Tmem65 is critical for the structure and function of the intercalated discs in mouse hearts

Abstract

The intercalated disc (ICD) is a unique membrane structure that is indispensable to normal heart function, yet its structural organization is not completely understood. Previously, we showed that the ICD-bound transmembrane protein 65 (Tmem65) was required for connexin43 (Cx43) localization and function in cultured mouse neonatal cardiomyocytes. Here, we investigate the functional and cellular effects of Tmem65 reductions on the myocardium in a mouse model by injecting CD1 mouse pups (3-7 days after birth) with recombinant adeno-associated virus 9 (rAAV9) harboring Tmem65 shRNA, which reduces Tmem65 expression by 90% in mouse ventricles compared to scrambled shRNA injection. Tmem65 knockdown (KD) results in increased mortality which is accompanied by eccentric hypertrophic cardiomyopathy within 3 weeks of injection and progression to dilated cardiomyopathy with severe cardiac fibrosis by 7 weeks post-injection. Tmem65 KD hearts display depressed hemodynamics as measured echocardiographically as well as slowed conduction in optical recording accompanied by prolonged PR intervals and QRS duration in electrocardiograms. Immunoprecipitation and super-resolution microscopy demonstrate a physical interaction between Tmem65 and sodium channel β subunit (β1) in mouse hearts and this interaction appears to be required for both the establishment of perinexal nanodomain structure and the localization of both voltage-gated sodium channel 1.5 (NaV1.5) and Cx43 to ICDs. Despite the loss of NaV1.5 at ICDs, whole-cell patch clamp electrophysiology did not reveal reductions in Na⁺ currents but did show reduced Ca²⁺ and K⁺ currents in Tmem65 KD cardiomyocytes in comparison to control cells. We conclude that disrupting Tmem65 function results in impaired ICD structure, abnormal cardiac electrophysiology, and ultimately cardiomyopathy.

Fig. 1. Tmem65 KD leads to dilated cardiomyopathy, fibrosis, and increased mortality.

a Schematic illustration for establishing a Tmem65 KD mouse model. b Kaplan–Meier survival curve showing all Tmem65 KD mice (red line, n = 30 total mice; blue dotted line, n = 16 males; green dotted line, n = 14 female) died within 7 weeks, while scrambled control mice (blue line, n = 29) remained healthy. Greater than 50% of Tmem65 KD mice died 3 weeks post injection. Experiments were performed in mice of both sexes. c M-mode recordings of mouse left ventricles showed depressed cardiac contractility in a Tmem65 KD mouse heart compared to a scrambled control heart. d Pulsed wave Doppler measurements at aortic orifice measuring blood velocity in a Tmem65 KD mouse heart and control mouse heart. The velocity of blood flow was reduced, or flow waveform was irregular or missing (yellow arrows) in Tmem65 KD mouse hearts. Echocardiography was performed in 6 female mice per group. e Morphology of Tmem65 KD and controls hearts. Tmem65 KD hearts were grossly larger compared to control hearts, including both atria and ventricles. Blood clots were always found in the left atrium (black arrow). Scale bar = 5 mm. f H&E staining of atria and ventricles of Tmem65 KD and control hearts. g Higher magnification of H&E-stained hearts showing many eosin-negative lesions, haematoxylin-positive nuclei. h Masson trichrome staining showing collagen deposits (blue color). Scale bar = 50 µm. i Quantification of cardiac fibrosis of cardiac tissues stained with Masson Trichrome dye. A nearly tenfold increase in fibrosis in Tmem65 KD hearts compared to control hearts. **P < 0.01. Histopathology was performed in 6 female mouse hearts. j Immunoblots demonstrating changes of Tmem65 protein levels in Tmem65 KD hearts. β-actin was included as a loading control. k Densitometry quantification of Tmem65 levels in j. Both long (25 kDa) and short (17 kDa) forms of Tmem65 were compared to the control samples. **P < 0.01, n = 4. All statistical analyses were performed by one-way ANOVA with Tukey’s post-hoc test. Data were expressed as mean ± standard error of the means.

Fig. 2. Tmem65 KD leads to eccentric hypertrophic cardiomyopathy and slowed cardiac conduction 3 weeks post injection.

