N-Acetylcysteine Attenuates Hyperalgesia in Rats with Diabetic Neuropathic Pain: Role of Oxidative Stress and Inflammatory Mediators and CXCR4

Objectives. CXCR4 plays critical roles in the development of diabetic neuropathic pain (DNP) in rats, and its mechanism is unknown. This study was aimed at evaluating the potential therapeutic value of the antioxidant N-acetylcysteine (NAC) against DNP in rats and how CXCR4 participates in the formation of DNP. Methods. Control or streptozotocin- (STZ-) induced diabetic Sprague-Dawley rats received vehicle or NAC for four weeks starting one week after STZ injection. Von Frey and Hargreaves Apparatus were used to analyze the behavioral changes of mechanical allodynia and heat hyperalgesia. CXCR4, p-CXCR4, interleukin- (IL-) 6, and tumor necrosis factor- (TNF-) α in the spinal cord and the prefrontal cortex were detected by western blotting. Plasma IL-6, TNF-α, superoxide dismutase- (SOD-) 1, SOD-2, and lipid peroxidation products malondialdehyde (MDA) and 15-F2t-Isoprostane were detected by ELISA. Results. The values of paw withdrawal threshold (PWT) and paw withdrawal latencies (PWL) were reduced in diabetic rats compared to control rats that were concomitant with significant increases of CXCR4, p-CXCR4, IL-6, and TNF-α protein expressions in the spinal cord and prefrontal cortex. The treatment with NAC decreased the IL-6 and TNF-α protein expression and further increased CXCR4 and p-CXCR4 in the spinal cord and the cortex of diabetic rats that were accompanied with enhancement of PWT and PWL. NAC also significantly attenuated or reverted the increases of plasma IL-6, TNF-α, SOD-1, SOD-2, MDA, and 15-F2t-Isoprostane in diabetic rats. Conclusion. It is concluded that NAC treatment could effectively alleviate DNP and that induction of CXCR4 and p-CXCR4 may represent a mechanism whereby NAC attenuates DNP.


Introduction
Diabetic neuropathic pain (DNP) is one of the most common chronic complications of diabetes, which brings suffering to diabetic patients [1,2]. The underlying mechanisms of DNP are complex, and both peripheral and central components of the sensory systems are reported to be involved in the progression and maintenance of neuropathy. Increased oxidative stress is a unifying mechanism in the causation of DNP [3]. Hyperglycemia-induced activation of the polyol and hexosamine pathway and formation of advanced glycation end products are known to increase the production of Reactive Oxygen Species (ROS) that contribute to nerve injury [4,5]. Excessive generation of ROS initiates a vicious circle by activating stress-sensitive pathways such as NF-Kb, p38 MAPK, and protein kinase C [6] or proinflammatory cytokines that contribute to diabetic complications [7,8]. Hyperglycemia induces higher levels of proinflammatory cytokines that have been shown to correlate with the incidence of neuropathy [9,10].
Chemokine receptor CXCR4 is a specific receptor for chemokine matrix-derived factor-1 (CXCL12). CXCR4/SDF-1 signaling has been implicated in the pathogenesis of neuropathic pain in a subset of animal models including diabetic neuropathic pain [11,12]. A study showed that the specific CXCR4 antagonist AMD3100 can reverse DNP in rat models of type II diabetes and that CXCR4 coordinates inflammation in diabetic dorsal root ganglia [13] that could contribute to the development of pain in diabetes [14]. Our previous research found that the values of PWT and PWL were reduced in diabetic rats that were concomitant with significant increases of both CXCR4 and TNF-α protein expression in the DRG of diabetes [15]. However, the role of CXCR4 in relation to ROS in the development of DNP has yet to be determined.
NAC, a precursor of GSH, confers its antioxidant effect via enhancing endogenous GSH/GSSG [16]. NAC also produces a broad range of effects, which include attenuation of pain [17] and confer neuroprotection after spinal cord injury [18] and peripheral nerve lesions [19]. Our previous research found that expression of the prooxidative NADPH oxidase and inflammatory markers increased in the heart of STZinduced diabetic rats, and these causes could be decreased by NAC [20,21]. However, the potential effects and mechanism of NAC on DNP have not been explored.
Pain conduction is regulated by up-and downregulated pathways, which are caused by peripheral and central nervous system sensitization. Central sensitization is an important pathogenesis of neuropathic pain, and the maintenance of neuropathic pain lies in central sensitization [22]. Therefore, we proposed to observe the changes of inflammatory factors and CXCR4 protein expression in the spinal cord as well as the prefrontal cortex of diabetic rats with DNP and to determine the effects of antioxidant treatment with Nacetylcysteine on the inflammatory mediators and CXCR4 in the spinal cord and prefrontal cortex and the impact on diabetic neuropathic pain.

