Methazolamide Attenuates the Development of Diabetic Cardiomyopathy by Promoting β-Catenin Degradation in Type 1 Diabetic Mice

Diabetes is a chronic disease characterized by high blood glucose level caused by insulin deficiency (type 1 diabetes mellitus [T1DM]) or insufficient insulin secretion (type 2 diabetes mellitus [T2DM]). Diabetic cardiomyopathy (DCM) is a common and severe complication of diabetes, which is characterized by cardiac functional and structural abnormalities in the absence of coronary artery disease or hypertension (1). DCM is characterized by myocardial hypertrophy, interstitial and perivascular fibrosis, and ventricular diastolic and systolic dysfunction; these manifestations ultimately contribute to heart failure (1–3). Despite the magnitude of DCM-related consequences in patients with diabetes (4), there is a lack of effective pharmacotherapy for DCM.

Carbonic anhydrases (CAs) participate in organismal pH regulation and ion transport by catalyzing the reversible conversion between carbon dioxide and bicarbonate (5). Several studies have demonstrated high CA levels in various heart diseases (e.g., myocardial infarction, cardiac hypertrophy, heart failure, valvular heart disease, and T2DM) (6–13). CA inhibitors have been shown to mitigate cardiomyocyte abnormality and dysfunction induced by various stressors. Benzolamide (BZ) and ethoxzolamide (ETZ) were shown to alleviate the reduction of left ventricular ejection fraction and onset of interstitial fibrosis remodeling in a rat myocardial infarction model (12). BZ ameliorated oxidative damage and ventricular systolic dysfunction following ischemia-reperfusion (14). ETZ impeded isoprenaline-induced hypertrophy in neonatal rat cardiomyocytes (NRCMs) (15). These findings indicate that CA inhibitors may play a protective role in the context of heart diseases; however, the underlying mechanisms of these effects are not well characterized.

Methazolamide (MTZ), a CA inhibitor, is used to treat various types of glaucoma owing to its suppressive effect on CAs in the ciliary body, which reduces aqueous humor production and intraocular pressure. MTZ reportedly can inhibit angiotensin II– or phenylephrine-induced adult cardiomyocyte hypertrophy (15). Furthermore, MTZ was shown to exhibit a hypoglycemic effect in streptozotocin (STZ)–induced diabetic rats and diabetic db/db mice; it also improves glucose tolerance in diet-induced insulin-resistant obese mice (16). MTZ was shown to reduce HbA_1c levels in diabetic db/db mice (16 ± 5% [151 ± 31 mmol/mol]) (16) and patients with T2DM (−0.39% [95% CI −0.82, 0.04]; −4.3 mmol/mol [−9.0, 0.4]) (17). To the best of our knowledge, no studies have investigated the role of MTZ in the treatment of DCM.

The Wnt/β-catenin pathway is highly conserved in animals and participates in cell fate specification, cell proliferation, and embryonic development (18). β-Catenin is a key protein in the Wnt/β-catenin pathway. Upon activation of Wnt signaling, β-catenin escapes from the β-catenin degradation complex, translocates to the nucleus, and interacts with transcription factor/lymphoid enhancer factor (TCF/LEF) to initiate gene transcription (18,19). The Wnt/β-catenin pathway reportedly plays an important role in myocardial hypertrophy and heart failure. High expressions of β-catenin have been observed in phenylephrine- or endothelin 1–induced cardiomyocyte hypertrophy; knockdown of β-catenin in NRCMs caused marked reduction in the cell surface area (20,21). In our previous study, heart tissues of patients with ischemic heart disease or idiopathic dilated cardiomyopathy showed activation of the Wnt/β-catenin pathway, nuclear accumulation of β-catenin, and enhancement of target gene expression (22). The Wnt/β-catenin pathway participates in type 1 DCM. In particular, β-catenin activation has been observed in STZ-induced DCM (23,24). However, no studies have investigated the effect of MTZ on the Wnt/β-catenin pathway, especially in the context of DCM.

Therefore, we hypothesized that MTZ prevents DCM by inhibiting canonical Wnt/β-catenin signaling. In this study, we investigated whether the potential cardioprotective effects of MTZ are associated with the regulatory effect of β-catenin; we also evaluated the underlying mechanisms in a T1DM mouse model and high glucose (HG)–treated NRCMs and adult mouse cardiomyocytes. Our results showed that MTZ potentiated the AXIN1–β-catenin interaction to increase degradation of β-catenin, thus reducing cardiomyocyte hypertrophy and improving cardiac dysfunction.

