2.1.1. Experimental animals.

Male mice of different genotypes (with an initial number of 20 in each group) were selected, and the specific grouping and background were as follows:

The wild-type control group: 8-week-old C57BL/6J mice (purchased from Cyagen Biosciences Inc.) were used. This strain is an inbred mouse line with a stable genetic background, exhibiting no spontaneous obesity or diabetic phenotypes. At 8 weeks of age, the mice were in their youth stage, characterized by stable basal metabolism and vascular function. Age-matched with mice in other experimental groups, the influence of age differences on experimental outcomes was effectively excluded.

The diabetes model group: 8-week-old db/db mice (purchased from Changzhou Cavens Laboratory Animal Co., Ltd.) were used. This strain develops spontaneous obesity and type 2 diabetes due to a mutation in the Leptin receptor, which meets the requirements for a diabetes model in this study.

The STING knockout diabetes group: STING knockout db/db mice (with the same background strain as the db/db mice) were used to verify the role of the cGAS/STING pathway.

The iNOS knockout diabetes group: iNOS knockout db/db mice (with the same background strain as the db/db mice) were used to verify the role of NO mediated by iNOS.

Animals were randomly assigned to groups using a random number table: after one week of acclimatization feeding, all 8-week-old male mice were first stratified according to genotype (C57BL/6J, db/db, STING knockout db/db, iNOS knockout db/db) and then allocated to corresponding experimental groups within each stratum via a random number table to ensure balanced initial physiological states across groups. The randomization process was conducted by personnel not involved in experimental operations, and the grouping results were sealed until unblinding after the completion of data collection. No significant differences were observed among the groups in fasting blood glucose, heart rate, or blood lipid levels (total cholesterol, triglycerides) prior to modeling (P > 0.05), indicating balanced and comparable baseline characteristics and eliminating interference from variations in basic physiological states on the experimental outcomes (Table 1).

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Comparison of baseline physiological parameters before modeling in each group of mice.

https://doi.org/10.1371/journal.pone.0350983.t001

All mice were housed in an SPF environment, maintained on a 12-h light-dark cycle, and allowed free access to standard diet and water. They were provided with nesting materials and toys; they were monitored daily by professionals, and were humanely euthanized immediately if signs such as a weight loss of over 10% or lethargy appeared. The experimental protocol was reviewed and approved by the Ethics Committee of Bishan Hospital, Chongqing Medical University (Approval No.: BSKJ2024063). All animal experiments were performed in strict accordance with the Guidelines for the Care and Use of Laboratory Animals in China. The body weight data of mice in each group are shown in Table 2. The body weights of the db/db model group and knockout groups were significantly higher than those of the control group (P < 0.01), consistent with the characteristics of a diabetic obesity model.

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Initial average body weight of mice in each group and changes in body weight after modeling.

https://doi.org/10.1371/journal.pone.0350983.t002

During surgery and invasive procedures, animals were anesthetized by intraperitoneal injection of 1% pentobarbital sodium (50 mg/kg), with the disappearance of the pedal reflex and a decrease in respiratory rate serving as indicators of anesthetic depth. After surgery, buprenorphine (0.05 mg/kg) was administered subcutaneously every 12 h for 48 h to relieve pain. At the end of the experiment, animals were euthanized in accordance with the euthanasia guidelines of the American Veterinary Medical Association. Cervical dislocation was performed under deep anesthesia (confirmed by the disappearance of corneal and pedal reflexes), and animals were euthanized immediately after the final assessment when tissue collection was required to minimize tissue degradation. Invasive procedures were performed by experienced personnel, and incisions were closed with sterile sutures and coated with antiseptic to prevent infection. This study strictly implemented blinding procedures: 1) Blinded Grouping: Animal grouping, numbering, and intervention operations were conducted by dedicated personnel. The technicians responsible for experimental assays and data analysts did not participate in the grouping process and were only aware of animal identification numbers, with no knowledge of group assignments. 2) Blinded Assessment: During procedures such as Western blot development, immunofluorescence image acquisition, transmission electron microscopy observation, and hemodynamic parameter recording, operators handled samples based solely on identification numbers and were not informed of the corresponding group assignments. 3) Blinded Statistical Analysis: After data compilation, blinding was lifted by a third-party statistician before statistical analysis was conducted to prevent subjective bias.

