Effects of hyperbaric oxygen on pain-related behaviours and nitric oxide synthase expression in a rat model of neuropathic pain

Department of Anesthesiology, Shengjing Hospital of China Medical University, Shenyang 110004, China Correspondence: Ling-xin Meng, No 36, Sanhao Street, Herping District, Shenyang 110004, China. Telephone 86-024-23380681, fax 86-024-23380681, e-mail kaishitengfei@163.com Neuropathic pain is caused by central or peripheral nerve injury, and represents a heavy burden to both society and family. Neuropathic pain is a complex process involving multiple factors. The basis of neuropathic pain is central nervous system hyperalgesia. Peripheral nerve injury induces a variety of immune-inflammatory factors, such as interleukin (IL)-1β, tumour necrosis factor-alpha (TNF-α) and IL-6, resulting in hyperalgesia through the activation of various intracellular signalling molecules (1,2). A satisfactory therapeutic method to treat neuropathic pain has not been developed due to the complexity of the disease. Therefore, finding a new and effective therapeutic method is an important issue in the field of neuropathic pain. Hyperbaric oxygenation (HBO) can enhance antioxidant activity, accelerate the clearance of free radicals, increase blood oxygen content, improve microcirculation and repair injured nerve tissue. In recent years, HBO has been used in the treatment of fibromyalgia, complex regional pain syndrome, spasmodic headache and postradiotherapy pain (3,4). However, little research has been performed on the effects of HBO on neuropathic pain. The purpose of the present study was to observe the effects of HBO on pain-related behaviours and nitric oxide synthase (NOS) expression in a rat model of neuropathic pain, and to identify its efficacy and mechanism of action in the treatment of neuropathic pain.


experimental process
Baseline mechanical withdrawal thresholds (MWTs) and thermal withdrawal latencies (TWLs) were measured in all rats one day before CCI.The rats in the pre-HBO group were exposed to HBO in the evening, and the operation (CCI or sham) was performed in all rats, except the rats in group C, the following morning.The rats in the post-HBO group were exposed to HBO in the evening following the operation.Pain-related behaviours were observed one, two, three, seven, 14, 21 and 28 days after CCI, starting 12 h after the HBO treatment of the post-HBO group (Figure 1).

cci model preparation
After the rats had been anesthetized by peritoneal injection of 1% sodium pentobarbital (40 mg/kg), CCI models were prepared using a modification of the method described by Bennett and Xie (5).In detail, a posterolateral incision on the right hindlimb was made and the sciatic nerve trunk was identified.The right sciatic nerve was loosely ligated with 4-0 silk thread to produce a slight pressure on the epineurium -namely, momentary muscle contraction in the sciatic nerve distribution region, followed by skin closure.After regaining consciousness, the rats were returned to cages.In group S, the sciatic nerve was exposed but was not ligated.hBo Rats in the pre-HBO or post-HBO groups were placed in HBO chambers 12 h before or after CCI, respectively.The chambers were infiltrated with pure oxygen for 10 min to allow the chamber oxygen concentration to exceed 90%, with the pressure at 0.25 MPa at a rate of 0.0125 MPa/min.The HBO treatment lasted 60 min under the condition of pure oxygen ventilation for 10 min.HBO was decreased to normal pressure within 30 min.samples Rats were anesthetized by peritoneal injection of 1% sodium pentobarbital (40 mg/kg) and intubation from the left ventricle to the ascending aorta was performed, followed by perfusion with a 0.9% saline solution until no red perfusate effused.After 0.1 M phosphate buffer (pH 7.4) containing 4% paraformaldehyde was perfused, the L4-L6 spinal cord segments were removed and fixed with 4% paraformaldehyde for future analyses.

