Analgesic Effect Of Cation-chloride Cotransportor Inhibitor In Postoperative Pain Model Of Rat

Postoperative pain treatment has been widely recognized as an important issue during the past two decades, and great progresses have been made on the understanding and treatment for postoperative pain, including of research on pathological mechanism, exploitation of novel analgesics, and usage of new analgesic techniques. But there are still41%patients suffering moderate to severe pain on the first day following operation, and about15%patients on the fourth day following operation. Patients suffering moderate to severe pain following operation are more likely to develop chronic pain syndrome, as reported, the incidence of chronic pain after some surgical procedures, such as inguinal herniorrhaphy, cholecystectomy, breast surgery, thoracotomy, and limb amputation is about10%-60%. Therefore, postoperative pain treatment is still an important clinical issue for adequate management, especially research and development of novel analgesic agents without side effect on consciousness and sensation to nonpain stimulation.Currently, inadequate treatment of postoperative pain is related to uncomplete knowledgement of pathophysiological mechanism. Accordingly it is significant to do research on postoperative pain mechanism, plasticity of nociceptive pathway, control of sensitation of nociceptive pathway, and exploitation of multimodal analgesia. The pathophysiological mechanism of postoperative pain is complex and unique, and is different to that of inflammation and nerve injury. The incision model of rat built in the mid-1990s by T.J. Brennan’s group, which the plantar skin, underlying fascia and flexor digitorum brevis muscle are incised, then the skin is closed with sutures. The somatic injury of incision model has some similarities to that of patients undergoing surgery, just like clinical postoperative pain mechanical hypergesia will continues several days then resolves gradually in incisional rats, therefore incision model of rats is a suitable animal model for research of postoperative pain mechanism and treatment, and so far widely be used.Two types of hyperalgesia, primary hyperalgesia and secondary hyperalgesia are caused by surgical injury. Primary hyperalgesia occurs for both thermal and mechanical stimuli applied to damaged tissues close to the site of injury. The underlying mechanism involves peripheral sensitization of primary afferent nociceptors by algogenic mediators locally released. Although inflammation certainly participates in incisional pain, its cause and its role are different from these in other models of tissular injury. The time course of inflammatory cytokine secretion in a rat model of postoperative pain does not coincide with the onset of mechanical hyperalgesia, and the bradykinin antagonists have no analgesic effect on incisional pain, so these results support that mechanism of incisional pain is distinct with inflammational pain. In contrast, ischemia may play an important role and local acidosis parallels postoperative pain behaviors and hyperalgesia. Low PH activates several ion channels susceptible to transduce pain, i.e. acid-sensing ion channels, vanilloid receptors, purinergic receptors, and potassium channels. Surgical injury also induces hypersensitivity in adjacent tissues, called secondary hyperalgesia and observed only for mechanical stimuli applied to uninjured tissues surrounding the wound. Secondary mechanical hyperalgesia is considered a consequence of central sensitization and results from enhanced response of dorsal horn neurons to peripheral inputs, with magnitude and duration related to the degree of tissue injury. The development and maintenance of secondary hyperalgesia are caused by different mechanisms because secondary hyperalgesia is mediated via peripheral afferents, but when fully developed it becomes independent of peripheral activity originating from the wound. Sensitization of central pain pathways involves both facilitated excitatory synaptic responses and depressed inhibitions, thereby amplifying responses to noxious and innocuous stimuli. There is plasticity in perioperative pain pathway, such as sensitization of sensory neuron. Better understanding of spinal plasticity, including of excitatory and inhibitory pathway, contributes to