The Nitrone Free Radical Scavenger NXY-059 Is Neuroprotective when Administered after Traumatic Brain Injury in the Rat
Abstract
Reactive oxygen species (ROS) play a significant role in the secondary injury processes following traumatic brain injury (TBI). Inhibition of ROS has consistently demonstrated neuroprotective effects in experimental TBI models. NXY-059, a nitrone free radical trapping compound, has shown neuroprotection in ischemic stroke models but has not been tested in experimental TBI. In this study, a continuous 24-hour intravenous infusion of NXY-059 or vehicle was started 30 minutes after severe lateral fluid percussion brain injury (FPI) in adult rats. Sham-injured animals received identical infusions and served as controls. Visuospatial learning was assessed using the Morris water maze on post-injury days 11 to 14, with a probe trial conducted on day 18 to evaluate memory. The animals were sacrificed on day 18, and the extent of hemispheric brain tissue loss was measured. Rats treated with NXY-059 after brain injury showed significant improvement in visuospatial learning compared to vehicle-treated controls. NXY-059 treatment also significantly reduced the volume of brain tissue loss compared to injured controls. These findings suggest that post-injury treatment with NXY-059 reduces brain tissue damage and improves cognitive function, supporting a major role for ROS in TBI pathophysiology.
Key words: learning and memory, neuroprotection, animal studies, NXY-059, oxidative stress, traumatic brain injury
Introduction
Traumatic brain injury (TBI) remains a major global public health concern. Advances in neurointensive care over the past two decades have improved outcomes for TBI patients. Despite these advances, many survivors suffer from lasting impairments including motor deficits, behavioral changes, and cognitive dysfunction, especially memory and learning problems. Currently, there are no pharmacological treatments with proven clinical efficacy to target specific components of the secondary injury cascade after TBI, despite numerous promising preclinical trials. One critical mechanism in the secondary injury cascade is oxidative stress, characterized by excessive production of reactive oxygen species (ROS) and reactive nitrogen species (RNS), coupled with depletion of the body’s endogenous antioxidant systems.
Following trauma, oxidative stress results from a multifactorial cascade involving increased intracellular calcium levels triggered by glutamate, leading to activation of enzymes such as xanthine oxidase, nitric oxide synthase (NOS), phospholipase A2, and cyclooxygenase. Other major sources of ROS overproduction in acute brain injury include the mitochondrial respiratory chain and endothelial cells of the cerebral microvasculature. Excessive ROS production causes lipid peroxidation of cellular and mitochondrial membranes, protein oxidation, DNA damage, and activation of immediate early gene responses.
To counteract oxidative stress, inhibition of ROS has been proposed as a therapeutic approach after TBI. Many antioxidants have been tested in animal models with encouraging results. Some compounds, such as polyethylene glycol-superoxide dismutase (PEG-SOD) and the lipid peroxidation inhibitor Tirilazad, advanced to clinical trials but failed to demonstrate significant benefit in human patients.
Nitrone ROS scavengers, initially used as detection agents for ROS in tissues, react directly with ROS to form stable adducts, providing neuroprotection in injured brain tissue. Various nitrones such as a-phenyl-tert-butyl nitrone (PBN), sodium-2-sulfophenyl-N-tert-butyl nitrone (S-PBN), and disodium 2,4-disulfophenyl-N-tert-butylnitrone (NXY-059) have been extensively studied for focal and global cerebral ischemia, consistently showing neuroprotective effects.
There is strong preclinical evidence that PBN and S-PBN protect the brain in TBI animal models. A key pharmacokinetic difference among these compounds is their ability to cross the blood-brain barrier (BBB): PBN readily crosses the BBB, whereas S-PBN and NXY-059 have poor BBB penetration. It is therefore believed that S-PBN and NXY-059 exert neuroprotection primarily at the level of the cerebral endothelium. This study investigated post-injury treatment with NXY-059 aimed at clinically relevant plasma concentrations following lateral fluid percussion brain injury in rats. The main outcome measures were cognitive function and morphological brain damage evaluated up to 18 days post-injury.
