AC16 cells were maintained in a humidified incubator at 37 °C with 5% CO₂. The cells were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM; Gibco, USA) supplemented with 10% fetal bovine serum (FBS; Sigma, USA), 100 U/mL penicillin, and 100 μg/mL streptomycin (Biosharp, China). The culture medium was replaced every 1-2 days.

To mimic ischemic stress, cells were rinsed with Phosphate Buffered Saline (PBS) and incubated in glucose-free DMEM (Procell, Cat. No. PM150270) in a tri-gas incubator maintained at 37 °C with 1% O₂, 5% CO₂, and 94% N₂. For AC16 cells, OGD was performed for 6 or 9 h to induce stress responses or apoptosis/end-point assays, as indicated. For primary neonatal rat ventricular myocytes (NRVMs), OGD was performed under the same conditions for 6 h or 10 h, as specified in the relevant experiments.

NRVMs were isolated from 1-2-day-old Sprague-Dawley rats using a protocol adapted from previous studies. Briefly, ventricular tissues were minced and digested using a serial trypsinization procedure. First, 40 mL of 0.125% trypsin solution was prepared by mixing 15 mL of 0.25% trypsin (Gibco, USA) with 25 mL of 0.05% trypsin (Gibco, USA), the tissues were incubated at 4 °C in a solution containing 10 mL of 0.125% trypsin overnight, followed by repeated digestion at 37 °C for 7 min per cycle until complete tissue dissociation was achieved. The supernatants were collected and centrifuged at 1,500 rpm for 10 min. Red blood cells were removed using red blood cell lysis buffer (Biosharp, China), and the remaining cell pellet was resuspended in DMEM supplemented with 10% fetal bovine serum. To reduce fibroblast contamination, the cells were pre-plated for 90-120 min to allow preferential fibroblast attachment. Cardiomyocytes remaining in suspension were then collected, resuspended in F12 medium (Gibco, USA) containing 10% FBS, and seeded for subsequent experiments.

A focused CRISPR-Cas9 loss-of-function screen was performed to identify ubiquitin-related regulators of cardiomyocyte survival under OGD conditions. A pooled lentiviral human ubiquitin-related single guide RNA (sgRNA) library targeting 903 ubiquitin-related genes in the lentiCRISPR v2 backbone was used, with 10 sgRNAs per gene and 9,030 sgRNAs in total. AC16 cells (1.33 × 10⁷) were transduced at a low multiplicity of infection (MOI < 0.3) to favor single-sgRNA integration in individual cells, achieving approximately 500-fold library coverage per sgRNA. 24 h after transduction, cells were selected with puromycin (1 µg/mL) for 3 days to generate stable edited input populations. Before OGD selection, library coverage was confirmed to remain above 500-fold per sgRNA. For positive selection, the pooled cell population was subjected to OGD under conditions calibrated to yield approximately 30% cell survival. Genomic DNA was extracted from both input and OGD-selected cells. The sgRNA cassettes were amplified by Polymerase Chain Reaction (PCR) and sequenced on an Illumina platform. Enrichment and depletion analyses were performed using MAGeCK to identify sgRNAs and corresponding ubiquitin-related genes whose knockout influenced cell survival. The sgRNA sequencing data have been deposited in the National Center for Biotechnology Information (NCBI) Sequence Read Archive (SRA) under BioProject accession number PRJNA1423684.

For lentiviral shRNA-mediated knockdown, three independent shRNA constructs targeting human JOSD2 (sh-1, sh-2, and sh-3) and a scrambled shRNA control (sh-Scr) were cloned into lentiviral vectors. The shRNA target sequences were as follows: sh-1, 5’-CAGGTAGATGGCATCTACTAT-3’; sh-2, 5’-GCATCTACTATAATCTGGACT-3’; and sh-3, 5’-GCAACTATGATGTCAACGTGA-3’. HEK293T cells were transfected with the shRNA constructs together with the packaging plasmids psPAX2 and pMD2.G using polyethyleneimine (PEI; Yeasen, Cat. No. 40816ES03). Viral supernatants were collected at 48 and 72 h after transfection, filtered through a 0.45-µm filter, and used to infect AC16 cells in the presence of 5 µg/mL polybrene. Stable transductants were selected with puromycin, and knockdown efficiency was confirmed by reverse transcription quantitative polymerase chain reaction (RT-qPCR) and Western blotting before further experiments.

