【作者】 陈晔凌；

【导师】 罗爱林；

【作者基本信息】 华中科技大学 ， 麻醉学， 2013， 博士

【摘要】 背景慢性疼痛由于治疗困难并带来沉重的经济和社会负担,已成为世界性的公共卫生难题。近来欧洲的一项流行病学研究显示,19%的欧洲成年人受到不同程度慢性疼痛的困扰,其中40%未能得到充分的治疗,严重影响其工作和生活质量。发达国家每年为此耗费上万亿美元,这也从另一方面说明目前治疗慢性疼痛的方法还存在很大局限性。神经病理性疼痛和炎性疼痛是慢性疼痛的两种主要类型。炎性疼痛在组织损伤和炎症得到控制后往往能恢复正常,而神经病理性疼痛则可能在基础疾病治愈后仍然存在并持续很长时间,更具有难治性。接受大手术如乳腺和胸科手术、截肢、冠脉搭桥术等的病人中,由于手术造成的神经损伤,10-15%可能在术后出现慢性疼痛。在神经损伤后,神经修复和神经病理性疼痛常常同时存在,由于二者的发生都基于神经元、胶质细胞和免疫细胞的相互作用,且有不少细胞因子均可以同时调控神经修复和慢性神经病理性疼痛,因此在寻求治疗神经病理性疼痛的新方法时也必须考虑到对神经修复和再生的影响。脑源性神经营养因子(BDNF)及其高亲和力受体TrkB对外周神经损伤后神经元的存活及轴突再生有重要意义,失去内源性BDNF的作用可影响轴突生长和髓鞘化,而局部注射BDNF可以促进神经的再生；同时BDNF也是诱发神经病理性疼痛的重要启动因子之一,在多种疼痛动物模型中都可以见到BDNF在DRG和脊髓中表达上调,在兴奋性突触传递中,BDNF可以产生和CCI模型相似的“损伤性足迹”,使用BDNF对mEPSC产生的影响也与CCI模型类似,阻断BDNF的作用能有效抑制多种疼痛模型产生的痛觉过敏现象。考虑到BDNF作用的多重性,BDNF可能成为研究神经修复和神经病理性疼痛关系的重要位点之一。本研究首先通过分级坐骨神经缩窄的方法建立大鼠分级神经病理性疼痛模型,并评价该模型中坐骨神经损伤程度与脊髓中枢敏化的情况,证实了外周神经损伤程度与动物术后疼痛行为学和脊髓中枢神经元、星形胶质细胞的活化之间存在良好的相关性。随后通过检测ATF-3和BDNF在背根神经节(DRG)中的表达情况了解分级神经病理性疼痛模型中DRG初级感觉神经元的活化状态及与BDNF表达量之间的关系。在后期的药物干预实验中,通过鞘内注射TrkB-Fc或抗BDNF抗体来阻断BDNF在中枢的作用,观察其对DRG神经元的活化、外周神经修复以及神经病理性疼痛的发生发展的影响,来探讨BDNF在神经病理性疼痛和神经修复的过程中发挥的作用。研究方法与结果1.分级慢性神经病理性疼痛动物模型的建立及评估方法：45只雄性SD大鼠(体重200～250g)随机分为5组,每组9只,用分级坐骨神经缩窄的方法建立动物模型,即：对N0、N1、N2和N4组大鼠在右侧坐骨神经中段用4-0铬肠线分别做0、1、2、4个疏松的单结环扎,并在切口皮下分别埋植4、3、2、0段5mm长的4-0铬肠线,C组为空白对照,不接受手术。在术前和术后第3、7、10、14天使用von Frey丝up-and-down法测定并计算大鼠右后爪50%缩爪阈值。术后第15天处死动物,其中30只大鼠(每组6只)取右侧坐骨神经制作石蜡病理切片,行Luxol fast blue (LFB)髓鞘染色及神经横断面髓鞘计数；取脊髓腰膨大作组织冰冻切片,进行免疫组织化学实验,观察c-Fos和GFAP在手术侧脊髓背角的表达和分布,测定手术侧脊髓背角浅层c-Fos阳性神经元计数和GFAP累积光密度值；剩余15只大鼠取右侧脊髓腰膨大组织提蛋白,行Western blot蛋白电泳测定脊髓腰膨大中c-Fos和GFAP的相对表达量。结果：大鼠右后爪50%缩爪阈值在C组和N0组术前术后无明显改变,在N1、N2和N4组大鼠术后较术前明显下降,于术后7～10天降至最低并保持稳定,术后第14天50%缩爪阈值N0>N1>N2>N4组(P<0.05)。术后第15天大鼠右侧坐骨神经横断面切片LFB染色显示,C组和N0组大鼠坐骨神经环扎远端与近端相比,无明显病理学改变,N1、N2和N4组大鼠环扎远端的神经纤维出现了不同程度的脱髓鞘改变。坐骨神经环扎远端髓鞘计数和远端髓鞘计数占近端髓鞘计数百分比(D/P ratio, DPR) No>N1、N2>N4组(P<0.001)。术后14天大鼠50%缩爪阈值与其DPR值间Spearman相关系数为0.901(P<0.001)。术后第15天大鼠手术侧脊髓腰膨大背角浅层c-Fos阳性神经元计数C、N0<N1<N2<N4组(P<0.001),GFAP累积光密度值C、No<N1<N2<N4组(P<0.05),Western blot蛋白定量结果与此相符。手术侧脊髓腰膨大背角浅层c-Fos阳性神经元计数和GFAP累积光密度值与术第14天大鼠50%缩爪阈值间Spearman相关系数分别为-0.910和-0.859(P<0.001)。2.分级慢性神经病理性疼痛模型中DRG神经元活化的检测方法：于术后第15天使用第一部分中N0、N1、N2和N4组已建立模型的大鼠的DRG组织进行实验。其中取每组各6只大鼠右侧L3-L6的DRG用4%多聚甲醛固定后作冰冻切片,用免疫荧光双标法对ATF-3/NeuN进行染色,测定DRG中ATF-3阳性神经元比例；行免疫荧光组织化学实验检测BDNF在DRG神经元中的表达和分布情况。另12只大鼠(每组3只)取右侧L3～L6DRG组织提蛋白,行Western blot蛋白电泳测定DRG中BDNF的相对表达量。结果：坐骨神经缩窄引起以ATF-3表达增多为特点的DRG神经元的活化,术后第15天手术侧L3～16DRG中ATF-3阳性神经元比例N0<N1<N2<N4组(P<0.001),与坐骨神经缩窄引起的神经脱髓鞘病变程度一致。与No组比较,N1、N2和N4组大鼠术后第15天L3-L6DRG神经元中BDNF表达升高,主要集中在小到中型神经元内,Western blot结果提示,DRG中BDNF的相对表达量也出现梯度增加的趋势,即N0、N1<N2<N4组(P<0.05)。3.干预BDNF的作用对损伤后慢性神经病理性疼痛及DRG神经元活化的影响方法：对正常成年雄性SD大鼠(体重200-220g)进行L3/4鞘内置管,5天后选择置管成功且不伴有神经损伤症状的42只大鼠进行后续实验。其中18只大鼠随机分为6组,每组3只,分别单次鞘内注射rhBDNF0、0.01、0.1、1、10、100ng,用von Frey丝测定注射后2小时、6小时、1、2、4、7天大鼠后爪50%缩爪阈值的变化；于注射后第7天处死大鼠,取脊髓腰膨大作冰冻切片,检测c-Fos和GFAP在脊髓背角的表达和分布变化。