【作者】 王伟；

【导师】 姜晓丹； 王伟民；

【作者基本信息】 南方医科大学 ， 神经外科学， 2010， 博士

【摘要】 皮质发育畸形(Malformations of cortical development, MCD)是指皮层发育过程异常和参与形成皮层的细胞异常所造成的脑畸形,是药物难治性癫痫的常见病因。自上世纪70年代Taylor最早描述局限性皮质发育不良病理变化的病理特点以来,皮质发育畸形逐渐得到广大神经科学工作者的重视。特别是上世纪90年代初期磁共振技术在临床的普遍应用,使得皮质发育畸形成为癫痫特别是难治性癫痫患者外科治疗中非常重要的一大类疾病。围绕着皮质发育畸形的研究是目前癫痫研究的热点。癫痫外科常见的皮质发育畸形主要分为：结节性硬化,局限性皮质发育不良,偏侧巨脑回,胚胎发育不良性神经上皮肿瘤,节细胞胶质瘤,神经节细胞瘤等。尽管20多年来针对上述疾病的研究取得了长足进步,有关的症状学、神经影像、电生理、分子生物学和外科治疗特点等仍然有许多问题尚未完全明了本研究拟通过通过回顾性分析,研究局限性皮质发育不良的临床、结构影像学、代谢影像学和病理学特点,通过手术病例的随访了解局限性皮质发育不良不同亚型的神经影像和病理的关系,以及与手术疗效的关系。回顾性分析胚胎发育不良性神经上皮肿瘤的神经影像、手术方式和治疗的选择与预后的关系。通过基因芯片检测的方法研究Ⅱb型局限性皮质发育不良、单纯型胚胎发育不良性神经上皮肿瘤和节细胞胶质瘤(WHOⅠ级)的基因表达差异。本研究论文主要包括四部分内容。皮质发育畸形(Malformations of cortical development, MCD)是指皮层发育过程异常和参与形成皮层细胞的异常所造成的脑畸形,是药物难治性癫痫的常见病因。目前MCD的具体发病率并不明确,有学者认为大约占所有癫痫患者的3%,而在儿童难治性癫痫患者中则占20/%-45%左右。大约75%的MCD患者表现为癫痫,而MCD所导致的癫痫通常情况下药物治疗效果差,因此是癫痫外科治疗中仅次于海马硬化的第二大类致痫性疾病。皮质发育畸形分类的分类方法众多,目前临床上比较常用的是Barkovich分类法,主要根据皮层发育的过程分为四大类：1由于神经元或胶质细胞异常增生或凋亡导致的皮质发育畸形；2由于神经元迁徙异常导致的皮质发育畸形；3由于皮层形成异常(包括晚期迁徙异常)导致的皮质发育畸形；4其他皮层发育畸形。每一大类又包含若干疾病。而癫痫外科常见的皮层发育畸形主要包括局限性皮质发育不良、结节性硬化、节细胞胶质瘤、胚胎发育不良性神经上皮肿瘤和神经节细胞瘤等。其中局限性皮质发育不良是药物难治性癫痫中最为常见的疾病,虽然目前发现其影像学检查有一定的特点,但是不同的亚型之间是否存在特异性差别,尚不得而知；由于病理上变化较轻患者的影像学上也不甚明显,如何提高这部分患者的影像诊断水平、其手术后癫痫控制的长期疗效如何,目前都还没有明确答案。至于胚胎发育不良性神经上皮肿瘤,至今还尚不明确其影像学特点、术前诊断、术后长期结果、以及是否能对癫痫发作不频繁的患者进行单纯随访；此外,不同类型的皮层发育畸形基因表达是否存在差异、皮层发育畸形患者导致瘤样增生的分子生物学机制,也是目前尚没有合理解释的问题。由此,本课题从事了致痫性皮层发育障碍性疾病的相关研究,力求在明确相关发病机制、临床特点的基础上,为临床更系统有效地诊断、治疗致痫性皮层发育障碍探寻客观依据。第二部分致痫性局限性皮质发育不良不同亚型的症状学、结构影像学、代谢影像学和影响手术疗效的因素目的：研究局限性皮质发育不良的临床、结构影像学、代谢影像学和病理学特点,通过手术病例的随访了解局限性皮质发育不良不同亚型的神经影像和病理的关系,以及与手术疗效的关系。材料方法：回顾性分析29例经手术病理证实的局限性皮质发育不良患者。术前评估包括症状学,磁共振,脑电图和PET。全部29例患者均在脑电监测条件下行癫痫灶切除术,其中12例在全麻唤醒麻醉下手术,8例合并海马硬化的患者行前颞叶切除术。研究结果采用SSPS13.0软件处理,统计方法采用t检验和x2检验,当P<0.05时认为有统计学差异。结果：术后病理按Palmini分类,Ⅱb型局限性皮质发育不良13例,Ⅱa型局限性皮质发育不良11例,Ⅰb型局限性皮质发育不良3例,Ⅰa型局限性皮质发育不良2例。病例分析结果显示,不同亚型局限性皮质发育不良的临床发作、过去史和家族史无明显差异,局限性皮质发育不良Ⅰ型的发病年龄(P<0.05)和手术年龄(P<0.001)早于局限性皮质发育不良Ⅱ型。磁共振(MR)特点：灰白质边界不清(29例),灰质增厚(23例),Flair或T2序列的高信号(13例)的病例最后病例报告均为Ⅱb型局限性皮质发育不良。T2序列脑回间弓状纤维的消失(29例)。在进行了PET检查的26例患者中,有24例显示病变区域葡萄糖低代谢。未见病变葡萄糖代谢增高。手术疗效：术后随访15～53个月,平均31个月。Eagel分级Ⅰ级72.4%(21/29),EagelⅡ级20.7%(6/29),EagelⅢ级6.9%(2/29)；其中Ⅱa型局限性皮质发育不良EagelⅠ级占81.8%(9/11),Ⅱb型局限性皮质发育不良EagelⅠ级占84.6%(11/13),2例Ⅰa型局限性皮质发育不良术后Eagel分级分别为Ⅱ级(50%)和Ⅲ级(50%),3例病例分别呈Ⅰb型局限性皮质发育不良术后EagelⅠ级(33%)、EagelⅡ级(33%)和EagelⅢ级(33%)各1例。EagelⅠ级在局限性皮质发育不良Ⅱ型为20例(20/24,83.3%),明显优于(p<0.005)局限性皮质发育不良Ⅰ型1例(1/5,20%)。合并海马硬化的双重病理组8例患者术后7例(87.5%)EagelⅠ级,而单纯性局限性皮质发育不良21例患者术后14例(66.7%)EagelⅠ级,二者相比无统计学差异(p=0.262)。位于颞叶的局限性皮质发育不良17例患者有15例(88.2%)术后达到EagelⅠ级,而非颞叶组的12例患者、其中仅6例(50%)术后达到EagelⅠ级。颞叶的局限性皮质发育不良手术疗效优于非颞叶组(p<0.05)。12例在全麻唤醒麻醉手术后EagelⅠ级10例(83%),而17例常规手术后Eagel I级11例(64.7%)。术后暂时性偏瘫3例,2例患者在3～6个月完全恢复,1例轻偏瘫；轻度运动性失语1例,6周逐渐恢复。无手术死亡和其他持续的神经功能障碍。结论：1、弓状纤维消失是影像学诊断局限性皮质发育不良的比较敏感的证据；2、T2和Flair高信号是诊断局限性皮质发育不良Ⅱb型的可靠依据；3、相对于Ⅰ型局限性皮质发育不良,Ⅱ型局限性皮质发育不良手术疗效较好；4、手术疗效与是否合并海马硬化无关；5、位于颞叶的局限性皮质发育不良疗效优于由于颞叶以外的病变；6、PET检查有助于局限性皮质发育不良的术前诊断；7、全麻唤醒下切除导致致痫性皮质发育不良手术疗效优于常规全麻手术。