For lentiviral overexpression, full-length human JOSD2 cDNA (NM_001270639.2) was amplified by PCR and cloned into the pHAGE-3 × Flag overexpression vector. Virus production and transduction were performed as described above. Stably transduced cells expressing JOSD2 or the empty vector control were selected with puromycin, and overexpression was confirmed by RT-qPCR and Western blotting.

Total RNA was extracted using TRIzol reagent (Vazyme, Cat. No. R411-01), and 1 µg of total RNA was reverse-transcribed into cDNA using the ReverTra Ace qPCR RT Kit (TOYOBO, Cat. No. FSQ-301) with random hexamers. Gene expression was measured by SYBR-based quantitative PCR and normalized to ACTB for human samples or Actb for mouse samples. Relative expression levels were calculated using the 2^-ΔΔCt method. Primer sequences are listed in Supplementary Table 1.

Heart tissues or cultured cells were lysed in Radio Immunoprecipitation Assay (RIPA) buffer (Beyotime, China) supplemented with protease and phosphatase inhibitors. Protein concentrations were determined using a Bicinchoninic Acid (BCA) assay. Equal amounts of protein were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS PAGE) and transferred onto Polyvinylidene Fluoride (PVDF) membranes (Millipore, USA). After blocking with 5% skim milk, the membranes were incubated overnight at 4 °C with primary antibodies, followed by incubation with Horseradish peroxidase (HRP)-conjugated secondary antibodies. Antibody sources and dilutions are listed in Supplementary Table 2. Protein signals were detected using ECL reagent (Epizyme, China), and densitometric analysis was performed using ImageJ, with β-actin used as the loading control.

After the indicated treatments, cells were harvested, washed, and stained with Annexin Fluorescein Isothiocyanate (FITC) and propidium iodide (PI) according to the manufacturer’s instructions. Flow cytometric analysis was performed using a FACScan instrument, and the percentages of viable, early apoptotic, and late apoptotic/dead cells were quantified using FlowJo software.

Cells grown on coverslips were fixed with 4% paraformaldehyde and permeabilized with 0.2% Triton X-100. Apoptotic cells were labeled using a Tetramethyl-Rhodamine-5-dUTP (TMR) Red TdT mediated dUTP Nick End Labeling (TUNEL) Detection Kit (Servicebio, Cat. No. MPC2503023) according to the manufacturer’s instructions. Nuclei were counterstained with 4’,6-diamidino-2-phenylindole (DAPI). Fluorescence images were acquired, and TUNEL-positive cells were quantified.

Cells were seeded in 96-well plates at a density of 1 × 10⁴ cells per well. After the indicated treatments, CCK-8 reagent (Dojindo) was added according to the manufacturer’s instructions. Following incubation at 37 °C for 4 h, absorbance at 450 nm was measured using a microplate reader.

Total RNA was extracted, and poly(A)+ RNA was enriched for library preparation. Libraries were sequenced on a DNBSEQ (DNA Nanoball Sequencing) platform in 150-bp paired-end mode. Raw FASTQ reads were quality-trimmed using SOAPnuke and aligned to the human reference genome (GRCh38.p14) using HISAT2 v2.2.1. Gene-level read counts were obtained using RNA-Seq by Expectation Maximization (RSEM), and differential expression analysis was performed using DESeq2 v1.34.0. Differentially expressed genes were identified using thresholds of |log2 fold change| > 0.58 and an adjusted P value (Benjamini-Hochberg false discovery rate) < 0.05. All samples were prepared and sequenced in a single batch, and no obvious batch effects were observed. Principal component analysis (PCA), heatmaps, and volcano plots were generated from normalized counts. Gene Ontology (GO) enrichment analysis was performed to identify biological processes affected by JOSD2 silencing. The raw sequencing data have been deposited in the NCBI SRA under BioProject accession number PRJNA1422660.

Plasmids encoding Flag-tagged JOSD2, HA-tagged LKB1, or HA-tagged AMPK were transfected into HEK293T cells. At 24-48 h after transfection, cells were lysed in immunoprecipitation buffer supplemented with protease and phosphatase inhibitors, and the lysates were clarified by centrifugation. The clarified lysates were incubated with anti-Flag or anti-HA magnetic beads overnight at 4 °C. After washing, the bound proteins were eluted with SDS loading buffer and subjected to Western blot analysis. In addition, endogenous co-immunoprecipitation was performed in AC16 cells to confirm the interaction between endogenous JOSD2 and LKB1. Briefly, cell lysates were incubated with an anti-JOSD2 antibody, and the immune complexes were captured with Protein A/G beads, washed, and analyzed by Western blotting to detect co-precipitated LKB1. Input and IgG controls were included to verify specificity.