The mouse experiments in this study were approved by the Ethics Committee of Guangzhou Medical University (Guangzhou, China) and were performed in accordance with the Animal Research: Reporting of In Vivo Experiments guidelines. Male C57BL/6J mice (aged 7–8 weeks; weight 20 ± 2 g) were housed in a controlled environment (temperature 22 ± 3°C; humidity 50–70%; 12-h light/dark cycle) and provided ad libitum access to food and water.

After acclimatization for 1 week, mice were administered daily intraperitoneal injection of STZ 45 mg/kg body weight (BW) per day (#S0130; Sigma-Aldrich, St. Louis, MO) for 5 consecutive days. Mice in the control group were administered an equivalent volume of citrate buffer (0.1 mol/L, pH 4.5). Mice with a random blood glucose level ≥16.7 mmol/L were presumed to have T1DM. Mice with a blood glucose level <16.7 mmol/L were administered injection of STZ 45 mg/kg BW for another day. Following 4 weeks of T1DM modeling, mice were administered MTZ (#S4039; Selleck Chemicals, Houston, TX) by intragastric gavage at doses of 10, 25, or 50 mg/kg BW per day for 16 weeks. The control mice were administered MTZ (50 mg/kg BW) or an equivalent volume of vehicle via daily intragastric gavage. MTZ was prepared fresh in 2% DMSO (#D8418; Sigma-Aldrich) (30% PEG300, 2% Tween 80, and 66% double-distilled H₂O).

Random and fasting blood glucose levels were monitored using a glucometer (OneTouch Ultra Mini Blood Glucose Monitoring System; Johnson & Johnson, New Brunswick, NJ) with blood samples collected through the caudal vein at 2-week intervals. After 16 weeks of MTZ treatment, an intraperitoneal glucose tolerance test was performed in mice after fasting for 16 h. The mice were administered intraperitoneal injection of 20% glucose solution (2 g/kg BW); blood glucose levels were measured at 0, 30, 60, and 120 min after glucose injection using a glucometer with blood collected through the caudal vein. The area under the blood glucose curve (AUC) was measured in accordance with the trapezoidal rule. For collection of urine samples, mice were housed in metabolic cages for 24 h with free access to food and water. Urine glucose was measured using a urine analyzer (Cobas u 601 urine analyzer; Roche AG, Basel, Switzerland).

After 16 weeks of MTZ treatment, all mice underwent cardiac function assessment using a VisualSonics Vevo 2100 Imaging System (VisualSonics Inc, Toronto, Ontario, Canada) equipped with an MS-400 (15 MHz) linear array ultrasound transducer. With the mice under isoflurane anesthesia, the left ventricle was identified in parasternal long-axis and short-axis views, and several left ventricular indices were measured in M-mode at the papillary muscle level. Hemodynamic indices were measured from pulse-wave Doppler images. Global longitudinal strain (GLS) was measured from speckle tracking echocardiography in B-mode.

NRCMs were isolated as described previously (25). Briefly, the heart tissues of 1–3-day-old Sprague-Dawley rats were minced and then digested with pancreatic enzymes and type II collagenase. NRCMs were dissociated by continuous washing, collected after attachment and centrifugation, and cultured in 10% FBS-supplemented DMEM containing 4.5 g/L glucose. To establish an in vitro diabetic model, NRCMs were cultured in 1% FBS-supplemented DMEM containing 1 g/L glucose for 14 h and then treated with glucose (33 mmol/L) for 48 h. For subsequent experiments, NRCMs were simultaneously treated with glucose (33 mmol/L)/MTZ (25, 50, 100, 200 μmol/L; #S4039; Selleck Chemicals) for 48 h. SKL2001 (40 μmol/L; #S8320; Selleck Chemicals), Wnt3a (200 ng/mL; #5036-WN-010; R&D Systems, Minneapolis, MN), MG132 (1 μmol/L; Sigma-Aldrich), chloroquine (10 μmol/L; Sigma-Aldrich), or bafilomycin A1 (2.5 nmol/L; Sigma-Aldrich) was used for further studies. Transfections were performed with negative control (NC) or AXIN1 siRNA using Lipofectamine RNAiMAX reagent (#13778150; Invitrogen, Carlsbad, CA), in accordance with the manufacturer’s instructions. The siRNA sequences are shown in Supplementary Table 1.

Adult mouse cardiomyocytes were isolated and cultured as described previously, with modifications (26). Briefly, the hearts of C57BL/6J mice were perfused with calcium-free digestion buffer with 300 U/mL collagenase II (Worthington Biochemical Corporation, Lakewood, NJ) for 3 min and changed to 28.57 μmol/L calcium-added digestion buffer for 10–15 min until hearts became soft, flaccid, and pale. Cardiomyocytes were harvested, subjected to gradient recalcification, and then cultured in 2.5% FBS-supplemented minimum essential medium with insulin-selenium-transferrin and blebbistatin for further studies. Adult mouse cardiomyocytes were treated with glucose (33 mmol/L) to establish an in vitro diabetic model. For subsequent experiments, adult cardiomyocytes were subjected to 24 h of glucose (33 mmol/L) in the absence or presence of MTZ (100 μmol/L) or 24 h of Wnt 3a (200 ng/mL) without or with MTZ (100 μmol/L).