development of more selective and effective analgesic agents for perioperative pain.’Balanced analgesia’should provide effective perioperative pain relief with the combination of both analgesic and antihyperalgesic drugs.Cation-chloride co-transporters (CCCs) are a class of membrane proteins that transport sodium, potassium, and chloride ions into and out of cells. CCCs, firstly found in1980, belongs to solute carrier family12(SLC12) gene family. The CCC family in mammals consists of nine members encoded by the genes Slcl2al-9. The CCC proteins are glycoproteins with apparent molecular weights of120-200kDa. Seven out of the nine CCCs described so far are plasmalemmal ion transporters, and in terms of function they fall into three categories:two members are Na-K-2C1cotransporters (NKCC1and NKCC2), one is a Na-Cl cotransporter (NCC), and four are K-Cl cotransporters (KCC1-4). The physiological roles of the remaining two CCCs (CIPl/SLC12a8and CCC9/SLC12a9) are yet unknown. Sodium-potassium-chloride co-transporter1(NKCC1) and potassium-chloride co-transporter2(KCC2) are two major CCC subtypes that regulate chloride balance within the nervous system. Under physiological conditions, NKCC1transports C1-into the cell whereas KCC2transports Cl-out of the cell. The increasing researches have shown that CCCs play a fundamental role in spinal adjustment of pain pathway through inhibiton mediated by GABAa receptors and glycine receptors.γ-amino-butylic acid (GABA) and glycine are major inhibitory neurotransmitter in central nervous system, and play an important role on controlling of neuronal excitability, information processing, neural plasticity, and neural network synchronization and so on. In spinal cord, GABAergic neurons and its fibers have its own local circuit. GABA plays a role through its receptor, and there are three subtypes of GABA receptors, respectively GABAA, GABAB, and GABAc. GABAA and GABAc belong to ionotropic receptors, become to Cl-, HCO3-channel after activation. GABAA receptor is a ligand-gated ion channel that extensively distributes in central nervous system, coupled with chloride channel. When activation, chloride channel open and flow direction depend on chloride concentration between internal and external of neuron. In the vast majority of mature neurons, KCC2can maintain low intracellular chloride concentration by transport chloride out of the cell, therefore chloride equilibrium potential (Eci) is lower than neuronal membrane potential (Vm). After activation of GABAA receptor, chloride ion flow into the cell, causing postsynaptic membrane hyperpolarization, therefore produce Inhibitory postsynaptic potential (IPSP). But during the neurogenesis and early postnatal period, NKCC1expression dominates in immature neuron, and KCC2expresses weakly, which making chloride ion gathering into the cell. When GABAA receptor activated, chloride ion flows out the cell, causing postsynaptic membrane depolarization, and producing excitatory effect. The excitatory effect of GABA plays an important role on synaptic formation and neural plasticity etc. Glycine receptor (GlyR) is alike to GABAa receptor, is also a ligand-gated ion channel, primarily distributes in spinal cord and brainstem of the central nervous system. Glycine receptor (GlyR) is a pentamer constituted by aand βsubunits, and chloride channel is constituted by five subunits. As GABA, following glycine receptor activation in immature neuron, chloride channel opens and chloride ions flow out the cell, causing postsynaptic membrane depolarization, and producing excitatory effect. Moreover in mature neurons, chloride ions flow into the cell following glycine receptor activation, causing postsynaptic membrane hyperpolarization, therefore produce Inhibitory postsynaptic potential (IPSP).Transmembrane chloride concentration is the pivotal factor to decide the polarity of response induced by GABAa receptor and glycine receptor. If intracellular chloride concentration is low, chloride ions flow into the cell following GABAA and glycine receptor activation, causing postsynaptic membrane hyperpolarization, as classical postsynaptic inhibition induced by GABA in the central nervous neurons. If intracellular chloride concentration is higher, chloride ions flow out the cell following GABAA and glycine receptor activation, causing