Animals and animal model

All animal care and experimental procedures were performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. In addition, this study received approval from Animal Care and Use Committee of Renmin Hospital of Wuhan University (20221207B). The cardiac-specific USP38 knockout and transgenic mice (C57BL/6 background) were obtained from the Cyagen Biotechnology using the Cre-loxp system technology. mice were housed in a specific pathogen-free (SPF) environment with a temperature under a 12-h light/dark cycle, with food and water available.

The AB surgery was performed as previously described (Wu et al. 2018). Briefly, mice were anesthetized with 3% pentobarbital sodium at a dose of 40 mg/kg. The thorax of the mouse was exposed, and the thoracic aorta was found and ligated with a 27G needle using a 7–0 silk suture. A 70% constriction was achieved when the blunt needle was withdrawn after ligation. A similar operation was performed without ligating the aorta for the sham operation group. The wild-type mice were divided into two groups: Sham groups, AB groups (n = 12 in each group). Mice with heart specific knockout of USP38 were divided into four groups: Flox-Sham groups, CKO-Sham groups, Flox-AB groups, CKO-AB groups (n = 20 in each group). Mice with heart specific transgenic of USP38 were divided into four groups: NTG-Sham groups, TG-Sham groups, NTG-AB groups, TG-AB groups (n = 20 in each group).

Cell culture and treatment

H9c2 cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM, Invitrogen Corporation, USA), supplemented with 10% FBS, streptomycin, penicillin, and incubated in a 5% CO2 atmosphere at 37 °C. Hypertrophy in vitro was induced in H9c2 cells by the addition of 1μM angiotensin II (Ang II). Recombinant adenoviruses were constructed to knock out and overexpress the target gene in H9c2. AdshUSP38 was generated using short hairpin RNAs (shRNAs) for USP38 knockdown, while AdUSP38 was constructed using a replication-defective adenoviral vector containing rat USP38 cDNA. AdshRNA and AdGFP served as controls for AdshUSP38 and AdUSP38 respectively. H9c2 cells were infected with AdshUSP38 and AdshRNA at a multiplicity of infection (MOI) of 40 for 24 h and with AdUSP38 and AdGFP at a MOI of 20 for 24 h. H9c2 cells infected with AdshUSP38 were inoculated on the 6-well plate and divided into four groups: PBS + AdshRNA groups, PBS + AdshUSP38 groups,Ang II + AdshRNA groups, Ang II + AdshUSP38 (n = 6 in each group). H9c2 cells infected with AdUSP38 were inoculated on the 6-well plate and divided into four groups: PBS + AdGFP groups, PBS + AdUSP38 groups, Ang II + AdGFP groups, Ang II + AdUSP38 (n = 6 in each group).

Echocardiography

After four weeks post-surgery, mice were anesthetized with 1.5% isoflurane, and their cardiac function was assessed using an ultrasound Doppler imaging system (VINNO6, Vinno Corporation, China). The B-Mode was utilized to locate the long-axis view, which was then switched to M-mode to record the myocardial movement over time. Parameters of cardiac function, including the left ventricular ejection fraction (LVEF) and left ventricular fractional shortening (LVFS), were measured (n = 4 in Sham and AB groups, n = 8 in Mice with heart specific knockout and transgenic of USP38 gruops).

Surface Electrocardiogram (ECG)

Mice were anesthetized with 1.5% isoflurane and placed on a heating plate. Similar to the surface-lead ECG (lead II), mice were implanted with subcutaneous electrodes to record electrocardiogram-related indicators, including RR, PR, QRS, and QT interval. Lab-Chart 8 Pro (AD Instruments) collected the mouse electrocardiogram for 5 min and analyzed the data. To correct for the effects of different heart rates, QTc was calculated using Bazett’s formula: QTc = QT/√RR (n = 8 in each group).

