Neuroprotective Potential of Nitroglycerin in Ischemic Stroke: Insights into Neural Glucose Metabolism and Endoplasmic Reticulum Stress Inhibition

Abstract

Background

Glyceryl trinitrate (GTN), also known as nitroglycerin, is predominantly recognized as a vasodilator for ischemic heart disease, and its potential neuroprotective properties in acute ischemic stroke remain under exploration. We sought to discover the therapeutic advantages and mechanisms of post‐recanalization GTN administration in acute ischemic stroke.

Methods and Results

A total of 118 male Sprague–Dawley rats were divided into groups: sham, transient/permanent middle cerebral artery occlusion (MCAO) with or without GTN treatment, and transient/permanent MCAO treated with both GTN and KT5823, an inhibitor of PKG. Acute ischemic stroke was induced by transient MCAO for 2 hours followed by 6 or 24 hours of reperfusion and permanent MCAO (28‐hour MCAO without reperfusion). The study assessed infarct volumes, neurological deficits, glucose metabolism metrics, NO, and cGMP levels via ELISA. mRNA and protein expression of key molecules of hyperglycolysis, gluconeogenesis, endoplasmic reticulum stress as well as signaling molecules (PKG, AMPK) were conducted via reverse transcription polymerase chain reaction and Western blotting, and cell death was assessed with TUNEL and ELISA. GTN significantly reduced cerebral infarct volumes, neurological deficits, and cell death only after transient MCAO. GTN led to a significant reduction in the expression of NO and cGMP levels, key glucose metabolism, endoplasmic reticulum stress‐related genes and proteins, and phosphorylated AMPK while boosting PKG expression, in transient MCAO but not permanent MCAO. The GTN‐induced reduction in glucose metabolites, lactate, and reactive oxygen species was exclusive to transient MCAO groups. Coadministration of GTN and PKG inhibitors reversed the observed GTN benefits.

Conclusions

GTN induced neuroprotection in transient MCAO by improving glucose metabolism and potentially controlling endoplasmic reticulum stress through the NO‐cGMP‐PKG signaling cascade to inhibit AMPK phosphorylation.

Nonstandard Abbreviations and Acronyms

AMPK

monophosphate‐activated protein kinase

GLUT1/3

glucose transporter 1/3

GTN

glyceryl trinitrate

PCK1/2

phosphoenolpyruvate carboxylase 1/2

PFK1

phosphofructokinase‐1

PKG

cyclic guanosine monophosphate (cGMP)‐protein kinase G

tMCAO/pMCAO

transient/permanent middle cerebral artery occlusion

Clinical Perspectives.

What Is New?

• Nitroglycerin exerts neuroprotective effects only upon reperfusion after acute ischemic stroke in rats.

• Nitroglycerin plays a crucial role in neuroprotection through neuronal mechanisms by inhibiting glucose metabolism and endoplasmic reticulum stress beyond the vasodilatory property.

What Are the Clinical Implications?

• Nitroglycerin, a Food and Drug Administration‐approved medication, is feasible for use in patients with acute ischemic stroke as an adjunctive neuroprotection therapy following reperfusion.

• When used in combination with mechanical thrombolysis, nitroglycerin presents promising potential to enhance treatment outcomes.

Acute ischemic stroke (AIS) remains a major cause of death and disability worldwide, despite significant advancements in revascularization techniques. ¹ The persistent high morbidity and mortality rates associated with AIS emphasize the urgent need for more effective therapeutic interventions to improve outcomes post‐revascularization. ²

Glyceryl trinitrate (GTN), also known as nitroglycerin, is primarily known as an organic NO donor and a vasodilatory agent and has been widely used to treat ischemic heart disease. Its therapeutic potential in ischemic stroke has recently garnered attention. ³ GTN, by affecting smooth muscle cells, helps lower blood pressure, increases arterial compliance, and reduces venous return, which is beneficial in AIS. ⁴ Clinical studies, including ENOS, RIGHT, RIGHT‐2, and MR ASAP, have shown that transdermal GTN patches moderately reduce peak systolic blood pressure (SBP), pulse pressure, and elevate heart rate, thus improving vascular oxygenation in AIS patients. However, these studies, ⁵ , ⁶ , ⁷ , ⁸ while confirming the safety of GTN, did not conclusively prove its efficacy in AIS. We hypothesize that the limited neuroprotection of GTN observed in these studies was attributable to its administration without concurrent revascularization, and that transdermal GTN patches may not have reached the optimal therapeutic levels required for neuroprotection in AIS. Our study aims to evaluate the administration of GTN in clinically relevant AIS models, both with and without reperfusion, to better understand its potential in this context.

