Cardiomyocyte-specific NHE1 overexpression confers protection against myocardial infarction during hyperglycemia

Abstract

Background

Acute hyperglycemia on admission is frequently observed during the early phase after acute myocardial infarction (MI), even without the history of diabetes mellitus. We previously reported that inhibiting Na⁺/H⁺ exchanger 1 (NHE1) activity post-MI may improve outcomes, but not in the setting of MI with acute hyperglycemia. However, the precise role of NHE1 in the pathophysiology of MI with acute hyperglycemia remains to be elucidated, and there are no effective strategies for its prevention or treatment.

Methods and results

We analyzed 85 post-MI patients, identifying acute hyperglycemia (glucose > 7 mM) in non-diabetic individuals, linked to elevated BNP, CK-MB, and reduced plasma Na⁺. Using retrospective cohort studies and MI with acute hyperglycemia mouse models, we demonstrated that hyperglycemia exacerbates myocardial injury by reducing extracellular Na⁺, increasing intracellular Na⁺, and elevating pH, suggesting NHE1 activation as inferred from the observed intracellular pH (pHi) shift. Cardiomyocyte-specific NHE1 ablation or pharmacological inhibition worsened cardiac dysfunction and fibrosis in MI with acute hyperglycemia, while NHE1 overexpression conferred protection. RNA sequencing and drug screening identified accelerated NHE1 activation via 3% NaCl and lithospermic acid (LA) as a novel strategy to mitigate cardiomyocyte necroptosis, alleviating ischemic injury in MI and ischemia reperfusion models. Hypoxia-hyperglycemia and necroptosis induction models in NHE1-knockout, NHE1-overexpressing, and MLKL-overexpressing cardiomyocytes revealed that NHE1 activation, unlike its protective role in oxygen–glucose deprivation, promotes MLKL degradation via autophagosome-lysosomal pathways, reducing cardiomyocyte death. MLKL knockout and MLKL-NHE1 double knockout mice confirmed that MLKL ablation counteracts NHE1 inhibition’s detrimental effects.

Conclusions

Activation of myocardial NHE1 promotes MLKL autophagic degradation, mitigating cardiomyocyte necroptosis and acute hyperglycemia-exacerbated MI, highlighting NHE1 as a hyperglycemia-dependent cardioprotective target. Moderate NHE1 activation may represent a novel therapeutic strategy for MI with acute hyperglycemia.

Graphical abstract

NHE1 activation emerges as a novel therapeutic strategy to reduce infarct size and preserve cardiac function in hyperglycemia-associated myocardial infarction (MI). Accelerated NHE1 activation via lithospermic acid (LA) and 3% NaCl infusion offers a groundbreaking approach for managing MI with acute hyperglycemia. For the first time, we demonstrate that cardiomyocyte NHE1 exacerbates cardiac injury by mediating MLKL degradation during MI with acute hyperglycemia. These findings not only unveil promising candidates for clinical cardiovascular therapy but also provide new insights into the treatment of diverse MI subtypes.

Research insights

What is currently known about this topic?

• Acute hyperglycemia on admission is common identified in patients with acute myocardial infarction (MI) and is independently associated with a poor prognosis, even without the history of diabetes mellitus.

What is the key research question?

• The precise role of NHE1 in the pathophysiology of MI with acute hyperglycemia remains to be elucidated.

What is new?

• The dual nature of cardiomyocyte NHE1 indicates that NHE1 activation promotes MLKL degradation and mitigates necroptosis induced by MI with acute hyperglycemia.

How might this study influence clinical practice?

• Moderate activation of NHE1 presumably serve as a novel therapeutic approach for reducing myocardial infarct size and protecting cardiac function following MI with acute hyperglycemia.

Background

Acute hyperglycemia on admission is commonly identified in patients with acute myocardial infarction (MI) and is independently associated with a poor prognosis, even without the history of diabetes mellitus [1]. Epidemiological data indicate that, up to 50% of patients with MI may experience acute hyperglycemia, with an incidence rate of 20–25% observed in non-diabetic individuals [2, 3]. Higher admission blood glucose levels have been identified as the major independent risk factor in patients with MI, with a significant increase in mortality observed in patients presenting with acute hyperglycemia [4, 5]. For each 1 mM increase in blood glucose levels, the mortality rate among non-diabetic subjects rises by 4%. There is a consensus regarding the blood glucose levels that define admission acute hyperglycemia, with fasting blood glucose inflection points for poor prognosis established at 5.6 mM for patients without diabetes and 10.6 mM for those with diabetes [4]. When admission blood glucose levels exceeded 11.1 mM, mortality was comparable between non-diabetic and diabetic patients with MI [2, 4, 6]. Simultaneously, patients exhibit a multitude of pathophysiological responses, including endothelial cell dysfunction, platelet activation, various inflammatory processes, increased levels of reactive oxygen species (ROS), electrolyte imbalances, aggravated myocardial stiffness and other clinical manifestations, collectively contributing to an increased mortality rate [5, 7, 8]. While acute hyperglycemia is closely associated with a fatal prognosis of MI, we currently lack suitable biomarkers for the development of effective treatment strategies.

Our previous study, in concert with many older studies, identified Na⁺/H⁺ exchanger-1 (NHE1) as a promising therapeutic target for ischemic heart disease [9,10,11,12]. Inhibiting NHE1 protects cardiomyocytes from ischemic injury by reducing autophagic cell death in MI without hyperglycemia. NHE1 is the most powerful regulator of intracellular pH (pHi) in the heart by extruding H⁺ and allowing subsequent Na⁺ influx in response to intracellular acidification [9, 13, 14]. Persistent or excessive NHE1 activation leads to calcium overload, promoting detrimental consequences such as ischemia–reperfusion injury, cardiomyocyte death, and calcium-activated pathological signaling [9, 15, 16]. A prevailing notion is that inhibition of NHE1 activity in response to ischemia in heart involved in MI contributes to a reduced risk of mortality [17]. Notably, NHE1 expressed on cardiomyocyte membranes has been identified as a key target for the cardioprotective effects of SGLT2 (sodium-dependent glucose transporters 2) inhibitors [18, 19], as demonstrated in our previous study [9]. However, in another study, compared with the MI group, NHE1 knockout failed to protect against mortality following MI with acute hyperglycemia [16]. The precise role of NHE1 in this process remains largely unclear.

In the present study, we utilized clinical data from patients with MI during acute hyperglycemia to elucidate the relationship between hyperglycemia, low plasma/serum sodium, and the exacerbation of cardiac damage. RNA sequencing and drug screening revealed that accelerating NHE1 activation via 3% NaCl and lithospermic acid (LA) represents a novel strategy to mitigate cardiomyocyte necroptosis, alleviating ischemic injury in both MI and I/R models. MLKL (mixed lineage kinase domain-like), a key effector of necroptosis [20, 21], was further investigated in relation to NHE1 in MI with acute hyperglycemia using mouse models. We generated cardiomyocyte-specific NHE1 knockout and overexpression mouse models, along with MLKL and MLKL-NHE1 double knockout mice, to demonstrate that NHE1 activation facilitates the degradation of MLKL via autophagosome-lysomal pathway, thereby reducing necroptosis and providing cardioprotection in vitro and in vivo during MI with acute hyperglycemia. Our findings suggested that cardiomyocyte NHE1 plays a novel protective role as a regulator of MLKL-autophagic degradation in MI with acute hyperglycemia, irreconcilably with the damage effect in MI. We believe these results may underscore NHE1 as a potential new target for improving MI with acute hyperglycemia and aid in the development of novel clinical applications.

Methods

Human studies and epidemiological database

Clinical characteristics of the human samples were listed in Table S1–S4. To limit possible confounding factors, we examined 85 MI patients and without a history of diabetes. All patients were divided into two groups, the normal group (normal glucose, serum glucose ≤ 7 mM at admission) and the high group (high glucose, serum glucose > 7 mM at admission). Distal coronary venous sinuses plasma from four patients who underwent PCI and six healthy controls were further collected for plasma sodium concentration measurement (ChiCTR-IDR-16007765).

The National Health and Nutrition Examination Survey (NHANES) designed to assess the health and nutritional status of the noninstitutionalized civilian population in the United States. In this study, we utilized data from NHANES 1999–2020. 1442 individuals without diabetes who had either coronary artery disease or MI was included in the analysis. Patients were divided into two groups: the normoglycemia group (plasma glucose ≤ 7 mM at admission) and the hyperglycemia group, which included 51 patients with levels greater than 7 mM. Finally, we obtained cardiac hs-Troponin I information for 14 patients with data from 2001 to 2004. Non-diabetic patients with coronary artery disease were diagnosed according to a combination of self-reported doctor diagnoses and standardized medical status questionnaires completed during individual interviews. Participants were asked, “Has a doctor ever told you that you have diabetes mellitus/coronary heart disease/heart attack?”. The participant answered “yes or no” to any of the above questions. The information of plasma glucose (mM) and Troponin I was collected from laboratory data in NHANES.

Animal studies

All experimental C57BL/6J mice (Male) were purchased from Shanghai laboratory animal center (SLAC, Shanghai, China) at the age of 8 weeks and housed in an air-conditioned environment with a 12 h light–dark cycle and provided with food (SYSE Bio-tech Co., LTD) and water ad libitum.