Methods
This study was designed as a follow-up and comparison to prior research using S-PBN after FPI, employing a similar Morris water maze protocol with latency measurements and a probe trial on day 18.
Animal surgery
Seventy-nine Sprague-Dawley rats weighing between 350 and 400 grams were used in the study. Five treatment groups, each consisting of 10 to 12 animals, were established. These included groups receiving fluid percussion injury (FPI) with NXY-059 treatment, FPI with vehicle treatment, sham-injury with NXY-059 treatment, sham-injury with saline treatment, and a naive control group. Animals were allowed free access to food and water throughout the experiment. Randomization was carried out by a colleague who was blinded to both the injury and treatment conditions. The study received approval from the relevant animal ethics committee and complied with applicable regulations.
Lateral fluid percussion brain injury
FPI was conducted according to established protocols. Rats were anesthetized with isoflurane and mechanically ventilated using a mixture of nitrous oxide and oxygen. Core body temperature was maintained at 37 degrees Celsius throughout the procedure. The tail artery was cannulated to allow monitoring of blood pressure and arterial blood gases, while the external jugular vein was cannulated for drug infusion. A 5-mm craniotomy was performed over the left parietal cortex, taking care to keep the dura intact. A Luer-Lok fitting was secured for connection to the FPI device. Severe brain injury was induced by delivering a brief, high-pressure saline injection into the closed cranial cavity, which rapidly increased intracranial pressure. After injury induction, the device was removed, and the scalp was sutured. Sham-injured animals underwent the same anesthesia and surgical procedures but did not receive the brain injury.
Physiological parameters such as arterial blood gases and blood pressure were monitored before and after the injury. After the procedure, the bone flap was replaced and animals were allowed to recover. Naive animals were handled similarly but did not undergo surgery. Animals were maintained for 18 days post-injury before being sacrificed. At the end of the study period, animals were euthanized and perfused with formaldehyde. The brains were then harvested, embedded in paraffin, and sectioned for further analysis.
Post-injury treatment
Thirty minutes after FPI, animals received a bolus dose of NXY-059 at 250 mg/kg or an equivalent volume of saline vehicle. They were then transferred to treatment cages and connected to an intravenous infusion system that allowed free movement. Infusion was initiated at approximately 500 ml/h, delivering a dose of 250 mg/kg per hour for a total of 24 hours. Following the 24-hour infusion period, animals were anesthetized using a mixture of isoflurane and air (4% isoflurane) for two minutes and maintained under anesthesia via a nose cone. A 200 µl blood sample was collected from the tail vein, anticoagulated with heparin, centrifuged at 1500×g for 10 minutes at 4 degrees Celsius, rapidly frozen on dry ice, and stored at –20 degrees Celsius. Plasma concentrations of NXY-059 were analyzed by reversed-phase high-performance liquid chromatography with ultraviolet detection at AstraZeneca, Sweden.
The rationale for the 24-hour intravenous infusion was based on previous studies with S-PBN following FPI. Given the low toxicity of NXY-059 observed in prior basic and clinical stroke studies, the infusion aimed to achieve a clinically relevant plasma concentration of 500 micromoles per liter.
Morris water maze
Visuospatial learning was assessed by a blinded observer using the Morris water maze paradigm. The maze consisted of a circular tank with a diameter of 1.4 meters and a submerged transparent platform measuring 12 by 12 centimeters, placed in the southwest quadrant of the tank, hidden 1 centimeter below the water surface. Animals were tested on their ability to find the platform from one of four designated starting points (west, north, east, and south). Testing involved 16 training trials over four days (four trials per day), conducted on days 11 to 14 after injury. Each trial began by placing the rat in the tank facing the wall at different entry points and activating a video tracking system. Trials ended when the animal located the platform or after 120 seconds had elapsed. If the platform was found, the rat was allowed to remain on it undisturbed for 15 seconds to observe visual cues around the pool. If the platform was not located within 120 seconds, the animal was manually placed on it for 15 seconds. Measures recorded included latency to find the platform, swim speed, and path length. The total mean latency was calculated by averaging latencies across the trial days. On day 18 post-injury, a memory retention trial was performed with the platform removed. Time spent in each quadrant was measured, focusing on the percentage of time spent in the correct quadrant and the number of times the animals crossed the previous platform location. Memory trials were analyzed over both the first 30 seconds and the full 60 seconds.
