Reactive oxygen species and lipid peroxidation inhibitors reduce mechanical sensitivity in a chronic neuropathic pain model of spinal cord injury in rats
Shayne N. Hassler,
Department of Neuroscience and Cell Biology, University of Texas Medical Branch, Galveston, Texas, USA
Address correspondence and reprint requests to Shayne N. Hassler, Department of Neurosciences and Cell Biology, University of Texas Medical Branch, 301 University Boulevard, Galveston, TX 77555-1043, USA. E-mail: firstname.lastname@example.org
Chronic neuropathic pain is a common consequence of spinal cord injury (SCI), develops over time and negatively impacts quality of life, often leading to substance abuse and suicide. Recent evidence has demonstrated that reactive oxygen species (ROS) play a role in contributing to neuropathic pain in SCI animal models. This investigation examines four compounds that reduce ROS and the downstream lipid peroxidation products, apocynin, 4-oxo-tempo, U-83836E, and tirilazad, and tests if these compounds can reduce nocioceptive behaviors in chronic SCI animals. Apocynin and 4-oxo-tempo significantly reduced abnormal mechanical hypersensitivity measured in forelimbs and hindlimbs in a model of chronic SCI-induced neuropathic pain. Thus, compounds that inhibit ROS or lipid peroxidation products can be used to ameliorate chronic neuropathic pain.
We propose that the application of compounds that inhibit reactive oxygen species (ROS) and related downstream molecules will also reduce the behavioral measures of chronic neuropathic pain. Injury or trauma to nervous tissue leads to increased concentrations of ROS in the surviving tissue. Further damage from ROS molecules to dorsal lamina neurons leads to membrane excitability, the physiological correlate of chronic pain. Chronic pain is difficult to treat with current analgesics and this research will provide a novel therapy for this disease.
Chronic pain affects 116 million people per year in the United States at a cost of over 635 billion dollars for treatment fees and lost productivity annually (Institute of Medicine 2011). Neuropathic pain, a chronic and difficult to diagnose and treat manifestation of pain, often does not respond well to commonly prescribed analgesic treatments (Murphy and Reid 2001; Vissers 2006). The lack of effective treatments can leave the patients in constant pain, leading to increased episodes of depression and suicide (Cairns et al. 1996; Widerstrom-Noga et al. 2001; Blair et al. 2003). Understanding the mechanisms behind chronic neuropathic pain will facilitate the development of targeted therapies for people suffering from chronic neuropathic pain.
Patients commonly develop chronic neuropathic pain by trauma to nervous tissue, either peripherally or centrally. Specifically, up to two-thirds of all spinal cord injured (SCI) people develop neuropathic pain syndromes (Finnerup and Jensen 2004; Werhagen et al. 2004). Our laboratory has developed a SCI animal model that consistently produces chronic neuropathic pain (Hulsebosch et al. 2000; Hulsebosch 2003) parallels the pathophysiology described in people with SCI (Bunge et al. 1993; Bunge 1994), and allows the rigorous study of cellular and molecular mechanisms of neuropathic pain after SCI in a controlled environment.
It has been reported that reactive oxygen species (ROS) play an important role in chronic neuropathic pain (Kallenborn-Gerhardt et al. 2013). ROS are highly oxidative molecules that naturally occur as a consequence of cellular energy production. Cellular stress or trauma results in higher than normal intracellular concentrations of ROS, which can overpower the homeostatic proteins and cause oxidative damage to the cell. Neurons are especially sensitive to ROS since neurons have greater energy demands to function as compared to glial and other cells in the central nervous system (Bell 2013).
We previously reported that downstream consequence of ROS, lipid peroxidation (LP) products, may also contribute to neuropathic pain in chronic SCI animals (Gwak et al. 2013). To better investigate the role that oxidation damage plays in chronic neuropathic pain, we examined four compounds that are known to reduce ROS and LP (Hall 1992; Khalil and Khodr 2001; Stefanska and Pawliczak 2008; Mustafa et al. 2010; Wilcox 2010). These four compounds are: (i) apocynin, a NADPH oxidase inhibitor; (ii) 4-oxo-tempo (also known as TEMPONE), a spin trap nitroxyl radical; (iii) U-83836E, a free radical scavenger that inhibits iron-dependent LP; and (iv) tirilazad, a potent peroxyl scavenger and membrane stabilizer. Each of these compounds was tested based on different mechanisms of action involving ROS and LP reduction products. We report that intraspinal administration of apocynin and 4-oxo-tempo significantly attenuated the abnormal mechanical hypersensitivity that develops following SCI in rats.
Materials and methods
Subjects were male Sprague–Dawley rats, 200–225 g, (Harlan Laboratories, Houston, TX, USA), and housed with a reversed day/night cycle of 12 h periods. Experimental procedures followed all National Institutes of Health Guidelines for the Care and Use of Laboratory Animals. Thirty-eight total animals were used in the experiments. For each experiment, 16 subjects were randomly divided into two groups for each trial (n = 8 per group), either the compound + vehicle + SCI group or the vehicle + SCI-alone group.
