S-nitrosoglutathione preserves vasodilation and attenuates myocardial ischemia-reperfusion injury

Abstract

Ischemia/reperfusion (I/R) injury exacerbates myocardial damage following acute myocardial infarction, underscoring the need for effective cardioprotective therapies. S-nitrosoglutathione (GSNO), an endogenous S-nitrosothiol, functions as a nitric oxide (NO) reservoir with vasodilatory and cytoprotective properties; however, its therapeutic potential in myocardial I/R injury remains incompletely characterized. We hypothesized that GSNO protects against I/R-induced cardiac injury by improving coronary vasodilation. The effects of GSNO were evaluated in an isolated Wistar rat heart model of regional ischemia followed by reperfusion using the Langendorff system. Reperfusion with GSNO-containing buffer (200 µM and 500 µM) enhanced coronary perfusion; however, only GSNO at 200 µM significantly reduced infarct size. GSNO also induced concentration-dependent relaxation in isolated coronary artery rings, confirming its potent vasodilatory activity. In vitro assays exposing cardiomyoblasts and endothelial cells to increasing GSNO concentrations revealed greater susceptibility of cardiomyoblasts to GSNO-induced cytotoxicity. Collectively, these findings demonstrate a dose-dependent cardioprotective effect of GSNO, likely mediated by sustained NO bioavailability and enhanced coronary vasodilation. This study supports GSNO as a promising NO-based adjunct therapy for mitigating myocardial I/R injury and provides important insights for optimizing its dosing strategy toward clinical translation.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-45498-x.

Introduction

Ischemic heart disease remains one of the leading causes of morbidity and mortality worldwide, with myocardial infarction (MI) representing its most severe clinical manifestation¹. Although timely reperfusion of the occluded coronary artery is essential to restore oxygen supply and limit infarct size, the abrupt restoration of blood flow triggers further cellular and molecular damage, a phenomenon known as ischemia/reperfusion (I/R) injury^2,3. This process is characterized by oxidative and nitrosative stress, calcium overload, mitochondrial dysfunction, and activation of inflammatory cascades, ultimately leading to cardiomyocyte death and adverse ventricular remodeling^4,5. Thus, developing adjunct therapeutic strategies to mitigate I/R injury remains a major goal in cardiovascular research.

Nitric oxide (NO) plays a fundamental role in cardiovascular homeostasis, regulating vascular tone, platelet aggregation, endothelial barrier function, and redox balance. Endothelial NO synthase (eNOS) derived NO maintains vasodilation through activation of soluble guanylate cyclase (sGC) and cyclic guanosine monophosphate (cGMP) signaling, while exerting anti-inflammatory and anti-apoptotic effects^6–9. Impaired NO bioavailability resulting from endothelial dysfunction, oxidative stress, or eNOS uncoupling contributes to vascular inflammation and the pathogenesis of atherosclerosis, hypertension, and ischemic heart disease. In the context of myocardial I/R, reduced NO signaling exacerbates vasoconstriction and tissue injury, whereas restoration of NO bioactivity improves coronary perfusion and limits infarct size^10,11.

Beyond its canonical cGMP-dependent pathway, NO modulates cardiac and vascular function via S-nitrosylation as S-nitrosothiols (RSNOs) regulate protein activity and mitochondrial stability. Enhancing RSNOs availability through pharmacological or genetic means confers cardioprotection by limiting reactive oxygen species (ROS) generation and preserving mitochondrial function^12,13. Experimental evidence indicates that increasing RSNOs availability either through ischemic preconditioning, inhibition of S-nitrosoglutathione reductase, or exogenous supplementation, attenuates tissue injury by maintaining mitochondrial integrity, a phenomenon also demonstrated in non-cardiac models using S-nitroso-N-acetylcysteine (SNAC)^14–17.

S-nitrosoglutathione (GSNO) is an endogenous low-molar mass RSNO and serves as a stable reservoir and carrier of NO bioactivity capable of releasing NO or trans-nitrosylating protein thiols. Compared to traditional organic nitrates, GSNO provides a more sustained and physiologically regulated source of NO, exhibiting vasodilatory, anti-platelet, and anti-inflammatory properties. Preclinical studies have shown that GSNO induces dose-dependent vasorelaxation in both conduit and resistance vessels, supporting its potential to preserve endothelial function and coronary flow under ischemic stress^18–21. Despite its promising pharmacological profile, the therapeutic potential of GSNO in acute MI remains underexplored. The effects of NO are highly dose-dependent: while physiological levels are cytoprotective, excessive NO can react with superoxide to form peroxynitrite, exacerbating oxidative injury^22,23. Hence, understanding the concentration-dependent effects of GSNO during cardiac I/R is essential for defining its therapeutic window.

In this study, we investigated the cardioprotective effects of GSNO in an isolated rat heart model of regional ischemia followed by reperfusion. GSNO was synthesized, characterized for its chemical stability and NO release kinetics, and administered at two concentrations (200 µM and 500 µM) during the I/R protocol. Functional outcomes were assessed through infarct size quantification, and coronary flow, resistance and vasodilatory response in isolated artery. We hypothesized that GSNO, at an optimal dose, would preserve coronary vasodilation and attenuate myocardial I/R injury by sustaining NO bioavailability. Our findings provide new insights into the dose-dependent actions of GSNO and reinforce its translational potential as a NO-based therapeutic for myocardial protection.