Lesion and hemispheric volumes
Brain analysis included assessment of multiple anatomical levels from bregma 0 to –7 mm using sections stained for microtubule-associated protein 2 (MAP-2). Digital images of the sections were obtained and analyzed with specialized software to measure cortical volume in the injured area. The perimeters of both the contralateral and ipsilateral hemispheres were traced by an evaluator blinded to injury and treatment conditions, and hemispheric areas were calculated using calibrated image analysis. Since previous studies indicated negligible contralateral tissue loss after lateral FPI, the contralateral hemisphere served as an internal control to account for inter-animal variability in brain size. Hemispheric tissue loss was calculated as a percentage of the contralateral hemisphere volume using the formula \[(Vc – Vi) / Vc] × 100, where Vc is the volume of the contralateral hemisphere and Vi is the volume of the ipsilateral hemisphere. Hemispheric volumes were integrated over the rostro-caudal distance of 7 mm.
Immunohistochemistry
Paraffin-embedded sections were rehydrated through xylene and graded alcohols. Antigen retrieval was performed by heating the sections in 0.01 M citrate buffer at pH 6.0 in a microwave oven for 10 minutes. Endogenous peroxidase activity was blocked by incubation in phosphate-buffered saline containing 0.1% hydrogen peroxide for 30 minutes. After rinsing, sections were incubated in 10% normal horse serum for 30 minutes. Primary antibody against MAP-2 (mouse monoclonal, diluted 1:200) was applied and incubated overnight at 4 degrees Celsius in a humidified chamber. Following primary antibody incubation, biotinylated horse anti-mouse secondary antibody (1:200 dilution) was applied for 30 minutes at room temperature. The avidin-biotin complex (ABC) was then applied for 30 minutes, followed by staining with diaminobenzidine (DAB) for 2 to 3 minutes. After rinsing in tap water, sections were counterstained with Mayer’s hematoxylin for 10 seconds, then dehydrated through graded alcohols and xylene before being mounted with a synthetic mounting medium.
Statistical analysis
Group comparisons for Morris water maze trials were performed using the non-parametric Kruskal–Wallis test, followed by Mann–Whitney U-tests if the overall test was significant. Hemispheric tissue loss ratios and lesion volumes were analyzed using analysis of variance (ANOVA) with Bonferroni/Dunn post hoc testing. Statistical analyses were carried out using StatView software. A p-value of less than 0.05 was considered statistically significant. Data are reported as means with standard deviations.
Results
Animals
Out of 39 animals subjected to severe fluid percussion injury (FPI), 12 died or were excluded due to injury-related complications, including mortality and apnea lasting more than 60 seconds, resulting in an injury-induced mortality rate of 30%. No adverse effects or increased mortality associated with NXY-059 infusion were observed. Seven additional animals were excluded due to catheter-related complications, and two animals had undetectable plasma levels of NXY-059. In total, 21 animals were excluded, leaving 56 animals for final analysis distributed among naive controls, sham-injured vehicle-treated, sham-injured NXY-059-treated, brain-injured vehicle-treated, and brain-injured NXY-059-treated groups.
Physiological Data
No significant differences were found between the injured treatment groups regarding FPI pressure, apnea duration, or post-injury weight changes. Before injury, arterial blood gases were similar across sham-injured and brain-injured groups. After injury, animals subjected to trauma exhibited significantly higher arterial oxygen partial pressures compared to sham-injured groups.
Plasma Concentration of NXY-059
At 24 hours post-injury, brain-injured animals treated with NXY-059 had a mean unbound plasma concentration of 475 ± 208 µmol/l, compared to 553 ± 366 µmol/l in sham-injured animals treated with NXY-059. No statistically significant difference was observed between these treated groups. NXY-059 was undetectable in the plasma of vehicle-treated sham-injured or brain-injured animals.