Fresh hearts were fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned. The tissue sections were stained with hematoxylin-eosin or Masson’s trichrome for histopathological examination.

For Western blot assay, protein extracts were separated by SDS-PAGE and transferred to polyvinylidene difluoride membranes. The membranes were incubated successively with primary antibodies, horseradish peroxidase–conjugated secondary antibodies, and Pierce ECL Western Blot substrate for detection.

For immunoprecipitation, the protein extracts of hearts or NRCMs were cleared using Protein A/G PLUS-Agarose (#sc-2003; Santa Cruz Biotechnology, Dallas, TX) and incubated overnight with primary antibodies at 4°C. On the 2nd day, the suspension was incubated with Protein A/G PLUS-Agarose for 4 h, and the beads were washed three times with immunoprecipitation assay buffer. The proteins were collected and subjected to Western blot. The antibodies used are listed in Supplementary Table 2.

Fresh hearts were fixed in 4% paraformaldehyde, embedded in optimal cutting temperature compound, sectioned, and blocked with 3% BSA for 1 h. NRCMs were fixed with 4% paraformaldehyde, permeabilized with 0.1% Triton X-100, and blocked with 10% goat serum for 1 h. After tissue sections and cells had been blocked, they were incubated overnight with primary antibodies at 4°C and then incubated with fluorescence-conjugated secondary antibodies. The sections were coincubated with FITC-conjugated wheat germ agglutinin (GeneTex, Inc., Irvine, CA). Nuclei were stained using DAPI (Sigma-Aldrich). Confocal microscopy was used to analyze immunofluorescence, and the cell cross-sectional areas were measured by ImageJ software. The antibodies used are listed in Supplementary Table 2. For negative controls primary antibodies were omitted or substituted with rabbit serum under identical conditions.

Total RNA was extracted from heart tissues and NRCMs using TRIzol reagent (Ambion, Waltham, MA). Seven hundred nanograms of total RNA from each sample was reverse transcribed using 1 µL Evo M-MLV RTase Enzyme Mix and 1 µL RT Primer Mix (Oligo dT [18T] Primer and Random 6-mers) in a total volume of 20 µL, as per the manufacturer’s instructions (#AG11711; AG Bio, Hunan, China). Quantitative real-time PCR was performed by SYBR Green Premix Ex Taq (Takara Bio, Shiga, Japan) on an ABI StepOne Real-Time PCR system (Thermo Fisher Scientific, Waltham, MA). The primer sequences for each gene are listed in Supplementary Table 1. The gapdh was used as the reference gene for quantitative real-time PCR normalization.

Molecular docking analysis was used to explore the potential binding model and affinity of MTZ for β-catenin and AXIN1. The X-ray crystal structures of β-catenin (Protein Data Bank [PDB] ID: 2Z6H) and the AXIN1–β-catenin cocomplex structure (PDB ID: 1QZ7) were chosen and modeled; potential binding sites were then chosen as the docking centers. Ligand–receptor docking was carried out using the Surflex-Dock GeomX (SFXC) in SYBYL-X 2.1.1 software (Tripos, Princeton, NJ). Total score (unit: −log₁₀) was used to evaluate docking results, and the highest score was chosen as the binding affinity in the best binding mode. A two-dimensional diagram of the ligand–receptor interaction was generated using Discovery Studio Visualizer 2020 (BIOVIA; Dassault Systèmes, Radnor, PA).

Data are presented as mean ± SD. Data sets with more than two groups were analyzed using one-way ANOVA, followed by Tukey post hoc analysis when homogenous variance was present; in case of lack of homogenous variance, Welch’s ANOVA followed by Dunnett T3 post hoc analysis were performed. P values <0.05 were considered indicative of statistical significance. The above analyses were carried out with GraphPad Prism 8.0 and SPSS Statistics 18.0.

The data sets generated during and/or analyzed during the current study are available from the corresponding author upon reasonable request. No applicable resources were generated during this study.