observational items
Pain-related behaviours were observed one, two, three, seven, 14, 21 and 28 days after CCI between 09:00 and 10:00 on each day (1).Rats' gait, right hindlimb posture, autotomy and the bearing weight of the hindlimb were observed (2).MWT was tested using von Frey filaments, according to the method described by Dixon (6).The rat was placed in the test box and allowed to acclimatize for 30 min.Each von Frey filament was applied to the plantar surface of the hind paw for 6 s to 8 s to observe the hind paw withdrawal response (3).TWL was assessed to quantitatively determine rats' thermal sensitivity.Rats were placed on the glass surface of a thermal testing apparatus.The rats were allowed to acclimatize for 30 min before testing.A mobile radiant heat source located under the glass was focused onto the hindpaw of each rat.The paw TWL was recorded by a timer three times and the mean of these three trials was used in the present study.A cut-off time of 20 s was used to prevent potential tissue damage.
Fresh samples were fixed with 4% paraformaldehyde for 4 h, dehydrated according to standard protocols, embedded in paraffin and sectioned at a thickness of 40 μm.The slices were washed with phosphate-buffered saline (PBS) three times, and 3% H 2 O 2 and 30% blocking serum were added, respectively.After 10 min, antineuronal NOS (nNOS), endothelial NOS (eNOS) or inducible NOS (iNOS) antibodies were added, respectively, and slices were incubated at 4°C for 48 h.The slices were washed with PBS three times, and biotin-labelled secondary antibody (goat anti-rabbit IgG) was added at 37°C for 1.5 h.The slices were washed with PBS three times, and horseradish peroxidase-labelled streptavidin was added at 37°C for 1.5 h.The slices were washed with PBS, visualized with diaminobenzidine, dehydrated in an alcohol gradient, were made transparent with xylene, and were sealed with gum followed by microscopic observation.Brown-coloured cells were considered to be positive.In negative control trials, primary or secondary antibodies were replaced with PBS and no positive neurons were present.In each spinal cord, a slice was selected from every five slices to count the number of nNOS-, eNOS-or iNOS-positive neurons.nNOS-positive neurons stained brown in the cytoplasm and were oval in shape, with a clear outline.A few nNOS-positive neurons were spindle-or cone-like in shape.iNOS-positive neurons were triangular, round, oval and spindle-like in shape with one or several neurites, and were of various sizes smaller than that of nNOS-positive glial cells.eNOS-positive neurons exhibited multiple neurites, which form a network around them.
Micrography and image analysis were performed using the U-MCB Olympus system (Olympus Corporation, Japan).The analysis of slices was performed using the MetaMorph (Molecular Devices, USA)/Cool Snap (Roper Scientific, USA) +X/AX70 image analysis system.Four slices were taken from each rat to determine the number of positive cells.

statistical analysis
Data analysis was performed using SPSS version 11.0 (IBM Corporation, USA).Data were expressed as mean ± SD.ANOVA and Student-Newman-Keuls q tests were used to compare MWT and TWL between groups.A paired t test was used to compare nNOS, eNOS and iNOS.Differences were considered to be statistically significant at P<0.05.

General behaviours
With the exception of the rats in groups C and S, the toes of the affected foot moved closer to one another with dorsiflexion, eversion and non-weight-bearing three days after CCI.Rats in the CCI group also exhibited limping, with the knee joint contacting the surface of the testing apparatus and the palm of the right foot occasionally suspended in midair.Fourteen days after CCI, muscle atrophy was apparent, and light stimulation to the sole of the affected foot caused the rats to place the affected limb into their mouth or swing the affected limb freely.mWt There were no significant differences in MWT among the groups before CCI.MWT was stable in group C, but was slightly decreased in group S without a statistically significant difference compared with group C (P>0.05).MWT was significantly lower in the pre-HBO, post-HBO and CCI groups compared with group S at each time point (P<0.05; Figure 2), and the reduction of MWT was highest in the CCI group.MWT was higher in the pre-HBO group compared with the CCI group at each time point (P<0.05),demonstrating that HBO has preventive effects on neuropathic pain.MWT was higher in the post-HBO group compared with the CCI group one day to 21 days after CCI (P<0.05),demonstrating that HBO had therapeutic effects on neuropathic pain.Based on the two latter findings, HBO pretreatment was more effective than HBO post-treatment.tWL Changes in TWL were essentially the same as those in MWT.There were no significant differences in TWL among the groups before CCI.TWL was stable in group C. TWL was significantly lower in the pre-HBO, post-HBO and CCI groups than in group S at each time point (P<0.05; Figure 2), and the reduction of TWL was the highest in the CCI group.TWL was significantly higher in the pre-HBO group than in the CCI group at each time point (P<0.05),demonstrating that HBO had preventive effects on neuropathic pain, while TWL was significantly higher in the post-HBO group than in the CCI group only one day to 14 days after CCI (P<0.05).Therefore, HBO pretreatment was more effective than HBO post-treatment.nnos Twenty-eight days after CCI, the number of nNOS-positive neurons in the spinal cord was significantly increased in the CCI group and the pre-HBO and post-HBO groups with group S (P<0.05).The number of nNOS-positive neurons was significantly higher in the CCI group than in the pre-HBO and post-HBO groups (all P<0.05; Table 1 and Figure 3).inos Twenty-eight days after CCI, the number of iNOS-positive neurons in the spinal cord was significantly increased in the CCI, pre-HBO and post-HBO groups compared with group S (P<0.05); the number of iNOS-positive neurons was significantly higher in the CCI group than in the pre-HBO and post-HBO groups (P<0.05)(Table 2 and Figure 4).enos Twenty-eight days after CCI, the number of eNOS-positive neurons in the spinal cord was not significantly increased in each group.There were no significant differences in the number of eNOS-positive neurons among the CCI, pre-HBO and post-HBO groups (P>0.05,Table 3 and Figure 5).