Morris Water Maze
Brain-injured animals, regardless of treatment status, displayed longer latencies to reach the hidden platform in the Morris water maze at two weeks post-injury compared to sham-injured and naive animals. However, brain-injured animals treated with NXY-059 showed a significantly shorter total mean latency to reach the platform compared to brain-injured vehicle-treated controls. Consistently, NXY-059-treated brain-injured animals had shorter latencies than vehicle-treated brain-injured animals. No significant differences in swim speed were observed among treatment groups, indicating that motor function did not account for differences in performance. Memory assessment on post-injury day 18 revealed no significant deficits in brain-injured animals compared to sham-injured or naive controls in terms of time spent in the correct quadrant or number of platform crossings during probe trials.
Morphometric Analyses
Brain injury resulted in significant hemispheric tissue loss in vehicle-treated animals compared to sham-injured controls, with evidence of cortical cavity formation. Treatment with NXY-059 significantly reduced the extent of hemispheric brain tissue loss after FPI. Additionally, brain-injured animals treated with NXY-059 had significantly smaller cortical lesion volumes compared to brain-injured vehicle-treated animals.
Discussion
This study presents the first evaluation of NXY-059 as a neuroprotective agent in traumatic brain injury (TBI). Post-injury treatment with NXY-059 significantly improved cognitive performance and reduced the loss of injured brain tissue in brain-injured rats. Treatment began 30 minutes after FPI, achieving plasma NXY-059 levels around 500 µmol/l, approximately twice the concentration targeted in human patients during the SAINT-1 study for cerebral ischemia and slightly higher than levels in a clinical safety study.
Nitrone Radical Scavenger Concept in Acute Brain Injury
Nitrone reactive oxygen species (ROS) scavengers have been investigated as both research tools and potential therapeutic agents for ROS-mediated secondary brain damage. The lipophilic nitrone PBN showed beneficial effects after cerebral ischemia in gerbils. PBN crosses the blood-brain barrier (BBB) effectively and reduces hydroxyl radical levels while attenuating phospholipid breakdown following traumatic brain injury (TBI) in rats. It also decreases lesion volume in experimental ischemia and TBI.
In contrast, S-PBN and NXY-059 are hydrophilic compounds that penetrate the BBB poorly. S-PBN has free radical trapping properties, reduces excitotoxic injury, and lowers infarct volume after focal ischemia. It has also improved functional outcomes and reduced tissue and neuronal loss after experimental TBI without entering injured brain tissue, suggesting that its action occurs primarily at the blood-endothelial interface. NXY-059 may exert neuroprotection at the microvascular level in stroke models. Despite limited BBB penetration after intravenous administration, NXY-059 has demonstrated significant neuroprotective effects following cerebral ischemia in both rats and primates.
Functional Outcome Following NXY-059 Treatment in TBI
The existing data support the idea that nitrone ROS scavengers are promising candidates for TBI therapy. In this study, administration of NXY-059 attenuated cognitive deficits observed in the Morris water maze (MWM) task measuring visuospatial orientation two weeks after injury. These results are consistent with previous findings using S-PBN, indicating that ROS play a role in the development of cognitive dysfunction following TBI and that inhibiting ROS improves outcomes.
The fluid percussion injury (FPI) model causes hippocampal damage, which correlates strongly with cognitive deficits measured in the MWM. The hippocampus is vulnerable to oxidative stress after TBI, as evidenced by lipid and protein oxidative damage to mitochondria within hours of controlled cortical injury, decreased levels of endogenous antioxidants, and increased markers of oxidative DNA damage. Prior studies have shown that S-PBN significantly improves regional blood flow in the hippocampus several hours after FPI and that PBN protects hippocampal neurons bilaterally after severe cortical contusion. These findings demonstrate that nitrone scavengers protect the hippocampus from ROS-induced injury. Other radical scavengers, such as OPC-14117 and L-Carnitine, have also improved MWM performance after controlled cortical injury.
The memory probe trial conducted 18 days post-injury did not show significant differences among groups, similar to previous studies using S-PBN and PBN. This suggests that NXY-059 reduces learning impairments but may not fully prevent all TBI-induced memory deficits.