2.1.2. Construction of diabetic hypoglycemia model.

Diabetes-modeling screening: 8-week-old db/db mice with fasting blood glucose ≥11.1 mmol/L for 3 consecutive days were confirmed to have successful diabetes modeling; 8-week-old C57BL/6J mice served as a normal control, with consistent age with diabetic model groups.

Hypoglycemia induction: All model groups (including the db/db group, STING knockout group, and iNOS knockout group) were subjected to the following treatment: After fasting for 12 h (with free access to water), intraperitoneal injection of Novolin R insulin (15 IU/kg, batch number J20050019, Novo Nordisk, Denmark) was administered. Tail-vein blood glucose was measured every 20 min, and when blood glucose levels dropped to ≤3.5 mmol/L, hypoglycemia induction was deemed successful. This state was maintained for 1 h, followed by intraperitoneal injection of 20% glucose PBS solution (1 g/kg, R21823-500, Shanghai Shangbao Biotechnology Co., Ltd.) to restore blood glucose levels [20]. This procedure was performed once daily for five consecutive days. The control group (C57BL/6J mice) received injection of an equivalent volume of normal saline without inducing hypoglycemia.

During the hypoglycemia induction process, the insulin injection speed (0.1 mL/min) and the timing of glucose solution resuscitation were strictly controlled to ensure that the duration of hypoglycemia (1 h) and the rate of blood glucose recovery (blood glucose restored to ≥7.0 mmol/L within 30 min after resuscitation) remained consistent across all mouse groups. The modeling was conducted daily at a fixed time between 9:00 AM and 11:00 AM to minimize the influence of circadian rhythms on the experimental results.

2.1.3. Modeling outcomes.

Initially, there were 20 mice in each group, and the final survival and successful modeling situations were as follows:

In C57 control group, 18 mice survived (2 died of natural aging);

In db/db model group, 17 mice were successfully modeled (3 were excluded due to severe hyperglycemic complications);

In STING knockout group, 19 mice were successfully modeled (1 was excluded due to failed genotype identification);

In iNOS knockout group, 18 mice were successfully modeled (2 were excluded due to poor tolerance to hypoglycemia).

2.2.1. Primary cell isolation and culture.

1. a. Primary mouse coronary artery endothelial cells (used for hypoglycemia model research):

The time point for isolation was 24 h after the completion of the hypoglycemia model construction (that is, 24 h after the end of 5 consecutive days of hypoglycemia induction), and the cells were isolated from the hearts of 6 db/db diabetic mice (8 ± 2 weeks old). The specific steps were as follows:

The hearts were removed from the heparinized (1 U/g) and anesthetized mice and washed in 4°C pre-cooled Krebs-Ringer buffer (containing Ca² ⁺ /Mg² ⁺ , T10489, Shanghai Shangbao Biotech Co., Ltd.); the anterior wall of the left ventricle was incised, and the left/right atria, interventricular septum, epicardium, and the outer 1/4 of the left ventricle were removed, leaving the coronary artery-enriched area; the tissue was minced and added to 0.2% Type II collagenase (Worthington) pre-warmed at 37°C, and digested in a 37°C water bath for 15−30 min; 0.05% trypsin + 1% EDTA (Gibco) was added to continue digestion for 30 min; the digest was filtered through a 100 μm sieve and centrifuged at 100 × g for 10 min; digestion was terminated by adding 10% fetal bovine serum (Nanjing Saihongrui Biotechnology Co., Ltd.), and after a second centrifugation, the precipitate was resuspended in endothelial cell culture medium (EGM-2, ScienCell); during the first passage, positive selection with anti-mouse CD31 microbeads was used, and the purity was verified by flow cytometry to be > 90%.

Cultivation conditions: EGM-2 medium (containing 2% FBS, VEGF, EGF, FGF) was used, and the cells were cultured in an environment of 37°C and 5% CO₂, with medium change every 2 days.