NHE1^fl/fl mice were constructed by Biocytogen (Beijing). Expression of Cas9 was prevented by an upstream Lox-Stop-Lox sequence. The CRISPR/Cas9 mice were crossed with myosin heavy-chain 6 (Myh6)-Cre transgenic mice, and then the Myh6^icreERTNHE1^fl/fl strain was used. 8-week-old mice (Male) were received daily intraperitoneal injections of either vehicle (Corn oil, Beyotime) or tamoxifen (75 mg/kg, MCE) for five consecutive days to induce CreERT2 activity. All studies were performed at 1 week after tamoxifen injection.

MLKL^−/− mice were a kind gift from Professor Zhenyu Cai at Tongji university of Medicine and the NHE1^+/− were constructed by using CRISPR-Cas9 technology at Biocytogen. NHE1^+/− mice were crossed with the MLKL^−/− mice to produce NHE1^+/−MLKL^−/− and then NHE1^+/−MLKL^−/− mice were crossed with the same genotyped mice to produce both NHE1^−/−MLKL^−/− mice.

The NHE1 overexpression mice were constructed by AAV9 adenovirus. The 4 weeks C57BL/6J mice were divided into two groups randomly, which receiving empty vector-capsuled adeno-associated virus 9 (AAV9-Control) and another group received cardiac in situ injection of AAV9-NHE1 (Gene Pharma, Suzhou, China) with an aliquot of 10 μL virus (5 × 10¹⁰ vg) 4 weeks after AAV9 injection, mice were undergone surgery.

Sequences of genotyping primers are listed in Table S6. All animal experiment procedures were performed according to the guidelines of the Care and Use of Laboratory Animals published by the National Academy of Sciences and the National Institutes of Health. Animal protocols and performances were approved by the Animal Care and Use Committee of Tongji University (TJ-HB-LAL-2022–06).

The myocardial infarction with acute hyperglycemia model (Glu-MI)

The acute hyperglycemia model has been previously elucidated in detail [16]. All mice used in this study were male. Briefly, the 10-week-old mice were administered an intraperitoneal injection of glucose (2 g/kg) 15–20 min before the operation. Both baseline and post-operative blood glucose levels were measured from the tail tip.

Mice were allocated using simple randomization with a specific ID number before the procedure. Experimental mice were randomly assigned to different groups and undergo the Sham or MI surgeries as previously described [15]. In brief, following chest shaving and induction of 2% isoflurane anesthesia, mice underwent intubation for general anesthesia and ventilation using a rodent respirator. The heart was exposed through left-sided thoracotomy and MI was induced by permanently ligating the proximal coronary artery (LAD). Following ligation, the chest cavity was meticulously closed using 6–0 sutures, layers of muscle and skin were then sutured in a sequential manner and received an immediate subcutaneous injection of buprenorphine at 0.1 mg/kg, followed by a second injection 24 h later. Sham-operated mice underwent the procedure without opening the chest. Core temperature was continually monitored and stabilized at 37 °C, while electrocardiogram (ECG) recordings were taken to assess ST-segment elevation during coronary occlusion. Mice were then provided with standard care and feeding. At the designated time points, the mice were euthanized, and the tissues were collected for analysis.

The 3% NaCl solution was administered via tail vein injection immediately following MI, with a dose of 400 μL per mouse (prepared as a 6% NaCl solution and injected at 200 μL). And the NHE1 inhibitor cariporide and activator lithospermic acid was injected intravenously via the tail vein to mice following MI at a dose of 3 mg/kg and 50 mg/kg respectively.

For each of these MI models, mouse body weight (BW) was recorded on a daily basis before drug injection unless otherwise stated. All mice had serum collected and the right tibia length (TL) and heart weight (HW) were measured when the mice were sacrificed at day 7, the ratio of HW to BW (HW/BW) or to TL (HW/TL) was then calculated for each group. The heart was collected for protein extraction or fixed by OCT for frozen sections.

Statistical analysis

Statistical analysis was performed using GraphPad PRISM software (Version 9.0.0, GraphPad Software, San Diego, CA). For animal experiments, a minimum number of 4 mice per group were required. (1) Normally distributed data: two-group comparisons by unpaired t-tests; multi-group by ANOVA with Tukey's test (mean ± SD). (2) Non-normal data/small samples: Mann-Whitney U test for two-group comparisons. Data were shown as mean ± SD. p value < 0.05 was considered as statistically significant.

Results

Acute hyperglycemia induced Na⁺/H⁺ imbalances are associated with impaired cardiac function post MI

To investigate the role of acute hyperglycemia in the pathogenesis of MI, we recruited 85 patients who had an MI with only a single LAD branch blockage, and analyzed their blood parameters and cardiac function before percutaneous coronary intervention (PCI) (see Supplementary material, Table S1). Excluding patients with a history of diabetes, the remaining non-diabetic individuals were categorized into a group with MI (Normal group) and another group during MI with acute hyperglycemia (High group), determined by a random blood glucose threshold of 7 mM (Fig. 1A). Compared with the Normal group, plasma N-terminal pro-brain natriuretic peptide (NT-pro BNP) and creatine kinase-MB (CK-MB) levels were significantly increased in MI patients with acute hyperglycemia (High group), indicating that hyperglycemia aggravated MI injury (Fig. 1B and C). The imbalance of sodium and potassium in plasma play an important role in the pathogenesis and development of cardiovascular diseases and contribute to sudden cardiac death [22, 23]. Nevertheless, we unexpectedly discovered that within the normal ranges of serum Na⁺ (135–145 mM) and K⁺ (3.5–5.5 mM) concentrations, patients experiencing MI with acute hyperglycemia exhibited significantly lower serum Na⁺ levels compared with those in the normal group, while K⁺ levels remained unchanged (Fig. 1D). To examine whether sodium levels are associated with acute hyperglycemic-exacerbated MI injury, we analyzed data from the National Health and Nutrition Examination Survey (NHANES) spanning from 1999 to 2020. This analysis focused on 14 patients experiencing MI with acute hyperglycemia who had available data on both cardiac injury markers, specifically cardiac troponin I (cTnI), and serum sodium/potassium concentrations (see Supplementary material, Fig. S1A and Table S2). Our findings indicated that within the clinically established normal ranges forserum sodium and potassium, lower sodium levels are significantly correlated with elevated cTnI levels, demonstrating an inverse relationship (Fig. 1E). In contrast, no correlation was found between potassium levels and cTnI in patients experiencing MI with acute hyperglycemia (Fig. 1E).

Fig. 1

Acute hyperglycemia induced Na⁺/H⁺ imbalances are associated with impaired cardiac function post MI. A Graphical overview of human study. B–D Analysis of plasma N-terminal pro‐B‐type natriuretic peptide (NT-proBNP) and Creatine Kinase-MB (CK-MB) levels, serum sodium (Na⁺) and potassium (K⁺) concentrations between MI patients with normal glucose (n = 14) and those with high glucose (n = 34). E Correlation of serum levels of Na⁺ and K⁺ with the Cardiac troponin I (cTnI) in non-diabetic MI patients with admission blood glucose exceeding 7 mM (n = 14), collected from the NHANES database. F Correlation of plasma levels of Na⁺ and K⁺ with the EF in patients (non-diabetic patients with admission blood glucose exceeding 7 mM) post PCI (n = 433). G Schematic illustration showing the comparison of plasma Na⁺ concentrations from cardiac regions between healthy individuals (n = 6) and PCI patients (n = 4). CSD, coronary sinus ostium (proximal segment). H Schematic illustration of the modeling strategies for mouse Sham, MI, and Glu + MI groups. I Diagram of intracellular Na⁺ enrichment in mouse heart tissue. J Total intracellular Na⁺ levels in heart tissue isolated from Sham, MI or Glu + MI, normalized by tissue weight, n = 4 (MI, Glu + MI), n = 5 (Sham). RA: remote area; BA + IA: border and infarct area. K Analysis of serum Na⁺ concentrations in Sham, MI, and Glu + MI groups, n = 7 (Sham), 5 (MI) and 8 (Glu + MI). L Flowchart of intracellular H⁺ measurement in cardiac tissue. M–N Representative immunofluorescence M and quantification N of pHrodo™ Red (pHi) levels in heart tissue from Sham, MI and Glu + MI, n = 5 per group, scale bar 100 μm. Statistical significance in B, D and G was determined by two-tailed unpaired student’s-test, in C was by Mann–Whitney U test, while J, K and N were assessed using one-way ANOVA with Tukey’s test. Correlations in E and F were analyzed using Pearson’s correlation coefficient and simple linear regression analyses. All quantitative data are expressed as mean ± SD. NS, not significant

To investigate the relationship between sodium levels and cardiac function, we had analyzed 433 non-diabetes patients who had an MI with only a single LAD branch blockage and blood glucose higher than 7 mM, and correlated their serum Na⁺ concentration with ejection fraction (EF) after PCI (see Supplementary material, Table S3). There was a clear statistical positive correlation (R = 0.2067, p < 0.0001). In particular, increased serum Na⁺ concentration was significantly associated with an enhanced EF, whereas no significant correlation was observed between serum K⁺ concentration and EF (Fig. 1F). Additionally, we collected peripheral plasma samples from 6 healthy subjects and local plasma samples from the coronary sinus ostium (proximal segment) of MI patients with acute hyperglycemia at the time of PCI (Fig. 1G and see Supplementary material, Table S4). Using a sodium ion assay kit, we further confirmed that the plasma Na⁺ concentration in MI with acute hyperglycemia patients was significantly lower than that in healthy controls (Fig. 1G). Collectively, these results suggest that MI patients with acute hyperglycemia may experience exacerbated myocardial injury and impaired cardiac function as a consequence of low sodium concentration.