Results

Real time NO release from aqueous GSNO solution

The synthesized GSNO was confirmed by its two characteristic UV-Vis absorption bands (Figure S1). A weak band at 545 nm (n → π* transition, ε = 15.9 M⁻¹ cm⁻¹) appears on the tail of a stronger band at 336 nm (π → π* transition, ε = 922 M⁻¹ cm⁻¹)^24,25. Because the S-N bond in RSNOs has partial double-bond character, GSNO exists as two geometric isomers. The 545 nm band corresponds to the anti conformer, while a shoulder near 515 nm represents the syn conformer, which overlaps with the main 545 nm peak²⁶.

Real time NO release profiles for 200 µM and 500 µM GSNO in Krebs-Henseleit (KH) buffer (Fig. 1) show an immediate burst of NO upon injection of the solutions into the NOA reaction chamber. Within approximately 5 to 15 min, these transient peaks decline and stabilize into a nearly constant NO flux that persists for the remaining 150 min of monitoring. The initial bursts reflect NO already released from GSNO during solution preparation and accumulated prior to injection; these amounts are minimal relative to the total GSNO content. As nitrogen gas flows through the system, this free NO is rapidly removed, and the subsequent steady-state signal corresponds to NO generated directly from GSNO decomposition in real time. This constant flux represents the NO levels to which the isolated hearts are exposed during reperfusion with GSNO in the Langendorff system (see below). Based on the integrated curves in Fig. 1, after the initial NO burst the 200 µM GSNO solution produces NO at a flux of 3.45 ± 0.001 nmol mL⁻¹ min⁻¹, whereas the 500 µM GSNO solution produces NO at a flux 2.4 times higher: 8.20 ± 0.006 nmol mL⁻¹ min⁻¹.

Fig. 1.

Real time NO release from KH buffer containing GSNO 200 µM and 500 µM measured by ozone-based chemiluminescence over 150 min at 37 °C. After the initial bursts, the constant NO release rate for the GSNO 200 µM solution was 3.45 nmol mL⁻¹ min⁻¹ whereas the GSNO 500 µM solution exhibited a rate of 8.20 nmol mL⁻¹ min⁻¹. The corresponding cumulative NO release curves are provided in Supplementary Material (Figure S2).

Effect of GSNO on coronary flow and resistance, and infarcted area following myocardial I/R

Figure 2a provides a representative schematic of the isolated heart preparation in the Langendorff system, in which hearts were subjected to 35 min of regional ischemia followed by 90 min of reperfusion. Coronary flow and resistance were monitored throughout the protocol, from the pre-ischemic baseline to the end of reperfusion. Both GSNO concentrations (200 µM and 500 µM) significantly increased coronary flow during pre-ischemia compared to control. This elevation persisted throughout the 35-min ischemic period and led to significantly higher flows at time points 60 min and 90 min (Fig. 2b).

Fig. 2.

(a) Diagram of the Langendorff perfusion system. (b-g) Coronary flow, coronary resistance, heart rate, left ventricular developed pressure (LVDP), maximum rate of left ventricular pressure rise (dP/dt_max), and maximum rate of left ventricular pressure decline (dP/dt_min) in isolated hearts from control and GSNO-treated groups (200 µM or 500 µM). Coronary flow is shown in mL min^[-1, and coronary resistance as perfusion pressure (mmHg) normalized to flow. (h) Quantification of the myocardial area at risk. (i) infarct size after ischemia-reperfusion, expressed as a percentage of the total area at risk. (j) Representative TTC-stained myocardial slices, with the area at risk (red) outlined in yellow and the infarcted region (lighter shade area) outlined in blue. Evans blue marks non-ischemic myocardium, TTC stains viable tissue within the area at risk, and the infarcted area remains lighter in color. Data are mean ± SEM (n = 7–10/group). Coronary and hemodynamic variables were analyzed by two-way ANOVA (* p < 0.05 for 200 µM vs. Control; ∞ p < 0.05 for 500 µM vs. Control). Area-at-risk and infarct size were analyzed by one-way ANOVA with Dunnett’s post hoc test; *p = 0.003 vs. Control.

In accordance with this result, hearts treated with 200 µM and 500 µM GSNO showed significantly lower coronary resistance throughout the 35 min ischemic period and at time points 60 min and 90 min (Fig. 2c). The concomitant reduction in coronary resistance at these time points further supports enhanced vascular relaxation and suggests attenuation of endothelial injury in GSNO-treated heart. Heart rate remained unchanged among groups across all experimental phases (Fig. 2d).

To comprehensively evaluate myocardial functional recovery following I/R, we analyzed left ventricular developed pressure (LVDP) and the maximum rates of ventricular pressure rise and decline (dP/dt_max and dP/dt_min), which represent indices of systolic and diastolic performance, respectively. Treatment with 200 µM GSNO, but not 500 µM, significantly improved functional recovery, maintaining higher values of contractile indices LVDP and dP/dt_max (Fig. 2e-f) and more negative dP/dt_min values (Fig. 2g) compared with the control group.

Following the functional assessments, tissue damage was evaluated. All groups exhibited a comparable area at risk, indicating that the magnitude of regional ischemia was consistent across treatments (Fig. 2h).

