Approval for the study was granted by the Animal Welfare Ethical Review Board at the University of Bristol (ethical approval number UB/15/017). Procedures adhered to the Guide for the Care and Use of Laboratory Animals issued by the US National Institutes of Health (Guide, NRC 2011) and complied with the UK’s Animals Act of 1986 [11]. Adult male Wistar rats (14–16 weeks old; 250–260 g) were provided by Charles River Laboratories (Oxford, UK). The animals were acclimatised in a controlled environment (23 $\pm$ 1 °C, 60–70% humidity, 12-h light/dark cycle) with unrestricted access to water and standard rat chow for a duration of 4–7 days. Following acclimation, the rats were terminated through stunning followed by cervical dislocation, and the hearts were quickly harvested, rinsed in ice-cold perfusion buffer and retrograde perfused in Langendorff mode as per established protocols [10]. The hearts were randomly assigned to one of three groups: Control, perfusion-only (n = 6), or Ischemia/Reperfusion (I/R) (n = 7), where global ischemia was induced by aortic flow occlusion for 30 minutes, succeeded by a 10-minute reperfusion phase. An additional group underwent I/R and was treated with the receptor-interacting protein kinase 3 (RIP3) inhibitor GSK^′872 (250 nM) (S8465, Selleck Chemicals, Pittsburgh, PA, USA) throughout the reperfusion period (I/R+GSK^′872) (n = 7) [10]. After the experimental procedure, ventricular tissues were swiftly homogenised for mitochondria isolation. Left ventricular samples were also collected, frozen in liquid nitrogen and stored at –80 °C for later molecular analyses.

Heart tissues were homogenised in a lysis buffer comprising 200 mM Tris, 150 mM NaCl, 1 mM ethylenediaminetetraacetic acid (EDTA), 1 mM sodium orthovanadate (Na3VO4), 20 mM sodium fluoride (NaF), 0.5% Triton X-100 (pH 7.4) (Sigma-Aldrich, MO, USA), and a complete protease inhibitor cocktail (Roche, Mannheim, Germany). Protein precipitate was achieved overnight using 80% ice-cold acetone (Sigma-Aldrich, MO, USA). Following centrifugation, the pellets were solubilised in 8 M urea (Sigma-Aldrich, MO, USA). Protein concentrations were quantified using the Bio-Rad protein assay (Bio-Rad Laboratories GmbH, Colbe, Germany). A total of 100 µg of protein was reduced with 10 mM dithiothreitol (Sigma-Aldrich, MO, USA) in 100 mM ammonium bicarbonate (Sigma-Aldrich, MO, USA) at 37 °C for one hour. Alkylation was conducted using 15 mM iodoacetamide (Sigma-Aldrich, MO, USA) in 100 mM ammonium bicarbonate, protected from light for 30 minutes. Trypsin digestion (Promega, Wisconsin, USA) was performed at an enzyme-to-substrate ratio of 1:100 for overnight incubation at 37 °C. The resulting peptide mixtures (100 ng) were separated by Acquity M-Class Ultra-High-Performance Liquid Chromatography (Waters, Milford, MA, USA), utilising a nanoEase Symmetry C18 trap column (25 mm length, 180 µm diameter, 5 µm particle size) (Waters, Milford, MA, USA) for desalting before transferring to a nanoEase High Strength Silica T3 C18 analytical column (100 mm length, 75 µm diameter, 1.8 µm particle size). A 90-minute gradient of acetonitrile (5% to 35%) (Sigma-Aldrich, MO, USA) with 0.1% formic acid (Sigma-Aldrich, MO, USA) was applied at a flow rate of 300 nL/minute. The analysis was performed using a Synapt G2-Si quadrupole time-of-flight mass spectrometer with ion mobility (Waters, Milford, MA, USA). Data were captured in a data-independent fashion while maintaining a spectral acquisition rate of one second. Peak detection and processing were executed using Progenesis QI 4.0 software (Waters, Milford, MA, USA). Calibration and normalisation of peak intensities were based on median distribution [12].