a Morphology of Tmem65 KD and controls hearts. Tmem65 KD hearts were qualitatively larger than control hearts, especially the left atrium (black arrow). (scale bar = 5 mm). b Measurement of heart weight-to-tibia length ratio in Tmem65 KD and controls hearts. **P < 0.01, n = 6 per group. Statistical analyses were performed by one-way ANOVA with Tukey’s post-hoc test. Data were expressed as mean ± standard error of the means. c H&E staining of cardiac transverse sections. Scale bar = 500 µm. d Phalloidin (F-actin) staining of adult mouse cardiomyocytes isolated from Tmem65 KD or control hearts. Scale bar = 20 µm. e Quantifications of cell dimensions of isolated adult mouse cardiomyocytes. Both lengths (left panel) and width (right panel) significantly increased in Tmem65 KD cardiomyocytes. **p < 0.01, n > 90 cells. Three mice per group. Statistical analyses were performed by one-way ANOVA with Tukey’s post-hoc test. Data were expressed as mean ± standard error of the means. f Immunoblots of multiple hypertrophic markers including α-ACTININ, cTnT, FHL1, MYH7, and phosphorylated ERK1/2 with densitometry quantifications. **P < 0.01; *P < 0.05. Statistical analyses were performed by one-way ANOVA with Tukey’s post-hoc test. Data were expressed as mean ± standard error of the means. g qPCR showing increased transcript levels of natriuretic peptide A and B (**P < 0.01) in Tmem65 KD hearts. Experiments were performed in more than 3 mice of both sexes. Statistical analyses were performed by one-way ANOVA with Tukey’s post-hoc test. Data were expressed as mean ± standard error of the means. h Representative images of cardiac optical mapping. Transfer of voltage-sensitive dye di-4-ANEPPS was monitored in control and Tmem65 KD hearts at the Sinus rhythm (top images) or paced at 11 Hz (bottom images). i The conduction velocity of the right ventricular free wall of Tmem65 KD hearts (n = 6 hearts, 4 females and 2 males) was significantly lower than that of control hearts at the Sinus rhythm or paced at 11 Hz (*P < 0.05). All cells were isolated from male mouse hearts. For optical mapping, data were expressed as mean ± SEM. Differences between scrambled shRNA and Tmem65 shRNA groups at sinus rhythm and during pacing were assessed with a two-way ANOVA with Sidak’s multiple comparison test. Data were presented as mean ± standard deviation.

Fig. 3. Tmem65 KD is associated with ICD defects in mouse hearts.

a Immunoblots demonstrating changes of Cx43 protein levels in Tmem65 KD hearts. α-Tubulin was included as a loading control. b Immunofluorescence of Cx43 in control and Tmem65 KD hearts. Internalized Cx43 proteins were seen in Tmem65 KD mouse hearts co-localized with ICD marker NCAD in control hearts (scale bar = 20 µm). c Pearson correlation coefficient (PCC)-based quantification showing reduced correlation between Cx43 and NCAD in Tmem65 KD hearts. **P < 0.01. d Immunofluorescence of Cx43 and Cox4 in control and Tmem65 KD hearts. Co-localization of Cx43 and Cox4 was visualized (white arrows). Scale bar = 50 µm. e Pearson correlation coefficient (PCC)-based quantification showing increased correlation between Cx43 and Cox4 in Tmem65 KD hearts. f Representative transmission electron microscopy of Tmem65 KD (bottom panel) and control hearts (top panel). Aberrant intercalated discs (black arrows) and dismantled myofibers (yellow arrows) were found in Tmem65 KD hearts. Scale bar = 500 nm. g Quantification of ICDs in electron micrographs showing the tendency in increasing ICD length (upper panel) and a significant decrease in ICD curvature smoothness (bottom panel). N.S., Not Significant; **P < 0.01. h Immunofluorescence of DSG showing no changes of DSG localization with the ICD marker NCAD in Tmem65 KD hearts. i Immunofluorescence of PKP2 showing reduced localization with the ICD marker NCAD in Tmem65 KD hearts. Scale bar = 50 µm. j Immunofluorescence of desmin and F-actin in control (upper panels) and Tmem65 KD (lower panels) hearts. Reduced sarcomeric desmin (white arrows) or longitudinal desmin (yellow arrow) were found in Tmem65 KD hearts. Scale bar = 50 µm. Experiments were performed in mice of both genders. Experiments were performed in greater than 3 mice of both sexes. Statistical analyses were performed by one-way ANOVA with Tukey’s post-hoc test. Data were expressed as mean ± standard error of the means.

Fig. 4. Tmem65 KD leads to losses of NaV1.5 channels at ICDs.

a An immunoblot showing reduced NaV1.5 and β1 protein levels in Tmem65 KD mouse hearts when compared to control samples (n = 4 hearts). **P < 0.01. Statistical analysis was performed by one-way ANOVA with Tukey’s post-hoc test. Data were expressed as mean ± standard error of means. b Immunofluorescence of ICD regions in scrambled and Tmem65 KD hearts, showing the loss of ICD-bound NaV1.5 in Tmem65 KD hearts. β1 staining showed scattered puncta (white arrows) or aggregates (yellow arrows) also in Tmem65 KD hearts. c Immunoprecipitation assays using NCAD antibody and mouse cardiac lysates showed a reduced NaV1.5/NCAD interaction in Tmem65 KD mouse hearts. d Representative images of sodium current (I_Na⁺) in control vs. Tmem65 KD cardiomyocytes. e Current density-voltage (I-V) plot shows that I_Na⁺ density did not significantly differ between scrambled (black) and Tmem65 KD (red) cardiomyocytes. f Quantification of peak I_Na⁺ density in scrambled and Tmem65 KD cardiomyocytes. No statistical difference was found. g Voltage-dependence of I_Na⁺ activation and steady-state inactivation in cardiomyocytes with the least-square fits to the Boltzmann function. A statistical significance was found between Tmem65 KD and control cardiomyocytes in activation state, but not in inactivation state. All measurements are summarized in Table 4. N = 16 control cells and 25 Tmem65 KD cells for in I_Na⁺. All cells were isolated from male mouse hearts. For patch-clamp experiments, data were presented as mean ± standard deviation. The difference of current density was compared using unpaired student t-test.