AIS leads to a significant reduction in cerebral blood flow, triggering metabolic disturbances and cellular damage in the brain. ⁹ A key aspect of this pathology is the disruption of cerebral glucose metabolism relative to oxygen utilization, accompanied by intensified endoplasmic reticulum (ER) stress. ¹⁰ These changes result in an increased glycolytic flux and gluconeogenesis, leading to elevated lactate and reactive oxygen species (ROS) levels that are detrimental to the survival of cells in the ischemic penumbra. ¹¹ Thus, our study aimed to investigate whether GTN improves cerebral glucose metabolism and alleviates ER stress, thereby exerting a neuroprotective effect in stroke. By bridging the gap between the known metabolic dysregulation in stroke and the potential therapeutic effects of GTN, we seek to provide new insights into AIS management.

METHODS

This manuscript follows the Animal Research: Reporting of In‐Vivo Experiments reporting guideline and the data that support the findings of this study are available from the corresponding authors per request.

Subjects

The study used a total of 118 adult male Sprague–Dawley rats (280–300 g, Vital River Laboratory Animal Technology Co., Ltd., China). All animal experiments were approved by the Institutional Animal Investigation Committee at the Capital Medical University and were performed according to relevant guidelines and regulations. The rats were accommodated and cared for during research in a uniform animal facility, following a 12‐hour light/dark cycle. Pain mitigation measures and judicious animal use were implemented to their utmost efficacy. The subjects were randomly grouped: (1) A sham‐operated middle cerebral artery occlusion (MCAO) for RNA/protein analysis (n=5) and immunofluorescence (n=5); (2) 2 hours MCAO group followed by reperfusion of 6, 24, or 48 hours for RNA/protein analysis (n=6×3) and immunofluorescence staining at 24 hours (n=6); (3) 2 hours MCAO and GTN administration (injected intraperitoneally at the onset of reperfusion) followed by reperfusion at 6, 24, or 48 hours for RNA/protein analysis (n=6×3) and immunofluorescence at 24 hours (n=6); (4) 2 hours MCAO with both administration of GTN and PKG inhibitor KT5823, followed by reperfusion for 6, 24, or 48 hours (n=6×3); (5) permanent MCAO (pMCAO) for RNA/protein analysis (n=6) and immunofluorescence (n=6) at 28 hours; (6) pMCAO with GTN for RNA/protein analysis (n=6) and immunofluorescence at 28 hours (n=6); (7) pMCAO with both GTN and PKG inhibitor KT5823 for RNA/protein analysis (n=6). All procedures and data analyses were performed in a randomized and blinded fashion.

Model of MCAO

MCAO was performed using previously established procedures. ¹² The rats underwent a 2‐hour right‐sided MCAO with 4‐0 nylon suture at the origin of the middle cerebral artery. Reperfusion was achieved by removing the filament at the corresponding time. Subsequently, rats were euthanized at 6 or 24 hours following reperfusion. A permanent MCAO model was established through a 28‐hour right MCAO without reperfusion. The temperature in the rectum was consistently maintained within the range of 36.5°C to 37.5°C.

Administration of GTN and PKG Inhibitors

In ischemic rats, GTN in saline was administered intraperitoneally at onset of reperfusion, at dosages of 5 or 10 mg/kg. KT5823 (0.12 μg/μL), the PKG inhibitor, was injected in lateral ventricular space 1 hour before onset of reperfusion in transient MCAO (tMCAO) and 1 hour after onset of ischemia in pMCAO. A dose‐dependent analysis was performed in the cell culture model to determine the highest effective doses of these drugs.

Neurological Deficits and Infarct Volume

Neurological deficits in the rats were examined before surgery, up to 48 hours of reperfusion, or after 28 hours in the pMCAO group. MCA‐supplied territories (cortex and striatum) of the rats were dissected into 2‐mm‐thick slices using a brain matrix and stained with 2,3,5‐triphenyl tetrazolium chloride (TTC, Sigma, USA).

Determination of Lactate and ROS in Brain

Lactate and ROS levels were measured as described by the authors. ¹³ A multimode detector (BioTek‐Synergy HT, USA) was used to detect fluorescence, which emitted light at 595 nm when excited at 535 nm. This detection process was performed for 15 minutes at a temperature of 37 °C.

TUNEL Assay and Quantitative Analyses

The TUNEL assay was used to investigate DNA fragmentation. This assay was conducted using a commercially available kit (In situ Cell Death Detection Kit, Fluoresce, USA) following the protocol. ¹⁴ Positive TUNEL staining was detected using a fluorescence microscope (DM4000, Leica, Germany). The TUNEL+ cells were manually counted using the cell counting tool in ImageJ. The values obtained from all sections within each group were blindly collected for subsequent statistical analysis. Images were randomly acquired from 4 distinct regions, and the statistical outcomes were indicative of cell death rates.