1. b. Primary mouse cardiomyocytes (used for myocardial injury research):

From C57BL/6J mice (without hypoglycemia induction), the steps were as follows:

After the heart was minced, it was placed in a petri dish containing 0.075% trypsin (Gibco) and digested at 4°C for 24 h (to inhibit fibroblast proliferation); the digested tissue was then transferred to 0.05% Type II collagenase pre-warmed at 37°C, and digested by magnetic stirring at 120 rpm and 34–35°C for 30 min until the viscous material disappeared; DMEM-F12 medium (T10299, Shanghai Shangbao Biotech Co., Ltd.) containing 10% FBS was added, and the mixture was centrifuged at 1000 rpm for 10 min; the precipitate was resuspended in DMEM-F12 medium containing 15% FBS and Brdu (67 μL/20 mL), and filtered through a 200-mesh sieve.

The culture conditions were as follows: DMEM-F12 + 15% FBS + Brdu was used, and the culture dishes had been pre-coated with 10 μg/mL laminin to promote adhesion.

2.2.2. Construction of hypoglycemia cell model.

1. a. Experimental groups:

The high glucose group (HG) was prepared with DMEM containing 30 mM glucose (4.5 g/L basic DMEM plus 1 M glucose solution).

The low glucose group (LG) was prepared with sugar-free DMEM containing 1 mM glucose (sugar-free DMEM plus 1 M glucose solution to adjust the concentration).

1. b. The model was established as follows:

The coronary artery endothelial cells were cultured in high-glucose medium (30 mM) for 48 h (to simulate the diabetic environment), then treated with low-glucose medium (1 mM) for 2 h, and subsequently recultured in high-glucose medium (30 mM) for 6 h. This procedure was repeated (three times), with a total duration of 24 h. The duration of low-glucose treatment was ≤ 2 h to avoid irreversible cell damage. Three replicate wells were set for each group, and the experiment was performed in triplicate [21]. During alternating high and low glucose treatments, the medium replacement volume was precisely controlled using an accurate pipette (2 mL per well). Uniform cell culture conditions across all groups were ensured by real-time monitoring of the CO₂ concentration (5%) and temperature (37°C) inside the incubator. The duration of low glucose treatment (1 mM) was strictly limited to 2 hours, which was verified through preliminary experiments to be sufficient for inducing cellular stress responses without causing irreversible damage.

The technical route of this study is illustrated in Fig 1.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Flowchart of the technical route.

https://doi.org/10.1371/journal.pone.0350983.g001

3.1.1. Subcellular localization detection (immunofluorescence).

Cardiomyocytes/endothelial cells were seeded on poly-L-lysine-coated coverslips at 5 × 10⁴ per well (24 h). They were fixed with 4% paraformaldehyde (PFA) (Catalog number 30525-89-4, Jiangsu Wenru Technology Chemical Co., Ltd.) (25°C, 20 min), followed by three washes with PBS (3 min each). Samples were treated with 0.1% Triton X-100 (in PBS) (15 min) to achieve permeabilization, followed by three PBS rinses. Blocking was carried out adopting 5% BSA (0.1% Triton X-100) [Yishijiu (Lianyungang, Jiangsu) Biotechnology Co., Ltd.. (25°C, 1 h). Overnight incubation at 4°C was performed with first antibodies (Ab), including cGAS (Catalog number YSH605050, Shanghai Yishenghe Biotechnology Co., Ltd., 1:200), STING (Shanghai Yuanye Biotechnology Co., Ltd., 1:200), and mitochondrial marker (Catalog number CS22668, AAT Bioquest, 1:500). Secondary Ab were applied (37°C, 1 h) (light-protected), including Alexa Fluor 488/594-conjugated secondary Ab (Thermo Fisher Scientific China, 1:500). The nuclei were stained with DAPI (5 μg/mL) (Shanghai Shangbao Biotechnology Co., Ltd.) (5 min), followed by washing with PBS and mounting with an anti-fade reagent (ProLong Gold) [Yishijiu (Lianyungang, Jiangsu) Biotechnology Co., Ltd... Sample visualization was carried out under a confocal microscope (Catalog number SZ61TR, Leica). Five fields of view were randomly selected from each group, and the colocalization coefficient (Pearson’s coefficient) of cGAS/STING with the mitochondrial marker was calculated by Image J software. The experiment was independently repeated three times. Nuclear fluorescence intensity of NF-κB p65 was quantified using ImageJ software, and the relative expression was normalized to the control group.

