【作者】 张洪涛；

【导师】 余晓斌； 詹晓北； Ten Feizi； 柴文刚；

【作者基本信息】 江南大学 ， 发酵工程， 2011， 博士

【摘要】 本论文以一株产β-1,3-葡聚糖(热凝胶)工业化生产菌株农杆菌ATCC 31749(Agrobacterium sp. ATCC 31749)为研究模型,从提升β-1,3-葡聚糖的生产效率入手,采用比较基因组学技术构建了电子传递链及β-1,3-葡聚糖合成相关的代谢途径,运用定量PCR(qRT-PCR)、和胞内核苷酸分析考察了不同溶氧(DO)条件下β-1,3-葡聚糖合成相关基因的RNA转录丰度变化,从而阐明溶氧影响Agrobacterium sp. ATCC 31749基因转录水平、碳代谢流以及其适应外界环境的生理机制,在此基础上,提出并验证了用于强化β-1,3-葡聚糖生物合成的二阶段溶氧和pH控制的策略。对不同糖苷键连接的葡聚糖进行水解,构建了包括β-1,3-葡聚寡糖在内的具有多种构型的葡聚寡糖糖库,并合成拟糖脂和制备寡糖糖芯片,借助糖芯片技术研究了糖与免疫系统信号蛋白之间的作用关系,挖掘出能够被信号蛋白识别的最小葡聚寡糖单元,探索功能糖-免疫信号蛋白之间的构效关系规律,为以免疫信号蛋白为靶点的高效功能糖单元的筛选和设计提供实验基础、理论依据与新思路,为基于代谢工程手段的功能糖生物合成提供改进方向。主要研究结果如下:1)构建了Agrobacterium sp. ATCC 31749的电子呼吸链和β-1,3-葡聚糖(热凝胶)合成代谢途径。在Agrobacterium sp. ATCC 31749全基因组还没有测序的情况下,通过设计兼并引物和定量PCR技术,获得假定的NADH氧化还原酶、琥珀酸脱氢酶、辅酶Q生物合成蛋白、细胞色素d末端氧化酶、细胞色素bo末端氧化酶、细胞色素cbb3-型氧化酶、细胞色素caa3-型氧化酶和苹果酸脱氢酶基因序列,将获得的基因序列与NCBI数据库中基因信息进行比对分析,完成了对所获得基因的解析,在鉴定假定基因与预期完全一致的基础上构建Agrobacterium sp. ATCC 31749电子传递链网络。进一步从基因组水平对获得的基因序列与已测序的4种Agrobacterium属菌株之间进行比对,发现了所有检测基因仅与Agrobacterium tumefaciens C58中的对应基因有98%以上的高度近似性,这表明Agrobacterium sp. ATCC 31749与Agrobacterium tumefaciens C58基因组具有高度相似性,基于此结论,采用比较基因组学构建了热β-1,3-葡聚糖(热凝胶)合成代谢途径。2)溶氧可以调控胞内与β-1,3-葡聚糖(热凝胶)合成相关基因的转录水平。借助qRT-PCR技术解析DO水平与β-1,3-葡聚糖(热凝胶)合成相关基因的mRNA丰度之间的相关关系。不同DO水平下与热凝胶合成相关的基因(①葡萄糖用于热凝胶合成与细胞壁多糖合成的基因;②TCA循环途径中酶的基因;③电子传递链中蛋白的基因)的转录水平分析表明,所考察基因的转录水平随溶氧的增加而增强,DO在50%时所考察基因的转录水平最高,细胞内cyoA、catD、fixN、icd、sdh B、mdh、glmM和galU基因的转录水平是低溶氧条件下(5%)的3-6倍。3)溶氧通过调整胞内的核苷酸水平来调整碳代谢流,强化β-1,3-葡聚糖(热凝胶)的合成。在通过溶氧来控制β-1,3-葡聚糖(热凝胶)发酵过程的基础上,系统地研究了溶氧变化对β-1,3-葡聚糖(热凝胶)合成及其相关前体核苷酸物质丰度的影响,阐明了溶氧水平对改变碳代谢流分布和加快糖代谢流用于β-1,3-葡聚糖(热凝胶)合成的分子机制。与低溶氧条件相比(15% DO),高溶氧条件下细胞内AMP、UMP、UDP-葡萄糖、NADH、UTP分别是低溶氧条件下的1.4、7.9、4、3和1.5倍;与5%溶氧相比,在45%和60%溶氧浓度下,底物比消耗速率和热凝胶比合成速率增幅均超过2倍。4)提出并验证了用于强化β-1, 3-葡聚糖(热凝胶)合成的二阶段溶氧与pH控制发酵策略。在热凝胶生产过程中用两阶段溶氧调控策略(20 h到50 h溶氧为60%,pH 5.6;51 h到120 h,溶氧为40%,pH 5.6)可以提高葡萄糖消耗速率和热凝胶合成速率,使热凝胶产量达到42.8 g/L。与恒定转速控制发酵相比,热凝胶的产量、生产强度、底物转化率分别提高28%,30%和20%。这一结果表明二阶段溶氧与pH控制策略可以有效地加速碳源的消耗,增强碳代谢流用于热凝胶的合成。5)基于糖芯片技术,确认了与免疫系统功能蛋白Dectin-1和DC-SIGN结合的葡聚寡糖最小功能单元,探索了不同结构葡聚寡糖与功能蛋白间的识别及构效关系。对来自源不同微生物的寡糖和多糖进行部分水解,获得了一系列不同糖苷链连接的寡糖分子,用质谱与核磁技术对寡糖分子的结构进行解析,合成了拟糖脂探针,并用于制备糖芯片。寡糖的二级质谱谱图表明:对于不同糖苷键连接的葡聚寡糖,其ES-CID-MS/MS具有不同的断裂方式;而对于相同糖苷键连接的葡聚寡糖,其单糖指纹图谱具有特定特征,即1,2-连接的寡糖,其单离子峰的碎片峰规律是:0,4A, B和C型离子峰在每一个寡糖单元内均全部出现,相邻峰C型离子峰间具有相同的差值规律(102, 42, 18),而1,3-连接的寡糖仅仅出现了C型离子峰。功能蛋白Dectin-1特异识别主链聚合度大于10的β-1,3-葡聚寡糖。功能蛋白DC-SIGN广泛识别来自于细菌与真菌的葡聚寡糖,尤其对主链为β-1,3-四糖在1和3位有β-1,6-分支的β-1,3-葡聚寡糖具有极高的识别能力。本研究只是初步表明DC-SIGN可以和β-1,3-葡聚寡糖作用,目前在Feizi ,针对DC-SIGN和β-1,3-葡聚寡糖识别的特异性如何?以及DC-SIGN和其他寡糖探针的作用强度如何等问题正在进一步的分析研究。以上结果为以Dectin-1和DC-SIGN为靶标的β-1,3-葡聚寡糖(热凝胶寡糖)的功能开发提供了方向。更多还原

【Abstract】 This dissertation selected Agrobacterium sp. strain, ATCC 31749 as a model system to explore an efficient fermentation strategy for the production ofβ-1,3-glucan (curdlan). There has been very limited genomic information on the Agrobacterium sp. ATCC 31749. The ETCs (Electron Transfer Chains) and metabolic network were constructed based on the comparative genomic method. The effects of DO (dissolved oxygen) gradients on the metabolic characteristics, gene expression profiles of Agrobacterium sp. were studied. Based on these results, an efficient curdlan production process (two-stage