The activity of lactate dehydrogenase (LDH) was assessed by analysing the effluent collected from both groups during the entirety of the 10-minute reperfusion [10]. Samples were mixed with a buffer containing 100 mM triethanolamine and 100 µM reduced beta-nicotinamide adenine dinucleotide ($\beta$-NADH), followed by the addition of 0.1 M sodium pyruvate. The absorbance at 340 nm (A340) was monitored using a spectrophotometer over 5 minutes at 37 °C.

The mitochondrial fraction was isolated in accordance with previously established protocols [10]. Ventricular tissue was placed in chilled sucrose buffer consisting of 300 mM sucrose, 10 mM Tris-HCl, 1 mM ethylene glycol bis (2-aminoethyl ether)-N,N,N^′,N^′-tetraacetic acid (EGTA) (pH = 7.2) (Sigma-Aldrich, MO, USA) and thoroughly homogenized. The resulting homogenate was quickly diluted with an isolation buffer containing 5 mg/mL bovine serum albumin (Sigma-Aldrich, MO, USA) and centrifuged at 2000 $\times$g for 2 minutes to eliminate cellular debris. The supernatant was then subjected to a second centrifugation at 10,000 $\times$g for 5 minutes to collect a crude particulate fraction, which was then re-suspended in the isolation buffer mentioned above. This mixture underwent further centrifugation at 10,000 $\times$g for another 5 minutes to obtain a pellet representing the mitochondrial fraction. This pellet was subsequently re-suspended in the isolation buffer for the assessment of mitochondrial swelling. This measurement was conducted at 37 °C under de-energized conditions by observing the reduction in light scattering, quantified as A520, in line with prior methodologies [10]. Mitochondrial swelling was induced by introducing 1 mM of calcium chloride (CaCl₂), resulting in a free buffered [Ca²⁺] of 80 µM [13]. This phenomenon is associated with the mitochondrial permeability transition pore opening, which can be inhibited by cyclosporin A [14, 15].

For immunoblotting assessment, left ventricular tissues were prepared using sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and Western blotting techniques as previously documented [10]. Briefly, after electrophoresis, proteins were transferred onto polyvinylidene fluoride (PVDF) membranes (Immobilon-P, Merck Millipore) and incubated with various primary antibodies, including apoptosis-inducing factor (AIF) (#4642, Cell Signaling Technology, Danvers, MA, USA), AIM2 (ab119791, Abcam, Cambridge, UK), caspase-1 (ab179515, Abcam, Cambridge, UK), Caspase-11 (#14340, Cell Signaling Technology, Danvers, MA, USA), Caspase-3 (#9662, Cell Signaling Technology, Danvers, MA, USA), Caspase-8 (#4790, Cell Signaling Technology, Danvers, MA, USA), CypD (# 78247, Cell Signaling Technology, Danvers, MA, USA), GSDMC (ab225635, Abcam, Cambridge, UK), GSDMD (#46451, Cell Signaling Technology, Danvers, MA, USA), GSDME (#88874, Cell Signaling Technology, Danvers, MA, USA), IL-1$\beta$ (ab9722, Abcam, Cambridge, UK), IL-33 (ab187060, Abcam, Cambridge, UK), inducible nitric oxide synthase (iNOS) (610432, BD Biosciences, Milpitas, CA, USA), NLRC4 (ab201792, Abcam, Cambridge, UK), NLRP3 (ab263899, Abcam, Cambridge, UK), PARP1 p25 (ab32064, Abcam, Cambridge, UK) and tumour necrosis factor (TNF) (ab66579, Abcam, Cambridge, UK). Following this, the membranes were incubated with Horseradish Peroxidase (HRP)-conjugated secondary antibodies: donkey anti-rabbit IgG (711-035-152, Jackson ImmunoResearch, West Grove, PA, USA), donkey anti-mouse IgG (115-035-174, Jackson ImmunoResearch, West Grove, PA, USA), donkey anti-rat IgG (112-035-175, Jackson ImmunoResearch, West Grove, PA, USA) and mouse anti-rat light chain specific IgG-HRP (112-035-175, Jackson ImmunoResearch, West Grove, PA, USA). Protein nitration was evaluated directly (35 µg protein/lane) without additional processing steps using a specific anti-NO2Tyr antibody (#9691, Cell Signaling Technology, Danvers, MA, USA) on a PVDF membrane. Whole-lane signal positivity was then used to calculate the relative expression of modified protein residues vs. the control group. Signals were detected using enhanced chemiluminescence (Crescendo Luminata, Merck Millipore, Burlington, MA, USA) and captured by a chemiluminescence imaging system (myECL imager, Thermo Scientific, Waltham, MA, USA). The positivity of signals across the whole lane was utilised to calculate the relative expression of modified protein residues compared to the control group. Detection of signals was accomplished through enhanced chemiluminescence (Crescendo Luminata, Merck Millipore, Burlington, MA, USA) captured by a chemiluminescence imaging system (myECL imager, Thermo Scientific, Waltham, MA, USA). Total protein staining with Ponceau S, assessed via scanning densitometry, served as the loading control for total tissue lysates [16]. The relative expression of target protein bands of interest was determined by adjusting the intensity against the overall lane protein staining.