3.1.2. Protein expression detection.

Following lysis in RIPA buffer containing protease inhibitors, quantification was carried out by BCA assay. 30 μg of protein was loaded for electrophoretic separation, followed by wet transfer to a PVDF membrane (100 V, 90 min). Overnight incubation (4°C) was carried out with first Ab, including cGAS (1:1000), STING (1:1000). Secondary Ab were applied (25°C, 1 h), including HRP-conjugated anti-rabbit/mouse IgG (1:5000) (Catalog number C2210, APPLYGEN). The proteins were visualized using ECL, and the gray values were determined with Image Lab software, with β-actin serving as the control for normalization [22]. The experiment was independently repeated three times, and the results were expressed as mean ± standard deviation (±s).

The quality control standards for cGAS/STING protein detection were as follows: the protein sample concentration was required to be ≥ 1 mg/mL; SDS-PAGE electrophoresis bands needed to be clear and without trailing; transfer efficiency was verified using pre-stained protein markers (with complete transfer of bands in the 25–150 kDa range); and the coefficient of variation for β-actin internal reference grayscale values was required to be ≤ 10% to ensure uniform protein loading.

3.1.3. Inflammatory factor detection.

ELISA for secreted proteins: The supernatant was collected after centrifugation (2,000 rpm, 10 min) and stored at −80°C. Overnight incubation (4°C) was carried out for coating Ab (IL-1β/IL-6/IL-8/ICAM1/GMCSF/IFN-β, Thermo Fisher Scientific China). Subsequently, samples/standards (50 μL per well) were applied (37°C, 2 h). The HRP-conjugated secondary Ab were applied (1 h), followed by TMB color development and measurement of the OD value at 450 nm. The concentrations of each factor (pg/mL) were calculated according to the standard curve, and the experiment was independently repeated three times.

qPCR for mRNA levels: Total RNA extraction was conducted adopting the Trizol approach, and the concentration was measured with a Nanodrop (A260/280 ≥ 1.8). cDNA was synthesized using the PrimeScript RT kit (Catalog number RR064A, Takara). The primer sequences are shown in Table 3, and the experiment was independently repeated three times.

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. qPCR primer sequences and reaction conditions.

https://doi.org/10.1371/journal.pone.0350983.t003

3.2.1. NO detection.

1. a. Detection of NO release by electrochemical method

Cells were seeded in a 6-well plate and cultured until reaching 80% confluence. The medium was then replaced with phenol red-free medium (Catalog number 11835030, Thermo Fisher Scientific, China) and incubation was performed at 37°C for 1 h. The medium was collected, and centrifugation was carried out at 4°C and 3000 rpm for 10 min, and the supernatant was reserved for later use. The ISO-NOP NO sensor (World Precision Instruments) was calibrated with a sodium nitrite standard curve ranging from 0.1 μM to 10 μM (R² ≥ 0.99). The samples were placed in a constant temperature bath at 37°C, and a 2 mm diameter NO electrode was inserted. After recording a stable baseline, L-arginine (1 mM, Catalog number 5256-76-8, Tai’an Heguang Fine Chemical Co., Ltd.) was added to stimulate NO release, and the peak changes within 5 min were continuously monitored. The NO concentration (nM/mg protein) was calculated by LabTrax 4.0 software, and the experiment was independently repeated three times.

1. b. Detection of NO metabolites by ELISA

The cell lysate or serum samples were added to Griess reagent (Yishijiu (Lianyungang, Jiangsu) Biotechnology Co., Ltd.) at a ratio of 1:1 and reacted in the dark at room temperature for 10 min. In a 96-well plate, standards (0–100 μM sodium nitrite), samples, and chromogen A + B (50 μL each) were added successively, and the mixture was reacted in the dark at 25°C for 10 min. The OD value at 540 nm was measured using a microplate reader. The NO concentration (μM) was calculated according to the standard curve, and the experiment was independently repeated three times.