As described previously [16], to mimic the clinical phenomenon of MI with acute hyperglycemia, we administered an intraperitoneal glucose solution (2 g/kg) and performed LAD ligation for mice 15 min after glucose injection to simulate the clinical phenomenon of acute hyperglycemia. We compared Na⁺ concentrations in peripheral blood and regional Na⁺ levels in heart tissue across the Sham, MI, and Glu + MI groups at various time points (Fig. 1H and I). The results showed that, 3 days post-MI, intracellular Na⁺ levels were significantly increased in the infarct/border zones of the Glu + MI group compared to the MI and Sham groups, while no changes were observed in the remote zones (Fig. 1J). Consistent with clinical findings, Na⁺ concentrations in the peripheral blood of mice in the Glu + MI group were significantly lower than those in both the sham and MI groups, with this reduction persisting for at least seven days (Fig. 1K and see Supplementary material, Fig. S1C). To further investigate the effects of acute hyperglycemia on cardiac tissue, we assessed tissue pH using pHrodo™ red staining [24] on the 3 day post-MI (Fig. 1M). Notably, the fluorescence intensity in the infarct/border zone of the MI group was significantly lower than that in the Glu+MI group, indicating a significant reduction in the intracellular H⁺ concentration (Fig. 1M and N). Taken together, our results suggest that acute hyperglycemia exacerbates MI injury by elevating intracellular Na⁺ concentrations while simultaneously changing H⁺ concentrations in cardiac tissues, implicating the involvement of proteins responsible for Na⁺ and H⁺ transport.

Deleting NHE1 in adult cardiomyocytes exacerbates cardiac injury in MI with acute hyperglycemia

Our previous studies have demonstrated that NHE1, a membrane protein that regulates Na⁺ and H⁺ transport, plays a critical role in determining the fate of cardiomyocytes [9, 16, 25]. To further investigate the role of NHE1 in MI with acute hyperglycemia, cardiomyocyte-specific inducible NHE1 knockout mice were generated by introducing a floxed allele of NHE1 with exon 2 flanked by loxP elements into the genome, followed by crossing it with mice carrying Myh6-CreERT2 (Fig. 2A) [16]. Myh6^iCre; NHE1^fl/fl (NHE1-cKO) and NHE1^fl/fl (F/F) mice were injected intraperitoneally (i.p) with tamoxifen (75 mg/kg) for 5 consecutive days to achieve gene ablation (Fig. 2B). NHE1 expression was effectively reduced in mouse heart tissue by PCR, while western blot and q-PCR demonstrated its successful knockout in adult mouse cardiomyocytes (AMCM) (Fig. 2C–E and see Supplementary material, Fig. S2A).

Fig. 2

Cardiomyocyte-specific deficiency of NHE1 exacerbates cardiac injury in MI with acute hyperglycemia. A Schematic showing the strategy for generating cardiomyocyte-specific knockout mice (Myh6^iCreNhe1^fl/fl, cKO; Nhe1^fl/fl, F/F). B Schematic representation of tamoxifen-induced NHE1 knockout in mouse cardiomyocytes with MI under acute hyperglycemia (Glu-MI). C Genotyping of WT, F/F, cKO and iCre;Nhe1^fl/+ by PCR using primer pairs specific to the Nhe1^loxp (loxp, 413 bp, wt, 319bp) and Myh6-iCre (iCre, 424 bp). D and E Western blot analysis and quantification of NHE1 protein levels in adult mouse cardiomyocytes (AMCM) from F/F and cKO mice, n = 3 per group. F Representative M-mode echocardiographic images of heart ultrasound recordings at baseline and on day 3 or 7 post MI with acute hyperglycemia (scale bar, 100 ms and 1 mm) of F/F and cKO mice. G Comparison of EF at baseline, day 3 and day 7 post MI with acute hyperglycemia in F/F and cKO mice (n = 6 for each group). H Representative Masson's trichrome staining in F/F and cKO mice at day 7 post MI with acute hyperglycemia, scale bar, 1 mm. I and J Quantification of cardiac fibrosis in total tissue area and LV wall thickness in F/F and cKO mice, n = 5 (F/F) and n = 6 (cKO). K and L Images and quantitative results of cardiac IgG immunostaining (red signals showing cardiac injury) in mouse hearts and the whole heart sections were stained with WGA (n = 3 per group). Scale bar, 500 μm. M Analysis of serum brain natriuretic peptide (BNP) levels between F/F and cKO mice at 7 day post-MI with acute hyperglycemia (n = 5 per group). P values correspond to one-way ANOVA with Tukey’s multiple comparisons test for G and two-tailed unpaired student’s t test was performed for E, I, J, L and M. All quantitative data are expressed as mean ± SD

To investigate the impact of NHE1 knockout in MI with acute hyperglycemia, we induced a model of MI by maintaining blood glucose levels at 20 ± 5 mM (Fig. 2B and see Supplementary material, Fig. S2B). One week after MI with acute hyperglycemia, the survival rate of F/F mice was 23.6%, whereas that of NHE1-cKO mice was 0% [16]. Although this difference was not statistically significant, it suggested that NHE1 knockout in this model does not confer a survival advantage and may even exacerbate myocardial injury. The hearts of NHE1-cKO mice were significantly enlarged, and the heart-to-body weight ratio in NHE1-cKO was significantly elevated compared with the F/F group, supporting the hypothesis that NHE1 loss exacerbates cardiac remodeling (Fig. S2C). To evaluate the effects of NHE1 deletion on cardiac function, echocardiogram analysis was performed on NHE1-cKO and F/F mice before the LAD and at day 3 and day 7 post MI with acute hyperglycemia (Fig. 2F). At the basal level, there were no significant differences in cardiac function between NHE1-cKO and F/F mice. After MI with acute hyperglycemia induction on day 7 (instead of day 3), NHE1-cKO exhibited remarkably decreased EF (Fig. 2G), fractional shortening (FS) (see Supplementary material, Fig. S2D), interventricular septum (IVS) (see Supplementary material, Fig. S2E) and significantly increased left ventricular internal diameter (LVID) (see Supplementary material, Fig. S2F) as compared with F/F mice, suggesting deteriorated cardiac function. Masson trichrome staining at day 7 post MI with acute hyperglycemia showed significantly increased fibrotic size from 31 to 54% (Fig. 2H and I) and decreased LV wall thickness from 0.56 mm to 0.32 mm in NHE1- cKO as compared with F/F mice (Fig. 2J). In the peri-infarct zone, areas of acute myocardial injury can be detected by staining for myocyte infiltration of endogenous IgG [26]. The presence of intracellular IgG is anticipated to occur exclusively in myocytes with significantly compromised membranes that allow permeability to large molecules such as serum proteins. Quantitative analysis of acute myocardial injury indicated a significant increase in the area of damaged myocardial cells in NHE1-cKO mice compared to the F/F group 7 days after MI with acute hyperglycemia (Fig. 2K and L). Furthermore, plasma levels of BNP, a key biomarker for heart failure, were significantly elevated in NHE1-cKO mice compared to F/F mice 7 days post-MI with acute hyperglycemia, indicating severe cardiac dysfunction (Fig. 2M).

In summary, these findings indicated that cardiomyocyte-specific loss of NHE1 in the context of MI with acute hyperglycemia results in cardiac dysfunction, increased fibrosis, and exacerbated myocardial injury, contrasting sharply with the protective role of NHE1 inhibition in MI with euglycemia [16]. Thus, our results suggested that NHE1 exhibits opposite functions in two different types of MI, highlighting the necessity for tailored therapeutic approaches.