AC16 cells expressing Flag-tagged JOSD2, HA-tagged LKB1, and Myc-tagged ubiquitin or ubiquitin mutants were treated with 20 μM MG132 for 6 h. HA-LKB1 was immunoprecipitated using anti-HA magnetic beads, and ubiquitination was detected by Western blotting with an anti-Myc antibody. To determine the specific ubiquitin linkage types regulated by JOSD2, linkage-specific ubiquitination assays were performed using Myc-tagged wild-type ubiquitin and K-only mutants (K6O, K11O, K27O, K29O, K33O, K48O, and K63O), in which only the indicated lysine residue is retained and all other lysines are replaced with arginine. Linkage-specific ubiquitination of LKB1 was assessed by HA immunoprecipitation followed by anti-Myc immunoblotting.

To achieve transient LKB1 knockdown, AC16 cells were transfected with LKB1-specific or control siRNAs using Lipofectamine 2000 according to the manufacturer’s instructions. Three independent siRNAs targeting LKB1 were designed and tested. The siRNA target sequences were as follows: si-LKB1-1, 5’-GTGTAGAACAATCGTTTCTGTTG-3’; si-LKB1-2, 5’-AAGGGTTTTTCCCTTCCTTTTGG-3’; and si-LKB1-3, 5’-CCGTCAAGATCCTCAAGAAGAAG-3’. Knockdown efficiency was evaluated by RT-qPCR.

Oxygen consumption rate (OCR) was measured using a Seahorse XFe24 Analyzer (Agilent Technologies) with the XF Cell Mito Stress Test Kit to assess mitochondrial respiration. Stable sh-Scr and sh-JOSD2 AC16 cells were seeded in XF24 plates at the optimal density, transfected with si-NC or si-LKB1, and subjected to OGD treatment. After OGD, the cells were incubated in Seahorse XF assay medium in a non-CO₂ incubator before OCR measurement. Basal OCR was recorded first, followed by sequential injections of oligomycin (1 μM), FCCP (1 μM), and rotenone/antimycin A (1 μM each). Respiratory parameters, including basal respiration, Adenosine Triphosphate(ATP)-linked respiration, maximal respiration, were calculated and normalized to total protein.

Male C57BL/6J mice aged 8-10 weeks were housed under specific pathogen-free (SPF) conditions with a 12 h light/dark cycle, controlled temperature (22 ± 2 °C), and humidity of 50%-60%, with free access to food and water. Mice were anesthetized with isoflurane and randomly assigned to the sham or MI groups. Unless otherwise indicated in the figure legends, six mice were included per group. MI was induced by permanent ligation of the left anterior descending (LAD) coronary artery with an 8-0 suture. Successful ligation was confirmed by immediate blanching of the anterior ventricular wall. Sham-operated mice underwent the same surgical procedure without coronary ligation. Postoperative analgesia was provided with buprenorphine (0.05 mg/kg, s.c.), and mice were closely monitored for 48 h.

At predetermined endpoints, mice were euthanized by intraperitoneal overdose of sodium pentobarbital (150 mg/kg). Cessation of respiration and loss of pedal reflexes were confirmed before tissue collection. Blood was immediately collected by cardiac puncture, and serum was separated and stored at -80 °C. Hearts were rapidly excised, perfused with ice-cold phosphate-buffered saline through the aorta, and dissected into infarct, border, and remote zones. Tissue samples were either snap-frozen in liquid nitrogen for molecular analyses or fixed in 4% paraformaldehyde for histological examination.

To minimize bias, investigators responsible for postoperative care, data acquisition, and data analysis were blinded to group allocation whenever feasible. All experimental procedures were approved by the Experimental Animal Welfare Ethics Committee of Zhongnan Hospital (Approval No. ZN2023239).

Recombinant adeno-associated virus 9 (AAV9) vectors carrying JOSD2 or control ZsGreen under the control of the cardiac troponin T (cTnT) promoter were produced, purified, and titrated by qPCR. Mice were injected via the tail vein with 1 × 10¹¹ viral genomes of AAV9-cTnT-JOSD2 or AAV9-cTnT-ZsGreen. Three weeks after injection, cardiac JOSD2 overexpression was confirmed by Western blotting.

