Decision letter: Improvement of muscle strength in a mouse model for congenital myopathy treated with HDAC and DNA methyltransferase inhibitors

Abstract

Article Figures and data Abstract Editor's evaluation Introduction Results Discussion Materials and methods Appendix 1 Appendix 2 Appendix 3 Data availability References Decision letter Author response Article and author information Metrics Abstract To date there are no therapies for patients with congenital myopathies, muscle disorders causing poor quality of life of affected individuals. In approximately 30% of the cases, patients with congenital myopathies carry either dominant or recessive mutations in the ryanodine receptor 1 (RYR1) gene; recessive RYR1 mutations are accompanied by reduction of RyR1 expression and content in skeletal muscles and are associated with fiber hypotrophy and muscle weakness. Importantly, muscles of patients with recessive RYR1 mutations exhibit increased content of class II histone deacetylases and of DNA genomic methylation. We recently created a mouse model knocked-in for the p.Q1970fsX16+ p.A4329D RyR1 mutations, which are isogenic to those carried by a severely affected child suffering from a recessive form of RyR1-related multi-mini core disease. The phenotype of the RyR1 mutant mice recapitulates many aspects of the clinical picture of patients carrying recessive RYR1 mutations. We treated the compound heterozygous mice with a combination of two drugs targeting DNA methylases and class II histone deacetylases. Here, we show that treatment of the mutant mice with drugs targeting epigenetic enzymes improves muscle strength, RyR1 protein content, and muscle ultrastructure. This study provides proof of concept for the pharmacological treatment of patients with congenital myopathies linked to recessive RYR1 mutations. Editor's evaluation The paper describes improvement in muscle phenotype of a congenital myopathy mouse model by a combined treatment with pharmacological inhibitors of Class IIa histone deacetylases and DNA methylases. The paper demonstrates in principle that there are treatment avenues to pursue but their application could be limited as phenotypic rescue appears to be restricted to particular muscle fiber types. https://doi.org/10.7554/eLife.73718.sa0 Decision letter Reviews on Sciety eLife's review process Introduction Skeletal muscle contraction is initiated by a massive release of Ca2+ from the sarcoplasmic reticulum (SR) via the opening of the ryanodine receptor 1 (RyR1), a calcium release channel, which is localized in the SR terminal cisternae (Rios and Pizarro, 1991; Endo, 1977; Fleischer and Inui, 1989). The signal causing the opening of the RyR1 is the depolarization of the sarcolemmal membrane, which is sensed by voltage-dependent L-type Ca2+ channels (dihydropyridine receptor [DHPR]) located in invaginations of the sarcolemma referred to as transverse tubules (TTs) (Rios and Pizarro, 1991; Franzini-Armstrong and Jorgensen, 1994). Communication between DHPR and RyR1 occurs in specialized intracellular junctions between TT and SR called Ca2+ release units (CRUs). Skeletal muscle relaxation is brought about by SR Ca2+ uptake via the activity of the sarco(endo)plasmic reticulum CaATPAses (SERCA) (MacLennan, 2000). Dis-regulation of Ca2+ signals due to defects in key proteins (RyR1 and DHPR) involved in excitation-contraction (EC) coupling (ECC) is the underlying feature of several neuromuscular disorders (Jungbluth et al., 2018; Treves et al., 2008; Lawal et al., 2020). In particular, mutations in RYR1, the gene-encoding RyR1, are causative of malignant hyperthermia (MH; MIM #145600), central core disease (CCD; MIM #11700), specific forms of multi-minicore disease (MmD; MIM # 255320), and centronuclear myopathy (Jungbluth et al., 2018; Treves et al., 2008; Lawal et al., 2020; MacLennan and Philips, 1992). A great deal of data has shown that RYR1 mutations result mainly in four types of channel defects (Treves et al., 2008). One class of mutations (dominant, MH-associated) causes the channels to become hypersensitive to activation by electrical and pharmacological stimuli (MacLennan and Philips, 1992). The second class of RYR1 mutations (dominant, CCD-associated) results in leaky channels