discussion
Neuropathic pain is caused by central or peripheral nerve injury, and characterized by spontaneous pain, hyperalgesia and allodynia.Neuropathic pain induces central nervous system changes that contribute to hyperalgesia (7-9).Peripheral nerve injury induces a variety  of immune-inflammatory factors, such as substance P, IL-1β, TNF-α, prostaglandin and IL-6, resulting in hyperalgesia through activation of intracellular signalling molecules (10).These signals are regulated by positive feedback that is influenced by several mediators.However, the exact mechanisms governing the pathogenesis of neuropathic pain remain unclear and effective therapeutic methods have yet to be developed.Peripheral neuropathic pain is the most common and severe, and treatment with opiates often provides little relief.Therefore, neuropathic pain continues to be an important issue in the field of pain management.What causes changes in NOS expression in neuropathic pain?In neuropathic pain caused by peripheral nerve injury, NOS in the injured nerve and corresponding spinal cord plays an important role in pain regulation.Some studies have shown that pain afferent nerve endings release excessive amounts of glutamate that activates N-methyl-D-aspartate receptors and NOS, which increases neuron excitability.The increased production of nitric oxide (NO) by NOS, in turn, further enhances the release of glutamate and leads to pain.(11)(12)(13).Other studies have indicated that NOS and NO mediate the analgesic effects of morphine and clonidine, and the inhibition of NOS activity can reduce their analgesic effects (9,(14)(15)(16).The cause of these conflicting findings remains unclear, and may be associated with different physiological actions of many NOS subtypes including nNOS, iNOS and eNOS.In neuropathic pain, NOS expression and activity in the spinal dorsal horn remain unclear.The purpose of the present study was to observe the effects of HBO on NOS (nNOS, iNOS and eNOS) expression in the spinal dorsal horn in a neuropathic pain rat model.HBO has preventive and therapeutic effects on brain injury caused by hypoxia, ischemia or carbon monoxide poisoning by regulating NO levels (17).HBO also can activate endogenous opioid peptides that produce analgesic effects (16).These studies suggest that HBO can not only improve oxygen supply and correct hypoxia, but also regulate the immune system to produce therapeutic effects.Therefore, we believe that HBO regulates NOS expression to relieve pain.

