【作者】 段志贵；

【导师】 梁宋平；

【作者基本信息】 湖南师范大学 ， 生物化学与分子生物学， 2012， 博士

【摘要】 本文采用电刺激方法采集大腹园蛛粗毒并进行了系统的分析。凝胶电泳及质谱数据显示,该蛛毒包含丰富的蛋白质或多肽成分,其分子量分布在4-45kDa的范围里。蜚蠊腹腔注射该粗毒后显示出明显的中毒症状,且半致死剂量为30.73μg/g。电生理实验结果显示,该粗毒不能阻断小鼠膈神经膈肌的神经肌肉传导。粗毒的酶活分析显示,该粗毒具有几种水解酶的活性,包括透明质酸酶和蛋白质酶活性。所有的实验结果显示,大腹园蛛粗毒含有丰富的生物学活性成分,对其主要的猎物-昆虫有着较强的毒性,而对哺乳动物的毒性极低。电刺激方法能够有效的避免唾液及其他组织液的污染,且能够重复利用蜘蛛资源。为了克隆大腹园蛛毒腺中编码毒素多肽和蛋白质的基因,本文采用了表达序列标签(EST)技术,首次构建了该蛛毒腺的定向全长cDNA文库。随机挑取原始文库的单克隆进行测序,共得到了886个EST序列。通过聚类分析分成246个族,其中49个族属于重叠群(每族都含2个及2个以上的EST),其他197个族是单态(每族仅含1个EST)。随后利用Emboss软件将所有EST翻译成氨基酸序列,根据动物毒素蛋白质的一般特点,设定毒素的筛选条件为：(1)具有4个或4个以上的半胱氨酸；(2)具有信号肽。结果发现,其中68.7%的EST属于的毒素样转录子；20.2%的EST属于细胞转录子；11.1%的EST属于新的转录子。将所有的EST序列去冗余后得到的毒素转录子编码200个开放读码框(ORF),即200个前体肽。根据前体肽的差异可以分成15个超家族,包含AVTX-Ⅰ至AVTX-ⅩⅩⅩ共30个家族。200个前体肽能被推导出150个非冗余成熟肽,且在现有的数据库中未发现高度相似性序列。不同的超家族含有半胱氨酸的数量差异明显,最小的仅为4个,最多的高达20个。此外,编码几丁质酶的基因也被检测到。所有结果显示,大腹园蛛蛋白质类毒素与所有已知的动物毒素存在较大的差别,是一些全新的毒素蛋白质。这些毒素序列的发现丰富了动物毒素的结构模式数据库。为了更加全面了解大腹园蛛的蛋白质类毒素的组成,本文通过活体电刺激的方法采集了大腹园蛛的粗毒,并采用组合的蛋白质组学策略来分析这种粗毒,包括两种不同的方法：(i)一维聚丙烯酰胺凝胶电泳后用nanoLC-MS/MS分析；(ii)双向电泳后采用nanoLC-MS/MS分析。凝胶图谱显示,该蛛毒所含蛋白质类成分的分子量主要分布在10到45kDa的范围中,其中三个高丰度的区域分别为10到14kDa、18kDa到26kDa和45kDa,且主要为碱性蛋白质。胶内酶解的多肽经串联质谱分析,得到的质谱数据通过跨物种搜寻的方法共鉴定到71种毒素蛋白质。除了典型的蜘蛛血蓝蛋白质外,几种水解酶被发现存在于这种粗毒中。而且这些水解酶可能与蛛毒的毒性有关。此外,大约有一半以上的高质量的MS/MS数据得不到明显的鉴定结果,这极可能是没有特定的蜘蛛毒素蛋白质数据库的缘故。为了更加全面分析蛋白质组学与转录组学数据,本文首次采用了de novo测序与cDNA文库相结合的方法来鉴定蜘蛛毒素的蛋白质类成分。利用Bruker公司的DataAnalysis4.0软件对来自2DE的胶内酶解样品的未匹配到任何已知多肽的高质量的MS/MS数据进行了denovo测序,二级质谱氨基酸残基的误差设定为±0.015。得到了约2000个高质量的de novo测序结果,去冗余后得到约200个。然后人工将这些序列对转录组得到的毒素样序列进行搜索。肽段完全匹配的限制条件设定为：(1)氨基酸序列完全相同且质量误差为±0.5；(2)对应的cDNA文库序列有合理的酶切位点。结果表明,来自cDNA文库测序的30个毒素样家族中的20个家族被明确鉴定,鉴定率达到66.67%。而且首次在蜘蛛毒素中大规模的检测到毒素突变体及谷氨酸残基的甲酯化修饰。该方法给无完整基因组数据库的物种的蛋白质组学鉴定提供了新的途径。更多还原