Transthoracic echocardiography was performed under light anesthesia with 2% isoflurane at the specified time points. Two-dimensional and M-mode images were acquired using a small animal ultrasound imaging system (Vevo 2100, VisualSonics, Toronto, Canada) equipped with a 30-MHz ultrasonic transducer. Left ventricular ejection fraction (LVEF) and left ventricular fractional shortening (LVFS) were calculated from short-axis M-mode tracings and averaged over three consecutive cardiac cycles.

Heart tissues were cut into approximately 2-mm-thick slices and incubated in 2% triphenyltetrazolium chloride (TTC) solution at 37 °C in the dark for 15 min. Viable myocardium was stained red, whereas infarcted areas remained white. The infarct size was quantified as the ratio of the infarct area to the total myocardial area.

Excised hearts were fixed, processed, and embedded for paraffin or frozen sectioning. For cardiomyocyte localization and protein expression analysis, sections were subjected to immunofluorescence staining with antibodies against JOSD2 and cTnT, followed by nuclear counterstaining with DAPI. Picrosirius Red (PSR) staining was performed to visualize and quantify interstitial fibrosis under polarized light, under which collagen fibers appear birefringent. For cardiomyocyte hypertrophy analysis, tissue sections were stained with wheat germ agglutinin (WGA) to visualize cell boundaries, and cardiomyocyte cross-sectional area was measured in defined regions.

Data are presented as the mean ± standard deviation (SD). Statistical analyses were performed using GraphPad Prism 10.0. For comparisons between two groups, statistical significance was assessed using an unpaired two-tailed Student’s t-test. For comparisons among multiple groups, one-way or two-way analysis of variance (ANOVA) was used, followed by Tukey’s multiple-comparisons test for post hoc analysis. A P value < 0.05 was considered statistically significant. Sample sizes for each experiment are provided in the figure legends. All experiments were independently repeated at least three times, and technical replicates, when performed, were not treated as independent biological replicates for statistical analysis.

To systematically identify ubiquitin-related regulators of cardiomyocyte survival under ischemic stress, we performed a pooled CRISPR-Cas9 loss-of-function screen in AC16 cells using a lentiviral human ubiquitin-related sgRNA library targeting 903 genes [Figure 1A]. Deep sequencing of two independent control libraries demonstrated near-complete sgRNA representation, with coverage rates of 99.89% and 99.93% [Figure 1B], and low Gini indices (0.0576 and 0.0595), consistent with uniform sgRNA abundance across the library [Figure 1C]. PCA of sgRNA abundance profiles showed clear separation between OGD-selected samples and controls, with tight clustering within each group, indicating a reproducible selection effect [Figure 1D]. Positive-selection analysis based on robust rank aggregation (RRA) prioritized candidate genes enriched in OGD-surviving cells relative to controls. The top-ranked candidates included ZBTB34, JOSD2, KLHL40, RNF185, UBE2L3, PSMB4, RAB40C, RING1, COMMD3, and KLHL26 [Figure 1E]; these candidates were subsequently subjected to integrated transcriptomic analysis and experimental validation.

To determine whether top candidates were transcriptionally responsive to ischemic stress, we measured messenger RNA (mRNA) levels of the top five genes in AC16 cells after OGD. RT-qPCR analysis showed that ZBTB34, JOSD2, and UBE2L3 were significantly upregulated following OGD, with JOSD2 showing the most pronounced induction, whereas RNF185 and KLHL40 remained unchanged [Figure 1F]. To evaluate in vivo relevance, we next assessed these five genes in murine hearts at one week after MI. Among them, only Josd2 expression was significantly increased in MI hearts compared with sham controls [Figure 1G]. Consistently, integration of our CRISPR screen hits with a published MI transcriptome time-course dataset (24 h, 48 h, and 1-week post-MI; GSE775) identified Josd2 as the sole overlapping gene across datasets [Figure 1H], establishing JOSD2 as a convergent candidate for further mechanistic investigation.