Cell Death by ELISA

The presence of DNA fragments associated with histones in the cytoplasm of ischemic brain tissues was determined using a photometric enzyme immunoassay (Rat cell death [Fas] ELISA kit, Shanghai Enzyme‐linked Biotechnology Co., Ltd, China). Each measurement was performed twice. The absorbance at 450 nm was detected using a multimode detector (BioTek‐Synergy HT, USA) as the final step.

Reverse Transcription Polymerase Chain Reaction

A real‐time reverse transcription polymerase chain reaction technique was used in this study as described previously. ¹⁴ Gene expression quantification was performed using the Applied Biosystems Prism 7500 real‐time polymerase chain reaction system. The specific primers used for the genes are shown in Table 1.

Table 1.

Primer Sequence

Name of primer	Sequences (5′–3′)
GLUT1a	F: CGCCCTGTCCAGACACA R: CCTCCTAGCCATTGTTCAGT
GLUT3b	F: AACACCGGAGTCATCAATGC R: GCTGCGCTCTGTAGGATAGC
PFK1c	F: ATTGAATATGCCGTCTCC R: CACAAGATACACACATGG
LDHd	F: CCGTTACCTGATGGGAGAAA R: AGCAGGGTAGAGCTGTGGAA
AMPKe	F: TGTGACAAGCACATTTTCCAA R: CCGATCTCTGTGGAGTAGCAG
PKGf	F: AGTGGTTTGAGGGCTTTA R: ATGTCCCAGCCTGAGTTG
PCK1g	F: CTCTTTGCGACAGGCAAGGT R: CCCCAGGCCTTTCAAGTTCA
PCK2h	F: ACCATGCCGTAGCATCCAAA R: TGGGGAGCTTCCGGATAAGA
BiPi	F: GGGGACAAACATCAAGCAGT R: ATAAACCCCGATGAGGCTGT
ATF4j	F: CGGCAAGGAGGATGCCTTTT R: GAGAGCCCAGGTAGGACTCA
CHOPk	F: GAGAATGAAAGGAAAGTGGCAC R: ATTCACCATTCGGTCAATCAGA
GAPDHl	F: CAAGAAGGTGGTGAAGCAG R: AAAGGTGGAAGAATGGGAG

This manuscript was sent to Rebecca D. Levit, MD, Guest Editor, for review by expert referees, editorial decision, and final disposition.

^a Glucose Transporter 1

^b Glucose Transporter 3

^c Phosphofructokinase‐1Lactate

^d Dehydrogenase

^e Monophosphate‐ Activated Protein Kinase

^f Cyclic Guanosine Monophosphate (cGMP) ‐ Protein Kinase G

^g Phosphoenolpyruvate Carboxylase 1

^h Phosphoenolpyruvate Carboxylase 2

ⁱ Binding Immunoglobulin Protein

^j Activating Transcription Factor 4

^k C/EBP Homologous Protein

^l Glyceraldehyde 3‐Phosphate Dehydrogenase

Western Blot Assay

Protein expression levels were quantified using a Western blot assay at 6 and 24 hours of reperfusion after tMCAO and pMCAO as described previously by the authors. ¹³ Primary antibodies include rabbit anti‐GLUT1 (glucose transporter 1) (1:500; Novus biotechnology), rabbit anti‐GLUT3 (glucose transporter 3) (1:500; Novus biotechnology), rabbit anti‐phosphofructokinase‐1 (PFK1) (1:10000; Proteintech), rabbit anti‐LDH (1:1000; Proteintech), rabbit anti‐PRKG1 (1:1000; Proteintech), rabbit anti‐pAMPK (1:1000; Affinity), rabbit anti‐PCK1 (1:1000; Cell Signaling Technology (CST), USA), rabbit anti‐PCK2 (1:1000; CST), rabbit anti‐Bip (1:1000; CST), rabbit anti‐Phospho‐PERK (1:1000; CST), rabbit anti‐Phospho‐eIf2α (1:1000; CST), rabbit anti‐ATF4 (1:1000; CST) and rabbit anti‐CHOP (1:1000; CST). The secondary antibody used was goat anti‐rabbit IgG conjugated with horseradish peroxidase (Catalog number sc‐2004, Santa Cruz Biotechnology). Quantitative analysis of protein expression was conducted using ImageJ 1.42 software, where relative image density from Western blot images was measured.

Double Labeling Immunofluorescence of BiP

The immunofluorescence was performed as we previously described. ¹⁴ The fluorescence microscope (DM4000, Leica, Germany) and a Leica TCS‐SP5 confocal microscope (×1000, Leica, Tokyo, Japan) were used to examine the double‐stained images. Immunological analysis was performed using the hemispheres on the same side of the body. Ten random regions of interest from both the cortex and striatum were assessed on each slide.