另外24只大鼠置管后随机分为对照组(C)、对照+注射TrkB-Fc组(C+TrkB-Fc)、对照+注射抗BDNF组(C+anti-BDNF)、N2模型组(N2)、N2模型+注射TrkB-Fc组(N2+TrkB-Fc)、N2模型+注射抗BDNF组(N2+anti-BDNF)、N2模型+第7天注射TrkB-Fc组(N2D7+TrkB-Fc)、N2模型+第7天注射抗BDNF组(N2D7+anti-BDNF)共8组进行实验(每组3只),分别于建立坐骨神经缩窄模型当天或术后第7天开始每日鞘内注射TrkB-Fc3μg或抗BDNF抗体5μg,测定药物对大鼠后爪50%缩爪阈值的影响,于术后第14天取大鼠手术侧坐骨神经、脊髓腰膨大和L3～L6DRG行病理学及免疫组织化学检查,观察用药后坐骨神经脱髓鞘病变程度,c-Fos和GFAP在腰膨大脊髓背角的表达,以及DRG神经元中ATF-3表达情况的变化。结果：对大鼠单次鞘内注射不同剂量的rhBDNF均可引起显著的痛觉过敏现象,痛阈下降持续至注射后7天(P<0.001),不同剂量rhBDNF引起的疼痛程度没有显著性差异；用药组脊髓腰膨大背角浅层c-Fos和GFAP表达均高于对照组(P<0.001),在0.1、1、10、100ng剂量组c-Fos和GFAP的表达高于0.01ng剂量组(P<0.05)。从坐骨神经缩窄手术当天开始鞘内注射TrkB-Fc3μg/天或抗BDNF抗体5μg/天阻断BDNF的作用可以抑制脊髓中枢神经元和星形胶质细胞的活化,减轻术后痛觉过敏的程度,同时抑制了DRG神经元中ATF-3的表达(P<0.001),但不影响环扎远端坐骨神经的脱髓鞘病变；而在坐骨神经缩窄手术7天后再注射TrkB-Fc或抗BDNF抗体则不产生上述作用,大鼠术后疼痛程度、脊髓中c-Fos和GFAP的表达以及DRG神经元中ATF-3的表达均与注射安慰剂无显著差异(P>0.05)。4.统计学分析所有数据均为计量资料,表达为均数士标准差(x±SD)。使用SPSS13.0进行数据分析,测痛数据使用重复测量数据的方差分析进行比较,其他数据使用单因素方差分析进行各组间及组内比较,双侧检验,P<0.05为统计学有显著差异。研究总结与结论1.研究总结：(1)本研究使用大鼠分级坐骨神经缩窄的方法成功建立了分级慢性神经病理性疼痛动物模型,病理学检查提示手术导致该侧坐骨神经环扎远端不同程度的脱髓鞘病变,与大鼠术后疼痛行为学改变存在显著的相关性,并伴有手术侧脊髓腰膨大神经元和星形胶质细胞不同程度的活化。(2)在分级慢性神经病理性疼痛模型中,大鼠手术侧腰段DRG神经元也伴有不同程度的活化,ATF-3表达呈现梯度性升高；免疫组织化学和Western blot结果发现BDNF在DRG初级感觉神经元中表达增加,与神经损伤、DRG活化和痛觉过敏的程度一致。(3)在坐骨神经缩窄早期使用TrkB-Fc或抗BDNF抗体阻断BDNF的作用可以抑制脊髓中枢敏化,减轻术后痛觉过敏的程度,同时抑制DRG神经元中ATF-3的表达,但不影响远端坐骨神经的脱髓鞘病变；但慢性神经病理性疼痛形成后再给予TrkB-Fc或抗BDNF抗体治疗则无上述效果。2.研究结论：大鼠分级坐骨神经缩窄的方法可以建立稳定的分级慢性神经病理性疼痛动物模型,在外周神经出现不同程度损伤的同时,伴有DRG神经元和脊髓中枢不同程度的活化,可用于研究外周神经损伤、修复和神经病理性疼痛的内在联系等。BDNF在外周神经损伤后的修复和神经病理性疼痛的发生中都具有重要作用,早期阻断BDNF在脊髓中枢的作用不仅可以抑制中枢敏化和神经病理性疼痛的发生,还抑制DRG神经元的活化,其具体机制及其下游反应仍需进一步研究。更多还原

【Abstract】 BackgroundChronic pain has become a worldwide public health problem because of the great economic and social costs associated with patient treatment. Recently, an epidemiological study in Europe showed that19%of the adult European population suffers from moderate to severe chronic pain and that40%of these patients receive inadequate analgesic treatment, which affects their social and working life. These observations not only reveal the high human and social economics cost (one trillion dollars per year in developed countries) but also highlight the current limitations of the available analgesic treatments. Neuropathic pain and inflammatory pain are the two principal types of chronic pain. Inflammatory pain arises as a consequence of tissue damage/inflammation and neuropathic pain from nervous system lesions. In contrast to inflammatory pain hypersensitivity, which usually returns to normal if the disease process is controlled, neuropathic pain persists long after the initiating event has healed. Major surgeries such as breast and thoracic surgery, leg amputation, and coronary artery bypass surgery also lead to chronic pain in10-50%of individuals after acute postoperative pain, in part due to surgery induced nerve injury.Regeneration and neuropathic pain usually coexist after peripheral nerve injury. Owing to the similar pathophysiological basis of communication of neurons, glia cells and immunocytes, the process of nerve regeneration and neuropathic pain might be modulated by some factors affecting both of them. Brain-derived neurotrophic factor (BDNF) is one of such a candidate, which is up-regulated