Overexpression of NHE1 in adult cardiomyocytes ameliorates cardiac damage in MI with acute hyperglycemia

After assessing the effects of NHE1 knockout on MI with acute hyperglycemia, we next sought to elucidate the role of NHE1-activation in MI with acute hyperglycemia in vivo through cardiomyocyte-specific NHE1 overexpression by in situ heart delivery of adeno-associated virus 9 (AAV9), driven by a cardiac troponin T (cTnT) promoter, at a cumulative dose of 5 × 10¹⁰ vg per mouse (Fig. 3A). Compared to the AAV9-Control group, the cardiac tissue of AAV9-NHE1-injected mice (5–6-week) demonstrated significantly elevated levels of NHE1 following a 4-week in situ injection (Fig. 3B and see Supplementary material, Fig. S3A-B). Furthermore, we investigated the pathophysiologic effects of cardiac-specific NHE1 overexpression by comparing AAV9-Control and AAV9-NHE1 mice subjected to MI with acute hyperglycemia characterized by blood glucose levels of 20 ± 5 mM (Fig. 3C and see Supplementary material, Fig. S3C). Interestingly, the ratio of heart weight to tibia length (HW/TL) (Fig. 3D) and ratio of heart weight to body weight (HW/BW) (Fig. 3E) all significantly decreased in AAV9-NHE1 mice during MI with acute hyperglycemia compared with AAV9-Control. Echocardiography was performed to monitor the progression of cardiac structural and functional changes (Fig. 3F). AAV9-NHE1 mice exhibited improved cardiac contractility 7 days after MI with acute hyperglycemia, as indicated by significantly increased EF (12% vs. 20%, p = 0.0380), FS (4% vs. 8%, p = 0.0059), IVS and significantly decreased LVID (Fig. 3G and see Supplementary material, Fig. S3D-E). We performed Masson’s trichrome and Picrosirius red staining to further evaluate myocardial fibrosis in heart tissue (Fig. 3H). Quantitative analysis revealed that AAV9-NHE1 hearts exhibited a significant reduction in fibrosis of approximately 20% and a notable increase in LV wall thickness of about 0.06 mm compared to AAV9-Control hearts (Fig. 3I and J). Notably, myocardial damage, as indicated by IgG accumulation, was reduced from around 38% to 20% in AAV9-NHE1 mice (p = 0.0380) (Fig. 3K and L). Furthermore, the serum BNP levels in NHE1-overexpressing mice were approximately 200 pg/mL lower than that in AAV9-Control mice, indicating that NHE1 overexpression was associated with improved cardiac function (Fig. 3M). These results collectively indicated that NHE1 overexpression protects against cardiac injury, fibrosis, and dysfunction in MI with acute hyperglycemia.

Fig. 3

Cardiomyocyte-specific overexpression of NHE1 alleviates cardiac damage caused by MI with acute hyperglycemia. A Schematic showing the strategy for cardiomyocyte-specific NHE1 overexpression (AAV9-Control, Con; AAV9-NHE1, NHE1) B Western blot analysis and quantification of NHE1 protein levels in extracts from heart samples of Con and NHE1 mice without MI with acute hyperglycemia (n = 3 for each group). C Schematic showing the experimental design for Con and NHE1 mice during MI with acute hyperglycemia. D and E The ratios of heart weight to body weight (HW/BW) and heart weight to tibia length (HW/TL) in Con and NHE1 mice (n = 10 per group). F Representative M-mode echocardiographic images of heart ultrasound recording at baseline and day 7 post MI with acute hyperglycemia (scale bar, 100 ms and 1 mm) for Con and NHE1 mice. G Comparison of EF at day 7 post MI with acute hyperglycemia in Con and NHE1 mice (n = 10 per group). H Representative Masson's trichrome staining (up) and Picrosirius red staining (down) in Con and NHE1 mice at day 7 post MI with acute hyperglycemia, scale bar, 1 mm. I and J Quantification of cardiac fibrosis in total tissue area and LV wall thickness in Con and NHE1 mice, n = 8 (Con) and n = 7 (NHE1). K and L Images and quantitative results of cardiac IgG immunostaining (red signals) in mouse hearts, whole heart sections were stained with WGA, n = 3 (Con) and n = 4 (NHE1). Scale bar, 500 μm. M Analysis of serum BNP levels between Con and NHE1 mice at 7 day post-MI with acute hyperglycemia, n = 8 (Con) and n = 7 (NHE1). Statistical significance in B, C, D, E, G, I, J, L and M were determined using two-tailed unpaired student’s t test. All quantitative data are expressed as mean ± SD

To assess whether the cardioprotective effects in NHE1 cardiomyocyte-specific overexpression mice were driven by increased NHE1 activity, we initially evaluated the effects of the preclinical NHE1 antagonist cariporide (Cari), a specific NHE1 inhibitor, on cardiac injury induced by MI with acute hyperglycemia (see Supplementary material, Fig. S4A). Cardiac function and structural changes were assessed at day 7 post-MI with acute hyperglycemia. Notably, compared with the AAV9-NHE1 group, the AAV-NHE1-cariporide group exhibited a significant increase in the ratio of HW/TL, indicating greater cardiac hypertrophy and adverse remodeling (see Supplementary material, Fig. S4B). Echocardiographic analysis revealed a marked reduction in ejection fraction (EF) in the cariporide-treated group (22% vs. 15%, p = 0.0027), indicating impaired cardiac function due to NHE1 inhibition. Histological analysis of cardiac fibrosis further demonstrated a significant increase in fibrotic area in the cariporide-treated group compared with the AAV-NHE1 group (20% vs. 40%, p = 0.0097), suggesting a worsened post-MI fibrotic response in the absence of NHE1 activation. These findings collectively indicate that the cardioprotective effects observed in NHE1-overexpressing mice are primarily mediated by NHE1 activation (see Supplementary material, Fig. S4C-E).

Previous studies have suggested that the cardioprotective effects of NHE1 overexpression in cardiac ischemic injury are mediated by the activation of endoplasmic reticulum (ER) stress in cardiac tissue [27]. To assess the potential contribution of ER stress to the cardioprotective effects of NHE1 overexpression, we analyzed the expression of ER stress markers, including Grp94, Grp78, Calr and P4hb in myocardial tissue of AAV-Con and AAV-NHE1 mice. qPCR analysis exhibited a slight but statistically insignificant increase in these markers (see Supplementary material, Fig. S4F). These findings collectively indicate that in the MI with acute hyperglycemia model, the cardioprotective effects of 1.5-fold NHE1 overexpression are primarily driven by increased NHE1 activity rather than ER activation. Based on these results, we propose that NHE1 activation represents a potential therapeutic target for MI with acute hyperglycemia.

NHE1 activators, instead of inhibitors, administration limits cardiac damage in mice during MI with acute hyperglycemia

To investigate the potential therapeutic possibility of NHE1 activators and inhibitors in pathological cardiac damage, we assessed the effects of NHE1 inhibition using cariporide in a mouse model of MI with acute hyperglycemia. Compared with the vehicle treatment (DMSO), we observed that the hearts of cariporide-treated mice were significantly enlarged, as evidenced by an increase in both the heart-to-body weight ratio and the heart weight to tibia length ratio in comparison to the DMSO group (Fig. 4B and see Supplementary material, Fig. S4A and S5A). Echocardiographic analyses at 1-week post MI with acute hyperglycemia showed significantly decreased EF, FS, IVS, LVPW, and significantly increased LVID in the cariporide-treated group as compared with vehicle controls (DMSO) (Fig. 4C, D and see Supplementary material, Fig. S5B-E). Histological analysis showed significantly aggravated fibrotic size and reduced LV wall thickness in cariporide-treated mice (Fig. 4E–G). Serum levels of BNP were found to be a significant elevation in cariporide-treated mice compared to DMSO (Fig. 4H). These findings were consistent with the results observed in NHE1 knockout models, collectively indicating that both genetic deletion and pharmacological inhibition of NHE1 result in cardiac dysfunction induced by MI with acute hyperglycemia.

Fig. 4

Administration of NHE1 activators, instead of inhibitors, resists pathological cardiac damage in mice during MI with acute hyperglycemia. A Schematic diagram showing the strategy for administering vehicle (DMSO) and NHE1 inhibitor (cariporide, Cari) to the mice. B The ratios of heart weight to body weight (HW/BW) in DMSO and Cari mice post 7 days MI with acute hyperglycemia, n = 10 (DMSO) and n = 8 (Cari). C and D High-frequency ultrasound echocardiography (HFUS) at day 7 after MI with acute hyperglycemia and the quantification of left ventricular ejection fraction (LVEF), n = 10 (DMSO) and n = 8 (Cari). E–G Representative Masson's trichrome staining and quantification of cardiac fibrosis in total heart area and LV wall thickness in DMSO and Cari mice at day 7 post MI with acute hyperglycemia, n = 7 per group, scale bar, 1 mm. H Analysis of serum BNP levels between DMSO and Cari mice at day 7 post-MI with acute hyperglycemia, n = 5 (DMSO) and n = 7 (Cari). I The experimental scheme of high-throughput screening of the activators of NHE1. J Schematic diagram showing the strategy for administering vehicle (Saline, 0.9% NaCl), LA and 3% NaCl to the mice. K Comparison of EF at baseline and day 7 post MI with acute hyperglycemia in Saline and 3% NaCl mice, n = 5 (Saline) and n = 8 (3% NaCl). L–N Representative Masson’s trichrome staining and quantification of cardiac fibrosis in total heart section area and LV wall thickness in Saline (n = 7), LA (n = 5) and 3% NaCl (n = 6) mice at day 7 post MI with acute hyperglycemia, scale bar, 1 mm. O Serum sodium concentrations from the Sham (n = 7), Saline (n = 7) and 3% NaCl (n = 6) post 7 day MI with acute hyperglycemia. P Correlation of serum sodium concentrations with EF post 7 day of MI with acute hyperglycemia (n = 13). Q Schematic diagram showing the strategy for I/R, Glu + I/R + vehicle (Saline) and Glu + I/R + 3% NaCl (3% NaCl) administration to the mice during I/R. R and S Representative Masson’s trichrome staining and quantification of cardiac fibrosis in Saline and 3% NaCl mice at day 7 post-reperfusion following 30 min of ischemia, with or without 15 min of acute hyperglycemia, n = 5 per group, scale bar, 1 mm. T and U Analysis of serum BNP levels and serum sodium concentrations from the Saline and 3% NaCl groups post 7 days reperfusion with acute hyperglycemia, n = 4 per group. Statistical significance in B, D, F, G, H and T, U were determined by two-tailed unpaired student’s t-test, while in K, M, N, O and S were determined by one-way ANOVA with Tukey’s test. The correlation in P was analyzed by Pearson's correlation coefficient and simple linear regression analyses. All quantitative data are expressed as mean ± SD. NS, not significant