Quantitative analysis revealed that infarct size reached approximately 65% in the Control group but was reduced to ~ 40% in hearts treated with 200 µM GSNO, whereas 500 µM GSNO did not confer significant cardioprotection (Fig. 2i-j). Representative TTC-stained ventricular slices (Fig. 2j) illustrate the area at risk in red (outlined in yellow) and the infarcted region in white (outlined in blue). The full set of TTC-stained slices used for quantification, comprising the five transverse sections from apex to base, is presented in the Supplementary Figure (Figure S3).

Dose-dependent effect of GSNO on cardiomyoblasts and endothelial cells viability

A progressive decrease in cell viability was observed for both cardiomyoblasts and endothelial cells exposed to increasing GSNO concentration (Fig. 3a-b). However, while in cardiomyoblasts cell viability was significantly reduced from 78 µM GSNO (Fig. 3a), this significance was observed only from 312 µM GSNO in endothelial cells (Fig. 3b). This differential response indicates that endothelial cells display greater resistance to the cytotoxic effects of increasing concentration of GSNO than cardiomyoblasts.

Fig. 3.

Cell viability of (a) H9c2 cardiomyoblasts and (b) HUVEC endothelial cells after 4 h exposure to increasing concentrations of GSNO or to culture medium (control, without GSNO). Viability was assessed using the CCK-8 assay. Data are expressed as mean ± SEM (n = 3). One-way ANOVA followed by Dunnett’s multiple comparisons test: *p < 0.05 ***p < 0.0009, ****p < 0.0001 vs. Control.

Dose-dependent vasorelaxant effect of GSNO in coronary artery rings

Figure 4a shows a representative schematic of the isolation of the septal branch of the left coronary artery and its mounting as a vascular ring in a wire myograph, in which a force transducer records changes in isometric tension over time (Fig. 4b). Under these conditions, GSNO induced a clear dose-dependent relaxation in coronary arterial rings (Fig. 4b-c). 

Fig. 4.

(a) Illustrative schematic depicting the isolation of the coronary artery and the preparation of the vascular ring for mounting in a wire myograph. (b) Representative traces showing the responses of coronary artery rings to cumulative concentrations of GSNO (1 nM-30 µM). (c) Concentration-response curves to GSNO in coronary artery rings. The pEC₅₀ value (negative logarithm of the agonist concentration producing half maximal response) is indicated within the figure. (d) Concentration-response curves to GSNO in coronary artery rings in the absence (Basal) or presence of the superoxide scavenger TEMPOL (100 µM). Data are presented as mean ± SEM (n = 3–4/group). One-way ANOVA followed by Dunnett’s multiple comparisons test: *p < 0.05 vs. −9 Log M; Two-way ANOVA followed by Bonferroni’s post hoc test: #p < 0.05 vs. Basal.

A significant relaxation was first detected at 1 µM (~ 34%), progressing to 99.8% at 10 µM and reaching complete (100%) relaxation at 30 µM GSNO. The calculated pEC₅₀ was 5.92, indicating that a concentration of approximately 1.2 µM (EC₅₀) elicited 50% relaxation in response to GSNO (Fig. 4c). Prior to dose–response experiments, tissue viability and experimental integrity were confirmed by assessing smooth muscle contractility to KCl and endothelial function through acetylcholine-induced relaxation (Figure S4). GSNO induced vasorelaxation was also evaluated in aortic rings (Figure S5), which similarly exhibited complete (100%) relaxation at 30 µM GSNO. Together, these findings demonstrate that GSNO is a potent vasodilator in both conductance (aortic) and resistance (coronary) vessels.

To investigate the role of basal superoxide in GSNO-induced relaxation, vascular rings were incubated with the superoxide scavenger TEMPOL. In coronary artery rings, pre-treatment with TEMPOL significantly attenuated the vasorelaxation induced by GSNO (Fig. 4d). A similar inhibitory effect of TEMPOL on GSNO relaxation was observed in aortic rings (Supplementary Figure S5f). These findings indicate that the reliance on basal superoxide for GSNO-mediated vasorelaxation is present in both conduit and resistance vessels.

Discussion

RSNOs are widely recognized as intermediates in NO signaling. GSNO, formed by S-nitrosation of glutathione (GSH), the most abundant intracellular thiol peptide, has been detected in biological fluids at concentrations ranging from 0.25 to 4.0 µM²⁷. Previous work from our group demonstrated that intravenous administration of another primary RSNO, S-nitroso-N-acetylcysteine (SNAC), produces marked systemic vasodilation in normotensive and hypertensive Wistar rats²⁸, consistent with findings reported by others^29,30. Similarly, we have shown that GSNO loaded hydrogels induce localized dermal vasodilation when applied to intact human skin, as well as to healthy and diabetic rat skin^31,32. The vasodilator action of GSNO in these cases can be assigned to its spontaneous thermal decomposition via bimolecular dimerization:

1

a reaction we have previously characterized in aqueous solutions and polymeric systems^33–36.