Statistical Analysis

Statistical analyses were performed using Graphpad Prism version 8.0 (Graphpad Software, USA). Group differences were assessed via one‐way ANOVA followed by Bonferroni's multiple comparisons test for comparison between multiple groups (>2 groups). P<0.05 was considered statistically significant. Data are reported as mean±SD or mean±SE.

RESULTS

Physiological Parameters

There were no significant differences in blood MAP, pO₂, or pCO₂ (Table 2) between these groups.

Table 2.

Physiological Parameters

	Stroke	GTN	KT5823
MAP (mm Hg)
Before MCAO	86.9±3.1	85.4±3.5	85.3±3.7
Onset of reperfusion	87.3±2.6	85.2±2.7	85.1±2.7
After reperfusion	87.2±2.5	85.4±3.4	85.4±3.1
pCO2 (mm Hg)
Before MCAO	43.5±2.1	44.5±3.3	43.5±2.7
Onset of reperfusion	44.2±1.8	44.2±2.2	44.1±2.1
After reperfusion	45.3±2.3	45.2±2.5	44.9±2.5
pO2 (mm Hg)
Before MCAO	134.6±5.8	136.7±6.0	135.4±5.5
Onset of reperfusion	135.3±4.9	135.5±5.1	135.5±4.9
After reperfusion	134.8±8.4	135.3±7.7	134.3±8.0

GTN indicates glyceryl trinitrate; MAP, mean arterial pressure; and MCAO, middle cerebral artery occlusion.

Reduced Neurological Deficits, Infarct Volume, and Cell Death After tMCAO

In the ischemic group (Figure 1A), infarct volumes, particularly in penumbral areas, were reduced in a dose–response manner to GTN, showing a ~30.93% (P<0.05) reduction with 5 mg/kg and a ~38.71% reduction with 10 mg/kg compared with the non‐treatment tMCAO with a ~49.84% infarct volume. In pMCAO, infarct volumes were only slightly, but not significantly reduced with GTN. Coadministration of GTN with KT5823 reversed the decreased infarct volumes (P<0.05), suggesting a potential signaling pathway through PKG (Figure 1B). Similarly, GTN reduced neurological deficits (P<0.05), while combined GTN/KT5823 demonstrated deteriorating neurological deficits (P<0.05). In pMCAO, neither sole GTN nor GTN with KT5823 changed neurological deficits (Figure 1C). Lactate and ROS were elevated at 6 and 24 hours of reperfusion, as well as in the pMCAO group. The elevated lactate and ROS levels were reversed by GTN after tMCAO following 6 or 24 hours of reperfusion (P<0.05), but not in the pMCAO group. After KT5823 and GTN, the inhibitive effects of GTN on lactate and ROS levels were abolished (P<0.05; Figure 1D). Cell death, quantified via ELISA, exhibited a significant elevation (P<0.01) in tMCAO at 6 and 24 hours after reperfusion. These elevations were notably reduced (P<0.01) by GTN. Additionally, a marked increase in cell death within the MCAO‐affected region was observed in tMCAO at 24 hours reperfusion and pMCAO via TUNEL (P<0.001). GTN significantly mitigated cell death only in tMCAO (P<0.0001; Figure 1E). When given 1H‐[1,2,4] oxadiazolo [4,3,‐a]quinoxalin‐1‐one (inhibitor of the vasodilator soluble guanylate cyclase, a key enzyme of NO signaling pathway) in ischemic rats treated with GTN, a significant reduction in infarct volume (Figures 2A and 2B, P<0.0001) and cGMP levels (Figure 2C, P<0.01) were still seen, suggesting that GTN exerts neuroprotective effects independent of NO‐induced vasodilation.

Figure 1. GTN exerts neuroprotective effects in tMCAO.

Illustrates how GTN reduced neurological deficits and infarct volume in tMCAO but not pMCAO rats. (A) TTC staining (48 h post‐reperfusion) shows a significant reduction in infarct size with GTN (*P<0.05 vs Stroke). (B) PKG inhibitor KT5823 before GTN administration suggests the role of PKG in GTN's efficacy (*P<0.05 vs Stroke, #P<0.05 vs Stroke+GTN). (C) Neurological scores (*P< 0.05 vs Stroke, #P<0.05 vs GTN). (D) Lactate and ROS (*P<0.05, **P<0.01, ***P<0.001 vs 1, #P<0.05, ##P<0.01, ###P<0.001 vs Stroke, & P<0.05 vs GTN). (E) Cell death further confirms the protective effect of GTN against I/R (*P<0.05, Sham). GTN indicates glyceryl trinitrate; I/R, ischemia followed by reperfusion for 6 or 24 hours; PKG, cyclic guanosine monophosphate (cGMP)‐protein kinase G; pMCAO, permanent middle cerebral artery occlusion; ROS, reactive oxygen species; tMCAO, transient middle cerebral artery occlusion; and TTC, 2,3,5‐triphenyl tetrazolium chloride.