In vivo electrophysiological study

The mice were anesthetized with 1% pentobarbital sodium, and then the chest was opened to expose the heart. The monophasic action potentials (MAP) of left ventricle were recorded using a platinum MAP electrode and stimulation procedures. A MAP electrode was positioned on the anterior free wall of the left ventricle to record epicardial MAP, accompanied by a paired platinum stimulating electrode on the basal surface of the right ventricle. The electrical stimulation program was performed in a method similar to that previously reported (Yang et al. 2020b, a). The heart was stimulated with a regular pacing cycle length (PCL). Action potential duration (APD) and APD alternans (ALT) were measured by S1-S1 pacing. When the PCL was 150 ms, APD90, APD50 and APD20 were defined as the average repolarization time of 90%, 50% and 20% of 6–8 consecutive MAP under 150 ms PCL (n = 8 in each group). ALT was measured by graded stimulation. PCL was decreased by 10 ms each step from 150 to 100 ms and then by 5 ms from 100 to 50 ms. When the difference between two adjacent MAP is at least 5%, it is considered that the ALT threshold. The ALT threshold is the longest S1-S1 pacing perimeter that induces APD90 alternation (n = 8 in each group). The S1-S2 (8- consecutive stimuli (S1,150 ms) followed by an extra stimulus (S2)) stimulation program was used to determine the effective refractory period (ERP) of the ventricle. When the pairing interval of S1S2 gradually shortens, S2 finally enters the refractory period until ventricular ERP (n = 8 in each group). Burst pacing (5 V, 50 Hz, 2 ms pulse, 2-s burst duration) was used to induce VAs. continuous VT or VF lasting > 2 s was recorded as VAs. The following are the number of groups for conducting burst pacing: Sham groups (n = 10), AB groups (n = 12), Flox-Sham groups (n = 16), CKO-Sham groups (n = 16), Flox-AB groups (n = 18), CKO-AB groups (n = 19), NTG-Sham groups (n = 17), TG-Sham groups (n = 16), NTG-AB groups (n = 19), TG-AB groups (n = 20).

Wheat germ agglutinin staining

The left ventricle tissue was fixed in 4% paraformaldehyde solution for 24 h. Subsequently, the tissue was dehydrated, embedded in paraffin, and cut into 3–5 μm slices. the cross-sectional area of the myocytes was observed through FITC-conjugated wheat germ agglutinin (WGA) staining (n = 4 in each group).

Immunofluorescence

Cx43 was assessed by immunofluorescence in ventricular tissue. Frozen heart sections were fixed and incubated with Cx43 (AF0137, affinity), and then incubated with the second antibody. Nucleus were stained with DAPI Staining Solution (Servicebio, China). Fluorescence images are collected using a fluorescent microscope (Nikon Eclipse C1, Japan) (n = 4 in each group).

Quantitative real-time PCR

RNA from left ventricle and H9c2 cells was extracted using reagent (Invitrogen), and the RNA was transcribed into cDNA with the PrimeScript RT reagent Kit (TaKaRa). qRT-PCR was performed On an Applied Biosystems VII7 instrument (Life Technologies, Carlsbad, USA) to determine USP38 mRNA levels. The primer sequences of USP38 and GAPDH are found in the Supplementary Table S1 (n = 3 in each group).

Western blotting

Cardiac left ventricular and cellular proteins were extracted on ice by adding phosphatase inhibitors to RIPA lysis buffer (Servicebio, China). Proteins were separated by electrophoresis in 10% SDS-PAGE gels and transferred to a polyvinylidene difluoride (PVDF) membrane (Millipore, America). The membrane was blocked with 5% skimmed milk. Subsequently, the membrane was exposed to the primary antibody and incubated overnight, and the secondary antibodies conjugated with HRP was added and incubated on the shaker for 1 h (n = 3 in each group).

Primary antibody against USP38 (17,767–1-AP) were purchased form proteintech. Primary antibody against phospho-CAMKII (GTX52342) was purchased from Gene Tex. Primary antibody against Kv4.2 (ab123543) and CAMKII (ab134041), were purchased form Abcam. Primary antibodies against Kv1.5 (PA5-77573), Kv2.1 (PA5-77580) and Kv4.3 (PA5-77583) were purchased from Invitrogen. Primary antibody against TBK1 (3504), p-TBK1 (5483), AKT (9272), p-AKT (4060) and the secondary antibody (7074) were purchased form Cell Signaling Technology. Primary antibody against GAPDH was purchased from Servicebio. Primary antibody against Cx43 (AF0137) was purchased from affinity.

Co-immunoprecipitation

H9c2 cells were collected and treated with immunoprecipitation (IP) lysis buffer, supplemented with PMSF and a cocktail. Following centrifugation of the cell lysate, the supernatant was collected. Anti-USP38, anti-TBK1, or anti-IgG antibodies were added to form immune complexes overnight. The resulting immune complexes were then incubated with magnetic beads at room temperature for 1 h. Subsequently, the magnetic beads were washed with loading buffer, denatured, and eluted for Western blotting experiments, the experiment was repeated three times.

Statistical analysis

Data were analyzed using GraphPad Prism software and presented as mean ± standard error (SEM) or percentage. To assess the normal distribution of the two data sets, we conducted the Shapiro–Wilk test. An unpaired t-test was used when the data sets adhered to a normal distribution. When the two data sets did not conform to a normal distribution, the data were analyzed using Mann–Whitney test. For multiple data sets conforming to a normal distribution, one-way ANOVA with Tukey’s post hoc analysis was used. Categorical data passed the Fisher's exact test. A level of P < 0.05 was considered statistically significant.