节点文献

HECT类泛素连接酶Smurf1活性调控、新功能及相关晶体结构研究

Regulation, Novel Function and Crystal Structure Research of Hect-type Ubiquitin Ligase Smurf1

【作者】 卢克峰

【导师】 贺福初; 张令强;

【作者基本信息】 中国人民解放军军事医学科学院 , 生物化学与分子生物学, 2010, 博士

【摘要】 泛素-蛋白酶体途径是真核细胞内最为重要的蛋白质降解系统,它通过选择性清除细胞内错误折叠的以及特定环境下需要降解的蛋白质,对维持细胞正常的生理功能发挥至关重要的作用。泛素化修饰的过程是在泛素活化酶E1、泛素结合酶E2和泛素连接酶E3的顺序作用下介导泛素结合到底物蛋白,最后泛素链标记的底物蛋白被蛋白酶体识别并在其中降解。整个过程中E3决定了底物蛋白识别的特异性。E3主要分为两大类:RING类E3和HECT类E3。人们对HECT类E3的功能和活性调控认识主要来自于对Nedd4家族的研究。Nedd4家族E3在TGF-β、EGF、IGF、VEGF、SDF-1以及TNF-α等生长因子或细胞因子介导的信号通路中具有重要的调控作用。作为Nedd4家族中的一员,Smurf1是在1999年筛选Smad相互作用蛋白时被发现可以降解TGF-β/BMP通路中的信号分子Smad1/5。后来又陆续发现了Smurf1的其它底物分子如RhoA、Runx2以及磷酸化形式的MEKK2等。Smurf1基因敲除和转基因小鼠骨表型分析都证明Smurf1在动物成体骨形成中发挥重要的负调控作用,强有力提示Smurf1是一个具有重要潜在价值的增强成骨细胞功能以治疗骨病的药物靶点。但是其调控分子,特别是可以调控其E3活性的调控分子一直没有被发现。2005年Science杂志的一篇关于大规模相互作用蛋白筛选的研究报道酪蛋白激酶相互作用分子CKIP-1可能与Smurf1具有相互作用,但未做深入研究,这对相互作用的生化与生理功能均不清楚。CKIP-1首先被发现是酪蛋白激酶CK2的一个相互作用蛋白,本实验室在人胎肝转录组的研究中也曾克隆鉴定到此基因,成为国际上最早克隆此基因的实验室之一,并对其功能与调控开展了近十年的研究。基于分子和细胞水平的研究表明它在细胞凋亡、细胞分化、细胞骨架调控以及募集ATM和CK2到质膜定位中发挥作用,但是CKIP-1在动物整体水平上的体内生理功能一直不清楚。我们对Smurf1与CKIP-1的相互作用及意义进行了深入的研究。首先证实CKIP-1可以特异的与Smurf1结合进而增强其E3活性,而与同家族的其它成员都没有相互作用。CKIP-1是第一个被发现的Smurf1活性增强因子。有意思的是,CKIP-1特异结合Smurf1分子中介导底物识别的两个WW结构域之间一段仅由15个氨基酸组成的连接区,而这段连接区在Smurf1与Nedd4家族其它成员之间存在序列差异,通过这种特异的靶向作用,CKIP-1上调Smurf1的E3活性、增强Smurf1与底物的结合力、进而促进Smurf1底物的泛素化降解。