Inhibition of tenascin C rescues abnormally reduced Na currents in dystrophin-deficient ventricular cardiomyocytes

Abstract

Cardiac arrhythmias significantly contribute to mortality in Duchenne muscular dystrophy (DMD), a severe muscle disease caused by dystrophin deficiency. Using the mdx mouse model for human DMD, we previously showed that the lack of dystrophin induces a significant loss of peak sodium current (I_Na) in ventricular cardiomyocytes. This provided a mechanistic explanation for ventricular conduction defects and concomitant arrhythmias in the dystrophic heart. The extracellular matrix protein tenascin C (TN-C), a major remodeling factor in the diseased heart, is strongly upregulated in DMD. The consequences of TN-C upregulation in the dystrophic heart, however, are unknown. Here, we tested if TN-C induces electrical remodeling in the dystrophic heart, and if inhibition of TN-C rescues peak I_Na loss in dystrophin-deficient ventricular cardiomyocytes. We found that cardiomyocytes from TN-C knockout (KO) mice had increased peak I_Na. The abnormally reduced peak I_Na in mdx myocytes was rescued to wild-type levels by additional TN-C KO, which was accompanied by enhanced Na_v1.5 channel expression. Further, peak I_Na in mdx myocytes was increased by treatment of mdx mice with TN-C siRNA. Twenty-four-hour incubation of wild-type myocytes with human recombinant TN-C reduced their peak I_Na, an effect which could be abolished by blocking antibodies specific for the α-7 integrin subunit. Our findings suggest that TN-C induces peak I_Na loss in the dystrophic heart, and that inhibition of TN-C expression rescues abnormally reduced peak I_Na in dystrophin-deficient ventricular cardiomyocytes. TN-C inhibition emerges as a strategy to counteract ventricular conduction impairments and arrhythmias in patients with DMD.NEW & NOTEWORTHY Dystrophin deficiency in cardiomyocytes leads to abnormally reduced Na currents. These can be rescued by inhibition of the expression of tenascin C.

Keywords: Duchenne muscular dystrophy; Na current; arrhythmias; cardiomyopathy; tenascin C.

Figure 1. Gene expression levels of TN-C in cardiac tissues from adult male wt and dystrophic mice and rats.

A: comparison of TN-C expression in left ventricular tissue from wt and mdx mice. B: TN-C expression in wt and DMD^mdx rat whole hearts. TN-C mRNA expression levels were normalized to the housekeeper HPRT1, and then to the mean relative expression of the wt (control group). Data are expressed as means ± SE [mice: n = 13 isolated hearts (one left ventricular tissue piece per heart) for both wt and mdx; rats: n = 8 isolated hearts for wt, n = 6 isolated hearts for DMD^mdx]. Each data point represents the mean of two technical replicates of one biological sample. Statistical analyses were performed with a Mann–Whitney U test (unpaired). ^**P < 0.01; ^***P < 0.001. DMD, Duchenne muscular dystrophy; TN-C, tenascin C; wt, wild type.

Figure 2. Comparison of the peak I_Na densities of ventricular cardiomyocytes derived from wt, dystrophin-deficient (mdx), TN-C KO, and mdx-TN-C double KO mice.

A: representative original I_Na traces, recorded from myocytes of the different mouse lines, elicited by the pulse protocol shown in the inset. B: peak I_Na density-voltage relationships for wt, mdx, TN-C KO, and mdx-TN-C double KO myocytes. V₅₀, the voltage at which half-maximum activation occurred, was −56 ± 1 mV for wt, −52 ± 2 mV for mdx, −56 ± 2 mV for TN-C KO, and −55 ± 1 mV for mdx-TN-C double KO myocytes (no significant difference between groups, P always > 0.24). C: comparison of the peak I_Na densities of myocytes from the different mouse lines at current maximum (−42 mV). ^*P < 0.05; ^***P < 0.001. wt vs. mdx: significantly different (P < 0.05) from −47 to −32 mV; mdx vs. mdx TN-C KO: from −42 to −32 mV; wt vs. TN-C KO: from −37 to −27 mV. n numbers represents the number of hearts (animals) used per group; the total number of cells is also given. D: comparison of the membrane capacitance values of the recorded myocytes as measure of cell size; ns, not significant. Data are expressed as means ± SE. I_Na, sodium current; KO, knockout; TN-C, tenascin C; wt, wild type.

Figure 3. Cardiac Na channel gene and protein expression in left ventricular tissue from wt, mdx, TN-C KO, and mdx-TN-C double KO mice.

