|Home | About | Journals | Submit | Contact Us | Français|
Cardiac myocytes from the mdx mouse, the mouse model of Duchenne muscular dystrophy, exhibit t-tubule disarray and increased calcium sparks, but a unifying molecular mechanism has not been elucidated. Recently, improper trafficking of junctophilin-2 on an altered microtubule network caused t-tubule derangements and calcium mishandling in a pressure-overload heart failure model. Mdx cardiac myocytes have microtubule abnormalities, but how this may affect junctophilin-2, t-tubules, and calcium handling has not been established. Here, we investigated the hypothesis that an inverse relationship between microtubules and junctophilin-2 underlies t-tubule disruptions and calcium mishandling in mdx cardiac myocytes. Confocal microscopy revealed t-tubule disorganization in mdx cardiac myocytes. Quantitative Western blot analysis demonstrated junctophilin-2 was decreased by 75% and showed an inverse hyperbolic relationship with α- and β-tubulin, the individual components of microtubules, in mdx hearts. Colchicine-induced microtubule depolymerization normalized junctophilin-2 protein levels and localization, corrected t-tubule architecture, and reduced calcium sparks. In summary, these results suggest microtubule-mediated misregulation of junctophilin-2 causes t-tubule derangements and altered calcium handling in mdx cardiac myocytes.
Loss of dystrophin causes Duchenne muscular dystrophy (DMD), an X-linked disease characterized by striated muscle dysfunction resulting in a life expectancy of only 20 years to 30 years (1). Heart failure is the cause of death for 20% to 25% of DMD patients (2); therefore research aimed at understanding the molecular and cellular phenotypes underlying the cardiomyopathy of DMD has been conducted. Altered calcium homeostasis marked by increased calcium sparks (3–5) and t-tubule disarray were documented in cardiac myocytes from the dystrophin-deficient mdx mouse (6,7), but a unifying molecular mechanism has not been identified.
Junctophilin (JPH)-2, the protein that links the plasma membrane of t-tubules to the ryanodine receptor, is essential for proper t-tubule structure and function (8). Recently, it was reported that JPH-2 mislocalization due to abnormalities in microtubule cytoskeleton caused pathological t-tubule remodeling and abnormal calcium handling in the pressure overload-induced heart failure model (9). However, the relationship between microtubules and JPH-2 in other models of heart failure or cardiomyopathy has not been analyzed. Because previous studies documented microtubule derangements in mdx cardiac myocytes (5,10), we tested the hypothesis that microtubule alterations cause JPH-2 misregulation and result in t-tubule disruptions and calcium mishandling in mdx mice.
Finally, to investigate the translational aspects of our hypothesis, we examined the cardiomyopathy of mdx mice via echocardiography and isoproterenol stress tests as previous studies showed mildly reduced systolic function (11–14) and excessive mortality with isoproterenol administration (10,15,16) in mdx mice. Because Zhang et al. (9) and Guo et al. (17) showed improvement in systolic function with normalization ‘of JPH-2, we hypothesized colchicine-induced JPH-2 normalization would reduce the severity of mdx cardiomyopathy.
Control C57BL/10 and mdx mice were purchased from Jackson Laboratories. All animals were housed and treated following the guidelines set forth by the University of Minnesota Institutional Animal Care and Use Committee.
Polyclonal antibodies for voltage-gated calcium channel (VGCC) (Sigma) and JPH-2 (ThermoScientific) and monoclonal antibodies for α-tubulin (Sigma), β-tubulin (Sigma), and dystrophin (Leica) were purchased from the identified vendors. Alexa-Fluor-488 or Alexa-Fluor-568-conjugated anti-rabbit antibodies were purchased from Molecular Probes. Infrared dye-conjugated anti-mouse and anti-rabbit antibodies were purchased from LICOR Biosciences.
Isolation of ventricular cardiac myocytes was performed as described previously (18).
Freshly isolated cardiac myocytes were fixed in 4% paraformaldehyde for 10 min at 37°C, washed with phosphate-buffered saline (PBS) 2 times for 5 min, incubated with AlexaFluor 488 conjugated Wheat Germ Agglutinin (Sigma) for 10 min at room temperature, and then washed in PBS 2 times for 5 min. Cells were mounted in Antifade (Molecular Probes) and imaged on Bio-Rad MRC 1000 scan head mounted on an upright Nikon Optishot microscope at the University Imaging Centers at the University of Minnesota. Z-stacks were collected and converted into a z-projection using ImageJ. T-tubule quantification was performed using the TTPower plugin on ImageJ as described (7).