Fig. 5. Tmem65-β1 interaction is required for preserving the perinexus in mouse hearts.

a Representative super-resolution microscopy showing Tmem65 (left panel, green) and β1 (mid panel, red) co-localized (right panel, yellow) in the enface orientation of the ICD in mouse hearts. Scale bar = 1 µm. Five images were taken from 2 mouse hearts for determining PCC. b Co-immunoprecipitation assays using β1, showing that Tmem65 was co-precipitated (*) from mouse hearts by β1 antibody (#). c Co-immunoprecipitation assays showing that Tmem65-FLAG (*) co-precipitated β1 (#) in transfected HEK293T cells. d Representative transmission electron microscopy showing an aberrant perinexal domain in the Tmem65 KD heart. Perinexus next to the Gap junction (black arrow) was highlighted in yellow. Scale bar = 100 nm. e Quantification of intermembrane distances at perinexal junctions and non-perinexal regions within ICDs in Tmem65 KD and control hearts. Tmem65 silencing led to increased perinexal intermembrane distance (right panel) in comparison to control hearts, but did not affect non-perinexal intermembrane distances (left panel). ***P < 0.01. Experiments were performed in mice of both sexes. Experiments completed in 4 Tmem65 KD hearts and 3 control hearts of both sexes. For the image analysis of the EM images, nested t-test comparisons between control and treatment were carried out using the log-transformed data set. Data were presented as mean ± standard deviation.

Fig. 6. Tmem65 KD leads to changes in APs and ion fluxes in mouse adult cardiomyocytes.

All measurements were recorded by whole-cell voltage clamping in cardiomyocytes, isolated from scrambled (black) and Tmem65 KD cells (red). a Representative images of calcium current (I_Ca) in cardiomyocytes. b Current density-voltage (I–V) plot shows that I_Ca density significantly differ between −20~40 mV voltage in scrambled and Tmem65 KD cardiomyocytes. c Quantification of peak I_Ca density in scrambled and Tmem65 shRNA cardiomyocytes. I_Ca was significantly reduced in Tmem65 KD cardiomyocytes in comparison to control cells (P < 0.01). d Voltage-dependence of I_Ca activation and steady-state inactivation in cardiomyocytes following the least-square fitting to the Boltzmann function. Statistical significance was only found in I_Ca activation and slop between Tmem65 KD and control cardiomyocytes. e Representative recordings of outward I_k⁺ in scrambled (black) and Tmem65 KD (red) mice. f Quantification for total outward currents (I_k, outward) was significantly higher in Tmem65 KD cells. *P < 0.05. g Quantification for, I_k, _slow1 was reduced in Tmem65 KD cardiomyocytes. *P < 0.05. h Quantification for I_{k, slow2}. i Quantification for remaining current, I_ss. j Representative recordings of 4-AP sensitive outward I_k⁺ following 4-AP administration in scrambled (black) and Tmem65 KD (red) mice. k Quantification for transient outward currents (I_to) was higher in Tmem65 KD cells. l Quantification of 4-AP-insenitive I_{to, fast} (I_to,f) showing increased current in Tmem65 KD cells. *P < 0.05. m Quantification of 4-AP-sensitive I_{to, slow} (I_to,s) showing increased current in Tmem65 KD cells. **P < 0.01. n Quantification for 4-AP sensitive I_{k, slow1} was significantly reduced in Tmem65 KD cells. All measurements are summarized in Table 4. n = 16 control cells and 25 Tmem65 KD cells for I_Ca measurements. n = 10 cells per group for I_k⁺ measurement. All cells were isolated from male mouse hearts. Data were presented as mean ± standard deviation. The difference of current density was compared using unpaired student t-test.

Fig. 7. Schematic illustration for the role of Tmem65 in mouse hearts and the summary of this study.

In control cardiomyocytes, Tmem65-β1 interaction is responsible for stabilizing the perinexus in the ICD and is needed for Cx43 and NaV1.5 localization to the ICD (image on the left). Conversely, reduced Tmem65 is associated with destabilized perinexus, reduced Cx43 and NaV1.5 at the ICD (image on the right). Desmosomes were also impaired in Tmem65 KD hearts. Specifically, the striated pattern of intermediate filament desmin and the localizations of desmosome proteins DSP and DSG2 to ICDs are markedly reduced in Tmem65 KD hearts.