To further investigate whether NHE1 activation therapy could serve as a potential pharmacological alternative to NHE1 gene overexpression in ameliorating cardiac pathological damage and dysfunction following MI with acute hyperglycemia, we established two screening strategies for agonist identification. Based on virtual screening of ligand-based drug design, we selected the previously reported NHE1 activator, rosmarinic acid [28], as a small molecule with established activity. Utilizing the pharmacophore model of rosmarinic acid, we searched a database of small molecules from Traditional Chinese Medicine for chemical structures with potential affinity for NHE1, and through this screening process, we identified lithospermic acid (LA) as possessing the closest binding energy (Fig. 4I). The calculated binding energy of rosmarinic acid was − 8.0 kcal/mol, while the well-established Traditional Chinese Medicine small molecule LA exhibited a comparable binding energy of − 7.7 kcal/mol (Fig. 4I). LA, primarily derived from the roots and rhizomes of Salvia miltiorrhiza, has been demonstrated to confer protection against ischemia–reperfusion injury [29]. However, the specific effects of LA on NHE1 remain unclear, prompting us to further investigate the role of LA in MI with acute hyperglycemia. Secondly, our experimental studies together with our preliminary human studies highlight the potential benefits of moderate higher sodium concentration in improving ventricular function. Given that NHE1 can be pharmacologically activated by the different concentration of NaCl administration [30,31,32,33] and plasma sodium concentration is reduced in MI with acute hyperglycemia patients (Fig. 1), we employed a 3% NaCl solution—commonly used for the treatment of hyponatremia (< 135 mM) [34,35,36]—in this model to investigate its potential cardioprotective effects.

We performed MI with acute hyperglycemia surgery in 10–12-week-old WT mice and then administered once tail vein injection of LA (50 mg/kg) or 3% NaCl or vehicle control (Saline) (Fig. 4J), which were optimized the dosage of NaCl and LA for validity. Compared with vehicle control group, mice treated with LA and 3% NaCl exhibited protection against MI with acute hyperglycemia, as indicated by a significant reduction in the ratios of HW/TL (see Supplementary material, Fig. S6A) and significant increases in EF and FS compared to vehicle controls (Fig. 4K and see Supplementary material, Fig. S6B-F). Histological analysis at 7 days MI with acute hyperglycemia revealed significantly decreased fibrotic scar size and LV wall thickness in LA and 3% NaCl-treated mice compared to the vehicle control group (Fig. 4L–N). Additionally, serum sodium concentration was significantly increased in mice treated with 3% NaCl (Fig. 4O). In accordance with the clinical data, a statistically significant positive correlation was observed in mice during MI with acute hyperglycemia. Specifically, elevated serum sodium concentrations were associated with improved EF (Fig. 4P). To assess the impact of 3% NaCl on MI in non-hyperglycemic patients, we constructed MI models and administered NaCl treatment to evaluate resulting changes. Treatment with 3% NaCl in MI mice demonstrated neither protective nor detrimental effects, indicating that 3% NaCl does not impact MI patients without acute hyperglycemia and is considered safe (see Supplementary material, Fig. S7). In conclusion, these results indicate that 3% NaCl or LA may be effective clinical treatments for MI with acute hyperglycemia, without adversely affecting MI patients, suggesting a favorable safety profile for this therapeutic approach.

To advance toward clinical translation, we further confirmed that acute hyperglycemia enhances ischemia-induced fibrotic scar size in an ischemia–reperfusion (I/R) mouse model (30 min/1 week), while treatment with 3% NaCl significantly reduces ischemia area with acute hyperglycemia-induced injury (Fig. 4R–S). Additionally, serum levels of BNP were found to be a significant decline in 3% NaCl mice compared to Saline (Fig. 4T). Peripheral blood sodium concentration was also significantly increased in mice treated with 3% NaCl (Fig. 4U). These findings suggest that NHE1 activation (3% NaCl and LA) exerts a broad-spectrum effect in mitigating ischemic myocardial injury associated with acute hyperglycemia.

NHE1 activator ameliorates heart injury by attenuating necroptosis

To elucidate the mechanism by which NHE1 activator mediates cardioprotection, we performed RNA-sequencing experiments with heart samples (infarct and border area) from both vehicle control and 3% NaCl-treated mice with Sham, MI or MI with acute hyperglycemia (Fig. 5A). The heatmap indicated that, compared to MI group, MI with acute hyperglycemia group significantly upregulates a number of genes in the infarcted area, including S100A8/a9, IL-1β, and the necroptosis related gene RIPK3 and MLKL (Fig. 5B). Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis was then used to assess functional changes. Notably, the pathways associated with immune response and cell fate, particularly those related to necroptosis and apoptosis were upregulated (Fig. 5C and see Supplementary material, Fig. S8A). While the administration of 3% NaCl downregulated the pathway of necroptosis and apoptosis, suggesting that NHE1 activation may confer a protective effect in MI with acute hyperglycemia by inhibiting necroptosis or apoptosis (Fig. 5D and See Supplementary material, Fig. S8B). We then confirmed the changes of these pathway key genes in heart tissue (infarct and border area) from MI with acute hyperglycemia mouse by treating with or without NHE1 activator by western blot analysis. Immunoblotting analysis demonstrated that the administration of the NHE1 activator significantly decreased the expression levels of MLKL and RIPK3, while having no changes in caspase-3 and LC3 (Fig. 5E and see Supplementary material, Fig. S9A). These findings suggest that necroptosis plays a critical role in MI with acute hyperglycemia and NHE1 activation potentially reverse this damage.

Fig. 5

NHE1 activator ameliorates heart injury by attenuating necroptosis. A Schematic illustration of heart tissue samples for RNA-Seq. B Heatmap showing the gene expression levels of distinct heart regions across different conditions. C KEGG pathway enrichment analysis of genes in the expression pattern in B (upregulation in the comparison of Glu+MI vs. MI) via hyper-geometric distribution. D Down regulated pathways in the comparison of Glu+MI+3% NaCl vs. Glu+MI+Saline via GSEA. E–K Western blot analysis of MLKL, p-RIPK3, RIPK3, Caspase 3 and LC3 in the hearts of the border and infarct area in Glu+MI mice treated with Saline and 3% NaCl (E), F/F and cKO (F), Con and NHE1 (G), n = 3 (E, G) and n = 4 (F). H Schematic of in vitro isolated adult mouse cardiomyocytes harvested from mouse hearts at day 7 post MI with acute hyperglycemia, treated with Saline and 3% NaCl for flow cytometry analysis. I–K Flow cytometric analysis (I) and quantification of surviving cells (J) and necroptotic cells (K) in adult mouse cardiomyocytes (n = 3 per group). Statistical analysis was performed using two-tailed unpaired student’s t-test. Data are presented as mean ± SD

It is well-established that necroptosis promotes adverse remodeling after MI [37,38,39]. However, whether knockout or activated/increased NHE1 reciprocally regulates necroptosis in MI with acute hyperglycemia remains largely unknown. To test this, western blot showed that the protein levels of MLKL and RIPK3 were significantly upregulated in the heart tissue of NHE1-cKO mice (Fig. 5F and see Supplementary material, Fig. S9B). Additionally, a significant decrease in these protein levels was observed in the cardiomyocyte-specific NHE1 overexpressing mice compared to their respective control groups (Fig. 5G and see Supplementary material, Fig. S9C), indicating the NHE1 activator ameliorates heart injury by regulating necroptosis in vivo. To further investigate the role of NHE1 activator in regulating necroptosis in cardiomyocytes, we isolated adult mouse cardiomyocytes from the heart after MI with acute hyperglycemia treated with or without 3% NaCl (Fig. 5H). Following staining with necroptosis markers Propidium Iodide (PI) and Annexin V, we conducted flow cytometry analysis (F5g. 5I). The results demonstrated that 3% NaCl treatment significantly reduced the proportion of necroptotic cardiomyocytes post MI with acute hyperglycemia, concurrently associated with an increase in the number of survival cells (Fig. 5J and K). Together, these results suggested that the activation or overexpression of NHE1 can rescue the cardiomyocyte necroptosis induced by MI with acute hyperglycemia, thereby providing cardioprotection.