after peripheral nerve injury to promote axonal growth and neuronal survival, and on the other hand is sufficient to trigger the onset of neuropathic pain and central sensitization by itself. Local administration of BDNF promotes axonal regeneration after axonotomy; and anti-BDNF antibody produces a dramatic reduction in the number of myelinated axons distal to the neuropathy and a decrease in the elongation of regenerating axons. In terms of excitatory synaptic transmission, both BDNF and CCI promote a similar "injury footprint"; a similar analysis of actions on delay neurons show that the actions of CCI and BDNF on mEPSC properties are similar; multiple evidences implicate microglia-derived BDNF in attenuation of Cl" mediated, GABA/glycine inhibition in the dorsal horn; inhibition of BDNF attenuate allodynia in different neuropathic pain models. Basing on the multiple effects of BDNF after peripheral nerve injury, we postulate that BDNF would be a key point to connect nerve regeneration and neuropathic pain.In the present research, we developed a rat model of graded neuropathic pain by graded sciatic constriction and assess the behavioral and pathologic changes of the sciatic nerve. We concluded that allodynia postoperatively and central activations of neurons and astrocytes in the ipsilateral lumbar spinal cord were well correlated with the degrees of peripheral nerve injury. In addition, ATF-3and BDNF expression in L3-L6DRG neurons were also detected to reveal the regenerative profile of the graded neuropathic pain model. In subsequent experiments, we evaluated the effects of rhBDNF, TrkB-Fc and anti-BDNF antibody on the development of neuropathic pain and the process of nerve regeneration, and explored the relationship between them.Methods and results1. Development and assessment of the graded neuropathic pain model in ratsMethods:45male SD rats were randomly allocated to5groups (group No, N1, N2, N4and C, n=9for each group) and received graded right sciatic constriction. Briefly, rats in group No, Ni, N2and N4received0,1,2and4loose ligations on the trunk of the right sciatic nerve with4-0chromic gut respectively; meanwhile another4,3,2and0segments of5mm-long chromic gut were put subcutaneously in the correspondent rats. No surgical treatment was taken in rats of group C. Mechanical allodynia was assessed with von Frey hairs before operation and on the3rd,7th,10th and14th day postoperatively, and the50%withdrawal threshold was calculated with up-and-down method. On the15th day postoperatively,30of the rats (n=6for each group) were sacrificed and the right sciatic nerve were dissected for transverse slicing and myelin staining with Luxol fast blue; the lumbar enlargement of the spinal cord were snap frozen at-80℃and sliced for immuno-histochemical staining of c-Fos and GFAP. The left15rats (n=3for each group) were sacrificed on the same day for quantitative analysis of c-Fos and GFAP in the ipsilateral lumbar spinal cord with