After 4 weeks of T1DM modeling in mice, the mice underwent intragastric gavage of MTZ (10, 25, or 50 mg/kg BW/day for 16 weeks) (Fig. 1A). MTZ did not reduce the blood glucose level in nondiabetic mice (Fig. 1B and Supplementary Fig. 1A and B). However, MTZ had a hypoglycemic effect in T1DM mice. The MTZ-related decrease in random blood glucose was first observed after 2 weeks of treatment in all MTZ groups; the decrease was statistically significant (and remained stable) after 6 weeks of treatment in all MTZ groups, compared with the T1DM control group (Supplementary Fig. 1B). The glucose-lowering effect of MTZ on fasting blood glucose was statistically significant after 8 weeks of treatment in all MTZ groups, compared with the T1DM control group (Supplementary Fig. 1A). After 16 weeks of treatment, MTZ at 25 and 50 mg/kg BW significantly reduced both random blood glucose and fasting blood glucose level in T1DM mice, compared with the T1DM control group (Fig. 1B). Moreover, a glucose-lowering tendency was observed with MTZ at 10 mg/kg BW, but the decrease was not statistically significant.

MTZ at 50 mg/kg BW reduced the total urine glucose concentration compared with the untreated control group; however, no significant differences in this respect were observed between the untreated control group and the 10 mg/kg or 25 mg/kg BW groups. Furthermore, there was no significant change in urinary excretion volume in any of the MTZ-treated T1DM groups (Supplementary Fig. 1C). Thus, the hypoglycemic effect of MTZ was not dependent on enhanced glucose excretion in urine, which is consistent with previous findings (16).

To examine the effect of MTZ on glucose tolerance, the mice were subjected to an intraperitoneal glucose tolerance test. After the test, the T1DM control group showed rapid increase in blood glucose level (Fig. 1C). AUC in the T1DM control group was much higher than that in the untreated control group; this AUC was significantly reduced by administration of MTZ at 50 mg/kg BW. These findings indicate that MTZ improved glucose homeostasis in T1DM mice.

Next, we sought to investigate the effect of MTZ on cardiac morphology and function in the setting of T1DM. Both heart weight (HW) and BW were decreased in the T1DM control group compared with the untreated control group (Fig. 1D). However, there were no differences in HW or BW between the MTZ-treated groups and the T1DM control group. The HW/BW ratio was not altered in any group. Next, hematoxylin-eosin and Masson staining were performed to analyze cardiac morphology (Fig. 1E). T1DM hearts showed disordered myocardial fibers, broken myofilaments, obscure cell boundaries, larger cells, and severe perivascular fibrosis. MTZ treatment in diabetic mice alleviated these abnormalities.

Cardiac function was assessed by echocardiography analysis. Compared with the untreated control group, T1DM hearts were characterized by reduced left ventricular anterior wall (LVAW) thickness and left ventricular posterior wall (LVPW) thickness, which were apparent in both systole and diastole (Fig. 2A and Supplementary Table 3). The change in left ventricular internal diameter (LVID) at systole differed among groups. These changes reflected the poor systolic function of T1DM hearts. The LVAW thickness and LVPW thickness at systole were greater in the MTZ treatment groups (10 and 50 mg/kg), but there were no significant differences with respect to LVID. MTZ-treated nondiabetic hearts showed reduced thickness of LVAW and LVPW, but these differences were not statistically significant. Pulse-wave Doppler images showed decreased E and A peak amplitudes in T1DM hearts, indicative of a tendency toward lower E/A ratios, compared with the untreated control group (Fig. 2B and Supplementary Table 3). In T1DM hearts, MTZ treatment (especially 10 mg/kg BW) increased the E peak amplitude; concurrently, MTZ further reduced the A peak amplitude, thereby increasing the E/A ratio. Compared with the untreated control group, T1DM hearts showed decreased isovolumic contraction time and myocardial performance index; these effects were ameliorated by MTZ treatment (25 mg/kg BW) (Supplementary Table 3). MTZ administration to nondiabetic mice did not affect the above cardiac indices. Speckle tracking echocardiography showed lower absolute value of GLS in T1DM hearts compared with the untreated control group, which was indicative of impaired LV global longitudinal function. MTZ treatment increased the absolute value of GLS in T1DM hearts. MTZ-treated nondiabetic hearts showed reduced GLS, but the decrease was not statistically significant (Fig. 2C). Overall, MTZ alleviated T1DM-induced systolic and diastolic dysfunction.

Next, NRCMs were treated with HG (33 mmol/L) to establish an in vitro diabetic model that was used to investigate the effects of MTZ on cardiomyocyte biology. Several hypertrophic features were detected. Western blot and α-actinin staining results demonstrated that the HG environment upregulated the protein level of atrial natriuretic polypeptide (ANP) and increased cardiomyocyte size (Fig. 3A and B). MTZ (especially 100 and 200 μmol/L) inhibited cardiomyocyte hypertrophy in the HG group: both the ANP level and the cardiomyocyte size were reduced (Fig. 3A and B). These findings indicate that MTZ may ameliorate HG-induced hypertrophy and rescue cardiac function in T1DM mice.