Quality control for NO release detection by the electrochemical method: The sensor calibration correlation coefficient was R² ≥ 0.995. The sample equilibration time before detection was ≥ 30 minutes. Each sample was tested in triplicate, and the mean of the peak values was taken as the final result. If the coefficient of variation among the three replicate measurements exceeded 15%, the sample was reanalyzed.

3.2.2. Mitochondrial function detection.

1. a. Mitochondrial oxygen consumption rate (OCR) and ATP production

Cardiomyocytes were seeded at a density of 8 × 10⁴ cells per well in the XF96 cell culture plate and cultured for 24 h at 37°C. The XF base medium (containing 10 mM glucose, 1 mM sodium pyruvate, 2 mM GlutaMAX, Catalog number 103575−100, Thermo Fisher Scientific, China) was used, and the cells were incubated at 37°C without CO₂ for 1 h. The XF96 Seahorse Analyzer was used for detection: Baseline measurement was performed three times → Oligomycin (1 μM, Catalog number O4876-5 mg, Sigma) was injected → Measurement was performed three times → FCCP (0.5 μM, Catalog number D1975, Sigma) was injected → Measurement was performed three times → Rotenone/Antimycin A (0.5 μM, Catalog Number DE0044/ S53076-1 mg, Lemeitan/Shanghai Yuanye Biotechnology Co., Ltd.) was injected → Measurement was performed three times. The Wave software was used for calculation: Basal respiration = Baseline OCR – OCR after rotenone; ATP production = OCR before oligomycin – OCR after oligomycin; Maximal respiration = OCR after FCCP – OCR after rotenone. Six replicate wells were set for each group, and the experiment was independently repeated three times.

1. b. Activity of respiratory chain complexes

Western blot detection was as follows: The abovementioned method was used. The primary antibodies were Complex I (NDUFB8, Catalog number ab251160, Abcam, 1:1000), Complex II (SDHB, Catalog number ZY756Mu015, HZbscience, 1:1000), and Complex V (ATP5A, Catalog number PA2000−1517, Aipudi, 1:1000). β-actin was used as the internal reference, and the relative expression levels were calculated.

The Clark oxygen electrode method was employed. Myocardial tissue homogenate was centrifuged at 600 g for 10 min (4°C) to obtain the supernatant, which was then centrifuged at 10000 g for 15 min (4°C) to obtain the mitochondrial pellet. The reaction system (1 mL) consisted of respiration buffer (225 mM mannitol, 75 mM sucrose, 10 mM KCl), 5 mM glutamate (substrate), and 0.5 mg mitochondrial protein. After incubation at 37°C for 5 min, 200 μM ADP was added to initiate state 3 respiration, and the OCR was recorded. State 4 respiration was defined as the OCR after ADP depletion. The respiratory control ratio (RCR) was calculated as state 3 oxygen consumption divided by state 4 oxygen consumption. The experiment was independently repeated three times.

3.2.3. Mitochondrial structural detection.

1. a. Observation by transmission electron microscope

Myocardial tissue (1 mm³) was taken and fixed in 2.5% glutaraldehyde (0.1 M PBS, pH 7.4) at 4°C for 24 h, followed by postfixation in 1% osmium tetroxide for 2 h. It was then dehydrated through an ethanol gradient (50% → 70% → 90% → 100%), embedded in epoxy resin (Epon 812), and polymerized at 60°C for 48 h. Ultrathin sections (70 nm) were double-stained with uranyl acetate and lead citrate and observed under a transmission electron microscope (model Tecnai G2 Spirit BioTWIN, FEI) at 20000 × magnification. Ten fields of view were randomly selected for each group and scored according to the Flameng criteria: 0 points (intact structure), 1 point (mild swelling, partial cristae disruption), 2 points (marked vacuolization), 3 points (complete loss of cristae). The results were expressed as the mean score.