Data are reported as means $\pm$ standard error of means (SEM) corresponding to the number of animals per group, while potential outliers were identified and removed from the statistical analyses by utilising Grubbs’ test. To evaluate differences between the groups with normally distributed variables, a Student’s t-test or one-way analysis of variance (ANOVA) with Tukey’s post-hoc test was employed. All statistical evaluations were conducted using GraphPad Prism 10.0 for Windows (GraphPad Software, San Diego, CA, USA). Differences between groups were considered significant when p 0.05.

In the case of the proteomic analysis, significance (p 0.01) (Fig. 1) was identified using a two-sided t-test. A heatmap displaying the top 50 differentially expressed proteins was generated utilising the tidyheatmaps package in R (https://www.r-project.org, v 4.3.1) [12].

Fig. 1.

Liquid chromatography with mass spectrometry (LC-MS/MS) proteome analysis. (A) Schematic representation of proteomic workflow; (B) Volcano plots comparing protein levels between C and I/R groups (significantly differentially expressed proteins were defined with $|$fold-change$|$ $>$1.5 and p 0.01); (C) Heatmap illustrating the differentially expressed proteins enriched in C and I/R groups. C, control group; I/R, ischemia/reperfusion group. Created with BioRender.com.

In the first 10 minutes of reperfusion, the LDH activity in the effluent was significantly increased, indicating a necrosis-like phenotype of cardiac damage. Likewise, the maximal extent of mitochondrial swelling as a marker of mitochondrial damage (sensitivity to Ca²⁺) was higher due to I/R. However, RIP3 inhibition significantly decreased LDH release and inhibited I/R-induced mitochondrial swelling (Table 1).

A liquid chromatography coupled with mass spectrometry (LC-MS/MS)-based untargeted proteomics approach identified 160 significantly (p 0.01) altered protein levels between the I/R and control (C) groups. More specifically, 12 proteins were significantly upregulated and 148 downregulated due to 10-minute reperfusion of previously ischemic hearts. Subsequently, group GO analysis was performed and the 50 most significantly altered proteins were categorised into a heatmap based on their involvement in a specific biological process (Fig. 1), such as cell death (GO:0008219), cellular component organisation or biogenesis (GO:0071840), cellular metabolic process (GO:0044237), negative regulation of protein sumoylation (GO:0033234), nitrogen compound metabolic process (GO:0006807), positive regulation of cellular process (GO:0048522), protein binding (GO:0005515), and response to stress (GO:0006950) (Fig. 1). The biological functions and biological action of the most changed proteins being related to cell death are indicated in Table 2 (Ref. [17]). Of note, none of this protein is a conventional protein of the signaling pathway of either of necrosis-like cell death modes.

While this study provides novel insights into I/R mediated cardiac damage and likely RIP3 involvement in the underlying mechanisms, several limitations should be acknowledged. First, as an ex vivo investigation, it captures an ultra-acute state, which may not fully mimic the complexity of in vivo conditions or encompass the broader spectrum of clinical scenarios. Additionally, the findings are predominantly derived from SDS-PAGE/Western Blot analyses, which, while informative, do not enable precise localisation or identification of the specific cardiac cell types affected.

Another limitation is the use of a single dose of the RIP3 inhibitor (GSK^′872). Moreover, the study would have benefited from complementary approaches, such as employing genetic knockout models, to provide more comprehensive mechanistic insights. These limitations underscore the need for further research to validate and expand upon the current findings.