Experimental Animals.
Male adult Sprague-Dawley with streptozotocin-(STZ-) induced diabetes and age-matched non-diabetic control rats were used. All rats (250 ± 10 g, 6-8 weeks) were obtained from and housed in the Laboratory Animal Service Center (University of Hong Kong) and received standard care in accordance with the principles of Animal Care of the University of Hong Kong. The committee on the Use of Live Animals in Teaching and Research approved the experimental protocols.

Induction of Diabetes and Antioxidant
Treatment. Type 1 diabetes was induced by tail injection of STZ (Sigma, USA) at the dose of 65 mg/kg body weight in 0.1 mol/L citrate buffer (pH 4.5) or citrate buffer alone as control under anesthesia with a combination of ketamine 67.7 mg/kg body weight and xylazine 6.77 mg/kg body weight [23]. Blood glucose was tested by a OneTouch UltraVue glucometer (Johnson, USA) 72 hours after the injection. Rats with a blood glucose level higher than 16.7 mmol/L 72 hours at STZ injection were deemed T1DM and subjected to subsequent experiments. Blood glucose and body weight were measured at baseline and monitored weekly during the experiment.
Some of the diabetic rats were treated with NAC (Sigma-Aldrich, St. Louis, MO, USA). Rats were randomly divided into three groups of n = 6 per group: control (C), untreated diabetic rats (D), and diabetic rats treated with NAC (D+N). NAC (Sigma-Aldrich) was dissolved in drinking water for 3 weeks' duration of treatment starting 1 week after induction of diabetes ( Figure 1). We used a dose of NAC at 1.5 g/kg/day; the optimal dosage was selected according to our previous experiment [24]. The daily dose of NAC (1.5 g/kg) was dissolved into 2/3 volume of the average daily amount of drinking water that was estimated based on preliminary study and previous studies to ensure that the total amount of NAC was taken by the rats before additional amount of water was supplied. Meanwhile, the plasma glucose, body weight, and water intake of rats were routinely monitored.

Behavioral Assessment.
Rats were subjected to behavioral assessments before the STZ injection (baseline) and at four weeks after STZ injection. To measure mechanical sensitivity, paw withdrawal threshold (PWT) of rats was assessed by the electronic Von Frey test (IITC) [25]. Before the experiments, the rats were placed in a transparent plastic dome with a metal mesh floor for nearly 30 minutes. The electronic Von Frey filament set has a calibrated range of bending force (floating 1~90 g) to record automatically pain threshold. A single filament was applied to the plantar surface 5 times with an interstimulation interval of 5 seconds. A positive response was defined as at least 1 clear withdrawal response in the 5   Journal of Diabetes Research applications. Licking of the paw was also considered as a positive response.
To quantitatively assess the thermal threshold of the hind paw, rats were placed on the glass surface of a thermal testing apparatus with acclimatization for 30 minutes before testing. A mobile radiant heat source (Hargreaves apparatus) located under the glass was focused onto each of both hind paws of each rat. A radiant heat source was applied individually to the middle of either hind paw for up to 60 seconds, with the paw withdrawal threshold (PWT) measured. The heating rate ramped from 30°C to 58°C over 60 seconds in a consistent fashion on each occasion [26]. The cut-off of 30 seconds was used to prevent potential tissue damage. The paws were inspected before and after thermal testing to ensure that no evidence of thermal damage was present. The withdrawal latency of both hind paws from three consecutive trials was averaged, and the mean value was used as the thermal threshold. There were 15-minute intervals in between trials.