NHE1 promote MLKL autophagic-degradation to confer cardioprotection

NHE1 downregulates necroptosis in cardiomyocytes subjected to MI with acute hyperglycemia. However, the precise mechanisms underlying this effect remain unclear. To further investigate this phenomenon, we established an in vitro cell culture model designed to mimic necroptosis in cardiomyocytes under conditions of MI and MI with acute hyperglycemic stress (see Supplementary material, Fig. S10A). The first model involved hypoxia combined with glucose deprivation (herein, OGD) and the second was high glucose combined with oxygen deprivation (herein, HG + OD), while the third was induced using TNF-α (T), SMAC mimetic SM-164 (S), and the pan-caspase inhibitor z-VAD (Z) (herein, TSZ), an inducer that can specifically induce necroptosis [40, 41]. To assess the effects of NHE1 activators on cardiomyocytes, we measured lactate dehydrogenase (LDH) release at various time points and doses in neonatal rat cardiomyocytes (NRCM) subjected to a necroptosis inducer (see Supplementary material, Fig. S10A). Our results indicated that treatment with 100 mM NaCl and 50 μM LA provided the most effective protection against necroptosis at 3 h post-induction (see Supplementary material, Fig. S10B-D). When cardiomyocytes exposed to HG + OD for 24 h or TSZ for 3 h, treatment with the NHE1 activation (NaCl or LA) significantly decreased the ratio of PI-positive cells in total cardiomyocytes and LDH release compared to the vehicle group (Fig. 6A–C), while the administration of the NHE1 inhibitor (cariporide) resulted in a significant increase in both parameters. In contrast, NaCl significantly increased the ratio of PI-positive cells in total cardiomyocytes and LDH release, while cariporide showed a protective effect in the OGD model (Fig. 6A–C). As anticipated, the NHE1 activation by NaCl, assessed by western blot in NRCM, showed upregulation of MLKL in the OGD model (Fig. 6D and see Supplementary material, Fig. S10E). Interestingly, in HG + OD model, NaCl significantly reduced the expression of MLKL, thereby improving cardiomyocyte survival rates (Fig. 6D and see Supplementary material, Fig. S10E and G). This finding is consistent with our previous animal studies, which indicate that NHE1 activation exacerbates injury in MI, while NHE1 activation in MI with acute hyperglycemia is cardioprotective. In contrast to the NHE1 activation by NaCl, our selected traditional Chinese medicine (TCM) small-molecule compound, LA, did not demonstrate a significant effect on MLKL expression (see Supplementary material, Fig. S10F and H), suggesting that it may influence necroptosis through alternative pathways. To exclude the influence of osmotic pressure, we utilized urea, which maintains equivalent osmotic pressure, as a control [42]. Our observations indicated that urea treatment did not alter the expression levels of RIPK3 and MLKL (see Supplementary material, Fig. S10I). To assess the cardiomyocyte specificity of NHE1 activation, we conducted validation experiments using the fibroblast cell line L929 [43]. Our results indicated that NHE1 activation did not modulate MLKL expression and failed to reverse necroptosis in the TSZ model within the fibroblast line (Fig. 6E and see Supplementary material, Fig. S11). Collectively, these findings suggest that the activation of NHE1 in cardiomyocytes under high glucose and hypoxia conditions specifically downregulates MLKL expression and reduces necroptosis.

Fig. 6

NHE1 promotes MLKL autophagic-degradation to confer cardioprotection. A Representative image of neonatal rat cardiomyocytes (NRCM) treated with vehicle (Vehi), NaCl (100 mM), LA (25 μM) and Cari (10 μM) during high glucose (35 mM)-oxygen deprivation (HG+OD), oxygen–glucose deprivation (OGD) stimulation and visualized using propidium iodide (PI). Scale bars, 100 μm. B Quantification of PI-positive cells in the total cell area (top, n = 5 for OGD, n = 3 for HG + OD) and cell viability measured by LDH assay (bottom, n = 3 per group) in NRCM treated with vehi, LA, NaCl, and Cari under HG + OD (24 h) and OGD (12 h) conditions. C Representative image and quantitative analysis of PI-positive cells in the total cell area (n = 6) and cell viability (n = 4) in NRCM following treatment with vehi, LA (50 μM), NaCl (100 mM), and Cari (10 μM) under TSZ (TNF-α, 50 ng/mL; SM-164, 5 μM and z-VAD, 25 μM) for 3 h in NRCM. D Western blot analysis of MLKL in NRCM treated with Saline and 100 mM NaCl during OGD (left) and HG + OD (right). E Western blot analysis of MLKL in L929 cells treated with 100 mM and 400 mM NaCl (left) for 30 min or with 100 mM NaCl during TSZ (right) for 60 min. Blots are representative of three independent experiments. F and G Representative images and quantification of PI⁺ cells in the total cell area and cell viability of NRCM infected with Lenti-shNHE1 and treated with TSZ for 3 h, n = 3, scale bars, 100 μm. H The ratio of intracellular pH to LDH release levels was measured by BCECF-AM (3 μM) and CellTiter-Glo assay, n = 4 per group. I Western blot analysis of MLKL in NHE1 overexpression and knockout in the cardiomyocytes with stable constructs, blots are representative of four independent experiments. J Transcriptional levels of MLKL in NRCM subjected to HG + OD for 6 and 12 h were compared between the Saline and 100 mM NaCl treatment groups, n = 3 per group. K and L Western blotting and quantification of MLKL were performed in MLKL-overexpressing cardiomyocytes treated with vehicle, 100 mM NaCl and then treated with 3-Methyladenine (3-MA, 10 mM, 12 h), Chloroquine (CQ, 25 μM, 12 h), Bafilomycin A1 (BafA1, 100 nM, 12 h), Rapamycin (RAPA, 5 μM, 24 h) and MG132 (20 μM, 8 h). Blots are representative of three independent experiments. Statistical significance was determined by One-way ANOVA with Tukey’s test. All quantitative data are expressed as mean ± SD

To further elucidate the specificity of NaCl and LA as NHE1 activation drug, we constructed the cardiomyocyte of NHE1 overexpression and knockdown (see Supplementary material, Fig. S12A). TSZ induces necroptosis, characterized by loss of cell membrane integrity, which can be identified by double positivity for PI and Annexin V [21]. Fluorescence staining analysis indicated that treatment with NaCl and LA significantly reduced the percentage of PI and Annexin V double-positive cells in WT cells, demonstrating resistance to TSZ-induced necroptosis, while the protective effects of the NHE1 activation were abrogated in NHE1 knockout cardiomyocytes (see Supplementary material, Fig. S12B and C). Knockdown of NHE1 in NRCM using lentiviral interference exacerbated TSZ-induced cardiomyocyte death (Fig. 6F), as evidenced by an increase in PI positive cells and reduced cell viability (Fig. 6G). Additionally, Western blot analysis revealed that NHE1 knockdown significantly increased the protein levels of MLKL, whereas NHE1 overexpression resulted in decreased in MLKL levels, compared to WT cardiomyocytes (Fig. 6I). Therefore, these data demonstrated that, in contrast to NHE1 deficiency, overexpression or activation of NHE1 significantly decreased the protein level of MLKL, thereby defending against cardiomyocyte death.

As NHE1 is a transmembrane protein that facilitates the exchange of hydrogen and sodium, intracellular pH is commonly employed as an indicator of NHE1 activation [13, 44, 45]. In WT cells, treatment with NaCl and LA resulted in a significant increase in intracellular pH (pHi), indicating that NHE1 was effectively activated by NaCl and LA to counteract TSZ-induced acidification, thereby exerting a protective effect (Fig. 6H and see Supplementary material, Fig. S12D). Therefore, the loss of NHE1 resulted in the inability of LA to alter pHi and diminished the capacity of NaCl to regulate pHi (Fig. 6H). These data suggest that, compared to NaCl, LA more specifically targets NHE1 and is more likely to function as a small-molecule compound modulating NHE1 activity (see Supplementary Fig. S14A). NaCl, on the other hand, may act on additional potential targets in cardiomyocytes. To explore the functional consequences of NHE1 activation under different glucose conditions, we also evaluated pHi in cardiomyocytes exposed to varying glucose concentrations. To investigate glucose-NHE1 interactions in MI with acute hyperglycemia, we performed two complementary in vitro experiments using NRCM: (1) Acute exposure to 5 mM (normoglycemia) or 22 mM (hyperglycemia) glucose for 30 min under normoxia/hypoxia; (2) Glucose fluctuation (60 min 5/22 mM under normaxia/hypoxia → 60 min 5 mM with reoxygenation). Under normoxic conditions, 22 mM glucose exposure did not significantly alter NHE1 activity compared to 5 mM controls. By contrast, hyperglycemia (22 mM glucose) during hypoxic conditions leads to relatively reduced NHE1 activity as compared to normoglycemic conditions (5 mM glucose) during hypoxic conditions. In the glucose fluctuation protocol, hypoxia/reoxygenation induced cellular alkalinization compared to normoxic conditions, yet no fold-change in NHE1 activity was detected between 5 and 22 mM glucose treatments under either oxygen state. This may be due to the inherent limitations of in vitro systems to precisely monitor dynamic NHE1 activity. These data demonstrate that acute changes in glucose concentration under normoxic conditions do not significantly alter NHE1 activity in vitro. However, under hypoxic conditions, NHE1 activity shows glucose concentration-dependent inhibition (see Supplementary material, Fig. S12E–F), suggesting that MI with acute hyperglycemia may exacerbate cardiac pH dysregulation through suppression of NHE1 activity. We further examined whether NHE1 activation affects ER stress in cardiomyocytes. We utilized NHE1-overexpressing cardiomyocytes to model enhanced NHE1 activity and compared them with WT group to assess changes in the expression of ER stress marker genes (Grp94, Grp78, Calr, and P4hb). Results indicated that NHE1 overexpression partially upregulated ER stress markers (P4hb and Grp78), suggesting a potential link between NHE1-mediated ER stress (see Supplementary material, Fig. S13A). These findings suggest that in the in vitro model of MI with acute hyperglycemia, NHE1 intrinsic activity is regulated. Beyond its role in modulating cardiomyocyte death, NHE1 may also function as a sensor under hyperglycemic conditions, potentially influencing cellular ER stress.