Figure 2. GTN exerts neuroprotective effects independent of NO‐induced vasodilation.

(A, B) A significant reduction in infarct volume (A, B, ****P<0.0001 vs tMCAO, ##P<0.01 VS tMCAO+GTN, &&&&P<0.0001 vs tMCAO) and (C) cGMP levels (####P<0.0001 vs tMCAO, &&P<0.01 vs tMCAO+GTN) when given ODQ (vasodilator soluble guanylate cyclase inhibitor) in ischemic rats treated with GTN, demonstrating that an inhibited vasodilatation did not block the GTN‐induced neuroprotection. cGMP indicates cyclic guanosine monophosphate; GTN, glyceryl trinitrate; NO, nitric oxide; ODQ, 1H‐oxadiazolo[4,3‐a]quinoxalin‐1‐one; and sGC, soluble guanylate cyclase.

Released NO, Activated NO‐CGMP‐PKG Signal, and AMPK Phosphorylation

As compared with baseline, levels of serum nitrite, a NO metabolite, were decreased in tMCAO (P<0.05, Figure 3A). After GTN administration, serum nitrite levels were increased at 1 and 6 hours (P<0.001), returning to normal levels at 24 hours. In pMCAO, serum nitrite levels were not significantly changed. Brain levels of cGMP (Figure 3B), a second messenger activated by NO, were decreased at 6 and 24 hours of reperfusion after tMCAO (P<0.01). GTN reversed these decreased cGMP levels (P<0.01 for 6 hours and P<0.05 for 24 hours). In the pMCAO, levels of cGMP were not changed with or without GTN. In contrast to the sham‐operated cohort, there was a notable reduction in mRNA (P<0.01) and protein (P<0.05) levels of PKG at 6 and 24 hours post‐reperfusion. After MCAO, PKG protein expression was decreased, while GTN significantly reversed the decrease in PKG mRNA and protein after reperfusion (P<0.01) but not in pMCAO. Furthermore, the increases in PKG mRNA and protein expression after reperfusion by GTN were reversed by KT5823 (P<0.01; Figure 3C). Rats subjected to tMCAO exhibited a substantial augmentation in AMPK mRNA and protein levels at both 6 and 24 hours following reperfusion, whereas no such increase was observed in the pMCAO group. This overexpression was reversed by GTN (P<0.01). These decreases in AMPK mRNA expression were reversed by KT5823 after reperfusion (P<0.05). P‐AMPK protein expression was significantly increased only at tMCAO (P<0.05; Figure 3D).

Figure 3. Activation of the NO‐cGMP‐PKG pathway.

The increase is shown in (A) NO and (B) cGMP levels in rat plasma post GTN treatment, with corresponding increase in (C) PKG expression and decrease in (D) p‐AMPK (*P<0.05, **P<0.01 vs Sham, #P<0.05, ##P<0.01, ###P<0.001 vs Stroke, &P<0.05, &&P<0.01 vs GTN), highlighting the role of NO‐cGMP‐PKG pathway in neuroprotection. cGMP indicates cyclic guanosine monophosphate; I/R, ischemia followed by perfusion for 6 or 24 hours; NO, nitric oxide; PKG, cGMP‐protein kinase G; and p‐AMPK, phosphorylation‐monophosphate‐activated protein kinase.

Reduced Hyperglycolysis

In contrast to the sham‐operated group, mRNA and protein levels of GLUT1 (Figure 4A) exhibited a significant increase at 6 and 24 hours of reperfusion but only slightly in pMCAO (P<0.01). These increases in GLUT1 mRNA and protein expression were decreased by GTN in tMCAO (P<0.01). The decrease in GLUT1 mRNA expression (P<0.05) after GTN administration after reperfusion was reversed by KT5823 and protein expression (P<0.01) was reversed at 24 hours of reperfusion. There was an increase observed in both GLUT3 mRNA and protein expression in both tMCAO and pMCAO (P<0.01), which was decreased by GTN only in tMCAO (P<0.01). KT5823 abolished this effect in tMCAO (P<0.05; Figure 4B). These findings indicate that GTN influences glycolysis by altering the molecular activities of pivotal enzymes, GLUT1 and GLUT3.

Figure 4. Impact on hyperglycolysis.

