作者姓名：相 扬   
  论文题目：G蛋白耦联受体介导的神经生长锥导向   
  作者简介：相扬，男，1977年2月出生，1999年9月师从于中国科学院上海生命科学研究院蒲慕明研究员，于2003年8月获博士学位（硕博连读生）。   
     
  中 文 摘 要   
  　

神经系统最令人吃惊的特征之一在于神经细胞之间以及神经细胞与其它组织之间联系的精确性。在人脑中，约1011个神经元彼此之间建立了约1014个突触联系。在发育过程中建立的如此巨大数量的精确联系是神经系统高效的接收、处理、储存和传递信息的结构基础。

伟大的神经解剖学家Ramon y Cajal在固定的组织切片中观察到了神经生长锥结构，并于1893年首次提出由靶细胞释放的扩散性信号能够形成浓度梯度，指导生长锥的生长方向.  Roger Sperry 1940～60年代的一系列经典实验证明成年神经系统的精确拓扑联系（topographical specificity）是由于发育过程中的高度精确性的轴突生长和突触形成造成的，并提出Chemoaffinity hypothesis，指出突触后神经元可以分泌化学信号指导突触前神经元生长锥寻找正确的靶细胞，并且突触前和突触后的神经元可以通过特异的化学标记(chemical labels)互相识别并形成突触联系。Sperry的实验在支持Cajal假说的基础上，进一步明确发育过程中轴突的生长及突触的形成具有内在决定性，并且有力地反驳了当时占主流地位的由Paul Weiss于1930～40年代提出的resonance hypothesis，后者认为轴突的生长和突触的形成大多是一些随机事件，成年脑中的精确联系是通过对随机形成的不恰当联系进行选择地消除形成的。Sperry的实验预见了生长锥导向因子的存在，并开启了生长锥导向分子机制研究的时代。

人们应用生化纯化、组织共培养技术以及在线虫、果蝇中的遗传学研究发现并鉴定了四类神经生长锥导向因子。它们分别是Netrin、Slit、Semaphorin和Ephrin。但是，神经细胞之间复杂的联系提示有其他导向因子存在并且发挥作用。

从广义上讲，神经生长锥导向是一种细胞趋化运动 (chemotaxis)，即细胞对外界特定化学导向因子的浓度梯度发生反应而进行定向的运动。常见的趋化运动还包括白细胞朝向炎症部位的趋化运动、阿米巴的趋化运动、恶性肿瘤细胞的转移以及神经元的迁移等。

阿米巴和白细胞的趋化运动作为研究细胞趋化运动的一个标准模型得到了广泛的关注。一个有趣的现象是：阿米巴和白细胞的趋化运动全部由细胞表面的G蛋白耦联受体介导。各种形式的的白细胞趋化都能被pertussis toxin阻断，表明Gi蛋白耦联受体介导了这一过程。而已发现的神经元迁移及轴突导向因子Netrin、Slit、Semaphorin和Ephrin则利用单次跨膜受体来传递信号。除了受体的差异外，神经元和白细胞迁移的环境，速度以及迁移时细胞的形态均不同，因此人们认为神经元迁移及轴突导向可能采用了完全不同与白细胞趋化的细胞及分子机制。

Chemokine作为一类分泌分子，通过G蛋白耦联受体传递信号。Chemokine的功能最初被鉴定为指导白细胞定向运动至炎症部位，也即白细胞的趋化因子。现在已知这类分子除广泛参与血液和淋巴系统的功能外，还参与器官发生、干细胞迁移、肿瘤转移等等。SDF-1是最早被鉴定的一种chemokine, 能够通过其G蛋白耦联受体CXCR4介导白细胞趋化。1998年，两个实验组分别敲除了小鼠SDF-1基因及其受体CXCR4基因，发现除了预期的血液和淋巴系统功能缺陷外，发育中的小脑EGL(external granule-cell layer)中的颗粒细胞在胚胎期发生了提前迁移。这一结果首次证明chemokine还参与神经系统的发育。随后的实验证明SDF-1由发育中小脑的脑膜表达并分泌，而CXCR4受体表达于EGL中的颗粒细胞， SDF-1可以成浓度梯度来吸引小脑颗粒细胞。像小脑一样，脑膜分泌的SDF－1能够指导海马齿状回颗粒细胞的迁移。表明SDF－1在发育中的神经系统中发挥广泛的作用。