Primary cardiac myocytes were fixed in 4% paraformaldehyde for 10 min at 37°C, permeabilized with 1% Triton X-100 in PBS, blocked in 5% BSA for 10 min 3 times, and incubated with primary antibodies overnight at 4°C. Sections were then washed and blocked with 5% bovine serum albumin for 10 min 3 times and then incubated with Alexa-Fluor-488- or Alexa-Fluor-568-conjugated secondary antibodies for 30 min at 37°C. Then, cardiac myocytes were washed with PBS and mounted in Anti-Fade Reagent (Molecular Probes).
Freshly isolated cardiac myocytes were loaded with Fura-2AM (a ratiometric Ca2+ indicator; 2 µM [Molecular Probes]) for 10 min at room temperature after a de-esterification period of 20 min in M199 medium. Cells were incubated with 10 µm colchicine or vehicle (double-distilled water [ddH2O]) for 2 h and then subjected to calcium analysis. Fura-2 fluorescence was measured using a spectrophotometer (Stepper Switch, IonOptix). Initially, Fura-2 was excited at 360 nm (the isosbestic point independent of Ca2+) and then continuously at 380 nm (Ca2+-dependent fluorescence). Emission was collected at >510 nm by a photo-multiplier tube. Ratiometric data were collected and analyzed online using commercial software (IonOptix).
Freshly isolated adult cardiac myocytes were plated on glass coverslips coated with 10 µg/ml laminin and in M199 medium and allowed to attach for 2 h. Then, colchicine (10 µM) or vehicle was added for 2 h. Cardiac myocytes were loaded with Ca2+ indicator Fluo 8-AM (5 µM, AAT Bioquest) for 10 min at room temperature followed by washout. For confocal imaging, we used an inverted confocal microscope (Leica TCS SP8, Germany) with a 40×, 1.3 NA oil-immersion objective. Fluo 8-AM was excited at 488 nm and the emission was collected at >505 nm. For Ca2+ spark imaging, line scan was performed at a speed of 1.43 ms/line for 1,000 lines.
Freshly isolated cardiac myocytes plated on laminin-coated coverslips were treated with 10 µM colchicine or vehicle for 2 h and were then loaded with 2 µM Fura-2AM for 10 min and de-esterified for 10 min at room temperature. After de-esterification, coverslips were mounted onto a perfusion chamber and perfused with modified Tyrode’s solution (in mM: 140 NaCl, 0.5 MgCl2, 5 HEPES, 5.5 glucose, 1.8 CaCl2, 5 KCl; pH 7.4) at 30°C. Myocytes were paced at 0.5 Hz for at least 20 s after which pacing was stopped and the perfusate was switched to Tyrode’s solution with 20 mM caffeine to measure SR calcium load. Finally, myocytes were again perfused with Tyrode’s solution and paced at 0.5 Hz to ensure they were still viable. No more than 2 myocytes were measured per coverslip. Transient data were collected by measuring the 360:380 ratio using the Ionoptix Calcium and Contractility System and data were analyzed using IonWizard software.
Confocal images were collected and equivalently processed using Adobe Photoshop Version CS6.
Whole protein extracts from isolated ventricles from mice were performed as described (19). Protein concentration was determined using a BCA protein assay kit (Pierce). Cardiac extracts (25 µg) were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred to nitrocellulose membrane. Membranes were washed/blocked in 5% milk in PBS for 1 h, and incubated with primary antibodies for 1 h at room temperature. Membranes were then washed/blocked twice for 10 min in 5% milk in PBS and then incubated with infrared secondary antibodies for 30 min at room temperature. Membranes were washed in TBS containing 0.1% Tween twice for 5 min. Imaging and quantification of Western blots was performed on an Odyssey Infrared Imaging system. SDS-PAGE was performed in parallel and stained with Coomassie brilliant blue stain and was imaging at the 700 nm wavelength on the Odyssey Imaging system to serve as the loading control as described (20).
Mice were treated with intraperitoneal injection of either filter-sterilized PBS or 0.5 mg/kg of filter-sterilized colchicine (Sigma) dissolved in PBS every other day for 2 weeks.
Echocardiographic analysis was conducted using a VisualSonics 2100 ultrasound machine with mice anesthetized with 2% to 3% inhaled isoflurane. Heart rate had to be >400 beats/min during echocardiographic analysis. Systolic and diastolic dimensions and fractional shortening were determined from M-mode images in the parasternal short-axis view at the level of the mitral valve.