Spinal cord injury procedures
The animals were anesthetized by intraperitoneal injection of sodium pentobarbital (40 mg/kg). Anesthesia is considered complete when there was no withdrawal response to noxious foot pinch. When the animal was fully anesthetized, its back was shaved, and a laminectomy was performed exposing spinal segment T10. We produced contusion spinal injury using the Infinite Horizon impactor (150 kdyne, 1 s dwell time). Following the injury, the musculature was sutured, the skin autoclipped and the animals allowed to recover from anesthesia. The animals were eating and drinking within 3 h of surgery. Antibiotic treatment began immediately after surgery with a subcutaneous injection of 0.3 mL of Baytril (22.7 mg/mL) followed by a second injection 7 h later; after which, Baytril injections were given twice daily for 7 days and once daily for 3 more days to prevent urinary tract infections. Bladders were manually expressed twice daily. Automatic bladder emptying is achieved in all spinally contused rats by 10 days post-contusion. Post-injury animals were housed in Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC)-approved facilities. Records were kept on each animal with respect to its day of injury, weight at the time of injury, the computer record of the impact parameters, drug regimen measurements, and samples taken.
Mechanical threshold to paw withdraw in the forelimbs and hindlimbs of the animals were tested using a range of von Frey filaments (North Coast, 0.2–25 g force) using a modified Dixon up-down method with the paw withdrawal threshold was calculated. During the von Frey assay, the animals were also monitored for supraspinal behaviors such as head turning, changes in body posture, avoidance, vocalizations, paw licks, and aggressive behavior toward the von Frey filament. These supraspinal responses are monitored to ensure non-reflexive pain behavior is being measured (Christensen and Hulsebosch 1997). Responses of the mechanical sensitivity assays were taken pre-surgery, for a baseline measure, and compared to responses measured every week, beginning 14 days after injury, to monitor the animals' development of hypersensitivity. Animals that did not display an increase in mechanical sensitivity of at least 40% of baseline values were excluded from the experiment. Of the 38 animals that had the SCI surgery, two animals were excluded for not meeting the criteria. Locomotor activity in open field conditions was ranked. There were no indications that the compounds tested had a sedative effect on the animals at the dosages tested, and the locomotor scores were unchanged.
The animals were given intrathecal lumbar injections with apocynin (0.01, 0.05, and 0.1 mg/kg; Tocris Bioscience, Bristol, UK), tirilazad mesylate (0.01, 0.05, and 0.1 mg/kg; Cayman Chemical, Ann Arbor, MI, USA), U-83836E (0.05, 0.5, and 1.0 mg/kg; Cayman Chemical), or 4-oxo-tempo (0.05, 0.5, and 1.0 mg/kg; Sigma-Aldrich, St. Louis, MO, USA) at L3–L4 of the spinal column to avoid damage to the spinal cord. The pH of each solution of inhibitors was set to 7.2–7.4 pH to match that of the animals' cerebrospinal fluid. The animals were anesthetized by inhalation anesthesia 4% isoflurane. Fifty microliters of vehicle (10% dimethylsulfoxide, in 0.9% Saline) or vehicle plus a test compound was injected during a 1-min period into the intrathecal space. Fifty microliters was chosen to ensure that the solution diffused to cervical segments. The animals were alert and mobile moments after the inhalation anesthesia were removed, displaying no signs of distress or neural damage.
Behavioral testing started 42 days post-injury, during the animal's nocturnal cycle, which is when the animals are most active. The animals were first tested before injection, and then 30, 60, and 120 min after injection. This design was repeated for each compound trial with a minimum of 1 day in between trials to allow for drug washout; washout was determined by behavioral analysis and knowledge of the inhibitors' half-lives. Animals were randomly assigned to either the vehicle or compound groups for each trial.
Statistical analyses were performed using SPSS software (ver. 14; SPSS Inc., Chicago, IL, USA). The significant value for each test was α < 0.05. A within-group repeated measures anova was used to analyze the data for the same group over time. Data are expressed and graphed as mean with standard error of the mean (mean ± SEM).
Figure 1 represents the changes in the forelimb paw withdraw thresholds (PWT, g) before SCI, before intrathecal injection, 30, 60, and 120 min after intrathecal injection. All forelimb withdraw thresholds values in the pre-injection groups were significantly different compared to baseline values. The 30, 60, and 120 min values for the vehicle groups were not significantly different from the pre-injection values. The apocynin group values were significantly different (α < 0.05) from their pre-injection value at 30 min (14.35 ± 0.96), and 60 min (12.19 ± 0.81) after injection for the highest dosage of 0.1 mg/kg. The 1.0 mg/kg 4-oxo-tempo values for 30 min (13.56 ± 0.92) and 60 min (10.26 ± 0.74) were significantly different from the 4-oxo-tempo pre-injection values. The tirilazad group was not significantly different from the pre-injection value 30 min after injection; however, 60 min after injection (9.32 ± 0.85) was significantly different for the 0.1 mg/kg dosage as compared to the pre-injection values.
