胰腺发育相关基因NeuroD/BETA2蛋白转导结构域的发现及功能研究

【作者】 陈静；

【导师】 卢大儒；

【作者基本信息】 复旦大学 ， 遗传学， 2006， 博士

【摘要】 第一部分蛋白质转导结构域(protein transduction domain，PTD)是一类具有跨膜活性的肽段，通常这类肽段在30个氨基酸残基范围内。神经分化因子1 (neurogenicdifferentiation 1，NeuroD1)，又叫β细胞E盒反式激活因子2(β-cell E-boxtransactivator 2，BETA2)属于B类碱性螺旋-环-螺旋(bHLH)家族，其对胰腺β细胞的分化及正常功能具有举足轻重的地位。本研究发现全长NeuroD-EGFP可高效进入肝癌细胞系PLC细胞中。进而对NeuroD蛋白转导结构域进行了分析，构建、表达并纯化了一系列NeuroD各突变体与EGFP融合蛋白，蛋白进胞实验结果表明富含碱性氨基酸的14肽，及完整的HLH结构域为NeuroD的蛋白转导结构域，HLH结构域的完整性对于该结构域介导蛋白进入细胞是必需的，且HLH PTD进入细胞是剂量依赖和时间依赖的。在与TAT PTD介导的蛋白转导效率进行对比后发现，HLH PTD转导效率与TAT PTD转导效率相当，甚至在有些细胞系中其转导效率高于TAT PTD。此外我们运用8种抑制剂，通过荧光显微镜观察和流式细胞仪检测，对HLH PTD跨膜机制进行了研究，结果表明除胞膜黏附和巨胞饮抑制剂外，细胞内吞、胞膜内陷及代谢抑制剂均未对HLH PTD跨膜效率造成显著影响。荧光和共聚焦显微镜观察结果显示HLH PTD可以高效介导与之融合的EGFP蛋白进入各种细胞系(LLC，HeLa，CHO，BHK及原代大鼠胰岛细胞)，并在胞浆和细胞核中均有定位。荧光素酶报告基因实验结果显示全长NeuroD蛋白在进入细胞后可高效启动NeuroD特异启动启动子，其启动效率显著高于对照组1.8倍(P＜0.05，n=3)。本研究是首次在一个蛋白家族中发现潜在的蛋白转导结构域，并对其转导机制和功能进行了系统的研究。而由于许多bHLH家族蛋白本身即为重要的转录因子或转录调节因子，因此也为体内／体外研究这些蛋白的功能，或进行特定疾病的治疗(可称之为“蛋白治疗”)提供了一种简便、安全的研究方法。第二部分I型糖尿病，又被称为胰岛素依赖型糖尿病，是由于机体对自身胰腺中的B细胞进行持续攻击，造成B细胞被大量破坏，从而导致胰岛素分泌的严重不足而引起的。目前国内外已有许多研究利用病毒载体介导胰腺发育相关基因(如Pdx1、Ngn3、NeuroD／beta2等)进入细胞，在体内、体外诱导各种干细胞向胰岛素分泌细胞分化达到治疗糖尿病的目的。然而，由于病毒载体存在潜在的免疫原性、病毒毒性、致突变性，这些均制约了病毒载体的进一步应用。本部分利用蛋白转导技术，将胰腺内分泌细胞发育相关的重要转录因子PDX-1及NeuroD蛋白体外导入到肝／干细胞(人肝母细胞瘤HepG2及肝卵圆细胞)中，并促使其向胰岛素分泌细胞分化。共聚焦显微镜观察结果显示PDX1-EGFP和NeuroD-EGGP蛋白可以高效进入HepG2和肝卵圆细胞中。RT-PCR结果显示经PDX1／NeuroD蛋白处理的HepG2细胞和肝卵圆细胞可以有效表达insulin，pdx-1，ngn3，neurod，pax4这些胰岛β细胞特异基因，葡萄糖转运蛋白glut2，前／胰岛素原加工酶PC2；但PC1／3未检测到表达；而肝细胞特异基因a1AT和肝卵圆细胞特异基因Thy1的表达量有所下调。荧光实时定量PCR结果显示，蛋白处理1周后的HepG2细胞其胰岛素表达量可达到人胰腺胰岛素表达量的1.14％；而处理后的肝卵圆细胞其胰岛素表达量达到小鼠胰腺胰岛素表达量的1.3％。此外细胞免疫荧光和western-blot实验也验证了处理后的细胞中内源性的胰岛素、NeumD和PDX-1的表达。从而为糖尿病胰岛移植提供了一种来源广泛的非胰岛细胞供体。此外，在糖尿病小鼠中腹腔注射NeuroD蛋白，在注射NemoD蛋白后第二天，小鼠血糖即可下降至正常水平，并维持正常血糖水平达4天，重复注射可再次降低血糖；且在此过程中小鼠体重逐渐增加。对小鼠肝脏总RNA进行RT-PCR检测，结果显示注射NeuroD蛋白3天后的糖尿病小鼠肝脏中有明显的胰岛素表达。这些结果均表明腹腔注射NeuroD蛋白可以激活肝脏中胰岛素表达，并有效降低糖尿病小鼠血糖。该方法相比起病毒或非病毒基因治疗来说不具有免疫原性、病毒毒性、致突变性、基因导入效率低、持续表达、启动表达和表达沉默等缺陷，因此为治疗I型糖尿病提供了一种安全、有效、简便易操作的治疗方案。更多还原