Eight-month to 10-month-old wild-type (WT) and mdx mice (n = 8 mice to 10 mice per treatment group) were treated with either PBS or colchicine for 2 weeks and then administered isoproterenol at a dose 10 mg/kg at 9am, 1pm, and 5pm via intraperitoneal injection for 3 consecutive days. Survival was assessed at 8-h intervals for the first 5 days and then daily thereafter.
All data are represented as mean ± standard error of the mean. Student t tests were used to compare 2 groups with assumed equivalent variance if sample size was <5 with an F-test to assess variance if sample size was >5 and 1-way analysis of variance with Tukey post-hoc test was used to compare means between 3 groups. Kaplan-Meier survival curve was used to determine survival differences after initial treatment with isoproterenol with log-rank test to assess differences in survival. Statistical significance was defined as p ≤ 0.05.
We first analyzed the t-tubule network in isolated cardiac myocytes from young (2- to 3-months old) and older (8- to 10-months old) WT and mdx mice. The 8- to 10-month time was used because that is the age when mdx mice begin to show signs of cardiomyopathy (12,14). When compared to 8-month to 10-months of age, WT and young mdx cardiac myocytes had no discernable difference in t-tubule architecture, but older mdx cardiac myocytes showed disrupted t-tubules (Figure 1A). When quantified using TTPower analysis, older mdx t-tubule structural integrity (39.2 ± 2.2 AU) was significantly different than WT (57.3 ± 1.3 AU) and young mdx mice (56.8 ± 2.3 AU) (Figure 1B).
To provide insight into the mechanism of t-tubule disruptions in older mdx mice, we investigated 2 key t-tubule proteins: the VGCC and JPH-2. Protein levels of VGCC were not altered in either young or older mdx heart extracts (Figure 1C to 1F). However, JPH-2 protein levels were markedly reduced by 75% in, older mdx hearts (WT: 1 ± 0.9 AU and mdx: 0.024 ± 0.4 AU) but not significantly altered in young mice (WT: 1.0 ± 0.1 AU and mdx: 1.3 ± 0.2 AU) (Figure 1C to 1F), a pattern that mirrored the t-tubule phenotype. Immunofluorescence analysis of isolated cardiac myocytes from older WT (74.2 ± 1.9 AU) and mdx mice (57.8 ± 1.7 AU) revealed altered localization of JPH-2 in mdx mice (Figure 1G and 1H). In summary, JPH-2 misregulation was associated with t-tubule disruptions in mdx hearts.
Because a previous report demonstrated that an altered microtubule cytoskeleton caused mislocalization of JPH-2 (9), we analyzed the relationship between α- and β-tubulin, the individual components of microtubules, and JPH-2 in mdx hearts. At a young age when JPH-2 levels were normal, α-tubulin (WT: 1.0 ± 0.1 AU and mdx: 1.2 ± 0.1 AU) and β-tubulin (WT: 1.0 ± 0.2 AU and mdx: 1.3 ± 0.1 AU) protein expression was not significantly different in mdx cardiac extracts (Figure 2A and 2B). However, in older mdx hearts when JPH-2 protein was reduced, α-tubulin (WT: 1.0 ± 0.1 AU and mdx: 2.7 ± 0.3 AU) and β-tubulin (WT: 1.0 ± 0.9 AU and mdx: 2.5 ± 0.2 AU) content was increased approximately 2.6-fold (Figure 2C and 2D). When the relationships between α- and β-tubulin and JPH-2 in older mice was analyzed, we observed a significant inverse hyperbolic relationships (r2 = 0.95 for both curve fitting models) (Figure 2E and 2F). These data suggested increased microtubule content was associated with reduced JPH-2 protein levels.
To further probe the relationship between microtubules and JPH-2 in mdx hearts, we treated older mdx mice for 2 weeks with colchicine to induce microtubule depolymerization and examined the effects on JPH-2 and t-tubules. Colchicine treatment effectively depolymerized microtubules in cardiac myocytes in vivo (Figure 3A) and was associated with a 7.6-fold increase in JPH-2 protein levels (mdx-PBS: 1.0 ± 0.08 AU and mdx-colchicine: 7.6 ± 1.0 AU) (Figure 3B and 3C). Moreover, JPH-2 localization patterns in mdx cardiac myocytes were significantly improved with colchicine treatment (mdx-PBS: 62.6 ± 1.6 AU and mdx-colchicine: 68.1 ± 1.9 AU) (Figure 3D and 3E). Finally, colchicine-mediated restoration of JPH-2 improved t-tubule organization in older mdx hearts (mdx-PBS: 44.5 ± 2.0 AU and mdx-colchicine: 52.5 ± 2.2 AU) (Figure 3F and 3G).