Because MTZ is a CA inhibitor, we examined CA expression in various groups of mice. CA1 and CA2 levels in the T1DM control group were significantly higher than those in the untreated control group (Fig. 4A). MTZ downregulated the CA levels in T1DM mice in a dose-dependent manner. However, no inhibitory effect of MTZ was observed in the control group.

To determine the mechanism by which MTZ mediates cardiac function, we explored several signaling pathways. As shown in Supplementary Fig. 2, AMPK and AKT signaling pathways were not affected by MTZ in any of the groups. However, we observed substantial changes in the Wnt/β-catenin pathway. In T1DM hearts, the Wnt/β-catenin pathway was activated with increased active β-catenin and total β-catenin levels (Fig. 4B), which is consistent with previous findings (23,24). The downstream target of the Wnt/β-catenin pathway, cyclin D2, was also strongly expressed. The expression of activated β-catenin and cyclin D2 were indicative of β-catenin activity. MTZ caused a dose-dependent decrease in the expressions of active β-catenin, total β-catenin, and the downstream target, cyclin D2. Immunofluorescence analyses confirmed increased protein level and nuclear accumulation of β-catenin (green) in the T1DM control group; these changes were repressed by MTZ treatment (Fig. 4C). Furthermore, MTZ tended to enhance β-catenin level and activity in control hearts, but the enhancement was not statistically significant.

We used an in vitro HG model to investigate whether MTZ directly affects β-catenin in cardiomyocytes. The results in NRCMs were consistent with the findings in mouse hearts. HG induced higher levels of CA1 and CA2 in NRCMs, which were repressed by MTZ at 50, 100, and 200 μmol/L (Fig. 5A). With respect to Wnt/β-catenin signaling, MTZ (especially 100 and 200 μmol/L) suppressed the HG-induced enhancement of β-catenin level (Fig. 5B). Immunofluorescence analysis showed enhanced protein level and nuclear accumulation of β-catenin (green) in HG-treated NRCMs; these changes were attenuated by MTZ treatment (Fig. 5C).

To further investigate the role of MTZ in the regulation of Wnt/β-catenin signaling, NRCMs were treated with Wnt3a, a Wnt/β-catenin signaling activator (Fig. 5D). As expected, β-catenin was activated by Wnt3a, which led to elevated β-catenin level and downstream cyclin D2 expression. MTZ treatment (100 μmol/L) reduced β-catenin level and activity in the Wnt3a group. Similar changes were observed in NRCMs treated with another activator, SKL2001 (Supplementary Fig. 3A). Furthermore, Wnt3a treatment enhanced the expression of the hypertrophy-related protein ANP in the HG group (Fig. 5E). Greater cardiomyocyte size was also observed in the HG+Wnt3a group (Fig. 5F). Notably, MTZ repressed these Wnt3a-dependent hypertrophic changes.

The same experiments were performed in adult mouse cardiomyocytes (Fig. 5G–J). Western blot results (Fig. 5G and I) showed that both HG or Wnt3a induced higher levels of active β-catenin, β-catenin, cyclin D2, and hypertrophy-related protein ANP in adult cardiomyocytes, which were decreased by MTZ treatment (100 μmol/L). Immunofluorescence images showed that MTZ attenuated HG- or Wnt3a-dependent nuclear accumulation of β-catenin (Fig. 5H and J). The results in adult cardiomyocytes were consistent with the findings in NRCMs. Collectively, these findings indicated that MTZ impeded β-catenin in the HG environment; this treatment may alleviate HG-dependent hypertrophy and cardiac dysfunction.

As mentioned above, other CA inhibitors may also have cardioprotective effects in such heart diseases (12,14,15). However, the effects of these inhibitors on hearts with diabetes are not clear. Therefore, the in vitro HG model in NRCMs was used to assess the effect of treatment with MTZ, acetazolamide (ATZ; 10 μm), BZ (10 μm), or ETZ (100 μm). The results showed that CA inhibitors reduced HG-induced enhancement of CA1 and CA2; MTZ, rather than other CA inhibitors, decreased HG-induced β-catenin level (Supplementary Fig. 4A and B). Moreover, HG-dependent hypertrophic changes (such as higher ANP expression and larger cardiomyocyte size) were repressed by several CA inhibitors, especially MTZ (Supplementary Fig. 4B and C). To further investigate the role of CA inhibitors on β-catenin, NRCMs were treated with Wnt3a and CA inhibitors (Supplementary Fig. 4D). The results showed that MTZ and ATZ treatment significantly decreased active β-catenin level; MTZ treatment dramatically decreased cyclin D2 expression; ETZ also showed a tendency for lowering cyclin D2, but the decrease was not statistically significant. Therefore, the cardioprotective effects of MTZ might be mainly mediated via inhibition of β-catenin, while ATZ, BZ, and ETZ mainly showed cardioprotective effects through CA inhibition.