To explore the potential involvement of JOSD2 in MI, we examined whether JOSD2 expression was altered in vitro under OGD conditions and in a mouse MI model. RT-qPCR and Western blot analyses revealed that both Josd2 mRNA and protein levels were significantly elevated with prolonged OGD exposure, in parallel with the upregulation of the cardiac stress marker natriuretic peptide A (Nppa) (encoding atrial natriuretic peptide, ANP) [Figure 2A-C]. In AC16 cells, OGD also markedly increased JOSD2 protein levels, accompanied by an increase in the apoptosis-associated protein BAX and a decrease in BCL-2 [Supplementary Figure 1A and B]. To assess the in vivo relevance of JOSD2 after MI, we evaluated regional expression in murine hearts at day 7 post-MI. Compared with sham controls, both the remote and border zones showed increased α-smooth muscle actin (α-SMA), a marker of myofibroblast activation and fibrotic remodeling, alongside elevated JOSD2, with the most prominent induction observed in the border zone [Figure 2D-F]. Moreover, in a post-MI time-course analysis, cardiac JOSD2 protein abundance increased significantly over time [Figure 2G and H]. Finally, immunofluorescence staining demonstrated substantial overlap between the JOSD2 signal and the cardiomyocyte marker cTnT, and revealed a marked increase in JOSD2 signal intensity in cardiomyocytes at one week post-MI compared with sham [Figure 2I].

To define the functional contribution of JOSD2 to ischemic stress-induced cardiomyocyte injury, we generated stable JOSD2-knockdown AC16 cells using lentiviral short hairpin RNAs (shRNAs). Among three shRNA constructs, sh-3 produced the strongest reduction in both JOSD2 mRNA [Figure 3A] and protein levels [Figure 3B and C] compared with sh-Scr, and was therefore selected for further experiments. We next assessed whether JOSD2 knockdown affects stress marker genes and apoptosis under basal conditions and after OGD. At baseline, JOSD2 knockdown did not alter the expression of cardiac stress markers NPPA and natriuretic peptide B (NPPB). Following OGD, however, JOSD2 knockdown significantly attenuated their induction compared to control [Figure 3D]. Similarly, although apoptosis-related proteins were unchanged at baseline, OGD triggered a clear pro-apoptotic shift in control cells that was reversed by JOSD2 knockdown, as shown by reduced BAX, cleaved caspase-3 (c-Cas-3), and cleaved caspase-9 (c-Cas-9), and increased BCL-2 [Figure 3E and F]. To directly quantify apoptosis, we performed flow cytometric analysis using sh-Scr and sh-JOSD2 AC16 cells under control conditions and after OGD for 6 h or 9 h. JOSD2 knockdown significantly decreased the apoptotic fraction at both OGD time points compared with sh-Scr cells [Figure 3G and H]. Consistently, TUNEL staining demonstrated that JOSD2 knockdown reduced the proportion of TUNEL⁺ cells under OGD [Figure 3I and J]. Moreover, Cell Counting Kit-8 (CCK-8) assays indicated improved cell viability in JOSD2-silenced cells under OGD stress [Supplementary Figure 2A]. To further characterize the transcriptional impact of JOSD2 silencing during ischemic stress, we performed RNA-seq. PCA showed clear separation between the two groups [Supplementary Figure 2B], and differential expression analysis identified broad transcriptomic changes [Supplementary Figure 2C]. Gene Ontology enrichment analysis of differentially expressed genes highlighted multiple apoptosis-related pathways upon JOSD2 knockdown [Supplementary Figure 2D]. RNA sequencing (RNA-seq) heatmap analysis revealed that, under OGD, sh-JOSD2 cells exhibited increased expression of anti-apoptotic genes and decreased expression of pro-apoptotic genes compared with sh-Scr cells, suggesting that JOSD2 knockdown attenuates apoptosis-related transcriptional responses [Figure 3K].