Because JPH-2 was shown to be an important regulator of the ryanodine receptor (21–23), we investigated how colchicine-mediated JPH-2 restoration affected calcium handling in isolated mdx cardiac myocytes treated acutely with colchicine. Acute colchicine treatment did not affect diastolic calcium, onset kinetics, or amplitude of calcium transients, but there was accelerated calcium decay at 25% to baseline (mdx: 59.4 ± 2.5 ms and mdx-colchicine: 51.1 ± 3.0 ms) but not at 75% to baseline (mdx: 219.9 ± 20.4 ms and mdx-colchicine: 202.2 ± 25.0 ms) (Figure 4A to 4F), which suggested improved ryanodine receptor gating. To further investigate ryanodine receptor activity, we measured calcium spark frequency and found acute colchicine treatment reduced calcium sparks by more than 50% (mdx: 0.84 ± 0.2 sparks/s/100 µm and mdx-colchicine: 0.31 ± 0.08 sparks/s/100 µm) (Figure 4G and 4H). Acute colchicine treatment caused a nonsignificant alteration in sarcoplasmic reticulum calcium load (Supplemental Figure 1), which would not explain the reduction in calcium sparks. All together, these data suggest that acute colchicine treatment improved ryanodine receptor gating in mdx cardiac myocytes.
Next, we determined if 2 weeks of colchicine treatment led to whole-organ level improvements in older mdx mice. M-mode echocardiography did not reveal any significant changes in cardiac geometry or systolic function after 2 weeks of colchicine treatment in mdx mice (Supplemental Figure 2), but both mdx groups showed signs of mild cardiomyopathy when compared to age-matched WT mice. Finally, colchicine treatment did not confer a significant survival benefit when older mdx mice were subjected to isoproterenol stress test (WT survival: 75%, mdx-colchicine survival: 40%, mdx-PBS survival: 10%) (Supplemental Figure 3).
Our main new findings are the following: 1) t-tubule disarray in mdx cardiac myocytes is associated with altered localization and decreased protein levels of JPH-2; 2) there is a significant inverse hyperbolic relationship between JPH-2 and α- and β-tubulin, the individual subunits of microtubules, in mdx hearts; 3) disruption of the JPH-2/microtubule relationship, using colchicine to depolymerize microtubules, not only normalizes JPH-2 localization, but we show for the first time that it restores JPH-2 protein levels which corrects t-tubule morphology; and 4) colchicine treatment can effectively reduce calcium spark frequency in isolated mdx cardiac myocytes, which we propose is mediated through stabilization of the ryanodine receptor. Collectively, these results provide evidence that microtubule-dependent misregulation of JPH-2 underlies t-tubule derangements and abnormal calcium handling in Duchenne cardiomyopathy, and implicates JPH-2 as a new molecular mediator in the pathophysiology of Duchenne cardiomyopathy.
Importantly, our results provide additional mechanistic insight into 2 secondary pathological cardiac phenotypes associated with dystrophin-deficiency: t-tubule disruptions and calcium mishandling. First of all, we show that normalization of JPH-2 protein levels and localization improves t-tubule architecture, a cellular phenotype that has not had a proposed molecular mechanism. Secondly, colchicine treatment reduces calcium spark frequency which was in agreement with results of Kerr et al. (24), which we propose is due to normalization of JPH-2 after microtubule depolymerization leading to improved ryanodine receptor gating. These results provide an additional molecular mechanism by which the ryanodine receptor gating is altered in mdx cardiac myocytes, in addition to the loss of a calstabin-ryanodine receptor interaction as previously defined (4). In summary, our results implicate JPH-2 as the molecular mediator of t-tubule abnormalities and abnormal gating of the ryanodine receptor in Duchenne cardiomyopathy.