【Abstract】 ;Protein transduction domains (PTDs), such as HIV Tat PTD, have been widely used as delivery tools into living cells. NeuroD (neurogenic differentiation 1), or named BETA2 (β -cell E-box transactivator 2), belonging to basic helix-loop-helix family, playcentral roles during the pancreas development.Here we reported for the first time that the helix-loop-helix (HLH) domain of basic helix-loop-helix (bHLH) family was a novel type of PTD, and the intact motif was required for translocation. Efficient internalization had been obtained with HLH domains derived from bHLH protein, NeuroD/Beta2, in various cell types including human liver tumor derived PLC cells, hamster ovary derived CHO cells, human cervix derived HeLa cells, human lung derived LLC cells, and primary rat islets, and the transduction was dose- and time-dependent. Confocal experiment showed that the internalized HLH-EGFP localized at cytoplasm and nuclear. Cellular uptake of HLH PTD was barely or slightly inhibited by the metabolic, phagocytosis, clathrin- or caveolar-endocytosis formation inhibitors, but significantly and substantially reduced by heparin and macropinocytosis inhibitor, which suggested important roles of cell surface glycosaminoglycans and macropinocytosis during the protein transduction. Furthermore, luciferase assay showed that addition of NeuroD-EGFP protein gave 1.8 fold transactivation, significantly higher than that of pCMV-EGFP control (paired-T test, P<0.05, n=3).Our results demonstrated a new motif of PTD different from previous models as cationic residues cluster or amphipathic helix. The HLH domain is also the character of the bHLH family, which implies a large number of PTDs could be discovered in this family to fit diffferent purposes, and some of them could be directly recruited to penetrate cell membrane according to their crucial roles in deveploment such as NeuroD and Ngn3.Type I diabetes, also known as insulin-dependent diabetes mellitus, is the result of insulin deficiency caused by the autoimmune destruction of the insulin-producing pancreatic β-cells. And it have been reported that pancreas duct, fetal liver, ES or MSC cells could differentiate into insulin producing cells by virus-mediated expression of pancreas development related genes, such as Pdx-1、 Ngn3、NeuroD in vitro or in vivo. However, potential immunogenisis, virus toxicity and mutant induced ability restricted further application of the virus gene delivery system.Here we explored whether liver progenitor cell, such as HepG2 and liver oval cells, could differentiate into insulin-producing cells by using protein transduction of the pancreatic duodenal homeobox 1 (PDX1) and/or NeuroD proteins. Confocal experiment showed that PDX1-EGFP and NeuroD-EGFP can efficiently get into HepG2 and liver oval cells. RT-PCR test indicated that protein treated liver oval cells activated multiple beta-cell genes, and Real-time PCR test showed that the treated cells produced and stored considerable amounts of insulin. Cytoimmunofluorescence and western-blot experiments demonstrated that endogenetic insulin, PDX-1 and NeuroD could significantly express in protein treated cells. This approach offers the potential of a novel source of cells for transplantation into patients with type 1 diabetes. Furthermore, after the intraperitoneal injection of NeuroD protein, the STZ induced diabetic ICR mice returned to normal glycemia, and it remained for 4 days, repeated injection could cause the similar results again. Body weight and quantity of insulin mRNA in the liver were assayed and proved the improvement of hyper-glycemic complication after protein administration. This approach offers a much more safe, efficiency and easy-operated method for IDDM treatment.更多还原