Protein Phosphorylation And Regulation Of Cytoskeleton I

脊椎动物神经元中神经丝蛋白（NFPs）的磷酸化调节神经丝的组装、运输和大轴突结构的稳定性。高分子量神经丝亚单位（NF-M和NF-H）羧基尾部区域KSP （lys-ser-pro）重复基序的磷酸化有助于增加轴突口径，从而提高神经纤维中脉冲的传导速度。在正常神经元中，这些KSP重复序列的磷酸化似乎在地形上局限于神经元的轴突室。尽管轴突细胞骨架系统的所有组成部分都是在细胞体中合成的，包括参与其磷酸化的激酶，但一些元素（如神经丝）的磷酸化主要发生在轴突。核周中KSP重复序列的异常磷酸化已被认为是几种神经退行性疾病中细胞体和近端轴突的病理性nf积累的机制。这些疾病包括肌萎缩性侧索硬化症（ALS）、帕金森病、伴路易体痴呆、糖尿病神经病变和伴路易体病理的阿尔茨海默病。因此，NFs的磷酸化及其调控似乎涉及多种生理功能和病理过程。然而，负责调节这种磷酸化的机制尚不清楚。本实验室的主要目的是了解神经元中NFs和其他细胞骨架蛋白磷酸化的机制。这可能为神经退行性疾病的潜在过程提供见解。我们之前的研究表明，在人类和大鼠NF-H中，KSP重复序列中的大多数丝氨酸/苏氨酸残基在体内被磷酸化，这表明脯氨酸定向激酶负责它们的磷酸化。在我们最近的研究中，我们已经表明，在体外和体内的神经元和转染细胞中，有一个丝裂原活化激酶家族（Erk1/2， SAPK）和神经元周期蛋白依赖性激酶5 （cdk5）可以磷酸化NF-M和NF-H的c端尾结构域的KSP重复序列。例如，非神经元细胞系共转染NF-M或NF-H，以及组成活性或被信号转导途径激活的激酶（MAP激酶），将诱导神经丝尾部区域KSP重复序列的强烈磷酸化。这些研究表明，体内脯氨酸定向激酶途径的激活是轴突nf磷酸化的原因。我们发现这些激酶的选择性取决于KSP基序的性质。例如，在大鼠NF-H中，大部分KSP重复序列（重复41次）作为KSPXXXK基序存在，主要通过ERK1/2激酶的激活而磷酸化，而整合素a1/b1介导的人NF-H cdk5磷酸化的激活与激活ERK1/2相比具有更高的功效。这是由于人类NF-H具有更多的KSPXK重复基序，cdk5的共识序列重复32次，而KSPXXXK重复11次。这些研究表明，不同的激酶级联可能在NFs中多个位点的顺序磷酸化中相互作用，这取决于物种和环境线索。类似的机制可能适用于其他细胞骨架蛋白，如人类tau蛋白；轴突中发现的微管相关蛋白。许多脯氨酸和非脯氨酸定向激酶已被证明在体内和体外磷酸化人类tau蛋白，在阿尔茨海默病神经元核周围成对螺旋细丝的异常聚集中产生过度磷酸化的tau蛋白。一些关于神经元磷酸化的地形调节的线索来自我们在伍兹霍尔海洋生物实验室对鱿鱼巨轴突的夏季研究。在巨大轴突的轴质中，我们已经证明了细胞骨架蛋白（微管蛋白和神经丝）与脯氨酸定向激酶（如cdc2样激酶和Erk1/2）以及非脯氨酸定向激酶（如CKI， PKA和CAMK）相关的活性多聚体复合物。这些复合物在这些轴突的细胞体中是不存在的；相反，我们发现了不同的、相对不活跃的细胞骨架蛋白和激酶复合物。这表明细胞骨架磷酸化的区室化及其调控取决于被分离到各自的区室中的磷酸化机器的性质。值得注意的是，我们通过证明类似的多聚磷酸化复合物可以通过相同的技术从大鼠脑裂解物中提取。我们继续对神经元周期蛋白依赖性激酶5 （neuroncyclin dependent kinase 5, cdk5）进行研究，以进一步了解其在神经细胞功能和调控中的作用。我们之前的研究表明，cdk5缺失小鼠表现出一种独特的表型，其特征是围产期死亡、由于异常神经元迁移导致的大脑皮层分层破坏、小脑叶状结构缺乏以及脑干和脊髓神经元变性。最近，为了挽救cdk5敲除表型，我们在cdk5缺失小鼠的大脑中以组织特异性的方式重建了cdk5过表达，并研究了其对致死表型的影响。与cdk5缺失小鼠不同，过度表达带有p35（细胞周期蛋白样cdk5激活因子）启动子的cdk5转基因（TgKO小鼠）是可存活和可生育的。由于p35主要在神经系统中表达，在这些小鼠中，cdk5的过表达仅限于神经元。相反，野生型小鼠在神经元和星形胶质细胞中表达cdk5。脑皮质层状结构和小脑叶状结构正常，脑干和脊髓未见退行性神经元。因此，我们能够逆转cdk5缺失小鼠的表型和相关的致死率。这项研究证实了cdk5在神经系统中的表达对胚胎发育和存活至关重要。在另一项研究中，我们为cdk5和MAP激酶途径之间的串扰提供了新作用的证据。Cdk5在体外和体内均通过磷酸化ERK1的上游调控因子MAP激酶激酶1 （MEK1）的Thr-286残基，下调其活性。使用缺乏明显cdk5活性的p35缺失小鼠，我们发现MEK1活性在体内受cdk5调节。这些观察结果为大脑中cdk5和MAP激酶途径之间的串扰提供了证据。这可能是调节生长、分化和凋亡细胞死亡过程中信号转导通路的另一种调节机制。

Phosphorylation of neurofilament proteins (NFPs) in vertebrate neurons regulates neurofilament assembly, transport and stability of large axonal structures. Phosphorylation of KSP (lys-ser-pro) repeat motifs in carboxy tail domains of high molecular weight neurofilament sub-units (NF-M and NF-H) contributes to increased axonal caliber, and, thereby, the conducting velocity of impulses in nerve fibers. In a normal neuron, phosphorylation of these KSP repeats seems to be topographically localized within the axonal compartment of neurons. Although all components of the cytoskeletal system of axons are synthesized in cell bodies, including the kinases involved in their phosphorylation, the phosphorylation of some elements such as neurofilaments occurs primarily in axons. The aberrant phosphorylation of the KSP repeats in perikarya has been