Figure 2 represents the changes in the hindlimb PWT (g) before SCI, before intrathecal injection, 30, 60, and 120 min after intrathecal injection. All hindlimb PWT values in the pre-injection groups were significantly different from baseline values. The 30, 60, and 120 min values for the vehicle group were not significantly different from the pre-injection values. The 0.1 mg/kg apocynin dosage group values of 30 min (17.14 ± 0.86), 60 min (17.11 ± 0.91), and 120 min (15.17 ± 0.87) were significantly different from the pre-injection values. The 1.0 mg/kg 4-oxo-tempo dosage group values of 30 min (18.15 ± 0.67), 60 min (17.05 ± 1.07), and 120 min (15.07 ± 0.87) were all significantly different from the 4-oxo-tempo pre-injection values. The 0.1 mg/kg tirilazad group values for 30 min (14.38 ± 0.48) and 120 min (14.23 ± 0.95) after injection were all significantly different from the pre-injection values. The 1.0 mg/kg U-83836E group value for 120 min after injection (14.57 ± 1.47) was the only value significantly different from its pre-injection values.
Neuropathic pain is difficult to treat, because patients do not respond well to commonly prescribed analgesics (Murphy and Reid 2001; Vissers 2006). Underlying mechanisms of neuropathic pain are important to determine to create treatments that target these mechanisms and treat neuropathic pain. This study tests four compounds that all reduce mechanical hypersensitivity in chronic SCI animals to different extents. However, the chemical mechanism of action is different for each compound and there may be off-target effects. By looking at the efficacy of each compound, insight into key processes can be gained in the ROS/LP pathway that affects chronic mechanical hypersensitivity. Apocynin and 4-oxo-tempo are effective and significantly reduced mechanical sensitivity in our model of chronic neuropathic pain. Apocynin, a NADPH oxidase inhibitor, reduces ROS by limiting the production of superoxides, which are the precursors of ROS. Apocynin has been shown to reduce inflammation in many other animal illness models as well as in animal models of nervous tissue damage (Impellizzeri et al. 2011; Ghosh et al. 2012; Seo et al. 2012; Valencia et al. 2012). The efficacy of apocynin is likely due to its mechanism, since reducing superoxide would have a broad effect on the downstream ROS production. 4-oxo-tempo and other nitroxide scavengers, such as TEMPOL have been used in treating animal models of hypertension (Wilcox 2010) and are potent scavengers of ROS radicals, particularly peroxynitrite (Carroll et al. 2000). The efficacy of 4-oxo-tempo is likely related to the broad effect that direct scavenging of oxidative radicals can have on chronic SCI animals, and may implicate peroxynitrite being a major contributor to generating chronic pain in SCI animals.
There was limited efficacy of both tirilazad and U-83836E, both of which are in the lazariod family of compounds. The mechanisms of both tirilazad and U-83836E target downstream LP process, by either scavenging the peroxyl radicals or stabilizing the bilayer membrane. The effects of tirilazad peaked at the 60 min measurements; this is most likely because the LP scavenging and membrane stabilization process has a longer time course when compared with the mechanisms of the other compounds. All the compounds seemed to persist longer in the hindlimbs when compared with the forelimb time course. This might have been due to the close proximity of the injection site to the lumbar enlargement, which subserves the hindlimbs. The limited activity of the lipophilic compounds may also be due to the limitations of solubility of the compounds. The experiments were limited to dosages that would remain soluble in the vehicle and at pH 7.2–7.4, which is the same as that of the animals' cerebrospinal fluid. It is possible that higher dosages of the lipophilic compounds may be more efficacious in reducing mechanical sensitivity in SCI animals; however, it is difficult to solubilize these compounds at higher dosages without altering the pH of the solution or using solvents that may adversely affect spinal tissue. By investigating the analgesic properties of these compounds, we are able to gain better insight into the mechanism that ROS and LP play in chronic neuropathic pain. A broad approach to reducing superoxides and other downstream effects of ROS via scavenging is more likely to produce an analgesic effect than that of inhibiting LP products which could also reduce mechanical hypersensitivity, but it requires longer intervals to be effective. This study demonstrates the novel finding that compounds that inhibit or reduce ROS or LP products reduce mechanical sensitivity in chronic spinal cord injury animals and should be considered when developing treatments for chronic neuropathic pain.
Acknowledgments and conflict of interest disclosure
This work was supported by grants from The Dunn and The West Foundations, Mr. Liddell, Mission Connect of TIRR and NIH grants NS 11255 and 39161. The authors have no conflicts of interest to declare.
All experiments were conducted in compliance with the ARRIVE guidelines.