In earlier work, we showed that GSNO solutions containing EDTA and kept in the dark are remarkably stable; for example, a 1000 µM GSNO solution at pH 7.0 and 37 °C has a half-life of approximately 54 days. Moreover, photodecomposition under typical laboratory lighting has minimal influence on the NO release kinetics of dilute GSNO solutions³⁷. In the present study, we monitored the decomposition of the 200 µM and 500 µM GSNO solutions at 37 °C under ambient laboratory light for three hours, longer than the total Langendorff reperfusion time. These measurements showed that GSNO decomposition under these conditions remains below 10% (Figure S6). Therefore, during the reperfusion experiments, the heart vasculature of the rats was exposed both to intact GSNO and to constant NO fluxes of 3.45 ± 0.001 nmol mL⁻¹ min⁻¹ and 8.20 ± 0.006 nmol mL⁻¹ min⁻¹, arising from GSNO decomposition in the 200 µM and 500 µM reperfusion solutions, respectively.

The principal finding of this study is that GSNO in a dose-dependent manner significantly reduces infarct size following regional I/R in the isolated rat heart, accompanied by improved coronary perfusion. This cardioprotective effect was not observed at the higher concentration used in the study, despite increased coronary flow, indicating a dose-dependent therapeutic window for GSNO efficacy. In vitro assays revealed that GSNO at approximately 200 µM was cytotoxic to cardiomyoblasts but did not affect the viability of HUVECs, highlighting cell type-specific responses. Furthermore, GSNO induced complete relaxation of coronary artery rings at 30 µM, indicating that this is the threshold concentration for maximum vasodilatory effect of GSNO in this vessel.

NO, whether produced endogenously or released from donors such as GSNO, plays crucial roles in vascular and cardiac homeostasis. In endothelial cells, NO is synthesized primarily by eNOS and diffuses into adjacent vascular smooth muscle cells, where it activates sGC, leading to increased cGMP levels and subsequent vasodilation. Additionally, endothelium-derived NO inhibits platelet adhesion, leukocyte infiltration, and oxidative stress, thereby contributing to vascular homeostasis^8,9. In cardiomyocytes, NO plays a key regulatory role in intracellular calcium handling, mitochondrial respiration, and redox balance^38–40. It exerts protective effects primarily through sGC-cGMP-PKG signaling and via S-nitrosylation of target proteins during I/R injury^38,39. However, when present at supraphysiological concentrations, NO or RSNOs may induce nitrosative stress, mitochondrial dysfunction, and cell death^29,41,42. In agreement with this, Horton et al. (2000) showed that high concentrations (1.5 mM) of the NO donors S-nitroso-N-acetyl-penicillamine (SNAP) and (Z)−1-[N-93-ammonio-propyl-N-(n-propyl)amino] diazen-1-ium-1,2 diolate (PAPA/NO) exacerbated tumor necrosis factor alpha (TNF-α) induced cardiac dysfunction and cellular injury in both isolated rat hearts (Langendorff model) and cultured cardiomyocytes⁴³. While low concentrations of these NO donors (≤ 300 µM) conferred cardioprotection, higher doses impaired contractility, increased creatine kinase release, reduced cell viability, and disrupted calcium homeostasis⁴³. These findings highlight the concentration-dependent dual role of NO in the heart: at physiological levels, it supports myocardial function, but in excess, it promotes cellular injury likely via mitochondrial impairment and oxidative/nitrosative damage.

Our findings reflect this duality. While 200 µM GSNO reduced infarct size in the perfused heart, it was cytotoxic to H9c2 cardiomyoblasts in vitro. This apparent discrepancy may stem from contextual differences between experimental models. Indeed, previous studies exploring mitochondrial S-nitrosylation and cellular signaling have demonstrated that significantly different, and often much lower, concentrations of NO donors can be biologically active depending on the specific cellular or subcellular system employed^44,45. However, in the whole-heart Langendorff model, the high-flow continuous perfusion requires higher concentration gradients to achieve effective myocardial tissue penetration. Furthermore, isolated H9c2 cardiomyoblasts represent a simplified in vitro system lacking perfusion and complex antioxidant support, whereas endothelial cells such as HUVECs possess a well-described antioxidant machinery that may confer tolerance to moderate GSNO exposure⁴⁶.

The choice of 200 µM GSNO in the Langendorff system was based on previous studies demonstrating its biological activity in isolated heart preparations. Konorev et al. (1996) employed this concentration of GSNO in isolated rat hearts subjected to cardioplegic ischemia and showed that GSNO promoted NO-dependent nitrosylation reactions, as evidenced by the formation of nitrosylmyoglobin detected by electron spin resonance⁴⁷. These findings indicate that GSNO at 200 µM effectively delivers bioactive NO under low-oxygen conditions relevant to global ischemia. The interaction of NO with intracellular targets, including myoglobin, is particularly relevant in this context, as NO binding to deoxymyoglobin occurs under hypoxic conditions and may modulate myocardial oxygen handling and cellular redox signaling. In the present study, retrograde perfusion with 200 µM GSNO significantly reduced infarct size. Moreover, GSNO at this concentration was non-toxic to HUVECs, the first cells exposed to GSNO-derived NO during retrograde perfusion, supporting its safety and endothelial tolerance in this experimental setting.