leading to depletion of Ca2+ from SR stores (Treves et al., 2008; Lawal et al., 2020). A third class of RYR1 mutations also linked to CCD causes EC uncoupling, whereby activation of the voltage sensor Cav1.1 is unable to cause release of Ca2+ from the SR (Avila et al., 2003). The fourth class comprises recessive mutations, which are accompanied by a decreased content of mutant RyR1 channels on SR membranes (Wilmshurst et al., 2010; Monnier et al., 2008; Zhou et al., 2007; Zhou et al., 2013). Patients with congenital myopathies such as MmD carrying recessive RYR1 mutations belonging to class 4 channel defects, typically exhibit non-progressive proximal muscle weakness (Jungbluth et al., 2005; Klein et al., 2012). This reduced muscle strength is consistent with the lower RyR1 content observed in adult muscle fibers that should result in a decrease of Ca2+ release from the SR (Wilmshurst et al., 2010; Monnier et al., 2008; Zhou et al., 2007; Zhou et al., 2013; Jungbluth et al., 2005). The decrease of RyR1 expression is also associated with moderate fiber atrophy, which may additionally contribute to the decrease of muscle strength. In addition to the depletion of RyR1 protein, muscles of patients with recessive RYR1 mutations exhibit striking epigenetic changes, including altered expression of microRNAs, an increased content of HDAC-4 and HDAC-5, and hypermethylation of more than 3600 CpG genomic sites (Zhou et al., 2006; Rokach et al., 2015; Bachmann et al., 2019). Importantly, in muscle biopsies from four patients, hypermethylation of one of the internal RYR1 CpG islands correlated with the increased levels of HDAC-4 and HDAC-5 (Rokach et al., 2015). In order to study in more detail the mechanism of disease of recessive RYR1 mutations, we developed a mouse model knocked in for two mutations, isogenic to those identified in a severely affected child with recessively inherited MmD (Klein et al., 2012). Such a mouse model carries the p.Q1970fsX16 + p.A4329D RyR1 mutations (Elbaz et al., 2019) and will henceforth be referred to as double heterozygous (dHT). Characterization of the muscle phenotype of dHT mice demonstrated that it faithfully recapitulates not only the physiological and biochemical changes, but also the major muscle epigenetic signatures observed in muscle biopsies from MmD patients. In the present study we treated dHT mice with drugs targeting epigenetic enzymes and evaluated the physiological effects of the treatment on muscle function as well as on muscle structure. Our results show that treatment of dHT mice with drugs targeting epigenetic enzymes rescues muscle strength, increases RyR1 protein content, and improves muscle morphology, that is, the treatment partially rescues CRUs (the intracellular sites containing RyR1) and mitochondria. This study provides proof of concept for the treatment of patients with congenital myopathies linked to recessive RYR1 mutations, with small molecules inhibiting DNMT and histone deacetylases. Results Effect of TMP269 and 5-aza-2-deoxycytidine on the in vivo muscle phenotype of dHT mice We first examined the pharmacokinetics and bio-distribution of TMP269, the class IIa HDAC inhibitor we selected for this study. After intraperitoneal (i.p.) injection of 25 mg/kg body weight of TMP269 dissolved in polyethylenglycol 300 (PEG300) (500 µl/kg) and N-methyl-2-pyrrolidone (NMP) (250 µl/kg), blood and/or skeletal muscles were collected at different time points and the content of TMP269 was quantified by liquid chromatography tandem mass spectrometry (LC-MS/MS) (Duthaler et al., 2019). The peak blood concentration of TMP269 was achieved approximately 1 hr after injection. The circulating levels of TMP269 decay within 12 hr (Appendix 1—figure 1A). Importantly, the class II HDAC inhibitor diffuses into skeletal muscle (Appendix 1—figure 1B), and as expected, its concentration profile in skeletal muscle follows that observed in blood. Although the level of TMP269 accumulating in skeletal muscle is lower compared to that present in blood, its concentration in muscle is adequate to induce an inhibitory effect on class IIa HDACs activity (Choi et al., 2018). An identical protocol was used to monitor the optimal dose of 5-aza-2-deoxycytidine (5-Aza), an FDA-approved