1. b. Detection of mitochondrial membrane potential

Cells were seeded into a confocal dish and incubated with JC-1 staining solution (5 μg/mL, Catalog number C2006, Shanghai Beyotime Biotechnology Co., Ltd.) at 37°C for 20 min, followed by three washes with PBS. Observation under a fluorescence microscope (model IX73, Olympus): Normal mitochondrial membrane potential appeared as red fluorescence (Ex/Em = 585/590 nm), while depolarization appeared as green fluorescence (Ex/Em = 514/529 nm). Detection by flow cytometer (model Accuri C6, Thermo Fisher Scientific), the red/green fluorescence intensity ratio was calculated. Three replicate wells were set for each group, and the experiment was independently repeated three times.

1. c. Detection of mitochondrial permeability transition pore (mPTP) opening

Cells were incubated with Calcein-AM (1 μM) and CoCl₂ (1 mM) at 37°C for 30 min (CoCl₂ quenched cytoplasmic fluorescence), and mitochondrial Calcein fluorescence (Ex/Em = 488/515 nm) was observed under a confocal microscope (model LSM880, Zeiss). A decrease in fluorescence intensity indicated mPTP opening. Five cells were randomly selected from each group, and the fluorescence intensity was quantitatively analyzed by Image J. The experiment was independently repeated three times.

3.3.1. Detection at the cellular level.

Cardiomyocyte viability: Detection was performed adopting the WST-1/Cell Viability kit. Primary cardiomyocytes in a 96-well plate (1 × 10⁴ per well) were cultured (24 h). The medium was exchanged, WST-1 reagent (10 μL/well, Shanghai Shangbao Biotechnology Co., Ltd.) was applied, and incubation (light-protected) (37°C, 2 h) was carried out. The OD value at 450 nm (reference 630 nm) was determined. The viability calculation equation is as follows:

(1)

Control was set as follows: Blank group was treated with cell-free culture medium + WST-1; Positive control was treated with 0.1% Triton X-100 (100% cytotoxicity). Six replicate wells were set for each group, and the experiment was independently repeated three times.

Cardiomyocyte function was assessed. Cardiomyocytes were seeded on laminin-coated glass-bottom culture dishes (35 mm) and cultured for 48 h until synchronized contraction was achieved. The IonOptix system was preheated, and the sampling frequency (240 Hz) and electrical stimulation parameters (1 Hz, 5 ms pulse width) were set. Contraction function was assessed, with indicators including contraction amplitude (μm), shortening fraction (%) = (diastolic length – systolic length)/ diastolic length × 100, and maximum contraction/relaxation rates (±dL/dt, μm/s). The IonWizard software was adopted to automatically track cell edge movement, and the average of 10 consecutive cycles was computed. Five cells were examined for each group, and the experiment was independently repeated three times.

3.3.2. Organ-level detection (Langendorff perfusion system).

Mice were heparinized (1 U/g) and then anesthetized. The thorax was opened, and the heart was rapidly excised and placed in 4°C Krebs-Henseleit buffer [Yishijiu (Lianyungang, Jiangsu) Biotechnology Co., Ltd.. (KHB, containing 118 mM NaCl, 4.7 mM KCl, 1.2 mM MgSO₄, 1.2 mM KH₂PO₄, 25 mM NaHCO₃, 11 mM glucose, 2.5 mM CaCl₂, pH 7.4). A perfusion cannula was inserted into the aortic arch, and the heart was perfused with oxygenated KHB at a constant pressure of 80 mmHg and 37°C (95% O₂/5% CO₂). Functional parameters were then monitored by inserting a pressure transducer (AD Instruments) into the left ventricular apex. The indicators measured included left ventricular systolic pressure (LVSP, mmHg), left ventricular end-diastolic pressure (LVEDP, mmHg), left ventricular developed pressure (LVDP = LVSP – LVEDP), and maximum contraction/relaxation rates (±dp/dt_max, mmHg/s). Baseline data were recorded after stable perfusion for 20 min. The heart was then perfused with glucose-free KHB + insulin (15 mU/mL) (30 min), returning to normal KHB. The changes in each parameter compared to baseline were computed adopting LabChart software. Six mice were assigned to each group.