To further assess the functional consequences of JOSD2 during ischemic injury, we generated stable JOSD2-overexpressing AC16 cells using lentiviral transduction. Successful overexpression of JOSD2 was confirmed by both RT-qPCR and Western blot analysis [Figure 4A-C]. We then examined the impact of JOSD2 overexpression on cardiomyocyte stress markers NPPA and NPPB under basal conditions and after OGD. In the absence of OGD, JOSD2 overexpression did not significantly alter the expression of these markers. However, following OGD, JOSD2 overexpression significantly exacerbated the induction of NPPA and NPPB compared to control cells [Figure 4D]. Next, we assessed apoptosis-related proteins BAX, c-Cas-3, c-Cas-9, and the anti-apoptotic protein BCL-2 across the experimental groups. Western blotting revealed that JOSD2 overexpression significantly upregulated pro-apoptotic proteins and downregulated BCL-2 expression under OGD, suggesting a shift toward a pro-apoptotic phenotype [Figure 4E and F]. To quantify apoptosis, we performed flow cytometric analysis of these cells after OGD. JOSD2 overexpression significantly increased the apoptotic fraction compared to control [Figure 4G and H]. Consistent with these findings, TUNEL staining also revealed a higher proportion of TUNEL⁺ cells in the JOSD2 overexpression groups under OGD [Figure 4I and J]. Additionally, CCK-8 assays demonstrated that while JOSD2 overexpression did not significantly affect cell viability under control conditions, it significantly suppressed cell viability under OGD [Figure 4K]. Taken together, these results indicate that JOSD2 overexpression exacerbates cardiomyocyte injury and apoptosis, particularly under ischemic stress, underscoring its potential role in promoting ischemic cell death.

To explore the mechanism by which JOSD2 regulates ischemic cell death, we analyzed transcriptomic profiles of JOSD2-knockdown cells after OGD. Transcriptomic profiling revealed that JOSD2 expression was closely associated with alterations in genes involved in canonical AMPK-downstream pathways, including lipid and glucose metabolism, cell growth, and autophagy, suggesting that JOSD2 may function by modulating AMPK activity [Figure 5A]. In line with this inference, Western blot analysis demonstrated that JOSD2 knockdown markedly increased phosphorylation of AMPK (p-AMPK) and its key upstream regulator LKB1 (p-LKB1) compared with control cells [Figure 5B and C]. Further co-immunoprecipitation experiments confirmed that JOSD2 interacts with LKB1, whereas no interaction was detected between JOSD2 and AMPK under the same conditions [Figure 5D and E]. Furthermore, endogenous co-immunoprecipitation (co-IP) experiments in AC16 cells confirmed that endogenous JOSD2 interacts with endogenous LKB1 under basal conditions [Figure 5F].

To examine whether JOSD2 regulates LKB1 through deubiquitination, we performed ubiquitination assays. JOSD2 overexpression reduced the overall ubiquitination level of LKB1 [Figure 5G], indicating that JOSD2 post-translationally regulates LKB1. To identify the types of ubiquitin chains removed by JOSD2, we performed linkage-specific ubiquitination assays. The results showed that JOSD2 preferentially reduced K27-, K29-, K33-, and K63-linked polyubiquitination of LKB1, whereas no obvious effect was observed on K6-, K11-, and K48-linked ubiquitination under the conditions tested [Figure 5H]. Collectively, these data demonstrate that JOSD2 interacts with and deubiquitinates LKB1, primarily at K27-, K29-, K33-, and K63-linked chains, thereby suppressing downstream AMPK phosphorylation and activation.

To further determine whether the protective effect of JOSD2 knockdown depends on LKB1, we designed three siRNAs targeting LKB1. RT-qPCR and Western blot analysis confirmed that siLKB1-1 achieved effective knockdown of LKB1 [Supplementary Figure 3A-C]. We then assessed apoptosis-related protein expression in sh-Scr and sh-JOSD2 AC16 cells with or without LKB1 silencing under OGD. Western blot analysis showed that JOSD2 knockdown reduced the expression of pro-apoptotic markers (BAX, cleaved caspase-3) and increased anti-apoptotic BCL-2, and these protective changes were largely reversed by LKB1 silencing [Figure 6A and B]. To evaluate the metabolic relevance of this LKB1-dependent effect, we performed Seahorse analysis to measure the OCR as an indicator of mitochondrial function. JOSD2 knockdown significantly increased OCR, basal respiration, maximal respiration, and ATP production after OGD exposure, indicating improved mitochondrial respiratory function, whereas this protective metabolic effect was largely abolished by LKB1 silencing [Figure 6C and D].

As a complementary pharmacological approach, we treated OGD-stressed cells with Pim1/AAK1-IN-1 (LKB1-in; also known as an LKB1/AAK1 dual inhibitor), which has reported inhibitory activity against LKB1^[23]. In line with the genetic silencing results, LKB1-in treatment largely reversed the anti-apoptotic effects of JOSD2 knockdown, as evidenced by restored expression of pro-apoptotic markers and decreased anti-apoptotic BCL-2 levels [Supplementary Figure 3D and E]. Consistently, flow cytometry, TUNEL staining, and CCK-8 viability assays all showed that the reduction in apoptosis and improvement in cell viability conferred by JOSD2 knockdown under OGD were abolished by LKB1 inhibition with LKB1-in [Figure 6E-H]. Collectively, these genetic, metabolic, and pharmacological data demonstrate that LKB1 is functionally required for the cardioprotective effect of JOSD2 depletion under ischemic stress.