DNA methyltransferase (DNMT) inhibitor (Kaminskas et al., 2005). Administration of TMP269 + 5-Aza for 15 weeks resulted in hypomethylation of 165 protein-coding genes (Supplementary file 1) in soleus muscles. Gene Ontology analysis showed that most of the hypomethylated genes belong to pathways involved in gene transcription, kinase activity, and membrane targeting (Appendix 2—figure 1). Importantly, administration of TMP269 + 5-Aza for 15 weeks increases the acetylation of Lys residues (Appendix 2—figure 2A and Appendix 2—figure 2—source data 1; Appendix 2—figure 2—source data 2; Appendix 2—figure 2—source data 3; Appendix 2—figure 2—source data 4; Appendix 2—figure 2—source data 5) and of H3K9 (Appendix 2—figure 2B and D and Appendix 2—figure 2—source data 1; Appendix 2—figure 2—source data 2; Appendix 2—figure 2—source data 3; Appendix 2—figure 2—source data 4; Appendix 2—figure 2—source data 5) in total homogenates from flexor digitorum brevis (FDB) fibers isolated from wild type (WT) and dHT mice, compared to that observed in fibers from vehicle-treated WT and dHT mice. This result unequivocally indicates that the inhibition of the deacetylation activity of class IIa HDACs by TMP269 occurs in the nuclei. In addition, Appendix 2—figure 2 also shows that in the absence of TMP269 + 5-Aza treatment, the level of H3K9 acetylation in myonuclei from FDB fibers isolated from dHT mice is 50% lower compared to WT. This result is consistent with: (i) higher deacetylation activity by class IIa HDACs in vehicle-treated dHT mice compared to TMP269 + Aza-treated dHT mice and WT mice; (ii) enrichment of class II HDACs in the nuclei of dHT mice. Furthermore, these results confirm that the drugs reach skeletal muscles where they exerted their biological function. We next conducted a series of experiments to determine how the drug treatment affects in vivo skeletal muscle function. To this end we injected 6-week--old WT and dHT mice with vehicle alone (PEG300 + NMP), with TMP269 alone (25 mg/kg), 5-Aza alone (0.05 mg/kg), or with the two drugs combined on a daily basis and investigated their effects on the in vivo skeletal muscle phenotype by analyzing forelimb grip force using a grip strength meter. Administration of each drug singly, namely TMP269 or 5-Aza alone, does not induce any change in the grip strength of WT (Figure 1A, top left and middle panels). The combined drug treatment did not affect the grip strength of WT mice (Figure 1A, top right panel). In dHT mice, TMP269 alone causes a small but significant increase of grip strength after 10 weeks of treatment (Figure 1A, low left panel). However, the increased grip strength in dHT mice (Figure 1A, lower right panel) was more evident by the combined drug treatment. In particular, this effect became apparent 4/5 weeks after starting the drug treatment and peaked at 10 weeks. The combined drug treatment rescues approximately 20% of muscle grip strength in dHT mice (see Figure 1—figure supplement 1 for raw data on the grip strength of individual dHT mice before and after treatment with TMP269 + 5-Aza). Based on these results, we performed all subsequent experiments only with the combined drug treatment (i.e. TMP269 + 5-Aza). Figure 1 with 1 supplement see all Download asset Open asset Treatment of double heterozygous (dHT) mice with TMP269 + 5-aza-2-deoxycytidine (5-Aza) improves in vivo muscle function as assessed using the grip strength test and voluntary running wheel. (A) Forelimb (two paws) grip force measurement of wild type (WT) (upper panels) and dHT (lower panels) mice treated with vehicle (WT, n = 9; dHT, n = 10), TMP269 (WT, n = 11; dHT, n = 6), 5-Aza (WT, n = 5, dHT, n = 10), and TMP269 + 5-Aza (WT, n = 10, dHT, n = 13). Grip strength was performed once per week during a period of 10 weeks. Each symbol represents the average (± SD) grip force obtained in the indicated number (n) of mice. Grip force (Force) values obtained on the first week were considered 100%. Black symbols, vehicle-treated mice, colored symbols, drug-treated mice. Statistical analysis was conducted using the Mann-Whitney test. *p < 0.05. (B) Spontaneous locomotor (dark phase) activity (left panel) and total running speed (right panel) measured over 20 days in 21-week-old dHT