Demonstrates the reduction by GTN in hyperglycolysis, evidenced by decreased mRNA and protein levels of (A) GLUT1, (B) GLUT3, (C) LDH, and (D) PFK1 in rats (#P<0.05, ##P<0.01, ###P<0.001 vs Stroke, &P<0.05, &&P<0.01 vs GTN), indicating GTN's regulatory effect on glucose metabolism. GLUT1 indicates glucose transporter 1; GLUT3, glucose transporter 3; GTN, glyceryl trinitrate; I/R, ischemia followed by perfusion for 6 or 24 hours; LDH, lactate dehydrogenase; mRNA, messenger RNA; and PFK1, phosphofructokinase‐1.

In tMCAO rats, LDH mRNA (P<0.001) and protein (P<0.05) levels were increased at 6 and 24 hours of reperfusion and were decreased by GTN (P<0.001) (Figure 4C). This effect was not observed in pMCAO. Again, the decreases in LDH expression were reversed by KT5823 in tMCAO but not pMCAO (P<0.05). In contrast to the sham group, there was a significant increase observed in both mRNA and protein expression of PFK1 (P<0.05; Figure 4D). Compared with the stroke group, PFK1 expression was decreased after reperfusion with GTN (P<0.01); however, this effect was not evident in pMCAO. The reductions in PFK1 mRNA and protein expression were reversed exclusively within the tMCAO group upon administration of KT5823 (P<0.05), with no such reversal observed in pMCAO.

Reduced Gluconeogenesis

In rats, both mRNA (P<0.001) and protein (P<0.05) expression of PCK1 and PCK2 (Figures 5A and 5B) were increased after tMCAO and pMCAO. These increases were reversed by GTN only in tMCAO (P<0.01). Furthermore, the effect was also reversed by KT5823 (P<0.05).

Figure 5. Influence of GTN on gluconeogenesis.

GTN decreases (A) PCK1 and (B) PCK2 expressions in rats, suggesting a reduction in gluconeogenetic processes (#P<0.05, ##P<0.01, ###P<0.001 vs Stroke, &P<0.05, &&P<0.01, &&&P<0.001 vs GTN). GTN indicates glyceryl trinitrate; I/R, ischemia followed by perfusion for 6 or 24 hours; PCK1, phosphoenolpyruvate carboxylase 1; and PCK2, phosphoenolpyruvate carboxylase 2.

Reduced ER Stress and Mediated Signaling

GTN markedly decreased the mRNA and protein expression of the initiation factor associated with ER stress, BiP (binding immunoglobulin protein), at both 6 and 24 hours of reperfusion (P<0.01; Figure 6A). Similarly, a notable decrease was observed in the protein expression of phosphorylated protein kinase R‐like endoplasmic reticulum kinase (p‐PERK) when compared with the ischemia/reperfusion group (P<0.05; Figure 6B). Protein expression of P‐eIF2α (phosphorylated α subunit of eukaryotic initiation factor 2) was significantly diminished (P<0.01; Figure 6C). For ATF4 (activating transcription factor 4), GTN markedly reduced both mRNAs at 6 and 24 hours (P<0.05) and protein expression at 24 hours (P<0.01; Figure 6D). Lastly, CHOP (C/EBP‐homologous protein), whose overexpression induces cell death, demonstrated significant decreases in both mRNA at 6 hours (P<0.05) and protein levels at 24 hours (P<0.01; Figure 6E). GTN appears to modulate ER stress and related signaling in ischemia/reperfusion.

Figure 6. GTN's role in ER stress reduction.

GTN modulates ER stress markers, including (A) BiP, (B) p‐PERK, (C) p‐eIF2α, (D) ATF4, and (E) CHOP (*P<0.05, **P<0.01, ***P<0.001 vs Sham, #P<0.05, ##P<0.01 vs Stroke), suggesting its effect on managing ER stress. ATF4 indicates activating transcription factor 4; BiP, binding immunoglobulin protein; CHOP, C/EBP‐homologous protein; ER, endoplasmic reticulum; GTN, glyceryl trinitrate; I/R, ischemia followed by perfusion for 6 or 24 hours; p‐PERK, phosphorylated protein kinase R‐like endoplasmic reticulum kinase; and p‐eIF2α, phosphorylated α subunit of eukaryotic initiation factor 2.

Reduced BiP Cellular Expression

The cellular expression of BiP was evaluated via confocal images, revealing its presence in neurons under normal, ischemic, and GTN‐administered conditions in rats. Stroke‐upregulated expression of BiP was reduced by GTN (Figure 7A). Following tMCAO and 24 hours reperfusion, BiP was expressed in 63.9% of neurons (P<0.0001) (Figure 7B), while this expression was reduced to 21.6% by GTN (P<0.0001). These results highlight that GTN inhibits the activation of ER stress following I/R, thus reinforcing mechanisms underlying GTN‐induced neuroprotection.