postsynaptic membrane depolarization, as excitory response induced by GABA in the primary afferent neurons in dorsal root ganglion. The expression and activity of NKCC1and KCC2protein are pivotal factors to change of transmembrane chloride concentration in neuron. Increasing reports have proved the influence of NKCC1and KCC2protein on pain modulation respectively in different inflammation models and neuropathic pain models. In the rat models of inflammation induced by capsaicin or formalin injection, intrathecal NKCC1inhibitor Bumetanide can decrease pain behavior and inhibit hyperalgesia. In the rats with chronic neuropathic pain caused by contusion to spinal cord, Bumetanide can decrease pain behavior and the hyperalgesia evoked by heat stimulation, and NKCC1expression is transiently increase at14d following injury accompany with downregulation of KCC2expression.In our previous study, we found that intrathecal furosemide reduced both primary and secondary hyperalgesia in the rat incision model, suggesting that changes in CCC function do indeed contribute to postoperative pain and sensitization. But abnormal exciting behavior was seen after intrathecal furosemide in intact and incision rats which might be caused by blocking of KCC2. Since bumetanide is a specific NKCC1inhibitor which has about500-fold greater affinity for NKCC1than for KCC2, in the present study, we evaluated the analgesic effect of intrathecal bumetanide in a rat incision model of postoperative pain by measurement of evoked and nonevoked pain behaviors, and changes in NKCC1and KCC2protein expression in dorsal root ganglion and dorsal horn of spinal corn at different time of incisional rats by multiple labeled immunofluorescence techniques. Our goal was to investigate the mechanism of postoperative pain modulation from NKCC1and KCC2, and mechanism of CCC inhibitors, finally gain insights into the possible effectiveness of NKCC1inhibitors for postoperative pain treatment.PART ONE Effect of intrathecal bumetanide on pain behavior in the incision model of ratObjective To investigate the analgesic effect of intrathecal specific NKCC1inhibitor bumetanide in a rat incision model of postoperative pain. Methods36adult male Sprague-Dawley rats, incision model was made using method of TJ Brennan’s group. Briefly, under inhalation anesthesia, the plantar aspect of the left hindpaw was sterilized and a1-cm longitudinal incision made through the skin and fascia, starting0.5cm from the proximal edge of the heel and extending toward the toes. After incising to the fascia, an underlying flexor digitorum brevis muscle was encountered. This underlying flexor muscle was elevated, divided and retracted. The skin was then apposed using two4-0antibacterial absorbable sutures. Rats were randomly divided into control group and bumetanide group (N=18) for behaviou measurement. For bumetanide group, the bumetanide solution100μg/20μl was intrathecally injected prior to incision. For control group, only aCSF vehicle20μl was intrathecally injected prior to incision. Refer to method of De la Calle, bumetanide was injected intrathecally under inhalation anesthesia. The back was shaved, and the drug was injected into the intrathecal space at the L3-L4interspace (with the positive indication being tail movement). A30-gauge needle (Hamilton7748-16) connected to a50μl syringe (Hamilton705LT) was employed, with the rat in the elevated lumbar position. Then rats were divided into three subgroups (N=6) for cumulative pain score, thermal withdrawal latency, and mechanical pain threshold measurement respectively. Rats were placed on an elevated plastic mesh floor (grid8×8mm) under a clear plastic cage for cumulative pain score measurement, and with an angled magnifying mirror the incised foot could be viewed. Following20min adaptation, each rat was closely observed during a1-min period repeated every5min for1h. Depending on the position in which the foot was found, a0,1,or2was given. If full weight bearing of the foot was present and the wound was blanched or distorted by the mesh, a score of0was recorded. If the area of the wound touched the mesh without blanching or distorting, a1was given. If the foot was completely off the mesh a score of2was recorded. The sum of the12scores (0-24) obtained during the1h session was used to assess pain in the incised