The findings thus far indicated an inhibitory effect of MTZ on β-catenin content and activity in the HG environment. To explore the mechanism by which MTZ influenced β-catenin, we examined the mRNA levels of β-catenin in heart tissues and NRCMs. Regardless of MTZ treatment, there were no significant changes in β-catenin mRNA levels, which indicated that MTZ did not influence the transcriptional regulation of β-catenin (Fig. 6A). β-Catenin can be degraded by ubiquitination and subsequent proteolysis in the canonical Wnt/β-catenin pathway, thus controlling intracellular β-catenin content (18,19). Thus, we used the proteasome inhibitor MG132 and multiple autophagy inhibitors (chloroquine and bafilomycin A1) in the subsequent experiment. MG132 alleviated the MTZ-dependent decrease in β-catenin, but autophagy inhibitors showed no influence on the MTZ-dependent inhibition of β-catenin (Fig. 6B and C). Immunoprecipitation in Wnt3a-treated NRCMs revealed slightly higher β-catenin ubiquitination level in the Wnt3a+MTZ group compared with the Wnt3a group (Fig. 6D). MG132 treatment further enhanced the β-catenin ubiquitination level. These findings indicated that MTZ repressed β-catenin by degradation rather than by transcription or autophagy.

In the canonical Wnt pathway, β-catenin is incorporated in a destruction complex with AXIN1, adenomatosis polyposis coli, glycogen synthase kinase 3β (GSK3β), and casein kinase 1α (CK1α) (18,19). β-Catenin undergoes phosphorylation by GSK3β and CK1α and subsequently ubiquitination and degradation by the proteasome. In this study, we detected the components of the destruction complex. MTZ treatment of T1DM hearts induced no significant changes in GSK3β and CK1α expression levels, but there was noticeable upregulation of AXIN1 (Fig. 7A and B). AXIN1 is a scaffold protein that binds to GSK3β and CK1α to provide a platform for β-catenin phosphorylation (27,28). To the best of our knowledge, no report has described the interactions among MTZ, AXIN1, and β-catenin. Thus, we used molecular docking analysis to determine whether MTZ interacted with β-catenin and AXIN1. β-Catenin reportedly binds to AXIN1 using an Armadillo repeat domain (29). Our analysis indicated that MTZ was docked in the Armadillo repeat domain of β-catenin (PDB ID: 2Z6H) with a total score of 4.0852, which suggested an interaction between MTZ and β-catenin (Fig. 7C and Supplementary Fig. 5A). MTZ was also docked in the interaction area between β-catenin and AXIN1 (PDB ID: 1QZ7) with a total score of 6.1111, implying that MTZ may be closely involved in the AXIN1–β-catenin interaction (Fig. 7C and Supplementary Fig. 5B). Therefore, we speculated that MTZ affected the AXIN1–β-catenin interaction rather than the phosphorylation of β-catenin by GSK3β and CK1α.

To investigate the effect of MTZ on AXIN1 and β-catenin, we performed immunoprecipitation analysis of mouse heart tissues. Our findings revealed weaker interactions between β-catenin and AXIN1 in T1DM hearts compared with the untreated control hearts; MTZ treatment increased the strength of this interaction in T1DM hearts (Fig. 7D). Immunofluorescence analysis of mouse hearts showed that colocalization of β-catenin (green) and AXIN1 (red) was less common in T1DM hearts than in untreated control hearts; MTZ treatment enhanced this colocalization in T1DM hearts (Fig. 7E).

In HG-induced NRCMs, MTZ treatment did not affect the protein levels of GSK3β and CK1α, but it caused AXIN1 upregulation (Fig. 8A and B). MTZ also did not affect Wnt3a- or SKL2001-induced protein levels of GSK3β and CK1α (Supplementary Fig. 3B and C).

Immunoprecipitation showed that MTZ treatment enhanced the AXIN1–β-catenin interaction in Wnt3a-treated NRCMs (Fig. 8C). These results were consistent with the findings in heart tissues. Furthermore, SKL2001 has been shown to disrupt the AXIN1–β-catenin interaction, thus enhancing β-catenin protein level and activity (30). Our results showed that MTZ treatment impaired SKL2001-induced β-catenin accumulation, suggesting that MTZ disrupted SKL2001-mediated AXIN1–β-catenin interaction (Supplementary Fig. 3A).