Figure 7. GTN and BiP expression.

(A, B) A decrease in BiP expression is seen in neurons post‐stroke with or without GTN treatment, with significant reduction in neuron staining (****P<0.0001 vs Sham), illustrating the neuroprotective effect of GTN in ER stress pathway at the cellular level. BiP indicates binding immunoglobulin protein; ER, endoplasmic reticulum; DAPI, 4',6‐diamidino‐2‐phenylindole; and GTN, glyceryl trinitrate.

DISCUSSION

Our study reveals that GTN exhibits neuroprotective properties in ischemic stroke involving reperfusion, but these effects are not observed in cases of ischemic stroke without reperfusion. Moreover, GTN is likely to aid in the regulation of ER stress and the rectification of anomalies in glucose metabolism, including the mitigation of excessive glycolysis and gluconeogenesis beyond the effect on vasodilatory properties. These beneficial effects are mediated through the release of NO, which activates PKG and subsequently inhibits AMPK phosphorylation (Figure 8). Our findings indicate that the effectiveness of GTN in neuroprotection is tied to the process of revascularization. In addition, penumbral areas are regions potentially salvageable by GTN therapy. GTN treatment reduces the infarct volume, indicating improved cell survival in these penumbral regions. During the ischemic phase, restricted blood flow leads to a critical depletion of essential substrates and an accumulation of metabolic waste products in the affected tissue. ¹⁵ Reperfusion is fundamentally important for cellular survival and recovery, as it restores blood flow, thereby delivering oxygen and nutrients necessary for cellular repair and recovery. This restored blood flow is vital for the effective delivery and action of GTN. In essence, reperfusion facilitates the efficient delivery of substrates and GTN, thereby potentially mitigating the extent of tissue injury. Earlier studies showed the neuroprotective potential of GTN in animal models pre‐stroke and post‐revascularization. ¹⁶ , ¹⁷ Our study has clarified GTN‐induced neuroprotection by comparing its effects in transient and permanent ischemia, emphasizing that reperfusion is necessary for neuroprotection. This is particularly relevant considering the high rates of large vessel revascularization but lower rates of ideal outcomes. ¹⁸ Thus, using GTN as an adjunctive neuroprotective strategy is both rational and advantageous.

Figure 8. Mechanisms of GTN‐induced neuroprotection in ischemic stroke.

Ischemic stroke reduces blood flow to the brain, causing an array of cerebral metabolic derangements and cell death. This mainly manifests as cerebral glucose metabolism relative to oxygen utilization and ER stress. GTN induces neuroprotection in tMCAO by improving glucose metabolism (reduced hyperglycolysis and gluconeogenesis), controlling ER stress, through the NO‐cGMP‐PKG signaling cascade to inhibit AMPK phosphorylation. AMPK indicates monophosphate‐activated protein kinase; cGMP, cyclic guanosine monophosphate; ER, endoplasmic reticulum; GTN, glyceryl trinitrate; NO, nitric oxide; PKG, cyclic guanosine monophosphate (cGMP)‐protein kinase G; and tMCAO, transient middle cerebral artery occlusion.

It is noteworthy that several key clinical studies have shown that the impact of GTN on microcirculatory flow and blood pressure does not necessarily correlate with neuroprotection. ⁵ , ¹⁹ , ²⁰ In addition, GTN did not significantly enhance cerebral perfusion in various regions of the brain in stroke patients. ²¹ These findings challenge the assumption that the neuroprotective effects of GTN are solely attributable to its vascular actions. Because GTN did not prove to be neuroprotective in those clinical studies, an alternative mechanism was explored in this research. Because the ischemic core is not salvageable, GTN can only protect the penumbra during reperfusion. Glycolytic oxidation is the primary source of brain metabolism, and abnormal glucose metabolism is the underlying mechanism of ischemia‐induced brain injury. ²² Gluconeogenesis is a multistep metabolic pathway catalyzed by the key enzyme, phosphoenolpyruvate carboxylase (PCK), to produce glucose from pyruvate. ²³ , ²⁴ Glycolysis and gluconeogenesis, crucial components of glucose metabolism, were observed in the brain during AIS because of reduced blood flow. ²⁵ In response to hypoxia, the expression of GLUT1, GLUT3, and phosphofructokinase‐1 (PFK1) was increased, leading to excessive rates of glycolysis. ²⁶ This high glycolytic flux results in elevated LDH (lactate dehydrogenase) levels, causing the onset of lactic acidosis and the excessive generation of ROS, leading to injury in the penumbra. ²⁷ In addition, as a compensatory pathway for glycogen supply, gluconeogenesis is also active after hypoxia. After AIS, the oxidative phosphorylation capacity is interrupted. This gluconeogenesis pathway is incomplete and instead leads to the accumulation of many intermediate products like lactic acid and ROS. ²⁸ In this study, the use of GTN after revascularization reduced LDH, PFK1, and PCK1/2 expression, suggesting that GTN interrupts hyperglycolysis and gluconeogenesis and, therefore, reduces cell death after ischemia–reperfusion.