DO combined with pH control) was developed. In the second part of the Thesis (described in Chapter 5) a library of glucan oligosaccharides, obtained from various types of glucans after their partial depolymerization, was constructed. These oligosaccharides were converted into neoglycolipid (NGL) probes for generating glucan oligosaccharide microarrays. Microarray binding experiments were then carried out with two proteins the human immune system to determine the specificities of the interactions, including the minimum chain length and glucose linkage requirement for oligosaccharides.The main results were described as follows:1) With the PCR and gene sequence methodology, ten genes sequences of Agrobacterium sp. ATCC 31749 were acquired and analysis as well as the genomic specificity identification between Agrobacterium sp. ATCC 31749 and other 1000 bacterial genomes were undertaken. As the genome sequence of Agrobacterium sp. ATCC 31749 was not available when this work was undertaken, the primers were designed based on the highly conservative region of the same gene from different bacteria. The DNA fragments acquired by PCR were sequenced. On that basis, the sequences of eleven gene fragments from Agrobacterium sp. ATCC 31749 were determined. These sequences were highly consistent with that from previous reports. The eleven genes from Agrobacterium sp. ATCC 31749 and the genes from other four Agrobacterium which is sequenced successfully were compared on the basis of homology. The genome of Agrobacterium sp. ATCC 31749 is highly consistent with the genome of Agrobacterium tumefaciens C58 at 100% value. The respiration chain and metabolic network in Agrobacterium sp. ATCC 31749 was constructed base on these in Agrobacterium tumefaciens C58.2) To determine the mechanism of DO modulation of mRNA translation, the effect of DO gradients on the expression of genes related to the ETCs, TCA cycle and curdlan synthesis in Agrobacterium sp. ATCC 31749 were investigated under nitrogen limited condition using qRT-PCR. The transcriptional levels of the genes governing TCA cycle (icd, sdh B and mdh), curdlan synthesis (glm M and gal U) and ETC system (cyo A, cat D and fix N) increased concomitantly with the elevation of DO. Whereas, the transcriptional level of cytochrome d (cyd A) exposed to different DO was changed slightly. At 50% DO level, the mRNAs of cyo A, cat D, fix N, icd, sdh B, mdh, glm M and gal U were 3-6 times higher than those under other DO conditions studied. The results suggested that DO can apparently affect the translational levels of the genes particularly related to the metabolic network of glucose and ATP regeneration have similar modulation profile under 50% DO.3) Based on the fermentation technology with DO control, the effect of DO gradients combined with pH on curdlan yield, intracellular nucleotide levels and glucose conversion efficiency into curdlan, and understanding of the mechanism of change of DO and the glucose metabolic network distribution to enhance the glucose metabolic flux for curdlan biosynthesis. The intracellular AMP, UMP, UDP-glucose, NADH and UTP at 60% DO is 1.4, 