foot. Thermal pain threshold was assessed using the hind paw withdrawal test to a thermal noxious stimulus. Briefly, a radiant heat source beneath a glass floor was aimed at the plantar surface which was0.5mm besides the wound of the hind paw. Paw withdrawl latency time (WLT) was recorded as the average of three trials with intervals greater than5min. For mechanical pain threshold measurement, after two days’ adaptation to the measuring cage, measurement was started on the third day using modified up-down method of Dixon. First, the rat was placed in the measuring cage for more than half an hour to acclimate. Then, baseline pain behavior was evaluated. To assess the hindpaw withdrawal threshold to a mechanical stimulus, von Frey filaments of logarithmic incremental stiffness (0.6-26.0g) were used, and50%probability withdrawal thresholds were calculated. The experiment was not begun until the baseline was above20(the experiment being performed on the next day after this criterion was satisfied). The three pain behavior tests were performed separately before incision (con),2h post-incision (2h), and second to sixth days post-incision (D2-D6). The results of50%withdrawal threshold and cumulative pain score are presented as median and interquartile range, and non-parametric tests were used. The data between testing days within group were analyzed with Friedman test, followed by wilcoxon matched pairs test, while the Mann-Whitney test was used between groups. The results of WLT were presented as mean±SD, and two-way repeated-measures analysis of variance (ANOVA) was used to compare the differences between groups. A P<0.05was considered statistically significant.Results Both in control and bumetanide group, the cumulative pain scores were significantly increased following incision (P<0.001). But after incision, the scores of bumetanide group were significantly lower than the control group ((P=0.004(D1)0.003(D2),0.003(D3),0.096(D4),0.007(D5),0.317(D6))) and continue to fifth day following incision. The thermal withdrawl latencies of bumetanide group were significantly higher than the control group (F=721.207, P<0.001).In incision control group, from2h to sixth day after incision, the thermal withdrawl latencies were significantly lower than basement (P<0.01). But in the bumetanide group, only from2h to the third day after incision, the thermal withdrawl latencies were significantly lower than basement (P=0.000,0.006,0.015). Both in control and bumetanide group, the50%withdrawal thresholds were significantly decreased from2h to sixth day following incision (all P<0.05). The basement thresholds of two groups showed no difference (P=0.902). But from2h to fifth day after incision, the thresholds of bumetanide group were significantly higher than the control group (P=0.003,0.003,0.002,0.026,0.022).Conclusion Intrathecal administration of high selective NKCC1inhibitor bumetanide could decrease postoperative cumulative pain scores, increase thermal pain threshold and mechanical pain threshold in the incision model of rats. It is suggested that bumetanide could reduce both rest pain and evoked pain postoperatively, could be a promising novel analgesic agent for postoperative pain.PART TWO Changes in NKCC1and KCC2expression in the dorsal root ganglion and dorsal horn of spinal cord in the incision model of ratsObjective To investigate the NKCC1and KCC2expression and its changes in the dorsal root ganglion and dorsal horn of spinal cord in the incision model of ratsMethods30adult male Sprague-Dawley rats, randomly divided into control group,2h,2d,3d,6d following incision group (N=6). Incision model was made using method of TJ Brennan’s group. Briefly, under inhalation anesthesia, the plantar aspect of the left hindpaw was sterilized and a1-cm longitudinal incision made through the skin and fascia, starting0.5cm from the proximal edge of the heel and extending toward the toes. After incising to the fascia, an underlying flexor digitorum brevis muscle was encountered. This underlying flexor muscle was elevated, divided and retracted. The skin was then apposed using two4-0antibacterial absorbable sutures. At different timepoint before and after incision(con,2h,2d,3d,6d post-incision), under deep anesthesia with5%sodium pentobarbital (about100mg/kg), rats were perfused through the ascending aorta with