔK210 mouse heart (Du et al., 2007). Cardiomyocytes from the dilated cardiomyopathy mice at 3 months old showed frequent Ca2+ waves (19.9 ± 4.6%, n = 6 mice), in normal Krebs solution (Fig. 6a, before drug). Ca2+ wave frequencies were measured in normal Krebs solution and then after 5 minute of incubation in Krebs solution containing vehicle alone (0.1% DMSO), 0.3 μM chloroxylenol, 3 μM methyl orsellinate, or 10 μM riluzole. None of these compounds inhibited the generation of action-potential-induced Ca2+ transients (Fig. 6A). The average Ca2+ wave frequency was increased 1.4-fold in control Krebs solution with vehicle (Fig. 6B-a), but the difference was not statistically significant. This increase is likely due to the natural progression of injury over time. Chloroxylenol (Fig. 6, A-b and B-b) and riluzole (Fig. 6, A-d and B-d) clearly reduced the occurrence of Ca2+ waves (average 3.3- and 25-fold compared with the pretreatment, respectively). Methyl orsellinate did not show statistically significant reduction in the proportion of cells showing Ca2+ waves (Fig. 6, A-c and B-c).
In Ca2+ transient measurements in mutant cardiomyocytes, RyR2 inhibitors suppressed Ca2+ waves but did not suppress the generation of action-potential-induced Ca2+ transients. Because it is difficult to analyze Ca2+ transients in cells that are prone to Ca2+ waves, we further examined the effects of these RyR2 inhibitors on Ca2+ transients in cardiomyocytes from WT mice. The peak amplitude of Ca2+ transients in WT cells was not appreciably altered by these inhibitors. The average amplitudes were only slightly increased by chloroxylenol (mean difference = +4.8%, 95% confidence interval (CI) = −1.3 – +11.0%) and methyl orsellinate (mean difference = +6.4%, 95% CI = +1.1 – +11.8%) and decreased by riluzole (mean difference = −7.5%, 95% CI = −1.8 – −13.3%). These results indicate that the RyR2 inhibitors found in this study effectively suppress Ca2+ waves at around their EC50s without greatly affecting the peak amplitudes and time courses of action-potential-induced Ca2+ transients (Fig. 6, C and D).
Discussion
In this study, we found that three compounds—chloroxylenol, methyl orsellinate, and riluzole—have marked RyR2 inhibitory effects, by searching a well-known compound library using ER Ca2+ signal-based assay. These compounds reversibly suppressed Ca2+ oscillations and elevated [Ca2+]ER in RyR2-expressing HEK293 cells and dose-dependently suppressed Ca2+-dependent [3H]ryanodine binding. Notably, these compounds suppressed Ca2+ waves in cardiomyocytes from CPVT model mice and dilated cardiomyopathy mice. The fact that the three unrelated compounds showed similar effects on RyR2-expressing cells supports the idea that RyR2 inhibitors are promising as antiarrhythmic drugs.
On the basis of the finding that [Ca2+]ER is inversely correlated with the channel activity of the RyR1 and RyR2 mutants in HEK293 cells (Murayama et al., 2015, 2016; Kurebayashi et al., 2022), we recently identified novel RyR1 inhibitors with the screening platform by time-lapse fluorescence measurement of [Ca2+]ER using R-CEPIA1er (Murayama et al., 2018). Using the same procedure, we here identified three potent RyR2 inhibitors with IC50s of less than 10 μM in the well-known compound library. Because this method targets RyR2 on the ER membrane in living cells, only compounds that can pass through the cell membrane are detected as hits. All the hit compounds identified by our method certainly inhibited [3H]ryanodine binding and suppressed abnormal Ca2+ waves without affecting action-potential induced Ca2+ transients in cardiomyocytes from the disease-linked mouse models. Thus, our method is highly efficient in identifying membrane permeable RyR2 inhibitors needed for clinical use.
Riluzole is a therapeutic agent for amyotrophic lateral sclerosis, and is known to act on various channels, including neuronal Na+ channel and glutamate receptors (Cheah et al., 2010). Although its effect on RyR2 may be one of its non-specific actions, it is characterized by a higher affinity for RyR2 than for RyR1 and RyR3. Interestingly, Radwanski et al. reported that riluzole suppressed diastolic Ca2+ release in cardiomyocytes from CPVT linked CASQ2-mutant mice (CASQ2-R33Q) (Radwanski et al., 2015) and suggested that neuronal Na+ channel contributes to arrhythmogenic diastolic Ca2+ release, thereby providing the mechanism behind its potential for antiarrhythmic therapy. Since the 10 µM riluzole they used has a strong RyR2 inhibitory effect, at least part of the riluzole’s effect on Ca2+ abnormalities in cardiomyocyte may be due to the RyR2 inhibitory action. Because therapeutic plasma concentration is reported to be approximately 1–2 µM (Sarkar et al., 2017), slightly higher doses of the drug may be required for effective suppression of arrhythmia.