A: Scn5a mRNA levels were normalized to the housekeeper HPRT1, and then to the mean relative expression of the wt (control group). Data are expressed as means ± SE. One way ANOVA revealed a significant difference between the four groups (P < 0.001). Post hoc comparison between two groups was performed with Holm–Sidak’s multiple comparisons test; ^*P < 0.05, ^**P < 0.01, ^***P < 0.001. n = 6 hearts (one left ventricular tissue piece per heart) per genotype. Each data point represents the mean of two technical replicates of one biological sample (one isolated left ventricle). B: representative Western blot of left ventricular tissue. Black arrowhead: Na_v1.5 α-subunit; white arrowhead: vinculin (loading control). C: densitometric quantification of Na_v1.5 intensities normalized to the respective band intensities of vinculin, and then to the mean relative intensity for the wt. n = 6 hearts (one left ventricular tissue per heart) per genotype. Each data point represents one isolated heart. A Kruskal–Wallis test (unpaired) revealed a significant difference between the four groups (P < 0.01). Post hoc comparison between two groups was performed with Dunn’s multiple comparisons test; ^*P < 0.05, ^**P < 0.01. KO, knockout; TN-C, tenascin C; wt, wild type.

Figure 4. Effect of TN-C siRNA treatment of mdx mice on the peak I_Na densities of ventricular cardiomyocytes.

A: original traces of I_Na recorded from myocytes derived from a vehicle (PBS)-(left) or siRNA-treated (right) mdx mouse. B: comparison of the peak I_Na density-voltage relationships of mdx myocytes derived from vehicle (PBS)- or siRNA-treated mdx mice. C: respective peak I_Na density comparison at current maximum (−42 mV). Data are expressed as means ± SE. ^*P < 0.05. mdx vehicle vs. mdx TN-C siRNA: significantly different (P < 0.05) from 62 to −37 mV. n numbers represents the number of hearts (animals) used per group; the total number of cells is also given. The membrane capacitance values of myocytes from vehicle- and siRNA-treated mice were similar (P = 0.33). I_Na, sodium current; TN-C, tenascin C.

Figure 5. Effects of 24 h incubation of wild-type (wt) ventricular cardiomyocytes with human recombinant tenascin-C (hTN-C).

A: peak I_Na density-voltage relationships of untreated and hTN-C (30 ng/mL)-treated wt myocytes. B: respective comparison of the peak I_Na densities at current maximum (−47 mV); ***P < 0.001. wt ctrl vs. wt 30 ng/mL TN-C: significantly different (P < 0.05) from −62 to −7 mV. C: representative Western blot (top) performed on isolated untreated and hTN-C (30 ng/mL)-treated wt ventricular cardiomyocytes. Black arrowhead: Na_v1.5 α-subunit; white arrowhead: vinculin (loading control). Densitometric quantification of Na_v1.5 α-subunit protein expression (membrane fraction) relative to vinculin (bottom), and then normalized to the mean relative intensity for the wt (n = 6 hearts for cardiomyocyte isolations; cells from each heart were plated onto two dishes: untreated and treated). A Wilcoxon matched-pairs signed rank test revealed a significant difference; *P < 0.05. D: peak I_Na density-voltage relationships of untreated and hTN-C (1 lg/mL)-treated wt myocytes. E: respective comparison of the peak I_Na densities at current maximum (− 47 mV); ^***P < 0.001. wt ctrl vs. wt 1 lg/mL TN-C: significantly different (P < 0.05) from −57 to −12 mV. F: comparison of the kinetics of fast inactivation (represented by decay half-times) at −47 mV between untreated wt control (ctrl) and hTN-C (30 ng/mL)-treated wt myocytes. ns, not significant. G: representative I_Na traces from an untreated control (ctrl) wt rat ventricular cardiomyocyte and from a 24 h hTN-C (30 ng/mL)-treated wt rat ventricular cardiomyocyte. H: peak I_Na density-voltage relationships of untreated and hTN-C (30 ng/mL)-treated wt rat myocytes. I: respective comparison of the peak I_Na densities at current maximum (− 47 mV); *P < 0.05. Rat wt ctrl vs. rat wt 30 ng/mL TN-C: significantly different (P < 0.05) from −57 to −7 mV. The cardiomyocytes originated from 10 to 11-wk-old wt Sprague–Dawley rats. J: original I_Na traces from an untreated control (ctrl), and a 24 h hTN-C (30 ng/mL)-treated wt mouse ventricular cardiomyocyte elicited by a 25 ms test pulse to −37 mV. The pulse protocol designed to test for steady-state fast inactivation is displayed in the inset. K: comparison of the voltage dependencies of steady-state fast inactivation between untreated (ctrl) and hTN-C (30 ng/mL)-treated wt mouse ventricular cardiomyocytes. V₅₀, the voltage at which half-maximum inactivation occurred, was -79 ± 2 mV for ctrl and −76 ± 2 mV for TN-C-treated myocytes and not significantly different (P = 0.16). In all experiments described in this figure, the membrane capacitance values of untreated and hTN-C-treated myocytes were similar (P always > 0.15). I_Na, sodium current.