The second higher concentration of 500 µM GSNO was chosen not to optimize efficacy but to prove a possible therapeutic window. The fact that GSNO 500 µM led to an increase in coronary flow and a consequent decrease in coronary resistance, statistically equal to those observed with GSNO 200 µM, shows that retrograde perfusion with GSNO 200 µM already produced the maximum possible vasodilation. Thus, these parameters remain unchanged with perfusion with GSNO solutions at higher concentrations. Crucially, the cardioprotection afforded by 200 µM GSNO extended to myocardial hemodynamics, significantly improving LVDP, dP/dt_max, and dP/dt_min. In contrast, despite maximizing coronary vasodilation, 500 µM GSNO failed to improve these functional indices. This functional failure aligns with the infarct size data, reinforcing that supraphysiological NO release may exacerbate contractile dysfunction via severe nitrosative stress, thus confirming the narrow therapeutic window. Consistent with this, the observation of reduced viability of H9c2 cardiomyoblasts and HUVEC endothelial cells to values below 80% when incubated with GSNO concentrations above 75 μm and 156 μm, respectively, indicates that the cytotoxic effect of high GSNO concentrations, such as 500 μm, may prevail over the beneficial effect of cardioprotection in I/R. These results support the concept of a narrow therapeutic window for RSNO-based interventions²⁹. Consistent with its role as an NO donor, GSNO induced full vasorelaxation in septal coronary artery at 30 µM. This vasodilatory response observed in the isolated artery indicates a direct effect of GSNO.

Mechanistically, GSNO likely operates through several complementary pathways. First, by acting as a NO donor or reservoir, GSNO may facilitate S-nitrosylation of mitochondrial proteins, thereby preserving mitochondrial function and suppressing ROS generation. This is supported by studies showing that inhibition of S-nitrosoglutathione reductase (GSNO-R) increases cardiac SNO levels and protects against I/R injury by reducing oxidative damage and infarct size^38,48. Second, GSNO may help maintain or restore endothelium derived NO signaling, preserving vasorelaxation, minimizing reperfusion-associated vasoconstriction, and supporting coronary perfusion. Indeed, GSNO has been shown to induce persistent vasodilation through NO-dependent pathways, even in endothelial denuded vessels, indicating its ability to serve as a depot for releasable NO or related species⁴⁹. Third, GSNO’s actions likely involve downstream cGMP signaling and inhibition of oxidative stress pathways, as demonstrated in isolated heart models, where GSNO improved post-ischemic recovery via enhanced cGMP formation and coronary flow⁴⁷. Additionally, GSNO may attenuate neutrophil infiltration and inflammatory injury during reperfusion, consistent with the broader literature on NO related cardioprotective mechanisms⁵⁰. It has already been shown that the perfusion of isolated rat heart subjected to I/R with GSNO solution accelerates the recovery of its normal contractile function⁵¹. However, it is well established that intact GSNO administered extracellularly is uncapable of crossing cell membranes to exert intracellular actions⁵². Instead, GSNO is thought to react with extracellular cysteine, forming S-nitrosocysteine (CysNO), which is transported into cells through L-type amino acid transporters (LAT)^52,53. Once inside the cell, CysNO can transfer its nitrosyl group to target proteins, generating high levels of S-nitrosated intracellular proteins and modulating their activity. These post-translational modifications occur independently of NO release and are therefore independent of sGC signaling⁵⁴. While circulating GSNO cannot cross endothelial or smooth muscle cell membranes within cardiac tissue, free NO produced endogenously or released from exogenous donors can readily diffuse across biological membranes and act in autocrine and paracrine signaling^55,56. Therefore, the increased coronary flow and reduced coronary resistance observed during reperfusion with 200 µM and 500 µM GSNO (Fig. 2), as well as the vasorelaxation responses of coronary rings (Fig. 4) and aortic rings (Figure S4), are attributable to NO diffusion from the GSNO containing solutions through the endothelium and into vascular smooth muscle cells (SMCs) as scheduled in Fig. 5.

Fig. 5.

Schematic diagram illustrating free NO generation from GSNO through its dimerization reaction, which produces oxidized glutathione (GS–SG), during the flow of GSNO solution through the coronary vasculature of Wistar rat hearts in the retrograde Langendorff reperfusion system. The released NO diffuses across the endothelial cells (EC) and enters vascular smooth muscle cells (SMC), where it activates soluble guanylate cyclase (sGC). Activated sGC converts guanosine triphosphate (GTP) into cyclic guanosine monophosphate (cGMP), initiating the signaling cascade that produces SMC relaxation and vasodilation, attenuating ischemia-reperfusion injury.

To further elucidate the vasodilatory mechanisms of GSNO, we investigated the role of basal superoxide in our vascular relaxation. Pre-incubation with the superoxide scavenger TEMPOL significantly attenuated GSNO-induced vasorelaxation in both coronary and aortic rings. This indicates that superoxide contributes to the optimal release of NO from GSNO, likely by acting as a reducing agent for transition metals (such as copper) that actively catalyze GSNO decomposition⁵⁷. Extrapolating this mechanism to the I/R scenario, we have that the early phase of reperfusion is characterized by an oxidative burst, leading to high tissue levels of superoxide⁵⁸. We postulate that this highly oxidative environment accelerates the decomposition of GSNO. At the cardioprotective concentration of 200 µM, this interaction is highly beneficial, ensuring a rapid and targeted delivery of bioactive NO exactly during the critical reperfusion window, thereby promoting vasodilation and limiting myocardial injury. However, when the heart is perfused with the much higher concentration of 500 µM, the massive release of NO in an environment already overloaded with superoxide likely drives the rapid and excessive formation of cytotoxic peroxynitrite. This excessive nitrosative stress would overwhelm endogenous antioxidant defenses, explaining the lack of cardioprotection and aligning with the cytotoxicity observed in vitro at high concentrations. Thus, the interplay between local superoxide levels and the concentration of the NO donor dictates whether the pharmacological outcome is protective or detrimental.