Western Blot Analysis.
After the completion of behavioral assessment, rats were then deeply anaesthetized with sodium pentobarbital (65 mg/kg), and the lumbar (L3 to L5) parts of the spinal cord and the prefrontal cortex were rapidly excised and frozen in liquid nitrogen. The tissues were homogenized using lysis buffer, sonicated, and centrifuged at 12000g for 20 min at 4°C. The protein concentration was quantified by bovine serum albumin and measured with the absorption of Coomassie brilliant blue in the spectrophotometer. Thereafter, the samples were frozen at -20°C for later use. Proteins were assessed by standard western blotting as described [27]. Briefly, equal quantities of protein were separated by SDS-PAGE and transferred to polyvinylidene difluoride membranes (PVDF, Millipore, Bedford, MA, USA). The membranes were blocked in 5% nonfat dry milk diluted with Tris-Buffered Saline Tween-20 (TBST) (in mmol/L: Tris-HCl 20, NaCl 150, pH 7.5, 0.1% Tween 20) at room temperature for 1 hour (h) and then probed with antibodies against TNF-α, IL-6, (Abcam, England), CXCR4, and p-CXCR4 (Sigma, USA) at 4°C overnight. After extensive washing, the membranes were incubated with secondary horseradish peroxidase-conjugated anti-mouse or anti-rabbit antibodies (diluted 1 : 2000; Amersham Biosciences, UK). The immunoblots were visualized using an enhanced chemiluminescence detection system (Amersham Pharmacia Biotech, Uppsala, Sweden). The protein bands were observed by an Immobilon Western Chemiluminescent HRP Substrate (Millipore, USA) and then processed by gray scanning using ImageJ (National Institutes of Health, USA). All the target proteins were normalized by β-tubulin and calculated as percentage of the control.
2.6. Statistical Analysis. All values are presented as means ± SD. One-way analysis of variance (ANOVA) was used for statistical analysis (GraphPad Software, Inc.) of data obtained within the same group and between groups, respectively, followed by Tukey's test for multiple comparisons of group means. P values less than 0.05 were considered to indicate statistically significant differences.

Results
3.1. General Characteristics. The baseline body weight and blood glucose did not significantly differ among rats in the three experimental groups (data not shown). As shown in Figure 2, STZ-injected rats had significant diabetic symptoms of hyperglycemia and weight loss. The blood glucose level of the diabetic rats increased but their body weight decreased compared to normal rats ( * P < 0:05, D vs. C). NAC treatment for three weeks had no significant effect on body weight and blood glucose in diabetic rats (P > 0:05, D+N vs. D). The values of PWT were significantly lower in the diabetes group up to 4 weeks of diabetes than those in the control group ( * P < 0:05, Figure 3(a)). At week 4 post STZ injection, the administration of NAC attenuated mechanical hyperalgesia ( & P < 0:05, Figure 3(a)).

Effect of N-Acetylcysteine
A significant decrease in the PWL was observed 4 weeks in diabetes rats than in the control group ( * P < 0:05, Figure 3(b)). At week 4 post STZ injection, the administration of NAC attenuated thermal hyperalgesia ( & P < 0:05, Figure 3(b)).

Effect of N-Acetylcysteine Treatment on TNF-α and IL-6
Protein Expression in the Spinal Cord and Prefrontal Cortex of Diabetic Rats. As shown in Figure 4, significant increases in TNF-α and IL-6 were observed in diabetic rats compared to nondiabetic rats both in the spinal cord and in the prefrontal cortex ( * P < 0:05, D vs. C). NAC treatment significantly decreased the proinflammation TNF-α and IL-6 levels compared to diabetic rats ( & P < 0:05, D+N vs. D).

Effect of N-Acetylcysteine Treatment on CXCR4 and p-CXCR4 Protein Expression in the Spinal Cord and the
Prefrontal Cortex of Diabetic Rats. As shown in Figure 5, significant increases in CXCR4 and p-CXCR4 were observed in diabetic rats compared to nondiabetic rats both in the spinal cord and prefrontal cortex ( * P < 0:05, D vs. C). NAC treatment further significantly increased the CXCR4 and p-CXCR4 levels compared to those in diabetic rats ( & P < 0:05, D+N vs. D).
3.5. Effect of N-Acetylcysteine Treatment on Plasma IL-6 and TNF-α Level of Diabetic Rats. As shown in Figure 6, significant increases in plasma IL-6 and TNF-α were observed in diabetic rats as compared to nondiabetic rats ( * P < 0:05, D vs. C). NAC treatment significantly decreased the plasma IL-6 and TNF-α levels compared to diabetic rats ( & P < 0:05, D+N vs. D).