phosphate-buffered saline (PBS) followed by paraformaldehyde at4℃. The L4-L5spinal cord segments and bilateral L5dorsal root ganglia were carefully removed. Each DRG was post-fixed for25min and spinal cord segments for14h in the same fixative. Fixed tissues were then sequentially dehydrated in10%,20%, and30%sucrose over three successive nights at4℃. Five group samples were arranged on the same blocks in optimal cutting temperature embedding medium (Tissue-Tek, Leica, Nussloch, Germany) and mounted on the same slides after sectioning. Transverse spinal sections (16μm) were cut on a cryostat (Leica CM1950) and processed for NKCCl or KCC2immunostaining. Briefly, slices were blocked in5%ChemiBLOCKERTM(Millipore2170, USA) and0.5%Triton X100in PBS for1hour at room temperature. Primary antibodies were diluted in the same buffer. For DRG immunostaining, fixed tissues were incubated in primary antibody (goat anti-NKCCl, Santa Cruz,#sc-21545; goat anti-KCC2, Santa Cruz,#sc-19420;1:20) for2h at room temperature. For spinal cord immunostaining, slices were incubated over2nights at4℃with the same antibody (1:100). After three washes in PBS, sections were incubated for90min with FITC-conjugated donkey anti-goat IgG (Jackson ImmunoResearch705-095-003,1:200), and4’,6-diamidino-2-phenylindole (DAPI,0.3μM) was used to stain the nuclei at the last10min. Slides were analyzed using a Leica microscope equipped with a Leica DFC425C CCD camera and Leica LAS software, and immunopositive neurons were counted for the whole dorsal root ganglion and dorsal horn by specially assigned person. No immunosignal was observed upon omission of primary antibody and preadsorption control using a fivefold excess of blocking peptides according to the manufacturer’s specification. The data were presented as mean±SD, group means were compared by analysis of variance (ANOVA) or Welch test followed by Levene’s post hoc test or Tamhane’s T2test for pair-wise comparisons. A2-tailed test was used and P<0.05was considered significant.Results In intact rats, NKCC1expressed weakly in DRG and DH. After incision, NKCC1increased in bilateral DRG, especially ipsilateral DRG, and the NKCC1immunoreactivity (IR) focus on the neuron membrane. The NKCC1IR positive neuron numbers of the ipsilateral DRG increased following incision (F=10.848, P=0.001). In ipsilateral dorsal horn, NKCC1IR increased in the deep layers, and also little NKCC1IR was found in the substantial gelatinosa (SG) whereas no was showed in control rats. The NKCC1IR positive neuron numbers of the ipsilateral DH increased following incision (F=219.996,P=0.000). No significant KCC2IR was found in DRG. In control rats, KCC2was predominantly expressed in the deep laminae of the DH, and expression appeared much stronger than NKCC1expression. There was little detectable KCC2expression in the SG. After incision, KCC2IR decreased in the deep laminae of the ipsilateral DH, although the KCC2IR increased in SG, and the number of KCC2IR neurons in the deep laminae was significantly lower than controls following incision (F=14.634,.P=0.000)Conclusion NKCC1expression was showed in the dorsal root ganglion and deep layers of dorsal horn of spinal cord in the intact rats. KCC2expression was showed in the dorsal horn of spinal cord in the intact rats, primarily in the deep layers. NKCC1expression was upgraded in the spinal level following incision, whereas KCC2decreasing in the deep laminae of dorsal horn. It suggested that NKCC1and KCC2participated in the mechanism of postoperative pain and hyperalgesia, and would be a promising therapeutic strategy for postoperative pain management.PART THREE NKCC1and KCC2expression in the A-fiber and C-fiber afferent neurons in the spinal cordObjective To investigate NKCC1and KCC2expression in the A-fiber and C-fiber afferent neurons in the dorsal root ganglion and dorsal horn of spinal cord.Methods30adult male Sprague-Dawley rats, randomly divided into control group,2h,2d,3d,6d following incision group (N=6). Incision model was made using method of TJ Brennan’s group. Briefly, under inhalation anesthesia, the plantar aspect of the left hindpaw was sterilized and a1-cm longitudinal incision made through the skin and fascia, starting0.5cm from the proximal edge of the heel and extending toward the toes. After incising to the fascia, an underlying flexor