proposed as a mechanism for pathological NF-accumulation in the cell bodies and proximal axons in several neurodegenerative diseases. These include Amyotrophic Lateral Sclerosis (ALS), Parkinson’s Disease, Dementia with Lewy bodies, diabetic neuropathy and Alzheimer's Disease with concomitant Lewy body pathology. Thus, phosphorylation of the NFs, and its regulation, seem to be involved in several physiological functions and pathological processes. However, the mechanisms responsible for the regulation of this phosphorylation are not well understood. The primary objective of this laboratory has been to understand the mechanism(s) of phosphorylation of NFs as well as other cytoskeletal proteins in neurons. This may provide insight into the processes underlying neurodegenerative disorders. Our previous studies have demonstrated that most ser/thr residues in the KSP repeats in human and rat NF-H are phosphorylated in vivo, suggesting that proline-directed kinases are responsible for their phosphorylation. In our recent studies we have shown that a family of mitogen activated kinases (Erk1/2, SAPK) and neuronal cyclin dependent kinase 5 (cdk5) can phosphorylate KSP repeats of NF-M and NF-H in their C-terminal tail domains in vitro and in vivo in neuronal and in transfected cells. For example, non-neuronal cell lines co-transfected with NF-M or NF-H, and kinases, that are either constituitively active or activated by a signal transduction pathway (MAP kinases), will induce robust phosphorylation of KSP repeats in neurofilament tail domains. These studies suggest that the activation of proline-directed kinase pathways in vivo are responsible for axonal NF-phosphorylation. We found that the selectivity of these kinases depends upon the nature of the KSP motifs. For example, in rat NF-H, where the majority of KSP repeats (repeating 41 times) are present as KSPXXXK motifs, are primarily phosphorylated by the activation of ERK1/2 kinases, while integrin a1/b1-mediated activation of cdk5 phosphorylation of human NF-H in a human cell line, has higher efficacy compared to activated ERK1/2. This is due to the fact that human NF-H has more KSPXK repeat motifs,, the consensus sequence for cdk5, repeating 32 times, than KSPXXXK (repeating 11 times). These studies suggest that different kinase cascades may mutually interact in the sequential phosphorylation of multiple sites in NFs, depending upon the species and environmental cues. A similar mechanism may be operative for other cytoskeletal proteins such as human tau; the microtubule associated protein found in axons. Numerous proline and non-proline-directed kinases have been shown to phosphorylate human tau in vivo and in vitro, producing hyperphosphorylated tau in the abnormal aggregates of paired helical filaments found in perikarya of Alzheimer neurons. Some clues as to the topographic regulation of phosphorylation in neurons have come