Discussion
In the present study, firstly, we found that the pain threshold decreased significantly after the establishment of diabetes. Furthermore, proinflammatory proteins TNF-α and IL-6 were significantly upregulated in the spinal cord, cortex, and plasma. Plasma SOD-1 and SOD-2 activities in diabetic rats also increased significantly as compared to those in control rats. CXCR4 and p-CXCR4 expression was significantly increased both in the spinal cord and cortex. Thirdly, NAC treatment could increase the pain threshold, at the same time, downregulate the expression of TNF-α and IL-6, and upregulate CXCR4 and p-CXCR4 expression, compared to the diabetic rats.
STZ impairs β-cell in the pancreas, and intravenous STZ at the dose of 65 mg/kg reliably and reproducibly induced type 1 diabetes in rats [30,31]. Hyperglycemia induced increases in the production of Reactive Oxygen Species (ROS) that contribute to nerve injury [32]. Excessive oxidative stress and overload of Ca (2+) entry are common features of neuropathic pain. In whole-cell patch clamp experiments, TRPM 2 currents in the DRG following diabetes induction with STZ were gated by H 2 O 2 and H 2 O 2 -induced TRPM2 gating was totally inhibited by NAC [33]. NAC may have a protective role on oxidative stress and calcium influx through regulation of the TRPM 2 channel in diabetic neurons. A study showed that NAC, given after the establishment of diabetes, may offer protection against the risk for stroke via enhancing platelet GSH and GSH-dependent methylglyoxal elimination and SOD-1 [34]. Ninety patients with DNP completed the eight-week course of the study, and the results showed that NAC significantly increased serum levels of SOD and glutathione peroxidase 1 and alleviated painful symptoms of diabetic neuropathy [35].
Our current research found that plasma SOD-1 and SOD-2 activities were increased in diabetic rats as compared  Journal of Diabetes Research to nondiabetic rats. These might be the consequence of hyperglycemia-induced overreaction of the oxidative stress system (SOD-1 and SOD-2) as a compensative response against oxidative stress, one of the key mechanisms of DNP. NAC increased pain threshold, and in the meantime, the effectiveness of NAC antioxidant therapy may have served to attenuate the compensative increases of SOD activities. The effectiveness of NAC antioxidant therapy may also be the major mechanism that resulted in overexpression of CXCR4 and p-CXCR4 in the spinal cord and cortex.
In the state of hyperglycemia, oxidases in various compartments, including the endothelium, media, and adventitia, produce ROS which render vascular smooth muscle cells (VSMCs) under oxidative stress [36]. The oxidative stress may injure VSMC or perturb the regulatory signal to induce the upregulation of inflammatory factors [37]. Our previous research found that IL-6 and TNF-α were increased in the plasma and the myocardium of STZ-induced diabetic rats [38]. The patients with diabetic kidney disease (DKD) showed higher IL-6 and TNF-α level than those without DKD [39]. The levels of IL-6 and TNF-α in the livers [40] and the spinal cord [41] of type I diabetic rats were also increased as compared to those of normal rats. There is an important discovery in our current experiment that IL-6 and TNF-α increased significantly not only in the spinal cord and plasma but also in the cortex. This suggests that the inflammatory response in the peripheral and central systems together leads to the DNP.
In our study, NAC significantly attenuated the increases of IL-6 and TNF-α levels in diabetic rats in the spinal cord, cortex, and plasma. In clinic, NAC is efficacious in improving neuropathic pain associated with diabetic neuropathy [35]. This further proved the important mechanism of DNP that hyperglycemia produced too much ROS with the subsequent presence of strong oxidative stress that induces acute inflammatory response.
CXCR4 is a G protein-coupled receptor that is involved in homing and chemotaxis in the hematopoietic and immune systems. The ligand of CXCR4 is a stromal cellderived factor (SDF)-1/CXCL12 that is associated with stem cell migration from the bone marrow to the cells expressing SDF-1 [42]. Intrathecal administration of SDF-1 induced hypersensitivity in naive rats. And repeated intrathecal administration of the CXCR4 antagonist, AMD3100, significantly suppressed the initiation and duration of neuropathic pain [11,43]. SDF-1/CXCR4 signaling also maintains central poststroke pain [44] and persistent abdominal pain in rats with chronic pancreatitis [45]. SDF-1/CXCR4 signaling contributes to bone cancer pain [46] and diabetic neuropathic pain [15,47].