Morphological Outcome and Pharmacokinetics
To assess morphological outcomes, cortical lesion volume and ipsilateral hemispheric tissue loss were quantified. Hemispheric tissue loss includes damage to the cortex as well as other regions such as the hippocampus. NXY-059 markedly reduced both the volume of injured brain tissue and the cortical cavity size, similar to effects observed with S-PBN treatment after FPI. Previous work indicates that bilateral hippocampal damage, which occurs in the FPI model, is necessary to produce deficits in MWM performance. Therefore, it is plausible that the reduction in ipsilateral hemispheric tissue loss is responsible for the improved cognitive effects observed with NXY-059 treatment.
The precise mechanisms underlying the observed neuroprotection are not fully understood. Notably, NXY-059 infusion was initiated after the peak of ROS production. Although blood-brain barrier penetration has been considered necessary for effective pharmacological neuroprotection, substances with poor ability to enter brain tissue have demonstrated neuroprotective effects in experimental models.
Plasma Levels and Treatment Time Window
Plasma concentrations achieved in this study (475–553 µmol/l) are similar to those shown to be effective in cerebral ischemia studies. Previous research has identified a plasma concentration of 400–450 µmol/l as optimal for protection after middle cerebral artery occlusion in rats. Treatment was started 30 minutes after TBI, while ischemia studies have demonstrated a therapeutic window of up to about four hours. Since this was the first evaluation of NXY-059 as a treatment for TBI, an early treatment paradigm was chosen. Early treatment in human TBI patients may be feasible due to the high safety profile of NXY-059 and its lack of effects on coagulation, potentially allowing administration even at the site of injury.
Mechanisms of Neuroprotection
Previous research has demonstrated that no detectable levels of S-PBN are present in injured brain regions following traumatic brain injury (TBI), suggesting that nitrone compounds primarily exert their protective effects at the blood-endothelial interface. Similarly, NXY-059 is believed to act within the vascular component of the neurovascular unit. Endothelial cells play a critical role as sources of reactive oxygen species (ROS) and are exposed to high oxygen tension as well as free radical-producing cells such as neutrophils and platelets. Treatment with S-PBN has been shown to increase blood flow in injured brain regions, which supports the hypothesis that nitrones influence the microvasculature after TBI. Furthermore, S-PBN reduces immune cell infiltration and downregulates adhesion molecules on endothelial cells following injury. NXY-059 has also been shown to protect brain endothelial cells in vitro after exposure to simulated ischemic conditions.
The leading hypothesis is that nitrones like S-PBN disrupt signaling between the injured brain and the peripheral immune system, likely through ROS-mediated mechanisms at the endothelial level. This disruption reduces the migration of T-cells to the injury site. This implies that targeting the neurovascular unit, specifically at the endothelial microvascular interface, may be a key mechanism underlying neuroprotection in TBI. However, more direct evidence is required to confirm that NXY-059 produces neuroprotective effects specifically at the vascular level.
Clinical Aspects
In a large phase III multicenter clinical trial involving patients with thromboembolic stroke, NXY-059 demonstrated significant neuroprotection by reducing disability at 90 days post-stroke onset. Despite these promising initial results, NXY-059 failed to show clinical efficacy in a subsequent large phase III clinical trial. This failure has dampened enthusiasm for the further commercial development of nitrone-based treatments. Nevertheless, extensive experimental evidence supports a robust neuroprotective effect of nitrones in various stroke and TBI models, including a 4-hour treatment window demonstrated in a primate stroke model. This strong preclinical data provides solid support for the pharmacological concept of neuroprotection via nitrone scavengers. Consequently, the nitrone scavenger approach remains scientifically valuable and warrants continued investigation in the context of acute brain injury Disufenton.
Acknowledgments
We express our gratitude to Sofi Forsberg, M.Sc., for her assistance and to Professor Richard Green for his important contributions. We also thank AstraZeneca R\&D in Södertälje, Sweden, for their support with NXY-059, plasma drug analyses, and the research grant that made this study possible.