因此，SDF-1，CXCR4信号除介导白细胞趋化，还指导了发育中小脑和海马神经元的迁移，说明神经元迁移和白细胞趋化可以共用相同的分子机制，同时说明G蛋白耦联受体参与了神经细胞运动的导向。

神经元迁移和轴突导向共用相同的导向因子，而已经鉴定的神经轴突导向因子Netrin-1，Slit，Semaphorin，Ephrin都通过单次跨膜受体传递信号，这促使我们去研究G蛋白耦联受体介导的轴突导向。在发育中的小脑，SDF-1的表达局限于脑膜，CXCR4在EGL的颗粒细胞表达，提示SDF-1可以扩散成梯度形式影响颗粒细胞的轴突导向。我们采用体外的生长锥转向分析技术研究了SDF-1对小脑颗粒细胞生长锥的导向作用。为平行实验，我们同时研究了GABAB受体, 另外一种G蛋白耦联受体在爪蟾神经生长锥导向中的作用。我们的实验结果如下：

1.本文首次证明G蛋白耦联受体能够介导神经生长锥导向。SDF-1的浓度梯度能够吸引或排斥培养的小脑颗粒细胞的生长锥。同样的，GABAB受体的激动剂baclofen的浓度梯度也能够吸引或排斥爪蟾脊髓神经元的生长锥，导向反应的极性(吸引或排斥)由胞内cGMP水平控制；

2.信号转导机制的研究表明，SDF-1和baclofen激活了生长锥表面的G蛋白耦联受体，通过Gi将信号传递给PLC，进而激活PKC和IP3信号通路。通常条件下，PKC通路介导了生长锥排斥反应，而IP3通路介导了吸引反应，由于PKC通路占据主导，SDF-1和baclofen排斥生长锥。升高胞内cGMP水平后，PKC通路和IP3通路均介导了吸引反应，SDF-1和baclofen吸引生长锥。

本文的主要结果提示：

1.G蛋白耦联受体能够介导阿米巴趋化，白细胞趋化，神经元迁移，生长锥导向等各种细胞运动形式，说明细胞运动机制高度保守；

2. SDF-1能够吸引或排斥生长锥，说明在体外chemokine可以指导生长锥找寻特异的靶细胞。当然，SDF-1是否能够作为一种新的神经生长锥导向因子发挥作用还需要体内实验证据；

3. SDF-1的受体CXCR4介导了白细胞趋化，神经元迁移，生长锥导向，提示这些运动形式共用相同的分子机制。因此，本文中阐明的PLC-PKC/IP3信号通路信号通路将有助于了解SDF-1介导的其他类型细胞运动的信号机制；

胞内第二信使水平调节几乎所有的轴突导向因子的作用，一个有趣的问题是：第二信使，例如本文中提到的cGMP能否调节神经元迁移和白细胞趋化的方向和速度。绝大多数的报道指出chemokine吸引白细胞，只有极少数情况下chemokine排斥白细胞。在某些自身免疫疾病和过度炎症反应中，排斥过多的白细胞具有重要的临床意义。同样，一些神经退行性疾病的干细胞移植治疗能否见效，很大程度上依赖于移植的神经干细胞能否迁移到达正确的靶位置，因此了解神经元迁移及其调节机制将大大改善这类疾病的治疗现状。一个简单的想法是研究cGMP对SDF-1浓度梯度引起的白细胞趋化和神经元迁移是否具有调节作用。

关键词：  G蛋白耦联受体、 生长锥导向、  chemokine

  
  Nerve growth cone guidance mediated by G protein–coupled receptors
Yang Xiang

ABSTRACT   
          The most remarkable characteristic of the nervous system is the precision and specificity of the connections formed between nerve cells or between nerve cells and their targets.  In human brain, there exist 1014 synapses between 1011 neurons, which formed the structural basis for the nervous system to receive, process, store and transmit information with high efficacy.  How these amazingly precise connections are formed during early development is a key question for neuroscientists.