Journal of Diabetes Research
In our study, CXCR4 and p-CXCR4 receptor expression increased significantly in the spinal cord and cortex in diabetic rats. This indicated that CXCR4 and p-CXCR4 overexpression may have contributed to the development of DNP in the absence of antioxidant treatment. However, what our study found especially interesting was that NAC treatment relieves DNP but further increases CXCR4 and p-CXCR4 protein expression both in the spinal cord and cortex. This result seems contradictory to what has been reported about a detrimental role of CXCR4 in pain development. It should be noted, however, among the chemokine receptors, CXCR4 stands out for its pleiotropic role as well as for its involvement in normal and abnormal development, including immune diseases, viral infections, and cancer [48]. SDF-1 can activate CXCR4-expressing cells and induce B cell proliferation and T cell recruitment. CXCR4 −/− mice display  Journal of Diabetes Research impaired vascularization in various organs, including the intestines, stomach, and heart [49]. A study using knockout mice has revealed the involvement of CXCR4 in cardiogenesis and brain development. Without CXCR4, the homozygous knockout mice die before birth. In an in vivo study, CXCR4-overexpressing MSCs were observed with reduced retinal damage and upregulation of rhodopsin and NSE protein and downregulation of inflammatory cytokines IL-6 and TNF-α [50]. Therefore, CXCR4 is a key molecule for normal development.
A study showed that ROS accumulation in prostate cancer cell lines resulted from activation of the NADPH oxidase (NOX) family of enzymes. NOX2 responded to and was regulated by the SDF-1α/CXCR4 signaling axis [51]. The mice treated with NAC presented markedly activated GSH-PX and SOD in mice with crystalline silica-induced pulmonary fibrosis and at the same time downregulation of oxidizing enzymes (NOX2, iNOS, SOD-2, and XO) [52]. In our study, three weeks of 1.5 g/kg NAC in water treatment reduced oxidative stress. It is possible that NAC treatment may have resulted in a downregulation of NOX2 and the reduction of oxidative stress, as a consequence of the upregulation of CXCR4 and p-CXCR4 expression. Thus, it is intriguing to hypothesize that increasing CXCR4 and p-CXCR4 expression may represent one of the major mechanisms whereby NAC attenuates DNP, although this hypothesis needed to be further tested.
There are clinical study findings showing beneficial effects of enhancing CXCR4 in diabetic patients. Diabetic Foot Ulcer (DFU) is the most common in patients who have diabetic peripheral neuropathy and angiopathy as well as a foot deformity. Diabetic foot early performance is mainly for cold foot, numbness, and other abnormal feeling and later will appear sensory decline or loss. Research showed that SDF-1, a specific ligand of CXCR4, was significantly decreased in both the circulation and tissue biopsies of patients with T2DM and infected DFU [53]. Contents and protein expressions of SDF-1α/CXCR4 protein expression in wound tissue of diabetic patients with DFU who received 8 weeks treatment of a Chinese ointment (Xiaodu Yuji Paste) were all significantly higher than those in the control group that was accompanied with improved angiogenesis and wound healing [54]. Therefore, we speculate that the increased expression of CXCR4 and p-CXCR4 in the early phase of diabetes as seen in our current study (e.g., 4 weeks after STZ induction) would protect blood vessels, but in the meantime may lead to abnormal nerve conduction, which eventually results in the decrease of pain threshold. Another study suggests that age-associated reduction in CXCR4 expression on BM MSC impairs hematopoietic niche activity with increased ROS production, driving an HSC aging phenotype. Thus, modulation of the SDF-1/CXCR4 axis in MSC by N-acetyl-L-cysteine (NAC) may improve their niche supporting activity and attenuate the HSPC aging phenotype [55]. The SDF-1/CXCR4 pathway plays an important role in pain conduction. Therefore, we speculate that hyperalgesia caused by the increased expression of CXCR4 in the early stage of diabetes is a protective mechanism. The CXCR4 inhibitor AMD3100 alleviates hyperalgesia by reducing pain transmission pathways. NAC, as an antioxidant, further increased the expression of CXCR4 and alleviated hyperalgesia through protective pathway. The exact mechanism of CXCR4 in the development of diabetic neuropathic pain merits further in-depth studies.