and WT littermates mice treated with vehicle or TMP269 + 5-Aza. Data points are expressed as mean (± SD; n = 4–5 individual mice). *p < 0.05 (Mann-Whitney test). The exact p value for day 20 is given in the text. Figure 1—source data 1 Grip strength in double heterozygous (dHT) mice is improved after 10 weeks of treatment with TMP269 + 5-aza-2-deoxycytidine (5-Aza). Each line represents the average grip strength (in grams) per mouse, calculated by averaging five measurements obtained from the same mouse. The graphs show raw data of grip strength before treatment (t = 0; gray symbols) and 10 weeks after treatment (red symbols). N = 7 for vehicle-treated dHT mice and N = 12 TMP289 + 5-Aza-treated dHT mice. https://cdn.elifesciences.org/articles/73718/elife-73718-fig1-data1-v2.zip Download elife-73718-fig1-data1-v2.zip We next assessed in vivo muscle function of WT and dHT vehicle-treated or drug-treated mice using the voluntary running wheel. We calculated the total running distance of WT (Figure 1B, top panels) and dHT (Figure 1B, bottom panels) mice injected with vehicle and compared it to that of mice treated for 15–18 weeks with TMP269 + 5-Aza. Three weeks of training improved running performance in both mouse groups. Nevertheless, on day 20 the total running distance of WT mice injected with vehicle alone was approximately 70% greater compared to that of vehicle-treated dHT mice: the total running distance of vehicle-treated WT and dHT mice was 186.27 ± 16.70 km n = 4 vs. 59.93 ± 21.38 km n = 5, respectively (mean ± SD, Mann-Whitney two-tailed test, calculated over the 20 days of running, *p < 0.05). On the other hand, treatment of dHT mice with TMP269 + 5-Aza has a remarkable effect on the total running distance (Figure 1B, lower panel). The beneficial effect begins 1 week after treatment commencement, and on day 20 the total running distance achieved by dHT mice injected with TMP269 + 5-Aza was two times higher compared to that covered by dHT mice injected with vehicle alone (Figure 1B, lower panel; Mann-Whitney two-tailed test; on day 20, p = 0.041): the total running distance for vehicle-treated and drug-treated dHT mice was 59.93 ± 21.38 km n = 5 and 125.90 ± 16.39 km n = 5, respectively (mean ± SD, Mann-Whitney two-tailed test, calculated over the 20 days of running *p < 0.05). The longer running distance was also associated with an increased median cruise speed of the drug-treated dHT mice compared to vehicle-treated dHT mice (Figure 1B, right panels) (Mann-Whitney two-tailed test, calculated over the 20 days running period, *p < 0.05). The epigenetic modifying drugs most likely affect a number of genes, which in turn leads to an improvement of the in vivo muscle performance of the dHT mice; the latter effect may result from an improvement of the mechanical properties of skeletal muscles and/or by an influence of the drugs on the metabolic pathways of muscles. In the next set of experiments, we investigated the mechanical properties of intact extensor digitorum longus (EDL) and soleus muscles from WT and dHT mice after injection of vehicle and of TMP269 + 5-Aza. Effect of TMP269+5-Aza treatment on isometric force development in muscles from WT and dHT heterozygous mice EDL and soleus muscles isolated from mice treated for 15 weeks with vehicle or TMP269 + 5-Aza were stimulated with a single 15 V pulse of 1.0 ms duration (Figure 2A, C, E and G) or by a train of pulses delivered at 150 Hz for 400 ms (EDL, Figure 2B and D) or 120 Hz for 1100 ms (soleus, Figure 2F and H) to obtain maximal tetanic contracture. The averaged specific twitch peak force induced by a single action potential in EDL from dHT mice injected with vehicle alone was approximately 37% of that obtained from EDLs from WT mice (64.92 ± 13.93 mN/mm2, n = 10 vs. 171.24 ± 29.32** mN/mm2, n = 8, respectively; mean ± SD; ANOVA followed by the Bonferroni post hoc test **p < 0.01; Supplementary file 2). The peak force developed after twitch stimulation of soleus muscles from dHT mice injected with vehicle alone was approximately 67% of that of obtained in soleus muscles from WT littermates injected with vehicle (67.55 ± 11.26 mN/mm2, n = 10 vs. 96.58 ± 25.78* mN/mm2, n = 8, respectively; mean ± SD; ANOVA followed by the Bonferroni