In summary, GSNO in a dose-dependent manner reduced infarct size and improved coronary perfusion in isolated rat hearts undergoing regional I/R. The concentration-dependent effects, cell-specific responses, and vascular bed dependent vasorelaxation suggest that GSNO’s cardioprotective potential is highly context-dependent. These findings support the development of GSNO as a potential adjunct pharmacological agent or incorporated into NO releasing biomaterials to mitigate I/R injury.

Materials & Methods

Materials

Unless otherwise stated, all reagents were purchased from Sigma-Aldrich, St. Louis, MO, USA. Analytical-grade water obtained from a Millipore Milli-Q Gradient filtration system (resistivity 18.2 MΩ cm at 25 °C) was used for all experiments. All reagents were used as received.

S-Nitrosoglutathione synthesis

S-nitrosoglutathione was synthesized through the S-nitrosation of glutathione (GSH) with sodium nitrite in acidic medium, according to the procedure described by Vercelino et al. (2013)⁵⁹. Briefly, 4.5 g of GSH and 1.0 g of NaNO₂ were dissolved in 23 mL of deionized water containing 1 mL of concentrated hydrochloric acid (37%, w/w) and ethylenediaminetetraacetic acid, EDTA (100 µM) added as a chelating agent. The solution was stirred continuously in an ice bath for 90 s, ensuring low temperature and protection from ambient light to prevent GSNO photodecomposition. GSNO formed was precipitated through the addition of 30 mL of cold acetone. The precipitate was collected by vacuum filtration, washed with 15 mL of cold acetone to remove residual salts and unreacted GSH, and then freeze-dried for 24 h. The resulting dry, reddish solid was stored at −20 °C in light-protected vials until use. Stock GSNO solution was diluted using phosphate-buffered saline containing 50 µM EDTA (pH 7.4).

Kinetic monitoring of GSNO decomposition in Krebs–Henseleit solution

The spectral evolution of KH-GSNO solutions (200 and 500 µM) was monitored over the 220–1100 nm range under ambient light for 3 h using a diode array spectrophotometer (HP 8453, Hewlett-Packard, Palo Alto, CA, USA), with air as the reference. GSNO decomposition kinetics were assessed by tracking the decrease in absorbance at 336 nm, characteristic of the S-N bond. Measurements were carried out in a quartz cuvette with a 1.0 cm optical path length, ensuring initial absorbance values within the optimal range for kinetic analysis.

The real-time NO release profile of KH-GSNO solutions was measured by ozone-based chemiluminescence using a nitric oxide analyzer (NOA Sievers 208i, GE Analytical Instruments). The instrument operated with an oxygen pressure of 6.2 psig and a cell pressure of 8.5 torr. For each analysis, 100 µL of KH-GSNO stock solutions (20 mM or 50 mM) were injected into the reaction chamber containing 10 mL of KH solution (pH 7.4), maintained at 37 °C and protected from light, resulting in final GSNO concentrations in the chamber of 200 µM and 500 µM, respectively. Measurements were performed continuously for 2.5 h to evaluate NO release kinetics.

Cultured cells

Murine cardiomyoblasts (H9c2, ATCC, CRL-1446) and primary human umbilical vein endothelial cells (HUVECs, ATCC, PCS-100-010) were evaluated in vitro. H9c2 were maintained in Dulbecco’s Modified Eagle Medium (DMEM, Gibco) and HUVECs in Roswell Park Memorial Institute (RPMI) 1640 medium (Gibco), both supplemented with 10% heat inactivated fetal bovine serum. Following centrifugation (1200 × g, 5 min), cell pellets were resuspended in fresh medium and seeded in tissue culture treated flasks at a density of 1–2 × 10⁵ cells/mL. Cultures were incubated at 37 °C in a humidified atmosphere of 5% CO₂ and ~ 95% relative humidity, with medium renewal every 48 h. Both cell lineages were passed at ~ 80% confluence using 0.25% trypsin-EDTA.

Cell viability assay

For the assay, 3.7 × 10⁴ cells were seeded in 250 µL of culture medium per well in 48 well plates and allowed to adhere for 24 h. Cells were then treated for 4 h with GSNO at concentrations of 20, 1.25, 0.625, 0.312, 0.156, 0.078, and 0.019 mM, freshly prepared in the respective culture media. Cell viability was tested by the Cell Counting Kit-8 (CCK-8, Dojindo) that is based on the reduction of the water-soluble tetrazolium salt WST-8 [2-(2-methoxy-4-nitrophenyl)−3-(4-nitrophenyl)−5-(2,4-disulfophenyl)−2 H-tetrazolium, monosodium salt] by cellular dehydrogenases, producing a water-soluble orange formazan dye. The intensity of the color formed is directly proportional to the number of viable cells.