Subsequently, NRCMs were transfected with AXIN1 siRNA. Compared with the NC group, AXIN1 siRNA significantly reduced the protein and mRNA levels of AXIN1 (Supplementary Fig. 6). In Wnt3a-treated NRCMs, AXIN1 siRNA attenuated MTZ-induced β-catenin degradation, thus increasing the β-catenin level (Fig. 8D and E). Collectively, these findings implied that MTZ enhances the AXIN1–β-catenin interaction to facilitate β-catenin degradation.

In this study, we found that MTZ reduced the plasma glucose level in T1DM mice and alleviated structural and functional impairment of cardiomyocytes. MTZ reduced T1DM- or HG-induced β-catenin level and activity, thereby ameliorating cardiac inadaptability and cardiomyocyte hypertrophy. Furthermore, MTZ potentiated the AXIN1–β-catenin interaction to increase β-catenin degradation rather than affecting β-catenin transcription (Fig. 8F).

According to our results and previous report, MTZ showed a significant hypoglycemic effect in diabetic mice, which was not related to its diuretic effect (16). Konstantopoulos et al. (16) showed that MTZ reduced hepatic glucose production, enhanced glycolysis and insulin sensitivity to exert hypoglycemic effect, which are important approaches for the treatment of T1DM and T2DM. Further studies are required to unravel the underlying mechanisms of hypoglycemic effects. Additionally, as a CA inhibitor, MTZ may be dependent on CA inhibition to reduce plasma glucose level. It has been reported that CA5, a mitochondria-specific isoform, facilitates gluconeogenesis by providing HCO₃⁻ substrate (31–35). In mitochondria, pyruvate carboxylase catalyzes HCO₃⁻ and pyruvate to form oxaloacetate, which is the first step of gluconeogenesis. Cytoplasmic isoenzyme CA2 may assist CA5-mediated gluconeogenesis by providing HCO₃⁻ in the cytoplasm (31,35). Thus, CAs may serve as important novel therapeutic targets in T1DM. However, it is unclear whether MTZ affects CA-induced gluconeogenesis in T1DM–MTZ targets CA1 and CA2. Overall, the MTZ-related hypoglycemic mechanisms require further investigation.

Hyperglycemia plays an essential role in the pathophysiology of DCM; therefore, glycemic control is pivotal for ameliorating myocardial maladaptation in DCM (2). In this study, we observed a cardioprotective effect of MTZ in T1DM mice. Therefore, the hypoglycemic effect of MTZ on DCM must be considered. The glucose-lowering effect of MTZ may work in concert with alleviation of cardiac dysfunction by modulating glucose metabolism. The effects and mechanism of MTZ-mediated cardioprotective effect with respect to a glucose-lowering effect require further investigation. However, in our in vitro study, MTZ was found to inhibit HG or β-catenin–induced cardiomyocyte hypertrophy, suggesting that the hypoglycemic effect was not the only mechanism of the cardioprotective effect of MTZ in DCM. Therefore, we presumed that MTZ may directly target the heart (especially cardiomyocytes) to exert a beneficial effect in the context of DCM.

The Wnt/β-catenin pathway is involved in the pathogenesis of cardiac diseases, especially DCM. β-Catenin reportedly combines directly with TCF/LEF and binds to the ANP promoter to promote ANP transcription (20). Phenylephrine enhances ANP expression and induces cardiomyocyte hypertrophy; these effects can be blunted by β-catenin knockdown (20). Furthermore, transduction of an adenovirus encoding a stabilized β-catenin mutant significantly increased NRCM size (21). In our previous study, we demonstrated both activation of β-catenin and enhanced nuclear accumulation of β-catenin in idiopathic dilated cardiomyopathy, ischemic heart disease, and murine desmin-related cardiomyopathy (22). Therefore, β-catenin presumably has a prohypertrophic effect. In this study, we showed increased nuclear accumulation of β-catenin in T1DM heart and HG-treated cardiomyocytes, which led to cardiac hypertrophy with larger cardiomyocyte size and higher ANP protein level. These results were consistent with previous findings. In our study, MTZ was able to repress the Wnt/β-catenin pathway by potentiating β-catenin degradation in T1DM hearts and NRCMs. MTZ-mediated reduction of β-catenin led to smaller cardiomyocyte size and lower ANP protein level, which contributed to the cardioprotective effect.