第三部分胚胎发育不良性神经上皮肿瘤的神经影像学特点和治疗方法选择目的：探讨胚胎发育不良性神经上皮肿瘤的临床表现、影像学诊断和治疗。材料方法：回顾性分析10例经病理证实和3例临床高度怀疑的胚胎发育不良性神经上皮肿瘤的临床表现、辅助检查、病理特点和中期随访结果。术前评估包括症状学、磁共振、脑电图和PET。皮层脑电监测下病变全切和扩大切除9例,部分切除+离断手术1例；全麻唤醒手术和常规麻醉手术各5例。3例患者抗癫痫药物治疗下行单纯随访。结果：MR特点：13例患者全部接受了磁共振扫描检查。对磁共振图像分析发现,病变全部以皮层为基底；病变呈三角形5例(38.46%),四方形5例(38.46%),不规则形3例(23.07%)；T1序列病变呈低信号3例(23.08%),呈等信号10例(76.92%)；全部13例病变(100%)在T2和Flair序列呈高信号；有9例(69.23%)出现T2序列假囊性征病变。周围水肿1例(7.69%)；有2例(15.38%)增强后点状强化；仅1例(7.69%)在Flair序列病变呈环状；病变自发出血1例(7.69%)。行CT扫描(含PET-CT)8例,显示病变呈低密度3例(37.5%)；等密度5例(62.5%)；病变内钙化2例(25%)；病变临近局部颅骨凹陷3例(37.5%)。8例患者行PET扫描,示踪剂为18氟-脱氧葡萄糖(18FDG),其中2例加作碳11标记的蛋氨酸PET,全部8例(100%)患者均显示病变低代谢；低代谢区域与病变一致的6例(75%)；低代谢区域大于病变的2例(25%)。手术疗效：手术切除组随访15～50个月,平均29个月。癫痫控制EngleⅠ级7例(70%),EngleⅡ级2例(20%),EngleⅢ级1例(10%)。扩大切除4例术后全部无癫痫发作,单纯病变切除的4例患者3例术后无发作,1例患者术后3月癫痫复发。1例病变位于运动前区的患者术后对侧轻偏瘫,3个月后完全恢复。余无特殊并发症。7例全切病例患者于术后6、12、24个月行MR复查,未见病变复发。其中1例部分切除患者分别于术后3、6、12、24、36个月行MR复查,病变与术后1周MR结果相比无变化。高度疑似患者3例,分别随访35、37和41个月,癫痫发作均<2次/年,患者无特殊不适,多次MR复查病变无变化。结论：1、磁共振T2序列假囊性征是胚胎发育不良性神经上皮肿瘤特征性表现；2、PET低代谢有助于胚胎发育不良性神经上皮肿瘤与其他高级别胶质瘤鉴别；3、胚胎发育不良性神经上皮肿瘤的手术目的是控制癫痫,癫痫的控制与病变及其周围致痫灶的切除程度有关；4、对于位于重要脑功能区的胚胎发育不良性神经上皮肿瘤,全麻唤醒手术有助于患者功能区的保护；5、对于高度怀疑胚胎发育不良性神经上皮肿瘤的患者,如病变位于重要脑功能区,癫痫发作不频繁,可行单纯药物治疗,密切随访,而不应当强行切除第四部分局限性皮质发育不良Ⅱb型、单纯型胚胎发育不良性神经上皮肿瘤和节细胞胶质瘤(WHOⅠ级)基因表达差异的基因芯片研究目的：探讨局限性质发育不良(Ⅱb型)、节细胞胶质瘤(WHO分级Ⅰ级)和单纯型胚胎发育不良性神经上皮肿瘤的基因表达差异。主要方法及结果：经手术病理证实的局限性质发育不良(Ⅱb型)、节细胞胶质瘤(WHO分级Ⅰ级)和单纯型胚胎发育不良性神经上皮肿瘤组织标本各Ⅰ份,标本切除后立即置于-173℃的液氮罐内冻存。TRIZOL法提取三个样本的总RNA,经1.2%甲醛变性胶电泳鉴定RNA完整性,试剂盒纯化RNA。RNA甲醛电泳显示28S、18S两条RNA条带,28S条带明显较亮,且12倍于18S条带,OD260/OD280>2.0,表明从3个标本中成功提取了RNA。将三个标本的RNA反转录为双链cDNA并以生物素标记和片段化,与预制的商业化Agilent人基因组基因芯片(GeneChip Human U133plusZ)杂交、并进行芯片扫描和分析。在对3个样品进行基因表达分析时,只有当差异表达的基因同时满足P<0.05和fold change>2时才认为结果有意义。将有差异表达的基因进行分析,发现胚胎发育不良性神经上皮瘤(dysembryoplastic neuroepithelial tumor, DNET)基因、神经节细胞胶质瘤(ganglioglioma, GG)基因与生物合成有关的差异表达基因为123个,与分子功能有关的差异表达基因为88个,与细胞成分有关的差异表达基因为37个(p<0.01)；DNET与FCD基因相比,与生物合成有关的差异表达基因为112个,与分子功能有关的差异表达基因为76个,与细胞成分有关的差异表达基因为33个(p<0.01)；GG与FCD相比,与生物合成有关的差异表达基因为81个,与分子功能有关的差异表达基因为57个,与细胞成分有关的差异表达基因为40个(p<0.01)。将上述结果结合基因的细胞功能进一步分析发现,与血管形成有关的BTG1基因和与Wnt信号通路有关的卷曲蛋白6(frizzled-6, FZD6)和卷曲蛋白7(frizzled-7,FZD7)基因在神经节细胞胶质瘤中表达下调明显；TSC1和TSC2基因在ⅠⅠb型局限性皮质发育不良、单纯型胚胎发育不良性神经上皮肿瘤和节细胞胶质瘤(WHOⅠ级)的表达差异不明显；纤维母细胞生长因子受体3(fibroblast growt h factor receptor3, FGFR3)基因在胚胎发育不良性神经上皮肿瘤表达明显上调；人类表皮生长因子3(Human epidermal-growth-factor receptor3, ERBB3)基因在神经节细胞胶质瘤中明显表达上调；而部分与钠钙离子通道和γ-氨基丁酸(Gamma-aminobutyric acid, GAB A)信号传导有关基因在GG和DNET表达下调明显。结论：1、与血管形成有关的基因BTG1和与Wnt信号通路有关的基因FZD6、FZD7在节细胞胶质瘤中表达下调,可能与节细胞胶质瘤内胶质细胞成分过渡增殖导致的成瘤性有关；2、TSC1和TSC2基因在Ⅱb型局限性皮质发育不良、单纯型胚胎发育不良性神经上皮肿瘤和节细胞胶质瘤(WHOⅠ级)的表达差异不明显；3、首次发现FGFR3基因在胚胎发育不良性神经上皮肿瘤上调明显,可能与胚胎发育不良性神经上皮肿瘤的皮层增厚有关；4、ERBB3基因在节细胞胶质瘤中明显表达上调,可能与其可以恶性变有关；5、与钠钙离子通道和GABA信号传导有关基因在GG和DNET表达下调明显,考虑与其高度致痫性有关。更多还原