这是第一次揭示Smurf1的WW结构域之间的连接区也可以介导与其它蛋白的结合并在Smurf1的活性调控中发挥非常重要的功能。这种活性调控机制不同于已经发现的HECT类型E3的调控机制,因而代表了一种全新的泛素连接酶活性调节机制。在E3研究中长期存在的一个问题是为什么序列相近的E3成员具有不同的底物特异性和调控机制。我们的精细分析表明Smurf1 WW连接区的单个氨基酸的突变即可以导致E3结合调控分子及底物选择的改变,从而解释了为什么结构相似的E3如Smurf1和Smurf2,虽然序列高度同源但底物识别有所区别以及活性调控方式不同的问题。在动物生理水平,CKIP-1基因敲除后小鼠骨量会随年龄增加而升高,成骨细胞功能增强,Smurf1的活性减弱。这不但是第一次真正阐明CKIP-1的生理功能,成为数量很少的成骨细胞功能调控成员中新的一员,更为重要的是证明CKIP-1对Smurf1的活性调控可以在生理水平发生而且对于骨形成具有重要的负调控作用,从而揭示了WW连接区介导的E3活性调控机制的真实性与重要性。CKIP-1与Smurf1的相互作用特别是介导结合的Smurf1的WW连接区成为具有重大筛选意义和应用前景的骨质疏松药物的潜在靶标。为发现更多Smurf1的调节分子和底物,我们以Smurf1的WW结构域为诱饵通过酵母双杂交技术筛选到了TRAF4为其新的相互作用蛋白。TRAF4是TRAF家族成员。TRAF家族分子作为胞内接头蛋白,可以介导NF-κB和JNK-AP1通路的信号转导,进而参与骨代谢、先天性和获得性免疫、炎症应激响应等多种生物学过程。我们证实TRAF4是Smurf1的一个新的底物分子。当扩展到其他TRAF家族成员时,我们发现Smurf1对其它TRAF分子都可以进行泛素化修饰,而且逆转它们在NF-κB通路中的作用。我们的研究把Smurf1的底物范围扩展到TRAF家族,表明Smurf1在免疫和炎症反应中也具有调控作用,从而增加了人们对Smurf1新功能的认识。Smurf1的C2结构域不但可以影响其自身泛素化,而且可以结合磷脂、从而对其定位和功能也具有重要的作用。我们对此结构域进行蛋白晶体结构解析并得到了精细的空间结构信息。Smurf1-C2由两组4个β片层形成的片层组和两端的loop组成,属于C2结构域家族中的class II类型。结构分析表明,Smurf1-C2结合磷脂的模式可能不同于PKC蛋白。对C2结构域的结构解析将有助于发现针对C2结构域并影响Smurf1整体分子功能的药物。综上,我们的研究结果第一次证明Smurf1的WW结构域之间的连接区对于E3复合体的形成及E3活性的调节发挥重要功能,并揭示了CKIP-1是第一个增强Smurf1活性的重要辅助因子,而且CKIP-1对Smurf1的这种活性增强作用在动物整体水平得到了验证。我们进一步的研究把Smurf1的底物范围扩展到TRAF家族,表明Smurf1可能通过调控TRAF家族在免疫和炎症反应中发挥作用。对影响Smurf1功能的C2结构域进行了晶体结构解析,得到了其精细的空间结构信息,有助于设计特异性的针对Smurf1的小分子药物。