Previous studies have demonstrated time-dependent activation of Wnt/β-catenin pathway in the hearts of STZ-induced diabetic rats, along with increased phosphorylation of GSK3β on Ser9 and increased protein levels of β-catenin, TCF7L2, and c-Myc (24,36). DCM is partially characterized by oxidative stress injury. Oxidative stress in diabetes upregulates the nuclear accumulation of β-catenin and downstream c-Myc protein level, as well as induces apoptosis, interstitial fibrosis, and cardiac dysfunction (23). A cardiac-specific β-catenin knockout in diabetic mice showed protective effects against oxidative stress injury in heart tissue (23). This is consistent with our findings in the hearts of STZ-induced diabetic mice, which showed enhanced activation of β-catenin and increased protein expression in the nucleus. Furthermore, we also observed enhanced downstream cyclin D2 expression and phosphorylation of GSK3β on Ser9. The inhibitory effect of MTZ on β-catenin thus alleviated T1DM-induced myocardial hypertrophy and cardiac dysfunction. Our findings further demonstrate the important role of β-catenin in DCM and indicate a cardioprotective role for MTZ in DCM. However, this study did not explore whether MTZ affects oxidative stress injury through inhibition of β-catenin in DCM.

AXIN1 is a scaffold protein that binds to GSK3β and CK1α to provide a platform for β-catenin phosphorylation (27,28). In this experimental study, we demonstrated that MTZ can directly mediate the interaction of AXIN1 and β-catenin. Additionally, the level of AXIN1 affects the formation of the β-catenin degradation complex, further mediating β-catenin content and activity (37). Following MTZ treatment, we observed increased AXIN1 protein expression in T1DM hearts and HG-treated cardiomyocytes, which is contrary to the findings in Wnt3a-treated cardiomyocytes. We presume that MTZ can facilitate AXIN1 expression or stabilize AXIN1 in the HG environment. However, additional studies are required to elucidate the effects of MTZ on AXIN1 level.

In this study, MTZ and other CA inhibitors reduced CA protein expressions in HG-treated NRCMs. Various CA inhibitors exhibit different sensitivity toward different CA isoforms. While MTZ mainly targets at CA1, 2, and 4, ETZ mainly targets at CA1. BZ and ATZ are widely used CA inhibitors. Notably, CAs can directly interact with sodium/hydrogen exchanger 1 (NHE1) to catalyze NHE1 activity and enhance its transport capacity, maintaining intercellular pH and Na⁺ content (38). In ischemic DCM, increased CA2 expression leads to NHE1 activation, promoting abnormal cell growth and apoptosis (10). In cardiac hypertrophic mouse models induced by guanylyl cyclase-A receptor knockout or in ischemia-reperfusion rat model induced by left coronary artery ligation, NHE1 is reportedly activated to increase Na⁺ influx, which facilitates Ca²⁺ influx and accumulation through the Na⁺/Ca²⁺ exchanger, resulting in several aberrant biological behaviors (39,40). In previous studies, BZ and ETZ were shown to inhibit CAs to affect NHE1 and pH and ameliorate oxidative damage to mitigate cardiomyocyte abnormality and dysfunction induced by stressors (12,14,15). Thus, we hypothesize that CAs participate in the onset of DCM by means of NHE1 and Na⁺/Ca²⁺ exchanger. The cardioprotective effect of MTZ in the setting of T1DM may be mediated through the CA–NHE1 axis as well; further studies are required to test this hypothesis.

In conclusion, this study demonstrated the hypoglycemic and cardioprotective effects of MTZ in T1DM mice. MTZ inhibited the protein level and activity of β-catenin, thus reducing myocardial hypertrophy and cardiac dysfunction. Moreover, MTZ repressed β-catenin by enhancing the AXIN1–β-catenin interaction, without affecting β-catenin transcription. Further research is required to investigate the mechanism of the MTZ-mediated hypoglycemic effect and cardioprotective effect. Our findings may help clarify the pathogenesis of DCM as well as the potential use of MTZ for treatment of DCM.

Acknowledgments. The authors thank the laboratory colleagues and collaborators for stimulating discussions. The authors also thank Medjaden Inc. for scientific editing of this article.

Funding. This work was supported in part by grants from the National Natural Science Foundation of China (81773720 and 81872869), the Natural Science Foundation of Guangdong Province, China (grant 2019A1515011848), and the High-level University Construction Fund of Guangdong Province (grant 06-410-2107206).

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. X.C. researched data and wrote the manuscript. Y. Li, X.Y., and W.Y. researched data and edited the manuscript. C. Li, Y.Z., Y.Lia., and X.Q. researched data and performed data analysis. Y.Q. and G.Z. contributed new reagents and analytic tools. W.Y., X.L., and C.Lu. participated in research design and edited the manuscript. J.-D.L. designed the research and edited the manuscript. N.H. designed the research, supervised experiments, and wrote the manuscript. X.C., Y. Li, X.Y., W.Y., and N.H. contributed to revising the manuscript. N.H. is the guarantor of this work and, as such, had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.