digitorum brevis muscle was encountered. This underlying flexor muscle was elevated, divided and retracted. The skin was then apposed using two4-0antibacterial absorbable sutures. At different timepoint before and after incision(con,2h,2d,3d,6d post-incision), under deep anesthesia with5%sodium pentobarbital (about100mg/kg), rats were perfused through the ascending aorta with phosphate-buffered saline (PBS) followed by paraformaldehyde at4℃. The L4-L5spinal cord segments and bilateral L5dorsal root ganglia were carefully removed. Each DRG was post-fixed for25min and spinal cord segments for14h in the same fixative. Fixed tissues were then sequentially dehydrated in10%,20%, and30%sucrose over three successive nights at4℃. Five group samples were arranged on the same blocks in optimal cutting temperature embedding medium (Tissue-Tek, Leica, Nussloch, Germany) and mounted on the same slides after sectioning. Transverse spinal sections (16μm) were cut on a cryostat (Leica CM1950) and processed for immunostaining. Briefly, slices were blocked in5%ChemiBLOCKERTM (Millipore2170, USA) and0.5%Triton X100in PBS for1hour at room temperature. Primary antibodies were diluted in the same buffer. For DRG immunostaining, fixed tissues were incubated in primary antibody for2h at room temperature. For spinal cord immunostaining, slices were incubated over2nights at4℃with the same antibody. The following antibodies combination and ultimate dilution were used:goat anti-NKCCl, Santa Cruz,#sc-21545; goat anti-KCC2, Santa Cruz,#sc-19420; diluted1:20for dorsal root ganglion and1:100for spinal cord; rabbit anti-neurofilament200(NF200)(Sigma, N4142), diluted1:400. After three washes in PBS, sections were incubated for90min with corresponding secondary antibodies combination away from light in room temperature. The fluorescence conjugated secondary antibodies used included of: FITC-conjugated donkey anti-goat IgG (Jackson ImmunoResearch705-095-003,1:200), AMCA conjugated donkey anti-goat IgG (Jackson ImmunoResearch705-155-147,1:300), TRITC conjugated donkey anti-rabbit IgG (Jackson ImmunoResearch711-025-152,1:300), FITC-conjugated lectin B4(Sigma L2895,3.3g/ml), and4’,6-diamidino-2-phenylindole (DAPI,5g/ml). Slices were washed with PBS for four times keeping in dark place, air drying, mounted with anti-fluorescence decay mounting liquid. Slides were analyzed using a Leica microscope equipped with a Leica DFC425C CCD camera and Leica LAS software. No immunosignal was observed upon omission of primary antibody and preadsorption control using a fivefold excess of blocking peptides according to the manufacturer’s specification.Results NKCC1expression was showed clearly in almost all neuron membranes of the DRG of rat at2days following incision by double staining of NKCC1and DAPI. NKCC1expression was showed in all C fiber afferent neurons in DRG by double staining of NKCC1and IB4—marker of C fiber afferent neurons, and also NKCCl expression was showed in most A fiber afferent neurons in DRG by double staining of NKCC1and NF200--marker of A fiber afferent neurons. Double staining of NKCC1and NF200in dorsal horn showed NKCC1expression in the A fiber neurons in the deeper laminae and laminae I. Double staining of NKCC1and IB4in dorsal horn showed that NKCC1mainly in deep layers where as IB4in substantia gelatinosa, only weak co-staining in substantia gelatinosa layer. Following incision, NKCC1positive signal increased in both deep and superficial layers, and more co-staining was seen in substantia gelatinosa layer. Double staining of KCC2and NF200in dorsal horn showed significant KCC2expression in some A fiber neurons in the deeper laminae. Double staining of KCC2and IB4in dorsal horn showed significant KCC2expression in deep layer and little expression in superficial layer; as IB4was only express in substantia gelatinosa layer some co-staining signal was seen.Conclusion There is NKCC1expression in A and C fiber afferent neurons in DRG, and A fiber neurons in dorsal horn of spinal cord. There is significant KCC2expression in some A fiber neurons in deep dorsal horn of spinal cord, and little KCC2expression in C fiber afferent neurons in substantia gelatinosa layer. These results suggest that NKCC1and KCC2play a role on spinal pain pathway modulation and formation of hyperalgesia and allodynia through different mechanism.