Western blot.Results:The50%withdrawal thresholds of the right hind foot dropped dramatically in group N1, N2and N4from the3rd day and maintained at the lowest level from the7th10th day to2weeks postoperatively compared to those in group No and C (P<0.001). On the14th day, the50%withdrawal thresholds presented a graded manifestation of No> N1> N2> N4(P<0.05). Graded demyelination (No<Ni and N2<N4, P<0.001) was also observed in the distal area of the ligated sciatic nerve, represented digitally by the myelin sheath counts and the DPR value (the ratios of the myelin sheath counts in the distal areas to those in the proximal areas). The Spearman correlation coefficient between DPR values and the50%withdrawal thresholds on the14th day was0.901(P<0.001). Both immuno-histochemistry and Western blot analysis revealed the graded increases of c-Fos and GFAP in the lumbar enlargement of the spinal cord on the15th day (No> N1> N2> N4). The Spearman correlation coefficients between the counts of c-Fos positive neurons, the integrated optical densities of GFAP in the superficial laminae of spinal cord and the50%withdrawal thresholds on the14th day were-0.910and-0.859respectively (P<0.001). 2. ATF-3and BDNF expressions in the lumbar DRG neurons in the graded neuropathic pain modelMethods:Animals in experiment1were reused in this procedure. On the15th day postoperatively, the ipsilateral L3-L6DRGs of rats in group No, Ni, N2and N4(n=6for each group) were fixed in4%paraformaldehyde and prepared for immuno-histochemical staining of ATF-3and BDNF. The ipsilateral L3-L6DRGs in another12rats (n=3for each group) were obtained on the same day for quantitative analysis of BDNF with Western blot.Results:Sciatic constriction obviously activated the ipsilateral lumbar DRG neurons represented by elevated ATF-3expression. The ratios of ATF-3positive neurons in the ipsilateral L3～L6DRGs showed a graded increase as No<N1<N2<N4(P<0.001) as correlated with the demyelination of the sciatic nerve. BDNF was also up-regulated in the lumbar DRG neurons in a roughly graded manner in this model (No、 N1<N2<N4, P<0.05). The majority of BDNF positive cells were small to medium primary sensory neurons.3. Effects of BDNF intervention on development of neuropathic pain and nerve regeneration after sciatic constrictionMethods:Adult male SD rats weighing200～220g were prepared for intrathecal cannulation through L3/4intervertebral space.5days later,42rats which received cannulation successfully without sensory or motor deficiency were selected for subsequent experiments.18of them received single intrathecal injection of rhBDNF in different dosages of0,0.01,0.1,1,10or100ng respectively (n=3for each group). Allodynia was recorded2h,6h,1d,2d,4d and7d after injection with von Frey hairs, and on the7th day the rats were sacrificed for detection of central sensitization in the lumbar spinal cord. The other24rats were randomized into8groups (n=3for each group):group C, group C+TrkB-Fc, group