After incubation with GSNO, 25 µL of CCK-8 reagent were added per well and incubated for 2 h at 37 °C. Absorbance was recorded at 450 nm using a microplate spectrophotometer (Epoch, BioTek), and results were expressed as the percentage of viable cells relative to the untreated control. Following the CCK-8 assay, H9c2 and HUVEC cells were washed with cold PBS, harvested, and lysed using RIPA buffer (RIPA Lysis and Extraction Buffer, Thermo Fisher Scientific). Total protein content in each sample was quantified using the BCA Protein Assay Kit (Thermo Fisher Scientific). To ensure higher accuracy and sensitivity, CCK-8 absorbance values were normalized to the total protein content obtained from each triplicate sample. This normalization minimizes variability related to cell seeding density and metabolic activity, improving the precision of the viability assessment.

Animals

A total of 37 male Wistar rats (8–10 weeks old, 250–350 g) were purchased from the Multidisciplinary Center for Biological Investigation (CEMIB), University of Campinas (UNICAMP), Campinas, SP, Brazil, and housed at the Multidisciplinary Animal Experimentation Unit (UMEA), Faculty of Medical Sciences (FCM), UNICAMP, in groups of three animals per cage with free access to standard feed and water. Temperature and light/dark cycles were automatically controlled at 22 ± 2 °C and 12 h, respectively. All procedures were approved by the Animal Use Ethics Committee of the University of Campinas (CEUA: 6388-1/2024) and conducted in accordance with the ethical principles of the Brazilian National Council for the Control of Animal Experimentation (CONCEA). Experiments were designed and reported in compliance with the ARRIVE guidelines 2.0. Wistar rats were selected due to their extensive use in cardiovascular research and the suitability of the rat heart for ex vivo perfusion in the Langendorff system. For isolated heart I/R experiments, animals were initially allocated at n = 10 per group based on an a priori G*Power calculation using infarct size as the primary outcome (effect size 25%, α = 0.05, power = 80%). Final group sizes were: control (n = 8), GSNO 200 µM (n = 7), and GSNO 500 µM (n = 7), due to predefined technical exclusions inherent to the Langendorff preparation, including cannulation failure, inadequate coronary perfusion, unstable baseline hemodynamics, or excessive anesthetic requirements. No exclusions were based on experimental outcomes. For vascular reactivity experiments, coronary artery and aortic ring preparations from three rats were used to assess GSNO-induced relaxation, with no exclusions.

Isolated heart perfusion

Rats were anesthetized with sodium thiopental (80 mg/kg, i.p.) and heparinized (2,500 IU/kg, i.p.) prior to heart excision. Following induction of anesthesia, animals were placed in the prone position for dorsal trichotomy. A bilateral thoracotomy was then performed to expose the thoracic cavity and allow rapid heart excision (within 3 min). The isolated hearts were mounted via the ascending aorta onto a perfusion apparatus (Radnoti LLC) and perfused with modified Krebs-Henseleit (KH) solution (in mM: 118 NaCl, 25 NaHCO₃, 11 D-glucose, 4.7 KCl, 1.22 MgSO₄·7 H₂O, 1.21 KH₂PO₄, and 1.4 CaCl₂·2 H₂O; all from Sigma, St. Louis, MO, USA), at 37^°C (pH = 7.4, gassed with 95% O₂ and 5% CO₂). Perfusion was performed retrogradely at a constant pressure of 70 mmHg, according to the classical Langendorff technique. Coronary flow was continuously measured using an in-line flow sensor placed in the perfusion line. Coronary vascular resistance (CVR) was calculated as the ratio between perfusion pressure and coronary flow (mmHg min mL⁻¹). After stabilization of cardiac contractions, a temperature probe was positioned in the pulmonary artery. Regional ischemia was induced by ligation of the left anterior descending coronary artery with a 4 − 0 cotton suture for 35 min, followed by 90 min of reperfusion after release of the ligature⁶⁰. After stabilization in KH for 5 min, hearts were treated with GSNO (200 or 500 µM) for 10 min during the pre-ischemia phase, which was continued throughout ischemia (35 min) and reperfusion (90 min). The concentration of 200 µM was selected based on previous ex vivo Langendorff studies establishing its biological relevance and effective NO delivery to the intact myocardium during ischemia⁴⁷. The higher concentration (500 µM) was included to evaluate the therapeutic window and potential dose-dependent cytotoxicity. Control hearts received KH solution without GSNO, following the same ischemia-reperfusion protocol. Hemodynamic data and coronary flow signals were continuously recorded as time-dependent curves. All parameters were acquired with the PowerLab^® system (ADInstruments, Colorado Springs, CO, USA) and analyzed using LabChart Pro^® software (ADInstruments)⁶¹.

Quantification of infarct area

At the end of the isolated heart perfusion protocol, the coronary artery was re-occluded, and 1% Evans Blue dye (Sigma-Aldrich, St. Louis, MO, USA) was perfused through the aorta to delineate the non-ischemic (non-risk) regions of the myocardium. The hearts were then frozen at −20 °C for approximately 30 min and sectioned transversely into 3 mm slices from apex to base. The slices were incubated in phosphate buffered saline containing 1% 2,3,5-triphenyltetrazolium chloride (TTC; Sigma-Aldrich, St. Louis, MO, USA) at 37 °C for 20 min to differentiate viable from necrotic tissue. Viable myocardium within the area at risk was stained red, whereas infarcted tissue remained pale (white). Slices were subsequently fixed in 10% neutral buffered formalin for 24 h to enhance contrast and tissue preservation.