Next, we provide further support for the hypothesis that JPH-2 regulates ryanodine receptor activity. Several studies have documented increased calcium spark frequency when JPH-2 is decreased or altered in localization (9,21–23), suggesting the ryanodine receptor is improperly gated when JPH-2 is disrupted. In particular, Zhang et al. (9) showed that microtubule depolymerization-induced normalization of JPH-2 localization patterns reduced calcium spark frequency in cultured mouse cardiac myocytes, a result very similar to our findings. However, unlike Zhang et al. (9), we did not observe changes in calcium transient amplitude with colchicine treatment. The discrepancy between our results and those of Zhang et al. (9) may be due in-part to SR load as JPH-2 knockdown causes reduced SR calcium content, but calcium transients normalized to SR load are actually increased (22). mdx mice have normal SR calcium content (25) and are not significantly altered by colchicine treatment (Supplemental Figure 1), which might explain why colchicine treatment did not increase calcium transient amplitude in mdx cardiac myocytes. Another explanation for the differences between our results and those of Zhang et al. (9) could relate to the timing of experiments as we analyzed calcium transients in freshly isolated cardiac myocytes whereas Zhang et al. (9) showed normalization of calcium transients in cells cultured for >48 h. Nonetheless, our results support a crucial role for JPH-2 in regulating ryanodine receptor activity.
Although we were able to document restoration of t-tubule organization with colchicine treatment, we did not observe organ-level cardiac improvements in mdx mice. One explanation is that t-tubule corrections do not always lead to improved whole organ function; a finding that was documented in previous publications. Firstly, t-tubule derangements precede left ventricular dysfunction in the thoracic aortic banded rat model (26). Moreover, while transgenic overexpression of JPH-2 could prevent pathological t-tubule remodeling and blunted the effects of aortic banding; it did not completely prevent the onset of heart failure (17). Another reason for the lack of improvement of cardiomyopathy in mdx mice treated with colchicine could be the relatively mild cardiomyopathy of mdx mice. Whereas Guo et al. (17) showed improvements in ejection fraction with JPH-2 overexpression, there was not a complete normalization of ejection fraction, which could explain why echocardiography did not reveal improvements in mdx mice treated with colchicine as the mdx mice had only mild reduced fractional shortening (Supplemental Figure 2). Finally, the lack of a significant improvement in isoproterenol stress test may be explained by the underlying pathophysiology that was corrected with colchicine treatment. Previously, membrane permeability was associated with increased death in mdx mice when treated with isoproterenol (10,15,16), and thus the nonsignificant improvement in survival with colchicine may have resulted from an inability to correct cardiac myocytes permeability in mdx hearts. In summary, there may be several reasons why colchicine treatment did not improve mdx cardiomyopathy.
Finally, our results add to several other publications that document a crucial role for JPH-2 in t-tubule integrity. The first link between JPH-2 and t-tubules was shown when knockout of JPH-2 resulted in embryonic lethality due to lack and cardiac contractility and improper t-tubule formation (27). Several other studies document that alterations of JPH-2 via miR-24 (28,29), RNA interference (21,22,30), or calpain-mediated protein cleavage (23) cause t-tubule derangements and rescue of JPH-2 protein levels or localization via transgenic overexpression (17), inhibition of miR-24 (28), calpain-inhibition (23), or microtubule depolymerization (9) prevent pathological t-tubule remodeling. Taken together with our new findings, JPH-2 is a critical mediator of t-tubule structure in cardiac myocytes.
Cardiomyopathy accounts for approximately one-quarter of deaths in Duchenne muscular dystrophy. Currently, our knowledge of the mechanistic pathophysiology of the Duchenne cardiomyopathy is incomplete, but these data provide a molecular explanation for the disruption of t-tubules and calcium mishandling in Duchenne cardiomyopathy.
Future studies that examine the effects of normalization of JPH-2 through multiple mechanisms including colchicine treatment, gene transfer, or micro-RNA inhibition in large animal models of Duchenne cardiomyopathy with a more severe phenotype could be conducted to gain further information about how JPH-2 affects outcomes in Duchenne cardiomyopathy. If results are positive, a clinical trial examining the effects of colchicine in Duchenne muscular dystrophy could be considered.
The authors would like to thank Drs. Amit Gaggar and Thenappan Thenappan for assistance in editing the manuscript.
This work was supported by the University of Minnesota Lillehei Heart Institute. Dr. Metzger has received grants from the National Institutes of Health (NIH) and MDA. Dr. Prins has received NIH T32 (HL069764) and F32 (HL129554) grants. Dr. Asp has received the NIH F32 (HL115876) grant. Dr. Wang has received the NIH RO1 (HL114760) grant.
All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
APPENDIX For supplemental figures, please see the supplemental appendix.