Table 1 Number of nNOS-positive cells in the spinal cord in each group 28 days after chronic constriction injury
HBO can enhance antioxidant activity, relieve nerve tissue edema, accelerate the clearance of free radicals, increase blood oxygen content, produce triphosadenine, promote the regeneration of capillary vessels and improve microcirculation (18).The mechanism of the neuroprotective effects may be that HBO effectively increases plasma oxygen tension and improves tissue oxygen capacity; relieves nerve tissue edema, improves nerve tissue hypoxia and promotes aerobic metabolism and the regeneration of capillary vessels; and inhibits inflammatory reactions, preventing leukocyte aggregation, and decreasing the secretion of TNF-α, IL-1 and IL-6 (16,19).
HBO has been used as an effective and noninvasive method for the treatment of spinal cord injuries and high-altitude sickness, and in immunosuppression and stem-cell research; however, it has yet to be applied to the treatment of neuropathic pain.The present study indicated that HBO effectively increased MWT and TWL, demonstrating that HBO has therapeutic effects on neuropathic pain.Our study provides a basis for HBO application in the treatment of neuropathic pain.
nNOS and eNOS are collectively referred to as constitutive NOS, and are regulated by Ca 2+ and present in normal tissue, while iNOS is not regulated by Ca 2+ .In neuropathic pain, iNOS levels are elevated in the spinal cord, and after transcription and translation, iNOS plays an important role in NO synthesis.Therefore, in peripheral nerve injury, the increased iNOS expression in the spinal cord can promote hyperalgesia.It has been determined that, in a rat model of neuropathic pain, cNOS levels are higher than iNOS, and cNOS has regulatory effects on neuropathic pain (20).nNOS not only is regulated by Ca 2+ , but also has many subtypes that are located in different cellular sites and play different roles.In general, nNOS and iNOS in the spinal cord may be harmful due to neurotoxic effects, while eNOS in the periphery is often beneficial after tissue injury.However, in neuropathic pain, the expression of nNOS, eNOS and iNOS in the spinal dorsal horn has not been investigated.The purpose of the present study was to investigate the effects of HBO on the levels of nNOS, eNOS and iNOS expression in the spinal dorsal horn in a neuropathic No statistically significant difference was observed when the pretreatment with hyperbaric oxygen (pre-HBO), post-treatment with hyperbaric oxygen (post-HBO) and chronic constriction injury groups were compared with the sham operation group In the present study, HBO effectively decreased the levels of spinal nNOS and iNOS expression, but failed to alter the level of eNOS; therefore, HBO likely relieved hyperalgesia through regulating the expression of spinal nNOS and iNOS.HBO increases MWT and TWL in a neuropathic pain rat model, demonstrating HBO has preventive and therapeutic effects on neuropathic pain.The expression of NOS-positive neurons in the spinal dorsal horn is increased in rats with CCI, suggesting that NOSpositive neurons in the spinal dorsal horn play an important role in the central mechanism of chronic neuropathic pain.HBO likely relieves pain through decreasing NOS expression.However, in the present study, HBO was administered only once; therefore, the optimal frequency of HBO treatment remains to be explored.

Figure 1 )
Figure 1) Experimental process.B Baseline values; C Blank control; CCI Chronic constriction injury; d Day; D Death (samples taken); HBO Hyperbaric oxygenation; S Sham operation.The time course refers to the time before CCI and after CCI; CCI is represented as // 05 when the pretreatment with hyperbaric oxygen (pre-HBO), posttreatment with hyperbaric oxygen (post-HBO) and chronic constriction injury groups are compared with the sham group; † P<0.05 when the pre-HBO and post-HBO groups are compared with the chronic constriction injury group.n=8 rats per group.nNOS Neuronal nitric oxide synthase

Figure 5 )Figure 4 )
Figure 5) Endothelial nitric oxide synthase immunohistochemistry in the spinal cord in each group (original magnification ×400).Arrows indicate endothelial nitric oxide synthase-positive neurons of various sizes.CCI Chronic constriction injury; Pre-H Pretreatment with hyperbaric oxygenation; Post-H post-treatment with hyperbaric oxygenation; S Sham operation

Table 2 Number of iNOS-positive cells in the spinal cord in each group 28 days after chronic constriction injury Group iNOS-positive cells per 10 4 cells
*P<0.05 when the pretreatment with hyperbaric oxygen (pre-HBO), post-treatment with hyperbaric oxygen (post-HBO) and chronic constriction injury groups are compared with group S; † P<0.05 when the pre-HBO and post-HBO groups are compared with the chronic constriction injury group.n=8 rats per group.iNOS Inducible nitric oxide synthase