The left ventricle was thus divided into three regions: (i) non-ischemic area (blue), (ii) viable tissue within the risk area (red), and (iii) infarcted area (white). The areas were quantified by planimetric analysis using ImageJ^® software (version 1.45; National Institutes of Health, USA). Infarct size quantification was conducted by an investigator blinded to group allocation and expressed as a percentage of the area at risk (infarct area/risk area × 100).

Vascular reactivity

Rats were euthanized with an overdose of isoflurane under deep anesthesia, and the thoracic aorta and interventricular septal branches of the left anterior descending coronary artery were carefully isolated and immediately placed in freshly prepared KH solution (in mM: 118 NaCl, 4.7 KCl, 25 NaHCO₃, 2.5 CaCl₂·2 H₂O, 1.2 KH₂PO₄, 1.2 MgSO₄·7 H₂O, 11 glucose, and 0.01 EDTA; all from Sigma-Aldrich, St. Louis, MO, USA). The arteries were cleaned of connective tissue and cut into transversal segments (2–3 mm).

Aortic segments were mounted under a resting tension (1.5 g) in an organ bath (Panlab Harvard Apparatus, Cornellà, Barcelona, Spain) and coronary arteries were mounted on a wire myograph (Danish Myo Technology, Model 610 M, Aarhus N, Denmark) using two 40 μm wires. Both arteries were maintained in KH solution (37 °C, pH = 7.4, gassed with 95% O₂ and 5% CO₂)⁶². Coronary arteries were stretched to optimal lumen diameter via internal circumference, wall‑length and wall‑tension normalization using the LabChart Pro‑DMT Normalization Module (ADInstruments). The internal circumference at L₁₀₀ was determined, and vessels were set to 0.9 × L₁₀₀⁶³.

Smooth muscle viability was assessed by contraction with KCl (80 mM for aorta; 60 mM for coronary) and endothelial integrity was confirmed by ≥ 50% relaxation to acetylcholine (10 µM) in arteries contracted with U46619 (concentration to achieve 70–80% of maximal KCl response). Rings exhibiting viability were washed with KH, pre-contracted with U46619, and then subjected to a first (basal) cumulative concentration–response curve to GSNO (1 nM-30 µM). After this initial curve, the vessels were thoroughly washed with KH to restore resting tension. To investigate the role of basal superoxide in NO release, the rings were then incubated with the superoxide scavenger TEMPOL (100 µM, Sigma-Aldrich, St. Louis, MO, USA) for 30 min. Following this incubation period, the vessels were pre-contracted again with U46619, and a second cumulative concentration-response curve to GSNO was performed. Relaxation responses to GSNO were expressed as a percentage of the initial U46619-induced contraction for each respective curve. Concentration-response data were fitted to a logistic equation, and nonlinear regression analysis was used to determine maximum response and logarithm of the concentration of agonist that produces a half-maximum response (pEC₅₀).

Statistical analysis

Data normality was assessed using the Shapiro-Wilk test. Time-course variables, including coronary flow, coronary resistance, and all hemodynamic parameters (dP/dt_max, dP/dt_min, heart rate, and LVDP), as well as comparisons between concentration-response curves in vascular reactivity assays (Basal vs. TEMPOL), were analyzed by two-way ANOVA followed by Bonferroni’s multiple comparisons test. Infarct size was analyzed by one-way ANOVA followed by Dunnett’s multiple comparisons test for normally distributed data, or by the Kruskal-Wallis test followed by Dunn’s post-hoc test for non-parametric data. Vascular reactivity was analyzed by one way ANOVA followed by Dunnett’s multiple comparisons test, comparing each GSNO concentration with the first concentration that did not induce relaxation. Cell viability was analyzed by one way ANOVA followed by Dunnett’s multiple comparisons test, using the GSNO-free group as control. All analyses were performed using GraphPad Prism 8.0 (GraphPad Software, La Jolla, CA, USA), and differences were considered statistically significant at p < 0.05.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

D.M.G. performed vascular reactivity assays, analyzed data, and drafted and revised the manuscript; E.A.P.J. performed nitric oxide release and UV-Vis assays, analyzed data, edited and co‑wrote the manuscript; E.I.L.G. conducted in vitro assays; A.P.D. revised the manuscript; A.C.S. edited the manuscript and contributed to conceptualization and supervision; M.G.O. secured funding, oversaw conceptualization and supervision, and edited the manuscript.

Funding

This work was funded by the São Paulo Research Foundation, FAPESP, grants 2022/14645-2 (to M.G.O.), 2025/00254-0 (to D.M.G), 2024/08075-4 (to E.A.P.J.) and, 2023/17022-9 and 2018/26080-4 (to A.P.D.).

Data availability

The data supporting the findings of this study are available from the Zenodo repository at [https://doi.org/10.5281/zenodo.18058045].

Declarations

Competing interests

The authors declare no competing interests.