from our summer studies on the squid giant axon at the Marine Biological Laboratory in Woods Hole. In axoplasm of the giant axon we have demonstrated active multimeric complexes of cytoskeletal proteins (tubulin and neurofilaments) associated with proline directed kinases such as cdc2-like kinases and Erk1/2 together with non-proline directed kinases such as CKI, PKA and CAMK. Such complexes are absent in the cell bodies of those axons; instead we found different, relatively inactive, complexes of cytoskeletal proteins and kinases. This suggests that the compartmentalization of cytoskeletal phosphorylation and its regulation depends on the nature of the “phosphorylating machines” that are segregated into the respective compartments. Significantly, we have followed up these studies by demonstrating that similar multimeric phosphorylating complexes can be extracted from rat brain lysates by identical techniques. We have continued our studies on neuronal cyclin dependent kinase 5 (cdk5) to further understand its role in nerve cell function and regulation. Our previous studies have shown that cdk5 null mice exhibit a unique phenotype characterized by perinatal mortality, disrupted cerebral cortical layering due to abnormal neuronal migration, lack of cerebellar foliation, and degeneration of neurons in the brain stem and spinal cord. Recently, in an attempt to rescue the cdk5 knockout phenotype, we have reconstituted cdk5 overexpression in a tissue specific manner in the brain in cdk5 null mice and examined its effect on the lethal phenotype. Unlike the cdk5 null mice, cdk5 that overexpress the cdk5 transgene with a p35 (the cyclin like cdk5 activator) promoter (TgKO mice) were viable and fertile. Since p35 is expressed primarily in the nervous system, in these mice cdk5 overexpression was limited to neurons. In contrast, wild type mice express cdk5 in neurons as well as astrocytes. The cerebral cortical layering pattern and cerebellar foliation was normal, and no degenerating neurons were seen in the brain stem and the spinal cord. We were thus able to reverse the phenotype and associated lethality of cdk5 null mice. This investigation confirms that cdk5 expression in the nervous system is critical for embryonic development and survival. In a separate study we have provided the evidence for a novel role for cross talk between cdk5 and MAP kinase pathway. Cdk5 down regulates the activity of MAP kinase kinase 1 (MEK1), the upstream regulator of ERK1 by phosphorylating its Thr-286 residue both in vitro and in vivo. Using p35 null mice which lack appreciable cdk5 activity, we show that MEK1 activity is regulated by cdk5 in vivo. These observations provide evidence for a cross talk between cdk5 and the MAP kinase pathway in the brain. This may serve as another regulatory mechanism that modulates signal transduction pathways during growth, differentiation and apoptotic cell death.