To elucidate how activated NHE1 affects necroptosis, we selected the small molecule MGCB, which directly regulates pHi, and nanoparticle-encapsulated NaHCO₃, which directly modulates intracellular sodium ion concentration, based on the characteristics of the NHE1 membrane protein in regulating these parameters. Our results indicated that both MGCB and NaHCO₃ effectively counteract TSZ-induced cell damage, as evidenced by LDH release (see Supplementary material, Fig. S14B). These findings suggested that NHE1 regulates necroptosis by enhancing intracellular Na⁺ levels and pHi, thereby promoting MLKL degradation.

Subsequently, to investigate how NHE1 alters intracellular sodium and hydrogen levels, leading to a reduction in MLKL protein levels. We performed quantitative PCR (qPCR) analysis. It revealed that, in contrast to the vehicle group, mRNA levels of MLKL were not decreased but rather increased following treatment with the NHE1 activation in cardiomyocytes exposed to hypoxia and high glucose for 6 and 12 h (Fig. 6J). This result indicated that NHE1 activation treatment does not influence the transcriptional levels of MLKL, implying a possible effect on the degradation of MLKL protein. The established pathways of protein degradation include ubiquitination modification, the lysosomal pathway and autophagy, among others [46,47,48]. To investigate the pathway underlying MLKL degradation, we established a cardiac cell line that overexpresses MLKL (see Supplementary material, Fig. S14C). And then we utilized several inhibitors and activators: MG132, an inhibitor of ubiquitination; Rapamycin (RAPA), which promote autophagy; BafA1, an inhibitor of autophagosome-lysosome fusion. These agents were employed to counteract the degradation of MLKL induced by the NHE1 activation in MLKL-overexpression cardiomyocytes (Fig. 6K). Interestingly, we observed that BafA1 administration, but not other inhibitors, significantly inhibited NHE1 activation-induced MLKL degradation in MLKL-overexpression cardiomyocytes (Fig. 6K–L). In summary, our results suggested that NHE1 mitigated myocardial injury induced by MI with acute hyperglycemia by modulating intracellular Na⁺ and H⁺ levels, thereby facilitating MLKL degradation via the autophagic-lysosomal pathway.

MLKL knockdown dampens injury caused by NHE1 deficiency/inhibition in MI with acute hyperglycemia

Pseudokinase mixed lineage kinase domain-like protein (MLKL) serves as a terminal-known obligate effector in the process of necroptosis [49]. To further investigate the role of NHE1 as an upstream regulator of MLKL in necroptosis in vivo, we performed NHE1 knockout and pharmacological inhibition in conjunction with MLKL knockout mice (Fig. S15A). Due to impaired growth and development in NHE1 knockout mice, characterized by ataxic gait, epileptic seizures, and a mortality rate of 60% [50, 51], these models were unsuitable for investigating MI with acute hyperglycemic conditions. Consequently, we opted to treat WT mice with cariporide, a pharmacological inhibitor, for subsequent comparisons (Fig. 4A). Although gross morphology of NHE1^−/− (NKO) mice was slightly smaller than WT and MLKL^−/− (MKO) mice and exhibited decreased postnatal growth with ataxic gait [50, 51], adult double NHE1^−/−MLKL^−/− (DKO) mice did not show significant differences compared to WT and MKO mice, as evidenced by representative images (Fig. 7A). Additionally, DKO mice did not exhibit defects such as ataxic gait, epileptic-like seizures or increased mortality observed in NKO mice (data not shown). Genotyping and western blot analysis confirmed the efficient knockout of the Nhe1 and Mlkl genes in DKO, NKO and MKO mice (Fig. 7B, C).

Fig. 7

MLKL knockout mitigates cardiac injury caused by NHE1 deficiency/inhibition in MI with acute hyperglycemia. A Gross morphology and heart size comparison of WT, MKO (MLKL^−/−), NKO (NHE1^−/−) and DKO (NHE1^−/−MLKL^−/−), scale bar 1 cm for mice and 1 mm for heart. B Genotyping of WT, MKO, NKO and DKO by PCR using primer pairs specific to the Nhe1 and Mlkl. C Western blot and quantification of NHE1 and MLKL protein levels in heart extracted from WT and DKO mice. D Schematic diagram showing the strategy of WT, Cari, MKO, DKO and MKO + Cari mice during MI with acute hyperglycemia. E Representative M-mode echocardiographic images (left) and Masson’s trichrome staining (right) of WT, Cari, MKO, DKO and MKO + Cari mice post 7 days of MI with acute hyperglycemia, scale bar, 100 ms and 1 mm for echocardiographic images and scale bar, 1 mm for Masson. F and G Comparison of EF and FS at day 7 post MI with acute hyperglycemia in WT (n = 5), Cari (n = 8), MKO (n = 7), DKO (n = 7) and MKO + Cari (n = 5) mice. H and I Quantification of cardiac fibrosis of total heart section area and LV wall thickness in WT (n = 6), Cari (n = 7), MKO (n = 4), DKO (n = 5) and MKO + Cari (n = 4) mice. J Analysis of serum BNP level between WT (n = 5), MKO (n = 5), DKO (n = 4) and MKO + Cari (n = 4) mice post 7 day MI with acute hyperglycemia. K Analysis of serum Na⁺ concentrations from the WT (n = 6), MKO (n = 7), DKO (n = 6) and MKO + Cari (n = 5) post 7 day MI with acute hyperglycemia. Statistical significance was determined by One-way ANOVA with Tukey’s test. All quantitative data are expressed as mean ± SD

As previously mentioned, we observed a progressive deterioration of multiple hallmarks of cardiac dysfunction and fibrosis in cariporide-treated mice after 7 days of MI with acute hyperglycemia, including EF, FS, fibrotic area, LV wall thickness and BNP levels (F7g. 7E–I). Interestingly, administration of cariporide to MKO mice resulted in significant increases in EF, FS and LV wall thickness, along with decreases in fibrosis area and BNP levels 7 days after MI with acute hyperglycemia (Fig. 7E–I). Subsequently, we constructed NHE1 and MLKL double knockout (NHE1^−/−MLKL^−/−) mice to investigate the effects of MI with acute hyperglycemia. Consistent with pharmacological inhibition, echocardiography and histological analyses revealed that MLKL deficiency mitigated the detrimental effects associated with either NHE1 inhibitor or loss in this model (F7g. 7F–I). However, DKO mice and MKO mice treated with cariporide exhibited less protection against NHE1 deletion or inhibition compared to MKO mice. This reduced protection may be attributed to the global NHE1 knockout in these mice and the targeted inhibition provided by cariporide. Additionally, the serum BNP level was decreased in MKO mice compared to the WT group, while this protective effect disappeared after inhibiting NHE1 (Fig. 7J). We also found that serum Na⁺ concentration was significantly elevated in MKO mice compared to the WT group, although this increase was attenuated in DKO mice (Fig. 7K). In conclusion, these results indicated that MLKL deficiency disrupts the progression of deleterious effects in NHE1 inhibitor and deletion mice following MI with acute hyperglycemia.

Discussion

Acute hyperglycemia on admission is a common identity during the early phase after acute MI, even without the history of diabetes mellitus. Previous studies have indicated that NHE1 regulates intracellular pH by exchanging intracellular H⁺ for extracellular Na⁺ to maintain cardiomyocyte function [52, 53]. However, the precise role of NHE1 in the pathophysiology of MI with acute hyperglycemia remains to be elucidated, and there are no effective strategies for its prevention or treatment. We present clinical data demonstrating that patients exprencing MI with acute hyperglycemia exhibit more severe cardiac injury and ionic imbalance compared to normoglycemia conditions, supporting the clinical relevance of our animal experimental models and in vitro studies. To explain the underlying mechanisms, we subsequently employed an MI with acute hyperglycemia mouse model to demonstrate that cardiomyocyte NHE1 activation attenuates ionic dysregulation and reduces necroptotic cell death. Then, through both neonatal rat cardiomyocytes and adult mouse cardiomyocytes, we provide novel insights into the biological significance of cardiomyocyte NHE1 in modulating necroptosis in response to hypoxia and hyperglycemia. We confirmed that hyperglycemia exacerbates myocardial injury by altering Na⁺ and pH levels, with NHE1 playing a key role in both humans and mice. Cardiomyocyte-specific ablation or pharmacological inhibition of NHE1 worsened cardiac dysfunction and fibrosis in MI with acute hyperglycemia, while NHE1 overexpression had a protective effect. Drug screening and RNA sequencing identified NHE1 activation, through 3% NaCl and LA, as a potential strategy to reduce necroptosis in cardiomyocytes and alleviate MI with acute hyperglycemia. Mechanistically, NHE1 activation promotes the degradation of necroptosis-related protein MLKL via autophagosome-lysosome pathways. Although NHE1 activation is harmful in MI without hyperglycemia, our study suggests its therapeutic potential in MI with acute hyperglycemia, highlighting its dual role and potential as a novel therapeutic strategy to reduce infarct size and protect cardiac function.