【Abstract】 Malformations of cortical development (MCDs) are brain malformations that result from abnormalities affecting the normal processes of cortical development and involving cells that participate in formation of the cerebral cortex under normal circumstances. It is not surprising, therefore, that MCDs are often associated with recurrent seizures, and that these seizures always be difficult controlled with medication. Since the term "focal cortical dysplasia" (FCD) was first used by Taylor in 1971 to describe a histological abnormality seen in surgical specimens from 10 patients with epilepsy, the MCDs gradually gained widespread attention in neuroscience workers. As the widespread clinical application of magnetic resonance image (MRI) in 90s last century, varies subtypes of MCDs became the second large category of epilepsy related surgery. The clinical and basic investigations for MCDs are becoming the "hot spot" in epilepsy research.The most commonest subtypes of MCDs be confronted with epilepsy surgery include tuberous sclerosis, FCD, hemimegalencephaly, dysembryoplastic neuroepithelial tumor, ganglioglioma and gangliocytoma. Although great improvement had been achieved in the last two decades, there are still lots of questions about the semiology, nuroimage, electrophysiology, molecular biology and surgical treatment need to be answered.In this research, we studied the clinical feature, structural imaging, metabolic imaging and pathological characteristics of FCDs with retrospective analysis method, and studied the MR- pathologic correlationn and long-term follow up results of different subtypes of FCD. Retrospective analysis of neuroimage, surgical approach, treatment options and prognosis about dysembryoplastic neuroepithelial tumor were proceeded. The gene expression differences in FCDⅡb type cortical dysplasia, dysembryoplastic neuroepithelial tumor simple type and ganglioglioma (WHOⅠ) were studied with gene microarray methods. The main content of this article are summarized as follows.Part 1 Background for researchMalformations of cortical development (MCDs) and their classification, common dysplastic lesions related to epilepsy surgery, and recent problems are briefly reviewed.MCDs are brain malformations that result from abnormalities affecting the normal processes of cortical development and involving cells that participate in formation of the cerebral cortex under normal circumstances. It is not surprising therefore that MCDs are often associated with recurrent seizures, always be difficult controlled with medication. The precise incidence of MCDs is not known. It is estimated that 3% of all epilepsy patients and 25～40% of intractable or medication-resistant childhood epilepsy are attributable to MCDs, and that at least 75% of patients with MCDs will have epilepsy. Now, the MCDs are the second largest group just less than hippocampal sclerosis in epilepsy surgery candidates.There are several classification systems for MCDs. The classification scheme published by Barkovich et al in 1996 