【Abstract】 Ubiquitin-proteasome pathway plays crucial roles in cellular physiological function through selective removal of misfolded and specific proteins. Protein ubiquitination process is catalyzed sequentially by ubiquitin-activating enzyme E1, ubiquitin conjugating enzyme E2 and ubiquitin protein ligase E3 to label substrate proteins with ubiquitin chains which promote their degradation in the 26S proteasome. Ubiquitin ligase E3s determine the substrate specificity during this process. Two main E3 families are characterized: RING (really interesting new gene) finger-type E3s and HECT domain (homologous to E6AP C-terminus)-type E3s. Knowledge of HECT type E3s is mainly obtained from research on Nedd4 family which exerts important functions in various signaling pathways including those induced by TGF-β, EGF, IGF, VEGF, SDF-1, TNF-αand other growth factors or cytokines.As a member of Nedd4 family, Smurf1 (Smad ubiquitination regulatory factor 1) was originally identified in 1999 as an E3 ligase of TGF-β/BMP pathway signaling molecules Smad1/5. RhoA, Runx2 and phosphorylated form of MEKK2 were subsequently found to be further substrates of Smurf1. Analysis of Smurf1 gene knockout and transgenic mice proved Smurf1 to be a negative regulator of bone formation, implicating that it might be a potential drug targets for bone diseases. However, whether and how its activity is regulated remains unclear.In 2005, one Science paper reported CKIP-1 (caseine kinase 2 interacting protein-1) as an interacting partner of Smurf1 in a high throughput screening for protein-protein interactions in the TGF-β/BMP signaling network. However, the biochemical and physiological functions of the interaction between CKIP-1 and Smurf1 has not been investigated. CKIP-1 was originally found to be an interacting protein of casein kinase 2. Our laboratory identified the CKIP-1 gene firstly in the establishment of human fetal liver transcriptome and secondly investigated its function and regulation for about a decade. Studies based on the molecular and cellular levels revealed that CKIP-1 is involved in regulation of cell apoptosis, differentiation, cytoskeleton reorganization and recruitment of ATM and CK2 to the plasma membrane. However, its physiological role in vivo is still unknown due to lack of research on animal levels.In this study, both the biochemical and physiological functions of the Smurf1-CKIP-1 interaction are thoroughly demonstrated. We first confirmed that CKIP-1 indeed specifically interacts with Smurf1 but not with other members of Nedd4 family, and enhances the E3 ligase activity of Smurf1. CKIP-1 functions as the first auxiliary factor to enhance the activation of Smurf1. Interestingly, CKIP-1 specifically targets the short linker region between the two WW domains of Smurf1 which mediate its interaction with substrates. The WW linker of Smurf1 consists of 15 amino acids and is distinct from that of other Nedd4 E3s. Through this manner, CKIP-1 enhances Smurf1 affinity with its substrates and promotes their ubiquitination and degradation. It is the first time to demonstrate that the short linker region between WW domains of Smurf1 could mediate the binding to its interacting protein; it is also the first time to show the importance of this short linker region for regulation of Smurf1 activity. This novel kind of regulatory mechanism is distinct from any known for HECT-type E3s. A long-standing question in understanding E3s is why closely related E3 members possess distinct substrate specificity and regulatory mechanism. Our results provide strong evidence that small differences in amino-acid sequence could accout for that specificity.We recently established the CKIP-1 gene knockout mice model. Phenotype analysis showed that CKIP-1-deficient mice undergo an age-dependent bone mass increase as a result of accelerated osteogenesis and decreased Smurf1 activity. These findings confirm the importance of the WW domains linker in complex assembly, regulating activity of HECT-type E3s and in bone mass control in vivo. CKIP-1, like Smurf1, ATF4 and Shn3, belongs to the small group of factors that regulate potential osteoblast activity. As far as we know, this is also the first time to characterize the physiological function of CKIP-1. The linker region of Smurf1 and the CKIP-1-Smurf1 interaction may thus serve as therapeutic targets for the treatment of bone diseases such as osteoporosis.In order to search for new substrates and/or regulators of Smurf1, we performed a yeast two-hybrid screening with the WW domains of Smurf1 as the bait and identified TRAF4 (TNF receptor associated factor 4) as one potential interacting partner. TRAF4 belongs to the TRAF family which functions as intracellular signaling adaptors to mediate NF-κB and JNK-AP-1 pathways and participate in bone metabolism, native and adaptive immunity and inflammation. Our further research showed that TRAF4 is a new substrate of Smurf1. When extended to other TRAF family members, we found that Smurf1 could promote their ubiquitination and reverse their negative or positive roles in NF-κB transcription activation. These results indicate novel functions of Smurf1 in immunity and inflammatory responses through targeting TRAF family members as its ubiquitination substrates.The C2 domain of Smurf1 not only affects its self-ubiquitination but also is important for its localization and function. Based on the crystal structure information of C2 We demonstrate that Smurf1-C2 belongs to classⅡtype and its molecular potential especially the regions for phospho-lipid binding is different from classical C2 domain of PKC. The knowledge of Smurf1-C2 structure will contribute to develop ment of clinical drugs that target C2 domain and affect functions of Smurf1.Taken together, we demonstrated for the first time that the short linker region between WW domains of Smurf1 exerts critical functions in complex assembly and in E3 activity regulation. CKIP-1 functions as the first auxiliary factor of Smurf1 and the interaction between CKIP-1 and Smurf1 is critical in bone mass control as indicated by gene knockout mouse models. Our research extends substrates of Smurf1 to TRAF family and indicates novel functions of Smurf1 in immunity and inflammation. Furthermore, we obtained the crystal structure of the C2 domain of Smurf1 which is beneficial for potential drug design.

节点文献中: