COMPONENTS AND KINETICS IN EXOCYTOSIS

我们继续对胞吐过程中膜融合的分子机制进行生物物理和生物化学的联合研究，胞吐作用是胰岛素、神经递质和免疫球蛋白分泌以及受精过程中的关键步骤。首先，用抗哺乳动物SNAP-25的单抗进行表达文库筛选，鉴定出一种紫斑链霉菌SNAP-25的同源物。克隆了海胆卵巢SNAP-25的单一亚型。该克隆编码212个氨基酸的蛋白质(GenBank登录号：AF061750)。当与来自几个物种的SNAP-25比对时，该SNAP-25在整个氨基酸序列中高度保守。重组蛋白已在大肠杆菌中以融合载体的形式表达，纯化后作为定量评估皮质囊泡中SNAP-25含量的标准。初步数据显示，每个小泡存在约1000个拷贝的SNAP-25。其次，从大脑皮层囊泡中免疫沉淀了合成素和VAMP，并用基质辅助激光解吸/电离(MALDI)质谱仪分析了肽图。免疫沉淀法从海胆中分离纯化得到的合成素，经消化后具有与海胆卵巢合成素基因预测的序列一致的多肽组成。从血管内皮细胞中分离出VAMP，并对其胰酶活性进行检测。虽然VAMP的肽图包含了许多从该序列预测的多肽，但出现了一个与海胆卵巢VAMP预测的多肽不对应的主要质量峰。从海胆卵巢文库中重新筛选出一个VAMP克隆，该克隆的氨基末端多肽与源后腐烂(PSD)片段化的团块一致。分离到几个阳性克隆，其氨基末端序列与从分离的CVVAMP获得的序列一致。测定了海胆卵巢VAMP的序列(GenBank登录号：AF151536)。此外，对文库进行筛选，寻找与数据库中报道的序列相对应的VAMP氨基末端序列。使用该序列没有获得阳性克隆，表明没有与已发表的氨基端序列匹配的海胆VAMP。用PSD裂解进一步分析该胰蛋白酶多肽，以确定其氨基酸序列和组成。PSD分析表明，该多肽含有氨基端N-乙酰丙氨酸，可能是氨基末端的多肽。虽然该肽的氨基末端序列与克隆的序列不同，但C末端的序列与克隆的氨基酸序列一致。已发表的VAMP cDNA序列与分离的VAMP的氨基酸序列之间的这种差异表明了确认表达蛋白的结构的重要性，而不是仅仅依靠克隆数据来确定序列。第三，利用差示干涉显微镜(DIC)和荧光显微镜结合膜电导记录研究了CV与平面磷脂双层膜的融合。在无钙条件下，CV与无蛋白的平面膜有很强的结合。用两种独立的方法检测钙诱导的CV融合：DIC显微镜可见的单个囊泡内容物的丢失；以及通过含量特定的染料的荧光增加检测到跨平面膜的囊泡内容物释放。结果表明，CV具有足够的钙敏感蛋白和与脂膜融合的对接机制，并且在天然皮质中，颗粒融合部位朝向质膜。从质膜上去除囊泡可以使融合部位接触到新的膜。

We have continued our combined biophysical and biochemical research on the molecular mechanisms of membrane fusion during exocytosis, the critical step in insulin, neurotransmitter, and immunoglobulin secretion, and in fertilization. First,expression library screening with a monoclonal antibody to mammalian SNAP-25 was used to identify a S.purpuratus SNAP-25 homologue. A single isoform of sea urchin ovary SNAP-25 was cloned. The clone encodes a 212 amino acid protein (Gen Bank accession no. AF061750). This SNAP-25 is highly conserved, throughout the whole amino acid sequence when aligned to SNAP-25 from several species. The recombinant protein has been expressed in E.coli as a fusion construct, purified, and used as a standard in quantitative evaluation of the amount of SNAP-25 in cortical vesicles. The preliminary data show that ~1000 copies of SNAP-25 are present per vesicle. Second, syntaxin and VAMP have been immunoprecipated from cortical vesicles and the peptide maps analyzed by matrix-assisted laser desorption/ionization (MALDI) mass spectrometry. The syntaxin immunoprecipated and purified from CV had peptide compositions after digestion that matched those predicted from the cDNA for urchin ovary syntaxin. VAMP from CV was isolated and the tryptic peptides were examined. Although the peptide map of VAMP contained many of the peptides predicted from the cDNA sequence, a major mass peak was present that did not correspond to the peptides predicted for urchin ovary VAMP. A stage-specific urchin ovary cDNA library was rescreened for a VAMP clone which had an amino terminal peptide consistent with the masses seen with post source decay (PSD)fragmentation. Several positive clones were isolated which contained an amino terminal sequence consistent with that obtained from the isolated CV VAMP. The sequence of the sea urchin ovary VAMP was determined (Gen Bank accession no. AF151536). In addition, the cDNA library was screened for sequences of the amino terminal of VAMP corresponding to that predicted for the cDNA sequence reported in the database. No positive clones were obtained using that sequence, indicating that there is no urchin VAMP that matches the published amino terminal sequence. Further analysis of this tryptic peptide by PSDfragmentation was used to determine the amino acid sequence and composition. PSD analysis indicated that the peptide could be the amino terminal peptide, since the peptide contained an amino terminal N-acetyl alanine. Although the amino terminal sequence of the peptide was distinct from the cDNA clone, the C-terminal sequence corresponded to that predicated from the cDNA clone. This discrepancy between the published VAMP cDNA sequence and the amino acid sequence of the isolated VAMP points out the importance of confirming the structure of the expressed protein, and not relying only on cloning data for sequence determination. Third, the fusion of CV to planar phospholipid bilayer membranes was studied by differential interference contrast (DIC) and fluorescence microscopy in combination with electrical recordings of membrane conductance. A strong binding of CV to protein-free planar membrane was observed in the absence of calcium. Calcium-induced fusion of CV was detected using two independent assays: loss of the contents of individual vesicles visible by DIC microscopy; and vesicle content discharge across the planar membrane detected by an increase in the fluorescence of a content-specific dye. The results indicate that CV have sufficient calcium-sensitive proteins and docking machinery for fusion to lipid membranes, and in native cortices granular fusion sites are oriented toward the plasma membrane. Removal of vesicles from the plasma membrane may allow fusion sites access to new membranes.