was accepted in worldwide, which includes four categories as below:1 malformations due to abnormal neuronal and glial proliferation or apoptosis; 2 malformations due to abnormal neuronal migration; 3 malformations due to abnormal cortical organization (such as late neuronal migration); 4 MCDs with no otherwise classified. Each category is re-divided in several subtypes further. The most common subtypes of MCDs in epilepsy surgery including focal cortical dysplasia, tuberous sclerosis complex, dysembryoplastic neuroepithelial tumor, ganglioglioma and gangliocytoma.Part 2 The semiology, structural imaging, metabolic imaging and the surgical outcome in different subtypes of focal cortical dysplasia (FCDs)ObjectiveTo explore the imaging characteristics and long-term outcome after epilepsy surgery of epileptogenic focal cortical dysplasia (FCD).Material and MethodsTwenty nine cases of FCDs confirmed by pathology were retrospectively analyzed. The preoperative evaluation included ictal symptom, MRI, scalp EEG and PET. Epileptogenic foci resection was applied with intra-operative EEG monitoring in all 29 cases,12 cases of which were under awaking craniotomy, anterior temporal lobectomy was applied in 8 cases which had accompanied hippocampal sclerosis. Datas are analyzed by SPSS13.0 software packet, and using t-test and x2 inspection. A value of p< 0.05 was considered statistically significant.ResultsThere are no statistics difference in ictal semiology, family history and past history in different subtypes of FCDs. The age of seizure onset (P<0.05) and the age of operation (P<0.001) are earlier in FCDⅠthan FCDⅡ. According to Palmini classification, pathology results confirmed FCDⅡb in 13 cases, FCDⅡa in 11 cases, FCDⅠb in 3 cases and FCDⅠa in 2 cases respectively.The characteristics of MR image included blurring of gray-white matter junction (23 cases), gray matter thickening (23 cases), blurring of fiber tracts connecting adjacent gyri (U-fibers) (23 cases) and increased signal intensity on T2 or Flair sequence (13case). All of them above are belong to the characteristics of FCDⅡb. PET scan was applied in 26 cases,24 of which were found as the decreased glucose metabolism.The follow-up period was 15～53 months, mean as 31 months. EagelⅠachieved in 21 (72.4%), EagelⅡin 6 (20.7%) and EagelⅢin 2 (6.9%) of all cases respectively. EagelⅠachieved in 9 of 11(81.8%) of FCDⅡa. EagelⅠachieved in 11 of 13(84.6%) of FCDⅡa. EagelⅡ(50%)and EagelⅢ(50%)were achieved in each of the 2 FCD la cases. EagelⅠ(33.3%), EagelⅡ(33.3%) and EagelⅢ(33%) were achieved in each of the 3 FCD Ib cases. The seizure control rate was favorable (P<0.005) in FCDⅡ(EagelⅠin 20 of 24cases,83.3%) than in FCDⅠ(EagelⅠin 1 of 5cases,20%). The seizure control rate was much more favorable (P<0.05) in FCDs located in temporal lobe group (EagelⅠin 15 of 17cases, 88.2%) than in no-temporal group (EagelⅠin 6 of 12cases,50%). EagelⅠachieved in 10 of 12 cases (83%) in the awaking craniotomy group and 11 of 17 cases (64.7%) in conventional surgery group. EagelⅠachieved in 7 of 8 cases (87.5%) in the dual-pathology group and 14 of 21 cases (66.7%) in non-dual-pathology group, the result of which had no significance for statistics.Temporary paralysis occurred in 3 patients, and 2 of them recovered in 3～6 months. Mild motor aphasia occurred in 1 patient case and gradually returned to normal level in 6 weeks. There was no mortality.Conclusion1 The blurring of fiber tracts connecting adjacent gyri (U-fibers) on MR image are sensitive evidence for the diagnosis of FCD;2 Increased signal intensity on T2 and Flair sequence are reliable evidence for the diagnosis of FCDⅡb;3 The seizure control rate is more favorable in FCDs located in temporal lobe group than in extra- temporal group;4 The surgical results in the dual-pathology group and non- dual-pathology group have no significance for statistics;5 The seizure control rate is more favorable in FCDⅡgroup than in FCDⅠgroup;6 PET scan contributes to preoperative diagnosis of FCDs;7 Epileptogenic foci resection with awaking craniotomy technique is superior to conventional surgery for higher seizure control rate.Part 3 Neuroimaging characteristics and treatment option for dysembryo-plastic neuroepithelial tumorObjectiveTo explore the clinical manifestations, neuroimage, treatment and surgical outcome of dysembryoplastic neuroepithelial tumor(DNET).Materials and MethodsThe semiology, neuroimage, pathological data and follow-up results of 10 postoperative and 3 suspected cases of dysembryoplastic neuroepithelial tumor were retrospectively analyzed. Preoperative evaluation included symptom study, MRI, EEG and PET. Epileptogenic foci resection was applied with intra-operative EEG monitoring in all 10 cases. Total lesion resection and the extended lesion resection were achieved in 9 patients, and "partial resection+dis-connection" in 1 case. Awaking craniotomy technique was applied in 5 cases. Three of suspected cases just were treated with antiepileptic drugs (AEDs) and follow-up. ResultsMagnetic resonance (MR) scan were performed in all 13 patients. The features of MR image included as below. The cortex was involved in all 13 cases. The lesions looked like triangle in 5 (38.46%), square-like in 5 (38.46%) and irregular shape in 3 (23.07%) cases respectively. The lesion had decreased signal intensity on T1 sequence in 3 (23.08%) cases, and iso-signal intensity on T1 sequence in 10 (76.92%) cases. All of 13 cases (100%) had increased signal intensity on T2 and the Flair sequence. MR scans disclosed pseudocyst appearance on T2 sequence in 9 (69.23%) cases. Peri-lesion edema observed in 1 of case (7.69%), and spontaneous bleeding occured in 1 of case (7.69%). One case had ring-sign on Flair sequence. Nodular enhanced occurred in 2 (15.38%) cases after gadolinium administration.Preoperative plain CT (including PET-CT) studies dada were obtained in 8 cases. Hypodense appearance observed in 3 (37.5%) and isodense appearance observed in 5 (62.5%) cases. CT scans disclosed calcifications in 2 (25%) cases. A deformity of the overlying skull was apparent in 3 (37.5%) cases. Preoperative PET-CT studies dada were obtained from 8 cases. Reduced glucose metabolism occurred in all 8 cases, in which the area of reduced glucose metabolism was consistent with the lesion in 6 (75%) and beyond the lesion in 2 (25%) cases.The follow-up period was 15～50 months in surgical group, mean as 29 months. EagelⅠachieved in 7 (70%). EagelⅡin 2 (20%) and EagelⅢin 1 (10%) respectively. Complete seizure control achieved in all 4 cases of extended lesion resection group, and 3 of 4 (75%) cases of lesion resection group achieved seizure-free, seizure relapse in 1 of patient at the 3rd month after the operation. Hemiparesis occurred in one patient with DNET located in the premotor cortex and completely recovered with in 3 months. No persistent complications occurred.Three cases of the suspected dysembryoplastic neuroepithelial tumor had been followed up for 35,37 and 41 months respectively. The seizure showed less than twice per year in all 3 of cases. Repeated MR scan proved that the lesions were stable and no headache and other complain occurred in all cases.Conclusion1 Pseudocyst appearance on T2 sequence of MR image is the characteristic performance of DNET.2 Reduced glucose metabolism of the lesion on PET scan contributes to preoperative diagnosis of DNET.3 The only purpose of surgery for DNET is seizure control. The seizure-free could only achieved in those patients who had been totally removed the DNET itself and nearby epileptogenic foci.4 Epileptogenic foci resection in patients suffered from DNET with awaking craniotomy technique is superior to the one with conventional surgery for brain function protection.5 For the suspected cases with DNET, if the lesions involved eloquent area and seizure attack is not frequent, it’s the better to wait and see with AEDs rather than reluctant lesion resection.Part 4 Gene microarray study on gene expression difference among focal cortical dysplasia (FCD)Ⅱb, dysembryoplastic neuroepithelial tumor (DNET) simple type and ganglioglioma (GG, WHO I)ObjectiveTo explore the gene expression difference among focal cortical dysplasia (FCD)Ⅱb, dysembryoplastic neuroepithelial tumor (DNET) simple type and ganglioglioma (GG, WHOⅠ).Materials and Methods Three of surgical specimens came from FCDⅡb, DNET simple type and GG respectively. After surgical removal, the samples were frozen and stored in a liquid nitrogen jar (-173℃) until use. The total RNA was isolated using TRIzol LS Reagent with Phase Lock Gel-Heavy tubes (Eppendorf), following the manufacturer’s