Stress hyperglycemia has been associated with adverse outcomes in individuals with critical illnesses, including those with acute MI, heart failure, and stroke [54]. However, a standardized definition of stress hyperglycemia in patients with MI is still not established. Most early studies defined hyperglycemia based on the first glucose value obtained at admission. In our study, patients exprencing MI with acute hyperglycemia were defined as those exhibiting an initial blood glucose measurement exceeding 7 mM. In diabetic patients, fasting blood glucose levels greater than 7 mM are defined as hyperglycemia. However, in our study, we were unable to differentiate between fasting blood glucose and postprandial blood glucose levels. Additionally, the stress hyperglycemia ratio (SHR) is defined as the ratio of admission glucose to the estimated average glucose, calculated from glycosylated hemoglobin A1c (HbA1c) [55], which was supposed to better identify acute hyperglycemia [6, 56]. Actually, admission glucose levels and conventional SHR, derived from admission glucose and HbA1c, may be affected by meal timing. Therefore, further research is needed to establish a more accurate definition of acute hyperglycemia in patients with MI.

The dual role of NHE1 may arise from its activation status under varying pathological conditions. Evidence suggests that the activation of NHE1 facilitates the exchange of cytoplasmic Na⁺ and H⁺, which subsequently leads to an increase in cytoplasmic Ca²⁺. This rise in Ca²⁺ levels activate calcium/calmodulin-dependent protein kinase II (CaMKII) and calcineurin (CaN), promoting cardiac hypertrophy in the context of I/R injury [57]. Empagliflozin (EMPA) inhibits NHE1, thereby preventing excessive autophagy and providing protection against cardiomyocyte death in MI [9]. Additionally, the release of glucocorticoids from the neuroendocrine system activates NHE1, leading to autophagic death of bone marrow B cells in the context of MI [25]. These findings elucidate that NHE1 is over-activated during MI or I/R without hyperglycemia, and its inhibition may confer cardioprotective effects. However, in our study, NHE1 was inhibited in MI with acute hyperglycemia due to an imbalance of Na⁺ and H⁺ flux across the cell membrane, resulting in its activation having a cardioprotective effect. This observation also explains why EMPA does not inhibit NHE1 but rather exerts cardioprotective effects through Beclin1 in the MI with acute hyperglycemia model. The activity of NHE1 may be influenced by glucose availability. So, we explored the interplay between glucose availability, NHE1 activity, and glucose metabolism, demonstrating that fluctuations in glucose levels alter NHE1 activity, which in turn modulates key metabolic pathways in cardiomyocytes. These findings suggest that NHE1 may function as a metabolic sensor under hyperglycemic conditions, linking ionic homeostasis with metabolic adaptation. However, it remains unclear how NHE1-mediated metabolic alterations influence downstream necroptosis pathways to exert cardioprotective effects. Further investigation is required to elucidate these mechanisms and their implications for hyperglycemia-associated cardiac injury. Furthermore, while ER stress was elevated in cardiomyocytes, it was not prominently observed in myocardial tissue. Moreover, the protective effects of NHE1 overexpression were abolished by cariporide, indicating that NHE1-mediated cardioprotection may primarily depend on its activity rather than solely on ER stress activation. In contrast, previous studies reported that in a Langendorff perfusion model using NHE1-overexpressing mice (cardiac-specific overexpression ~ 21-fold), treatment with the NHE1 inhibitor zoniporide (3 μM) did not alter the cardioprotective effects in an ischemia–reperfusion injury model [27]. The authors concluded that NHE1 overexpression-induced cardioprotection was mediated by elevated NHE1 protein levels (via ER stress activation) rather than its activity. The discrepancy between our findings and Ref. [52] likely stems from differences in NHE1 overexpression efficiency (~ 1.5-fold vs. ~ 21-fold), suggesting a threshold-dependent mechanism where moderate overexpression preserves activity-dependent effects, while extreme overexpression may induce compensatory pathways (e.g., ER stress). In conclusion, in different disease models, membrane proteins make corresponding adjustments according to changes in the cellular microenvironment, so this study provides a new indirect method for membrane protein dual regulation.

NHE1 has emerged as a promising target for drug development, leading to the development of multiple inhibitors, with cariporide receiving the most significant attention [58]. Extensive preclinical studies indicate that inhibition of NHE1 have demonstrated a significant protective effect against myocardial ischemia across various species, including porcine [17, 59], rabbits [12], dogs [60], rat and mice [61], predominantly through reduced intracellular accumulation of Na⁺ and consequently Ca²⁺. In contrast, numerous clinical investigations involving the NHE1 inhibitors cariporide and eniporide in patients with evolving MI and those at elevated risk for MI have yielded inconsistent and somewhat contradictory findings [58]. In the GUARDIAN trial, while there was no significant benefit of cariporide compared to placebo in the overall population, notable cardioprotective effects were observed specifically in a subgroup of high-risk patients undergoing coronary artery bypass graft surgery [62,63,64]. Conversely, the ESCAMI trial indicated no cardioprotective benefit from eniporide when administered as a late adjunct to reperfusion therapy in patients with evolving MI [65]. The mixed findings from clinical studies involving NHE1 inhibitors, in contrast to the promising preclinical data, highlight the complexity of the clinical condition in acute MI patients. Our findings partially elucidate the discrepancy in the effects of NHE1 inhibitors between MI patients and MI mice, attributed to the significantly greater complexity of MI patients.

While NHE1 is well-established as a pivotal target in cardiovascular disease[10, 63, 65, 66] and numerous inhibitors have been developed, there is a paucity of strong agonists that demonstrate effective targeting and specificity. The small-molecule compound rosmarinic acid has been shown to activate NHE1, thereby regulating pH and aiding in the maintenance or recovery of skin barrier functions [28]. Moreover, IgE has been reported to activate macrophages by enhancing NHE1 activity, thereby exacerbating atherosclerosis [67, 68]. However, the potential of these mechanisms in the treatment of MI with acute hyperglycemia has not yet been investigated. In this study, we identified lithospermic acid (LA) and 3% NaCl as specific activators of NHE1, enhancing its activity both in vivo and in vitro, and representing a novel therapeutic approach for improving cardiac function. In addition to NHE1, cardiomyocytes express other membrane proteins responsive to intracellular/extracellular pH and sodium dynamics, including SGLT1, Na⁺/Ca²⁺ exchanger (NCX), and proton pumps (e.g., H⁺/K⁺-ATPase). Studies have shown that in normoglycemic MI, NCX operating in reverse mode exacerbates Ca²⁺ overload and myocardial injury [69]. SGLT1 inhibition attenuates post-ischemic cardiac dysfunction by suppressing glucose-sodium cotransport, thereby reducing oxidative stress and inflammation [70]. Furthermore, dysregulated proton pump activity disrupts mitochondrial proton gradients, aggravating mitochondrial dysfunction [71]. These findings suggest that in MI with acute hyperglycemia, coordinated targeting of SGLT1, NCX, and ion pumps, alongside NHE1, may offer novel multi-target therapeutic strategies. A limitation of this study is that NHE1 activity was inferred indirectly from intracellular pH changes rather than measured directly. To conclusively establish its functional role under hyperglycemic conditions, future studies should employ direct assessment methods such as NH₄⁺ pulse-induced pH recovery assays in HEPES-buffered systems [18].

Cardiomyocyte necroptosis has significantly detrimental effects on cardiac function [38, 72]. MLKL is recognized as crucial therapeutic targets within the necroptosis pathway. Consequently, targeting MLKL represents a promising strategy for treating necroptosis-related diseases (see Supplementary material, Tables S5). For example, NSA was shown to block necroptosis in HT29 cells by covalently modifying MLKL at Cys86. However, the lack of corresponding cysteine residues in mouse MLKL hinders further development of drug candidates [73]. In this study, we proposed that targeting cardiomyocyte membrane protein NHE1 can regulate MLKL degradation and screen out more promising NHE1 activators for clinical translation. However, in fibroblasts, much higher doses of NaCl than we used (400 mM vs. 100 mM-herein) activated necroptosis mediated by NHE1-induced increase in cytosolic pH [43]. Our results demonstrated that NHE1 regulates necroptosis by influencing the degradation of MLKL in cardiomyocytes. Therefore, we propose that the mechanism of necroptosis inhibition involves not only the conventional suppression of MLKL phosphorylation and subsequent translocation to the plasma membrane for pore formation but also the enhancement of MLKL protein degradation. While our data implicate NHE1 activation in MLKL autophagic degradation, the underlying mechanisms require further exploration.

Conclusions

In summary, NHE1 activation presents a novel therapeutic strategy for reducing infarct size and protecting cardiac function following MI with acute hyperglycemia. Accelerating NHE1 activation via LA and 3% NaCl infusion represents a potential innovative approach for managing MI with acute hyperglycemia. Importantly, we have demonstrated for the first time that cardiomyocyte NHE1 exacerbates heart damage by mediating the degradation of MLKL during MI with acute hyperglycemia. Our findings not only highlight promising compounds for clinical therapy in cardiovascular diseases but also offer a new perspective on the treatment of various types of MI.