C+anti-BDNF, group N2, group N2+TrkB-Fc, group N2+anti-BDNF, group N2D7+TrkB-Fc, group N2D7+anti-BDNF. We detected the effects of TrkB-Fc and anti-BDNF antibody on allodynia after sciatic constriction, the central sensitization in spinal cord, the activation of neurons in the lumbar DRGs and the demyelination of sciatic nerve in two different timing of treatment.Results:A single intrathecal injection of different doses of rhBDNF evoked significant allodynia in healthy rats, and caused over-expressions of c-Fos and GFAP in the lumbar spinal cord. Early sequestering BDNF with TrkB-Fc or anti-BDNF antibody since the day of surgery alleviated allodynia after sciatic constriction and prevented central sensitization in the spinal cord; the ATF-3expression in DRG neurons was also inhibited by early intervention with TrkB-Fc and anti-BDNF antibody (P<0.001), while demyelination in sciatic nerve was not influenced (P>0.05). No detectable effects in the above aspects could be seen when delayed TrkB-Fc or anti-BDNF antibody was given to rats undergoing neuropathic pain (P>0.05).4. Statistical analysisAll data were expressed as mean±standard deviation (x±SD). Statistical analysis was performed with SPSS13.0. Results of50%withdrawal threshold were compared by repeated measures of multiple variances, and the other data were analyzed by one-way analysis of variance. P<0.05with two-tailed test was considered statistically significant.Summary1. Main results:(1) We successfully developed a graded neuropathic pain model in rats through graded constriction of the sciatic nerve. Demyelination of the injured nerve and central sensitization in the spinal cord were assessed and proved to correlate with allodynia postoperatively. (2) ATF-3and BDNF expressions in the ipsilateral L3～L6DRG neurons increased significantly in the graded neuropathic pain model in a corresponding graded manner.(3) Intrathecal BDNF injection evoked obvious allodynia in healthy rats. Early sequestering BDNF with TrkB-Fc and anti-BDNF antibody could inhibit central sensitization in the spinal cord and activation of primary sensory neurons in DRGs, alleviating neuropathic pain without deterioration of the injured nerve.2. Conclusions:The graded model of neuropathic pain can be developed by graded constriction of the sciatic nerve on a pathologic basis of different degrees of demyelination, accompanied by a graded activation of neurons and astrocytes in the lumbar enlargement of the spinal cord. At the same time, the corresponding DRG cells can be activated with over-expressions of ATF-3and BDNF to promote nerve regeneration. Early sequestering BDNF with TrkB-Fc and anti-BDNF antibody could inhibit central sensitization in the spinal cord and the activation of DRG neurons, alleviating neuropathic pain without deterioration of the injured nerve. The intrinsic mechanism underlying BDNF and activation of primary sensory neurons and the follow-up reactivities need further investigations.更多还原