instructions. RNA integrity was determined using a nano-drop spectrophotometer. RNA Integrity and gDNA contamination Test were performed by Denaturing Agarose Gel Electrophoresis. The intensity of the 28S band was about twice than that of the 18S band, and OD260/OD280>2.0. RNA labeled with the GeneChip Expression 3=amplification one cycle target labeling system. Each RNA sample was hybridized to Agilent GeneChip System. After the hybridization, the GeneChips were washed and stained on a fluidics station (Agilent) and then scanned in a confocal scanner (Agilent GeneArray Scanner) according to the Agilent GeneChip Expression Analysis Manual. The GeneChip Human U133plusZ (Agilent Inc.) comprised more than 40,000 of well-characterized human genes. Comparisons were performed among three samples, requiring at least P<0.05 and twofold change for a significant difference (Student’s t-test).ResultsAll the different expressed genes were under GO analysis, the results of which showed that there were 123 of genes overlap related biological process,88 of genes overlap related molecular function and 37 of genes overlap related cellular component in DNET vs GG. There were 112 of genes overlap related biological process,76 of genes overlap related molecular function and 33 of genes overlap related cellular component in DNET vs FCD. There were 81 of genes overlap related biological process,57 of genes overlap related molecular function and 40 of genes overlap related cellular component in GG vs FCD.According to the specific features, it was found that significantly lower level of expression of angiogenesis-related gene BTG1 and Wnt signaling pathway related genes FZD6 and FZD7 appeared in GG. No significant different expression of TSC1 and TSC2 genes were observed in GG, DNET and FCD samples. Overexpression of FGFR3 gene was observed in DNET. Significant overexpression of FGFR3 gene was observed in GG sample. Significant lower-expression of genes involved in sodium ionic channels, calcium ionic channels and GABA signaling were observed in DNET and GG samples.Conclusion1 Significant lower expressions of angiogenesis-related gene BTG1 and Wnt signaling pathway-related genes FZD6 and FZD7 were observed in GG samples, which might be related to the over-proliferation of glial cells causing tumorigenicity of GG.2 No significant different expressions of TSC1 and TSC2 genes were observed in GG, DNET and FCD sample.3 Overexpression of FGFR3 was observed in DNET, which might be the reason for gray matter thickening of DNET.4 Significant overexpression of FGFR3 was observed in GG sample, which might be related to the malignant transformation of GGs.5 Significant low-expressions of genes involved in sodium ionic channels, Calcium ionic channels and GABA signaling were observed in DNET and GG samples, which might be related to the epileptogenesis of those lesions.更多还原