To validate the pathological relevance of JOSD2 function in vivo, we established a cardiomyocyte-targeted overexpression model using tail-vein delivery of AAV9 under the control of the cTnT promoter (AAV-JOSD2). Cardiomyocyte-specific overexpression of JOSD2 was confirmed by immunoblotting three weeks after AAV9 injection [Supplementary Figure 4A-C]. We then subjected AAV-treated mice to MI or sham surgery following the protocol outlined in Figure 7A. At 14 days post-MI, AAV-JOSD2 mice exhibited a significantly greater reduction in LVEF and LVFS compared with AAV-ZsGreen controls, indicating worsened systolic dysfunction [Figure 7B and C]. Consistent with impaired early myocardial preservation, TTC staining at 1 day post-MI revealed that cardiomyocyte-specific JOSD2 overexpression significantly increased infarcted area after MI [Figure 7D and E]. Molecular assessment at 14 days post-MI further supported the presence of aggravated cardiac remodeling. RT-qPCR analysis showed that AAV-JOSD2 did not upregulate heart failure markers (Nppa and Nppb) in sham hearts, whereas JOSD2 overexpression markedly enhanced the induction of these markers after MI compared with AAV-ZsGreen controls [Figure 7F]. Histological analyses at 14 days post-MI further demonstrated more severe adverse remodeling in AAV-JOSD2 mice, characterized by increased infarct size, greater interstitial fibrosis, and enlarged cardiomyocytes [Figure 7G and H].

Mechanistically, consistent with our in vitro findings, JOSD2 overexpression in vivo suppressed activation of the LKB1-AMPK axis after MI, as evidenced by decreased p-LKB1/LKB1 and p-AMPK/AMPK ratios [Figure 7I and J]. In parallel, JOSD2 overexpression promoted a pro-apoptotic protein signature in post-MI hearts, increasing BAX, c-Cas-3, and c-Cas-9 while decreasing BCL-2 [Figure 7K and L]. Collectively, these data indicate that cardiomyocyte JOSD2 overexpression exacerbates infarction and adverse remodeling after MI, at least in part by suppressing LKB1-AMPK activation and enhancing ischemic cell death in vivo.

First, the main mechanistic validation of JOSD2-mediated regulation of the LKB1-AMPK signaling pathway in this study was conducted primarily in vitro using AC16 cells. Although AC16 cells can partially model cardiomyocyte injury and apoptosis under ischemic stress, they do not fully recapitulate the complex pathological and physiological environment of MI in the heart, including inflammatory infiltration, cell-cell interactions, microenvironmental changes, and metabolic remodeling. Therefore, more rigorous in vivo studies, particularly in animal models with cardiomyocyte-specific conditional gene modification, are needed to clarify the role and mechanism of JOSD2 in ischemic heart injury. Second, this study focused on the acute and relatively early stages of MI and did not include systematic evaluation of long-term repair and chronic remodeling processes beyond four weeks post-MI. Whether JOSD2 continues to participate in the progression of chronic heart failure and whether it affects long-term ventricular remodeling and cardiac function outcomes remain to be investigated. Such studies would help more comprehensively evaluate the therapeutic potential of targeting JOSD2 in chronic heart failure. Third, this study lacks clinical samples and clinical data. The expression pattern of JOSD2 in myocardial tissues or peripheral blood from patients with MI, its correlation with disease severity and prognosis, and its clinical translational relevance as a potential intervention target all need to be further evaluated in clinical samples. In summary, although this study provides functional and mechanistic evidence that JOSD2 is an important negative regulator of ischemic myocardial injury, its long-term pathological significance and clinical translational value require further investigation.

This study identifies the deubiquitinase JOSD2 as a previously unrecognized mediator of myocardial ischemic injury through CRISPR screening. JOSD2 exerts its detrimental effects by inhibiting the cardioprotective LKB1-AMPK pathway, highlighting JOSD2 as a promising upstream therapeutic target.