A model for structural similarity between different SNARE complexes based on sequence relationships.
A model for structural similarity between different SNARE complexes based on sequence relationships.
复制标题
DOI：
10.1016/s0962-8924(98)01285-9
复制
发表时间：
1998-12
影响因子：
19
通讯作者：
T. Weimbs;Keith E. Mostov;Seng‐Hui Low;Kay Hofmann
中科院分区：
文献类型：
--
作者：
T. Weimbs;Keith E. Mostov;Seng‐Hui Low;Kay Hofmann
在1998年6月出版的《细胞生物学趋势》中，Götte和Fischer von Mollard总结了最近的结果，更新了介导大多数（如果不是所有）细胞膜融合事件的SNARE机制的结构和功能（见参考文献1和其中的参考文献）。突触SNARE机制的结构已经被研究得非常详细，现在很清楚的是，核心结构涉及靶膜上的两个SNARE蛋白（t-SNARE syntaxin 1和SNAP-25）和囊泡膜上的一个SNARE（v-SNARE synaptobrevin/VAMP）。这三种蛋白质通过SNAP-25的两个结构域与突触融合蛋白和小突触泡蛋白/VAMP各一个之间的卷曲螺旋相互作用形成稳定的三聚体复合物。然而，SNARE复合物在细胞器而不是质膜上的结构仍然是一个谜。在这些复合物中没有发现SNAP-25样蛋白，相反，许多小的v-SNARE样蛋白与细胞内细胞器上的突触融合蛋白同源物相互作用。在某些情况下，有证据表明，一个复合物中存在不止一个v-SNARE样蛋白1。我们想补充一下我们最近的发现，即v-和t-SNARE在进化上彼此相关，并提出一个假说来解释SNAP-25同源物的缺失和细胞内SNARE复合物中一个以上的v-SNARE的存在。分析SNARE之间的序列关系一直很困难，因为仅有的保守区域显示出七肽重复的倾向，这发生在许多不相关的蛋白质中。使用基于序列分析的方法，我们最近证明了syntaxin和SNAP-25家族的t-SNARE的进化关系2。突触融合蛋白含有一个拷贝的保守的60个氨基酸的“t-SNARE结构域”，而SNAP-25蛋白有两个拷贝。这些结构域与介导SNARE蛋白之间相互作用的卷曲螺旋结构域相同。尽管小的v-SNARE样蛋白具有共同的特征，但它们的分类更加困难。我们现在已经扩展了我们的SNARE蛋白的序列分析，使用广义的配置文件，这是来自多重比对，并包含信息的序列的哪一部分是最高度保守的，该序列的哪些区域可能容忍删除或插入3。迭代细化的轮廓搜索是检测远距离序列相似性的敏感方法，并且与蛋白质线程方法不同，轮廓搜索发现的相似性通常反映基于发散而不是收敛进化的关系。我们使用高尔基体v-SNARE Bos 1 p序列和最近报道的相关蛋白质序列创建了一个配置文件。数据库搜索与此配置文件显示了几个新的和uncharacterized，但明显相关的，酵母和线虫序列，这是包括在配置文件。将所得的精细图谱（由中心保守区和C-末端膜锚组成）用于进一步的数据库检索。令人惊讶的是，得分最高的匹配是syntaxin和synaptobrevin/VAMP家族的成员，其中最佳匹配表明具有统计学意义的关系。这
In the June 1998 issue of trends in CELL BIOLOGY, Götte and Fischer von Mollard summarized recent results in an updated picture of the structure and function of the SNARE machinery that mediates most if not all cellular membranefusion events (see Ref. 1 and references therein). The structure of the synaptic SNARE machinery has been studied in most detail, and it is clear now that the core structure involves two SNARE proteins at the target membrane (the t-SNAREs syntaxin 1 and SNAP-25) and one SNARE at the vesicle membrane (the v-SNARE synaptobrevin/VAMP). These three proteins form a stable trimeric complex held together by coiled-coil interactions between two domains of SNAP-25 and one each of syntaxin and synaptobrevin/VAMP. However, the structure of SNARE complexes at organelles other than the plasma membrane has remained more of a mystery. No SNAP-25-like proteins have been identified in those complexes, and, instead, a multitude of small v-SNARE-like proteins interact with syntaxin homologues on intracellular organelles. In some cases, evidence suggests that more than one v-SNARE-like protein is present in one complex1. We would like to add to this discussion our recent finding that v-and t-SNAREs are evolutionarily related to each other, and suggest a hypothesis to explain the absence of SNAP-25 homologues and the presence of more than one v-SNARE in intracellular SNARE complexes.There are several homologues of v-and t-SNAREs involved in fusion to different membrane compartments in the cell. It has been difficult to analyse sequence relationships between SNAREs because the only conserved regions show a propensity for heptad repeats, which occur in many unrelated proteins. Using profilebased sequence analysis, we demonstrated recently the evolutionary relationship of t-SNAREs of the syntaxin and SNAP-25 families2. Syntaxins contain one copy of a conserved ‘t-SNARE domain’of 60 amino acids, whereas SNAP-25 proteins have two copies. These domains are identical to the coiled-coil domains that mediate interactions between SNARE proteins. A classification of the small v-SNARE-like proteins has been more difficult despite the fact that they share common features. We have now extended our sequence analysis of SNARE proteins using generalized profiles, which are derived from multiple alignments and contain information on which part of the sequence is most highly conserved and which regions of the sequence are likely to tolerate deletions or insertions3. Iteratively refined profile searches are a sensitive method of detecting distant sequence similarities, and, unlike protein threading methods, similarities found by profile searches typically reflect relationships based on divergent rather than convergent evolution. We created a profile using the sequence of the Golgi v-SNARE Bos1p and recently reported sequences of related proteins. Database searches with this profile revealed several new and uncharacterized, yet clearly related, yeast and nematode sequences, which were included into the profile. The resulting refined profile, consisting of the central conserved regions and the C-terminal membrane anchor, was used for further database searches. Surprisingly, the highest-scoring matches were members of the syntaxin and synaptobrevin/VAMP families, the best matching of which indicated a statistically significant relationship. This