More than one hundred years ago, the great Spanish neuroanatomist Ramon y Cajal discovered, in the fixed brain tissue, a structure called growth cone, a tiny and subtle tip of the nerve fiber.  In 1893, Cajal proposed, for the first time, that the growth cone may sense and respond to the chemical gradient in the environment, thus helping the nerve fiber to navigate toward its correct target.  Cajal also suggested that some chemical substance was secreted by the target cells and diffused to form a concentration gradient.  The classic experiment carried out by Roger Sperry in 1940-1960 provided solid evidence to support Cajal’s hypotheses.  Sperry showed that the regenerated retinal axons always innervated the original sites of termination in the optic tectum.  This experiment strongly disprove the resonance hypotheses proposed by Paul Weiss, who suggested that the nerve growth and synapse formation are largely random events, and the precision of connections between nerve cells emerged by selectively eliminating the inappropriate connections, only at later developmental stage.  Sperry’s studies also implied the role of some molecules in selective axonal growth and synapse formation and open an era of molecular biology of axon guidance.

In the following 20 years, scientists identified 4 families of molecules that can guide growing axons, by using genetic screen in Drosophila and C. elegans, and by using protein purification in vertebrate.  They are Netrins, Semaphorins, Slits and Ephrins.  However, there is also substantial evidence suggesting the existence of other guidance molecules.

The axonal guidance is one form of chemotaxis, which means that a cell (or part of the cell, e.g. a nerve growth cone) migrates in a specific direction by sensing the concentration gradient of guidance factors in the environment.  Chemotaxis also includes the migration of leukocytes toward inflammation sites, migration of amoeba toward food, metastasis of malignancy cancer cells and neuronal migration. Different forms of cell migration are believed to share similar mechanisms: they sense the spatial concentration gradient by comparing the concentration difference between the front and the back, they determine the migration direction by asymmetrically localizing some molecules in the cytosol, and they drive the directional movement by asymmetrical actin polymerization/depolymerization.   Thus, the studies of amoeba chemotaxis and leukocyte chemotaxis have served as a model for scientist to understand how cells migrate in response to a concentration gradient.  However, there exists some difference between amoeba/leukocyte chemotaxis and axon guidance/neuronal migration: amoeba/leukocyte chemotaxis can be blocked by pertussis-toxin, a specific blocker for G protein, suggesting that G protein-coupled receptors (GPCRs) sense and transmit guidance signals, while the established neuronal guidance cues, including Netrins, Semaphorins, Slits and Ephrins, signal through single transmembrane receptors.  Furthermore, the morphology, rate and environment of migrating neurons/axons are different from migrating amoeba/leukocyte, implying that different cellular and molecular mechanisms are used.

The idea that neuronal migration/axon guidance employs different mechanism from amoeba/leukocyte chemotaxis was recently disproved by an elegant study showing that SDF-1 (stromal-derived factor 1), a class of chemokine known to chemoattract leukocyte through its GPCR, CXCR4, is involved in neuronal migration during early development.  Knock out the gene cxcr4 or sdf-1, can cause premature migration of cerebellar granule cells.  Another study showed that slit, a repellent for migrating neurons/axons, can inhibit the leukocyte chemotaxis triggered by SDF-1.  These studies prompted us to ask whether GPCR can also mediate axonal guidance.  We addressed this question by studying SDF-1 induced turning of growth cones of cerebellar granule cells, using a growth cone turning assay.  This may have some physiological relevance since SDF-1 is secreted by cerebellar meninges and may function in a gradient fashion to affect axon guidance.  To generalize our findings to other GPCRs, we also examined the effect of activation of GABAB receptors in the growth cone of cultured Xenopus spinal neurons.