Figure 6. Acute application of TN-C to wt and mdx ventricular cardiomyocytes does not affect peak I_Na.

A: normalized peak I_Na of wt cardiomyocytes before, during superfusion with 30 ng/mL (top) or 1 μg/mL (bottom) hTN-C, and after washout. During the whole recording period I_Na was elicited by the pulse protocol shown in the inset at 1 Hz frequency. n = 26 cells from 4 wt hearts (30 ng/mL hTN-C) and 23 cells from 4 wt hearts (1 μg/mL hTN-C). B: respective experiment on mdx cardiomyocytes. n = 30 cells from 3 mdx hearts (30 ng/mL hTN-C) and 24 cells from 3 mdx hearts (1 μg/mL hTN-C). Data are expressed as means ± SE. I_Na, sodium current; TN-C, tenascin C.

Figure 7. Effect of different integrin subunit blocking antibodies on the TN-C incubation-induced reduction in peak I_Na density.

A: comparison of the peak I_Na density-voltage relationships of wt mouse ventricular cardiomyocytes either untreated (ctrl), 24 h hTN-C (30 ng/mL)-treated, 24 h hTN-C (30 ng/mL)- and integrin α-7 antibody (1:500)-treated, or only integrin α-7 antibody (1:500)-treated. B: respective peak I_Na density comparison at current maximum (−47 mV). ^*P < 0.05, ^***P < 0.001. ctrl vs. TN-C: significantly different (P < 0.05) from −52 to −12 mV; TN-C vs. TN-C ITGA7: significantly different (P < 0.05) from −57 to −12 mV. The membrane capacitance values of all groups were similar (P always > 0.26). C: Representative Western blot performed on isolated untreated, hTN-C (30 ng/mL)-treated, and hTN-C (30 ng/mL)-plus integrin α-7 antibody-treated wt ventricular cardiomyocytes. Black arrowhead: Na_v1.5 α-subunit; white arrowhead: vinculin (loading control). D: densitometric quantification of Na_v1.5 α-subunit protein expression (membrane fraction) relative to vinculin, and then normalized to the mean relative intensity for the wt control [n = 6 hearts for cardiomyocyte isolations; cells from each heart were plated onto three dishes: untreated, treated with TN-C, and treated with TN-C and integrin α-7 antibodies (1:500)]. A Kruskal–Wallis test (P = 0.0004) with Dunns’s multiple comparison revealed significant differences; *P < 0.05, **P < 0.01. E: comparison of the peak I_Na densityvoltage relationships of wt mouse ventricular cardiomyocytes either untreated (ctrl), 24 h hTN-C (30 ng/mL)-treated, 24 h hTN-C (30 ng/mL)- and integrin b-1 antibody (1:500)-treated, or only integrin b-1 antibody (1:500)-treated. F: respective peak I_Na density comparison at current maximum (−47 mV). *P < 0.05, **P < 0.01. ctrl vs. TN-C: significantly different (P < 0.05) from −72 to −7 mV; ctrl vs. TN-C ITGB1: significantly different (P < 0.05) from −72 to −7 mV. The membrane capacitance values of all groups were similar (P always > 0.71). G: comparison of the peak I_Na density-voltage relationships of wt mouse ventricular myocytes either untreated (ctrl), 24 h hTN-C (30 ng/mL)-treated, 24 h hTN-C (30 ng/mL)- and integrin α-5 antibody (1:500)-treated, or only integrin α-5 antibody (1:500)-treated. H: respective peak I_Na density comparison at current maximum (−47 mV). *P < 0.05, **P < 0.01. ctrl vs. TN-C: significantly different (P < 0.05) from −57 to −17 mV; ctrl vs. TN-C ITGA5: significantly different (P < 0.05) from −62 to −17 mV. The membrane capacitance values of all groups were similar (P always > 0.2). I_Na, sodium current; TN-C, tenascin C.

Figure 8. Summary of the findings of the present study and their potential impact.

A: in healthy ventricular cardiomyocytes, a large number of Na channels are expressed in the plasma membrane, which give rise to a big I_Na and consequently a high ventricular conduction velocity. B: in dystrophic cardiomyocytes, TN-C interacts with integrin α-7, leading to reduced Na channel expression, diminished I_Na, and finally slowed ventricular conduction. C: by blocking integrin α-7 activity in dystrophic cardiomyocytes, normal Na channel expression, I_Na, and ventricular conduction velocity are restored. Figure created with a licensed version of Biorender.com. I_Na, sodium current; TN-C, tenascin C.