AMPK is known to be a crucial regulator of glycol metabolism that helps to balance energy supply and demand. ²⁹ In conditions of reduced energy supply, such as post‐stroke hypoxia, the activation of AMPK occurs in response to an elevation in the AMP/ATP ratio, which subsequently fosters ATP production, potentially via enhanced glucose uptake. ³⁰ While AMPK assumes a pivotal role in the preservation of energy homeostasis under ischemic circumstances, its overactivation after stroke can exacerbate hyperglycolysis and gluconeogenesis, with detrimental metabolic effects. The finding that GTN reduced p‐AMPK levels in ischemia suggests that GTN‐activated metabolic signals downregulate AMPK.

ER stress frequently plays a role in a spectrum of pathological mechanisms, including neurodegenerative disorders, stroke, cancer, metabolic conditions, and inflammatory responses. ³¹ , ³² , ³³ , ³⁴ , ³⁵ , ³⁶ Our investigation suggests that GTN functions as a mediator in reducing the ER stress pathway, thereby presenting a promising anti‐cell death mechanism for neuroprotection following stroke.

Our study suggests that GTN's neuroprotective mechanism involves the release of NO. cGMP acts as a crucial intracellular second messenger influencing various pathophysiological processes. ³⁷ During the early reperfusion phase, the rise in cGMP correlates with substantially reduced infarct size, a phenomenon driven by PKG activation. ³⁸ In this study, cGMP and PKG levels were decreased after stroke, suggesting an impairment of the cGMP‐PKG pathway because of AIS. GTN treatment effectively reversed these effects. The selective PKG inhibitor diminished the protective impact of GTN, underscoring the role of PKG activation in neuroprotection. The effect of GTN on AMPK phosphorylation was removed by PKG signaling inhibition. These outcomes demonstrate GTN's therapeutic efficacy in stroke through its regulation of AMPK phosphorylation, which contributes to reduced lactic injury, thereby aiding in the preservation of brain function.

Earlier studies revealed conflicting results as to whether NO ameliorated or mediated damage in AIS. NO‐mediated neurotoxic effects via N‐methyl‐D‐aspartate receptors in neuronal cultures especially at high levels of NO. ³⁹ Studies further explored the possible therapeutic effects of inhibiting NO synthase using NG‐nitro‐L‐arginine (L‐NAME), ⁴⁰ , ⁴¹ revealing significant decreases in infarct volume when inhibiting NO synthase. It was recently reported that L‐NAME helped mitigate the overproduction of NO, thus decreasing neurotoxicity induced by NO overproduction. ⁴² However, L‐NAME therapy for stroke has not been used because of its non‐selectivity and associated adverse effects. ⁴³ , ⁴⁴ Other prominent studies found that inhaled NO significantly reduced brain injury in AIS by increasing penumbral blood flow and decreasing inflammation. ⁴⁵ , ⁴⁶

CONCLUSION

In conclusion, although the possibility that the discrepancy between the current results in rats and previous results in humans could reflect differences in underlying biology across species, this is the first report showing the GTN's neuroprotective effect upon reperfusion after AIS in which the NO‐cGMP‐PKG signals mediated AMPK phosphorylation. While the exact mechanism remains unclear, this study does indicate that reperfusion plays a crucial role in GTN‐induced neuroprotection and provides new insights into the potential neuroprotection strategy in stroke patients with revascularization.

In future studies, we aim to specifically investigate how GTN induces neuroprotection, whether through its effects on the vasculature/microcirculation, neurons, or a combination of both. In addition, the nitroglycerin dose used in rats does not precisely match the dose used in humans. Our preliminary study determined the rat dose based on efficacy, considering known doses from clinical trials in humans. In clinical settings, the dosage can be limited because of concerns like hypotension or other side effects. In future studies, we will aim to further refine and explore the dosage to associate it more closely with clinical settings, ensuring both efficacy and safety. By exploring these mechanisms and doses further, we hope to contribute more to the understanding of GTN's role in neuroprotection and the broader context of NO donors in neurological disorders.

Sources of Funding

This research received partial funding from the National Natural Science Foundation of China (82271332), the Beijing Tong Zhou District Financial Fund (2024), and the Yunhe talent Program of Beijing Tongzhou District (2024).

Disclosures

None.