Chloroxylenol is an antiseptic and disinfecting agent, and is most effective against gram-positive bacteria. The chemical structure of chloroxylenol is similar to those of 4-CMC and 4-CEP, well-known activators of RyR1, which stimulates Ca2+-dependent [3H]ryanodine binding with an EC50 of approximately 100 μM or higher (Zorzato et al., 1993; Westerblad et al., 1998). Chloroxylenol was found to suppress RyR2 with an IC50 of 0.3 μM, but it showed a trend of activation at a concentration of 30 μM or more, whereas it only activated RyR1 without any inhibitory effect. Because this compound distinguishes between RyR1 and RyR2 at low concentrations, it would be interesting to know the binding sites at low and high concentrations in RyR2 molecules. Methyl orsellinate also showed a dose-dependent biphasic effect similar to that of chloroxylenol (Fig. 2). Chloroxylenol and methyl orsellinate may act on the same binding site in RyR2.
Notably, all of these compounds suppressed Ca2+ waves in cardiomyocytes from RyR2 R420W and tnnt2 ΔK210 mice at around their IC50s, but only slightly affected peak amplitudes of Ca2+ transients in cardiomyocytes from WT mice. This important property can be explained by the mechanism of inhibition of the compounds (Fig. 3): they all shift Ca2+ dependence to the right and mildly suppress the maximal activity, leading to a greater suppression of Ca2+ release at lower [Ca2+]cyt than at higher [Ca2+]cyt. As a result, local Ca2+ waves/sparks generated at low [Ca2+]cyt may be effectively suppressed, but action-potential-evoked Ca2+ transients caused by massive Ca2+ influx via L-type Ca2+ channel may be less suppressed. Interestingly, these compounds were able to ameliorate the frequent Ca2+ release by gain-of-function mutant RyR2 to a level closer to that of the WT in HEK293 cells.
It is widely believed that drugs that suppress RyR2 have an antiarrhythmic effect, and indeed several drugs have been reported to modify RyR2 activity (Szentandrassy et al., 2022). However, in many cases, these drugs also have strong effects on other target proteins, and so this concept has yet to be definitively confirmed. Flecainide is an effective antiarrhythmic drug that mainly targets Na+ channels, but its action on RyR2 has been reported to contribute to its antiarrhythmic effect (Watanabe et al., 2009; Bannister et al., 2022; Salvage et al., 2022). A β-blocker, carvedilol, has also been reported to suppress abnormal Ca2+ release via RyR2 (Zhou et al., 2011). There is thus a need to clarify whether and how RyR2 inhibitors exhibit antiarrhythmic effects using strong RyR2 inhibitory effects.
EL20, which is an analog of tetracaine, has been reported to potently inhibit abnormal Ca2+ release from RyR2 mutant mice and induced pluripotent stem cell-derived cardiomyocytes (iPSC-CM) (Klipp et al., 2018; Word et al., 2021). This compound strongly inhibited single channel activity of RyR2 (IC50=8.2 nM) in the absence of calmodulin. However, in the presence of calmodulin, EL20 showed a much weaker effect (IC50=3–10 μM) on the single channels and on Ca2+-dependent [3H]ryanodine binding in cardiac microsomes. The unnatural verticilide exhibits antiarrhythmic activity and appears to be RyR2-specific, but its inhibitory effect on [3H]ryanodine binding is partial (Batiste et al., 2019). A group of compounds such as Rycal drugs that stabilize the association of FK binding protein 12.6 (FKBP12.6/calstabin2) with RyR2 has also been reported to ameliorate abnormal Ca2+ release in cardiomyocytes (Andersson and Marks, 2010). Compared with these drugs/compounds previously reported to suppress RyR2 activity, the compounds found in our study clearly suppress [3H]ryanodine binding, especially at Ca2+ concentrations of less than 10 μM. Importantly, the inhibitory effects of the three compounds were not mediated by associated proteins (FKBP12.6 and calmodulin) or phosphorylations. They inhibited FKBP12.6-unbound RyR2 in HEK293 cells that do not express FKBP12.6. They also appear to suppress both calmodulin-unbound and calmodulin-bound RyR2, as the amount of calmodulin in RyR2 expressing HEK293 cells is insufficient in molar ratio compared with the amount of RyR2 (Gao et al., 2023). Furthermore, they appear to inhibit phosphorylated and non-phosphorylated RyR2s in a similar manner, as they suppressed both phospho-null (RyR2-S3A) and phospho-mimetic mutants (RyR2-S3D) (Supplemental Fig. 1).