post hoc test *p < 0.05; Supplementary file 2). While the combined drug treatment does not affect the force developed by EDL muscles, we found that it induces a 25% increase of the twitch force in soleus muscles from dHT mice compared to that obtained from vehicle-treated dHT mice (84.61 ± 14.06 mN/mm2, n = 13 vs. 67.55 ± 11.26* mN/mm2, n = 10, respectively; mean ± SD; ANOVA followed by the Bonferroni post hoc test *p < 0.05; Figure 2G and Supplementary file 2). We next examined the force developed during tetanic contractures of EDL and soleus muscles stimulated by a train of pulses delivered at 150 and 120 Hz, respectively. The maximal specific tetanic force developed in EDL muscles from dHT mice injected with vehicle was approximately 20% lower compared to that of EDL muscles from WT mice (373.76 ± 73.16* mN/mm2, n = 10 vs. 452.97 ± 89.59 mN/mm2, n = 8, respectively; mean ± SD; ANOVA followed by the Bonferroni post hoc test *p < 0.05; Figure 2B and D and Supplementary file 2). We found no effect of the combined drug treatment on the maximal specific force developed by EDL muscles isolated from dHT mice (Supplementary file 2). The maximal specific tetanic force generation observed in soleus muscles from dHT mice injected with vehicle was 13% lower compared to WT (276.29 ± 40.04* mN/mm2, n = 10 vs. 315.86 ± 56.96 mN/mm2, n = 8, mean ± SD; ANOVA followed by the Bonferroni post hoc test *p < 0.05) (Figure 2F and H). Contrary to what we observed in EDL muscles, the combined drug treatment fully rescues the maximal tetanic force of slow twitch muscles. Indeed, soleus from dHT mice treated with TMP269 + 5-Aza for 15 weeks displayed a maximal tetanic force which was 18% higher compared to that of soleus from dHT mice injected with vehicle (334.78 ± 65.74* mN/mm2, n = 13 vs. 276.29 ± 40.04 mN/mm2, n = 10, mean ± SD; ANOVA followed by the Bonferroni post hoc test *p < 0.05) (Supplementary file 2). Figure 2 Download asset Open asset The mechanical properties of double heterozygous (dHT) treated with TMP269 + 5-aza-2-deoxycytidine (5-Aza) improve after 15 weeks of treatment. Mechanical properties of extensor digitorum longus (EDL) and soleus muscle from wild type (WT) and dHT mice treated with vehicle (WT, n = 8; dHT, n = 10) and dHT treated with TMP269 + 5-Aza, n = 13. (A) Representative traces of twitch and (B) maximal tetanic force in EDL (150 Hz) muscle from WT and dHT. Force is expressed as specific force, mN/mm2. (C) Statistical analysis of force generated after twitch and (D) tetanic stimulation of isolated EDL muscle. Data points are expressed as Whisker plots (n = 8–13 mice). Each symbol represents the value from a muscle from a single mouse. (E) Representative traces of twitch and (F) maximal tetanic force (120 Hz) of soleus muscle from WT and dHT mice. (G) Whisker plots of force generated after twitch and (H) tetanic stimulation of isolated soleus muscles. Each symbol represents the value from a muscle from a single mouse (n = 8–13 mice). *p < 0.05; **p < 0.01 (ANOVA followed by the Bonferroni post hoc test). The exact p values are given in Supplementary file 2. Fiber type composition and minimal Feret’s diameter of soleus muscles from dHT mice treated with TMP269+5-Aza In this set of experiments, we examined whether the ergogenic effect associated with the inhibition of epigenetic modifying enzymes is linked to fast-to-slow fiber transition and/or to changes of minimal Feret’s diameter. Treatment with TMP269 + 5-Aza causes no changes in the content of the fiber type composition of soleus muscles (Figure 3A and B, and Supplementary file 3). Similarly, the improved specific force cannot be attributed to a major shift of minimal Feret’s fiber diameter distribution (Figure 3C). Figure 3 Download asset Open asset Histology of soleus muscles from TMP269 + 5-aza-2-deoxycytidine (5-Aza) and vehicle-treated double heterozygous (dHT) mice. (A) Analysis of soleus muscles from wild type (WT) (vehicle-treated) and dHT (vehicle and TMP269 + 5-Aza-treated) mice using monoclonal antibodies specific for myosin heavy chain (MyHC) isoforms. Frozen muscle sections were stained with anti-MyHC I antibodies (slow fibers, blue), anti-MyHC IIa antibodies (fast fibers, yellow), and counterstained with anti-laminin antibodies (red). MyHCIIx (fast fibers) are unstained. (B) Bar plots of fiber type composition of soleus muscles. Left, mean (%, ± SEM) MyHC I fibers, middle, mean (%, ± SEM) MyHC IIa fibers, right mean (%, ± SEM) MyHC IIx fibers. (C) Minimal Feret’s distribution of type I and type II fibers. Data points are expressed as mean (± SEM). For WT vehicle treated a total of 2881 fibers from 3 mice were counted, for dHT vehicle treated a total of 2642 fibers from 3 mice were counted, for dHT TMP269 + 5-Aza a total of 2983 fibers from 3 mice were counted. Effect of TMP269+5-Aza treatment on calcium transients in single FDB fibers from WT and dHT mice We investigated resting [Ca2+] and calcium transients evoked either by a single pulse (Figure 4A and B) or by a train of action potentials (Figure 4C and D) in FDB fibers from WT and dHT mice treated for 15 weeks with vehicle or TMP269 + 5-Aza (4–6 mice per group). We used single FDB fibers (i) because they are a mixture of fast and slow twitch muscles and (ii) since intact single fibers from EDL and soleus muscles from 22-week--old mice are nearly impossible to obtain. We found that the resting [Ca2+] was similar in FDB fibers from WT and dHT vehicle- or drug-treated mice. The Fura-2 fluorescence values (F340/F380, mean ± SD) were 0.81 ± 0.09 (n = 75 fibers isolated from 4 mice), 0.77 ± 0.07 (n = 40 fibers isolated from 5 mice), and 0.81 ± 0.08 (n = 33 fibers isolated from 5 mice) in WT mice injected with vehicle, dHT mice injected with vehicle, and dHT mice treated with TMP269 + 5-Aza, respectively. We next investigated the Ca2+ transients in response to electrical stimulation in FDB fibers loaded with 10 µM of the low affinity Ca2+ indicator Mag-Fluo-4. In the presence of 1.8 mM Ca2+ in the extracellular solution, baseline fluorescence (mean ± SD fluorescence in arbitrary units) was similar in FDBs from vehicle-treated WT mice (1.13 ± 0.12, n = 91 fibers isolated from n = 4 mice), in vehicle-treated dHT mice (1.14 ± 0.12, n = 110 fibers isolated from n = 6 mice), and drug-treated dHT mice (1.16 ± 0.10, n = 155 fibers isolated from n = 5 mice) (Appendix 3—figure 1). In the presence of 1.8 mM Ca2+ in the extracellular solution, the average peak intracellular Ca2+ transient induced by a single action potential in FDB fibers from dHT mice injected with vehicle is approximately 30% lower than that observed in fibers from WT mice (∆F/Fo values were 0.98 ± 0.22, n = 110 fibers isolated from 6 mice vs. 1.38 ± 0.30, n = 91 fibers isolated from 4 mice, respectively; mean ± SD; Figure 4A and B, Supplementary file 4). Interestingly, treatment of dHT mice with TMP269 + 5-Aza causes a 23% increase of the peak Ca2+ transient compared to that observed in dHT mice injected with vehicle (*1.21 ± 0.28, n = 155 fibers isolated from 5 mice vs. 0.98 ± 0.22, n = 110 fibers isolated from 6 mice, respectively; ∆F/Fo values are expressed as mean ± SD; ANOVA followed by the Bonferroni post hoc test *p < 0.05; Figure 4A and B and Supplementary file 4). In the presence of 1.8 mM Ca2+ in the extracellular solution, the peak Ca2+ transient evoked by a train of pulses delivered at 100 Hz in FDB fibers from dHT mice injected with vehicle is approximately 25% lower compared to that of FDB fibers from WT mice (*1.22 ± 0.22, n = 92 fibers isolated from 6 mice vs. 1.62 ± 0.21, n = 63 fibers isolated from 4 mice, respectively; ∆F/Fo values are expressed as mean ± SD; Figure 4D and Supplementary file 4). When dHT mice are treated with TMP269 + 5-Aza for 15 weeks, the summation of calcium transients induced by a train of supramaximal pulses is 16% higher compared to that of dHT mice injected with vehicle alone (*1.42 ± 0.23, n = 78 fibers isolated from 5 mice vs. 1.22 ± 0.22, n = 92 fibers isolated from 6 mice, respectively; ∆F/Fo values are expressed as mean ± SD; ANOVA followed by the Bonferroni post hoc test *p < 0.05; Figure 4C and D and Supplementary file 4). Altogether, the increase of the peak calcium transient after either a single action potential or a train of pulses is consistent, at least in part, with the ergogenic effects caused by the combined treatment with TMP269 + 5-Aza on dHT mice. Figure 4 Download asset Open asset Electrically evoked peak Ca2+ transients in muscle fibers from treated