【Abstract】 The Araneus ventricousus venom was collected by electrical stimulation and systematically analyzed. Gel electrophoresis and mass spectrometry showed that the venom contained aboundant peptides and proteins with molecular weights ranging from4kDa to45kDa. Injection of the venom in cockroaches Periplaneta americana gave rise to obvious poisoned symptoms, with LD50value of30.73μg/g. Electrophysiological experiments showed that the venom could not block the neuromuscular transmission in isolated mouse phrenic nerve-hemidiaphragm and rat vas deferens preparations. Enzymatic analysis indicated that the venom possessed activities of several kinds of hydrolases including hyaluronidase and proteases. These results demonstrate that A. ventricosus venom is rich in bioactive components targeting insects, the spider natural preys, and the content of neurotoxins to mammals, if present, is low. The electrical stimulation method to collect the venom has the advantages of avioding contamination and repeated use of the valuable A. ventricosus venom resoures.In order to clone the genes encoding the proteins and peptides from the venom glands, a directional cDNA library of A. ventricosus venom glands was firstly constructed using the expressed sequence tag (EST) strategy. Clones from original cDNA library were randomly selected to sequence. Eight handreds and eighty-six ESTs from cDNA sequencing were grouped into49clusters and197singletons. All ESTs were translated into protein sequences using software Emboss. According to the usual properties of animal toxins, the parameters of toxin filter were set for:(1)4or more cysteines;(2) contain signal peptide. The result indicated that68.7%of total ESTs belong to toxin-like genes,20.2%are cellular transcripts and11.1%have no significant similarity to any known genes. Two handreds unique propeptides were produced by translating all ESTs and nonredundancing. These propeptides were grouped into15superf ami lies which contain30families from AVTX-I to AVTX-XXX. All precursors were processed to achieve150mature peptides with abundant cysteines. However, the number of cysteine is diversified from4to20among these families. And these propeptides were not identified by any database. Moreover, the gene ecoding chitinase was found in the cDNA library. All results demonstrated that these protein toxins from A. ventricosuswere some novel toxins which may be give novel model for understanding action mechanism of spider toxin.In order to analyze systematically the components of A. ventricosus venom, in this paper, combinative proteomic strategies were employed to analyze the venom collected from living spider A. ventricosus by electrostimulation. The combinative proteomic strategies including the following two different approaches:(i) one-dimensional SDS/PAGE plus nanoLC-MS/MS,(ii)2DE plus nanoLC-MS/MS. The images show that most proteins visualized on the gel have molecular masses in the range of10-45kDa. And three major positions correspondings to molecular weight of about10to20kDa,26kDa and43kDa, there were high abundance alkaline proteins. Peptides from in-gel digestion were analyzed by nanoLC-MS/MS. A total of71venomous proteins were identified using cross-species searching. Besides typtical spdier hemocyanins, several hydrolases were found in the venom, which were speculated to be related to insecticidal activity. In addition, beyond half of high-quality MS/MS spectra gave no signification search results, most likely due to the lack of specific database for spider venom.In this paper, we employ de novo sequencing to combine the proteomic and transcriptomic data. The gel spots from2-DE were carefully excised and digested with trypsin. Then the tryptic peptides were analyzed by nanoLC-MS/MS. The spectral data were searched against an online Mascot database. Then these data with no signification search results were sequenced using the software DataAnlysis4.0from Bruker company. The mass accuracy better than±0.015Da is required for decreasing the false positive rate. About2000sequences were produced by de novo sequencing. And about200unique sequences were collected by removing the duplication. These sequences were assigned against protein sequences from cDNA library. Efficient results must fill these parameters:(1) Mass accuracy of peptide is better than±0.5Da between theoretical and measurement value;(2) the peptide can be achieved by tryptic digestion. The results indicated that twenty of thirty families from cDNA library were matched by sequences from de novo sequncing. The identification rate is66.67%. Interestingly, abundant mutants and methyl esterification of glutamic acid from the venom were identificated by the method. The method gives a new approach for indentif ication of proteins from an organism whose genome is not known.更多还原