In addition to their use as antiarrhythmic drugs, RyR2 selective inhibitors are expected to be beneficial agents for the other diseases. In the hearts of HF and diabetic patients, various ion channels and regulatory proteins have been reported to be modulated. Among them, RyR2 can be modulated by association/dissociation of FKBP12.6 (calstabin2), calmodulin, triadin/calsequestrin, calcineurin, and transmembrane protein 38A (TMEM-38A) (Yuan et al., 2014; Santulli et al., 2018; Kansakar et al., 2021), as well as phosphorylation, oxidation, S-nitrosylation, and glycation of RyR2 (Kansakar et al., 2021; Kobayashi et al., 2021; Tian et al., 2021; Gambardella et al., 2022). Furthermore, RyR2 has also been implicated in diseases in the brain, such as Alzheimer’s disease and epilepsy (Lehnart et al., 2008; Lacampagne et al., 2017; Lieve et al., 2019) and in pancreas (Santulli et al., 2018). RyR2-selective inhibitors might help to treat these diseases.
Limitations and Perspectives
Chloroxylenol and methyl orsellinate have activating effects on RyR1, and riluzole has insufficient affinity for RyR2 inhibition, and thus we are not ready to test these compounds as antiarrhythmic agents on animals in their current form. In this regard, it is necessary to develop compounds with higher affinity and selectivity for RyR2 by structural modification of the three lead compounds. For that purpose, analysis of ultrafine structure of compound-bound RyR2 by cryo electron microscopy would be helpful (Santulli et al., 2018; Iyer et al., 2020; Kobayashi et al., 2022; Woll and Van Petegem, 2022). Our screening method should greatly contribute to the search for novel RyR2 inhibitors as a new category of antiarrhythmic drugs.
Data Availability
All data supporting the findings in this study are available upon request.
Acknowledgments
The authors are grateful to Montserrat Samso for the suggestion to create the RyR2-S3A and RyR2-S3D cell lines. The authors thank Ikue Hiraga, and Mirei Takahashi for their technical assistance and Yutaka Hirata for helpful advice on usage of the mutant mice. They are grateful for the Center for Biomedical Research Resources and Laboratory of Radioisotope Research, Research Support Center, Juntendo University Graduate School of Medicine. The authors also thank Edanz (https://jp.edanz.com/ac) for editing a draft of this manuscript.
Authorship Contributions
Participated in research design: Murayama, Kagechika, Sakurai, Kurebayashi.
Conducted experiments: Takenaka, Kodama, Murayama, Kurebayashi.
Contributed new reagents or analytic tools: Ishigami-Yuasa, Mori, Ishida, Suzuki, Kanemaru, Iino, Miura, Nishio, Morimoto, Kagechika.
Performed data analysis: Takenaka, Kodama, Murayama, Ishigami-Yuasa, Sugihara, Kurebayashi.
Wrote or contributed to the writing of the manuscript: Takenaka, Kodama, Murayama, Kurebayashi.
Footnotes
- Received May 9, 2023.
- Accepted August 29, 2023.
This work was supported by JSPS KAKENHI [Grants 19K07105 and 22K06652] (to N.K.), [Grants 19H03404 and 22H02805] (to T.M.), [Grant 22K15244] (to R.I.), the Practical Research Project for Rare/Intractable Diseases [Grant 19ek0109202] (to N.K.) from the Japan Agency for Medical Research and Development (AMED), Platform Project for Supporting Drug Discovery and Life Science Research Basis for Supporting Innovative Drug Discovery and Life Science Research (BINDS) [Grant JP19am0101080] (to T.M. and N.K.) and [Grant JP20am0101098] (to H.K.), the Cooperative Research Project of Research Center for Biomedical Engineering (to H.K.), an Intramural Research [Grant 2-5] (to T.M.) for Neurologic and Psychiatric Disorders from the National Center of Neurology and Psychiatry (to T.M.), and the Vehicle Racing Commemorative Foundation [Grant 6303] (to T.M.).
The authors declare no competing financial interests.
↵1M.T., M.K., and T.M. contributed equally to this work.
↵This article has supplemental material available at molpharm.aspetjournals.org.
Abbreviations
- 4-CEP
- 4-chloro-3-ethylphenol
- 4-CMC
- 4-chloro-m-cresol
- [Ca2+]cyt
- cytosolic Ca2+ concentrations
- [Ca2+]ER
- ER Ca2+ concentrations
- CI
- confidence interval
- CPVT
- catecholaminergic polymorphic ventricular tachycardia
- ER
- endoplasmic reticulum
- RyR
- ryanodine receptor
- TMDU
- Tokyo Medical and Dental University
- WT
- wild-type
- Copyright © 2023 The Author(s)
This is an open access article distributed under the CC BY-NC Attribution 4.0 International license.