double heterozygous (dHT) compound heterozygous (dHT) mice was rescued by TMP269 + 5-aza-2-deoxycytidine (5-Aza) administration. Enzymatically dissociated flexor digitorum brevis (FDB) fibers dissected from 4 to 6 mice per group, were loaded with Mag-Fluo-4 and electrically stimulated by field stimulation. Black line, vehicle-treated wild type (WT), gray line, vehicle-treated dHT, red line, TMP269 + 5-Aza-treated dHT. (A) Representative Ca2+ transient evoked by a single pulse (twitch) of 50 V with a duration of 1 ms. (B) Whisker plots of peak twitch. Each symbol represents results obtained from a single FDB fiber. (C) Representative Ca2+ transient evoked by tetanic stimulation by a train of pulses delivered at 100 Hz for 300 ms. (D) Whisker plots of peak transient induced by tetanic stimulation. Each symbol represents results obtained from a single FDB fiber. *p < 0.05 (ANOVA followed by the Bonferroni post hoc test). The exact p values are given in Supplementary file 4. TMP269+5-Aza rescues RyR1 expression in muscles from dHT mice The results obtained so far indicate that treatment with TMP269 + 5-Aza exerts a beneficial effect preferentially on slow twitch muscles. Nevertheless, the genome-wide effects linked to the combined inhibition of class II HDACs and DNMT may affect a number of processes underlying muscle strength, making it difficult if not impossible to dissect the exact mechanisms underlying the improvement of slow twitch muscle function observed in dHT mice. However, based on our previous results, we postulate that the improvement of muscle strength observed in soleus muscles may be explained, at least in part, by an increase of the key proteins involved in skeletal muscle activation. Figure 5A shows Ryr1 transcript expression in soleus muscles from WT and dHT mice. Treatment with vehicle alone does not rescue Ryr1 expression (Mann-Whitney two-tailed test, WT vehicle vs. dHT vehicle, p = 0.039), however treatment with TMP269 + 5-Aza for 15 weeks causes a significant increase in Ryr1 transcript levels (Mann-Whitney two-tailed test, dHT vehicle vs. dHT TMP269 + 5-Aza, p = 0.019). Cacna1s levels are not affected by vehicle or drug treatment. Hdac4 transcript levels are increased in soleus muscles from dHT vehicle-treated mice, compared to vehicle-treated WT mice (Figure 5A, Mann-Whitney two-tailed test, WT vehicle vs. dHT vehicle, p = 0.041). Hdac4 transcript levels decrease in muscles from TMP269 + 5-Aza-treated dHT mice compared to vehicle-treated dHT mice (Figure 5A, Mann-Whitney two-tailed test, dHT vehicle vs. dHT TMP269 + 5-Aza, p = 0.002). We also investigated RyR1 protein content in total homogenates from soleus muscles from WT and dHT mice. The RyR1 protein content of soleus muscles from dHT mice injected with vehicle is 46% lower compared to that of WT mice (Figure 5B and Figure 5—source data 1–3). The mean ± SD % intensity of the immunopositive band corresponding to RyR1 is 100% ± 7.30, n = 8 in WT vs. 54.93% ± 18.45, n = 6 in dHT, respectively (ANOVA followed by the Bonferroni post hoc test, *p = 0.039). The RyR1 protein content is partially re-established by treatment with TMP269 + 5-Aza. Indeed, the RyR1 protein content in drug-treated dHT mice increased and reaches a value of 88.03% ± 10.42*, n = 7, in contrast to the 54.93% ± 18.45, n = 6 found in vehicle-treated dHT mice (ANOVA followed by the Bonferroni post hoc test, *p = 0.037 for vehicle vs. TMP269 + 5-Aza-treated dHT mice; Figure 5 and Figure 5—source data 1–3). This recovery of RyR1 is consistent with the in vivo and in vitro muscle phenotype amelioration induced by the combined drug treatment. We also measured the effect of the combined drug treatment on the content of other proteins involved in skeletal muscle ECC including SERCA1, SERCA2, calsequestrin-1, and JP-45. As shown in Figure 5C and Figure 5—source data 4–11, the muscle content of these proteins is unaffected by the drug treatment. Figure 5 Download asset Open asset Treatment with TMP269 + 5-aza-2-deoxycytidine (5-Aza) reverses ryanodine receptor 1 (RyR1) loss in soleus muscles from double heterozygous (dHT) mice. (A) Real-time quantitative polymerase chain reaction (qPCR) on RNA isolated from soleus muscles isolated from vehicle-treated wild type (