7.9, 4, 3, and 1.5-times, respectively, under 15% DO. The curdlan yield improved with the elevation of DO. The highest curdlan yield was 43 g/L under 60% DO which was 2.85 times higher than that of 5% DO condition (15 g/L).4) The DO concentration had apparent effect on curdlan production through enhancement of mRNA translational levels and intracellular nucleotide levels. Baseed on these results, a two-stage curdlan fermentation process has been developed and verified. The more optimal DO for curdlan production was between 45% and 60 %. The two-stage DO combine with pH (0-20 h, the pH was 7.0 after 20 h, the pH was controlled at 5.6; for DO levels, from 20 h to 50 h, 60% DO was suitable, thereafter 40% DO) was used. The resulting curdlan yield, curdlan productivity and glucose conversion efficiency into curdlan were improved by 28%, 30% and 20%, respectively.5) Series of gluco-oligosaccharide neoglycolipid (NGL) probes from 2mer to 13mer, includingβ-1,3,α-1,4,β-1,4,α-1,6 andβ-1,6-linkage, have been prepared in Prof. Ten Feizi’s laboratory. To complement currently available gluco-oligosaccharides and have a comprehensive series of the for microarray analysis,α-1,2-glucofructosides, cyclicβ-1,2-glucan,α-1,3-glucan and lentinan (β-1,3/β-1,6-branched glucan) were partially depolymerized by acid hydrolysis and fractionated by gel-filtration chromatography. Mass spectrometry and NMR were used for linkage and sequence analyses of the oligosaccharide fractions. Electrospray tandem mass spectrometry analysis of all the gluco-heptasaccharides revealed that different linkage-types produced different characteristic fragmentation. Howeverαorβanomeric configuration does not affect the fragmentation behaviour. Theαandβconfigurations can be readily assigned by 1H-NMR. In this way, all the oligosaccharide sequences were determined. NGL probes ( a total of 63 probes) were then synthesized from oligosaccharides obtained from different glucans before construction of glucan oligosaccharide microarrays. The microarrays were applied to analyze the linkage specificities and the minimum oligosaccharide chain lengths recognized by several recombinant proteins of the immune system, among them are Dectin-1 and DC-SIGN. As predicted Dectin-1 specifically bound toβ-1,3-gluco-oligosaccharides, and did not bind to any of theα-1,2-,α-1,3-,α-1,4-,α-1,6-,β-1,2-,β-1,4-,β-1,6-,β-1,3-β-1,4-gluco-oligosaccharides. The interactions of Dectin-1 with branched gluco-oligosaccharides, in particular those withβ-1,3/β1,6-linkages require further investigation. A particularly potent preparation of DC-SIGN was found to bind to gluco-oligosaccharides derived from bacteria, fungi or plants. The binding of DC-SIGN to gluco-oligosaccharides is a novel finding. Further investigation is ongoing in the Feizi laboratory to define the fine specificity, including the glucosyl linkage preference of DC-SIGN and the binding strength to gluco-oligosaccharides compared with other known DC-SIGN carbohydrate ligands. The results indicated that the position and frequency ofβ-1,6-branch play an important role for DC-SIGN interaction with gluco-oligosaccharides.更多还原