基因组代谢网络模型方法模拟重组大肠杆菌生产羟基-L-脯氨酸和葫芦巴碱的研究
发布时间：2018-08-28 12:24
【摘要】：目的将基因组规模代谢网络模型与代谢工程(重组大肠杆菌生产羟基-L-脯氨酸和葫芦巴碱)结合,对大肠杆菌模型进行相关途径修改后,对比不同分析方法的模拟效果,以及预测可行的基因敲除策略,并优化M9培养基培养大肠杆菌的条件和尝试重组葫芦巴碱合成酶的表达生产。方法1.下载大肠杆菌BL21(DE3)的代谢网络模型:iB21_1397,并添加合成羟基-L-脯氨酸和葫芦巴碱的合成途径,形成两个新的模型。2.对未修改的iB21_1397模型进行基本的FBA(flux balance analysis)分析,以及必需基因预测,指导M9培养基的优化。3.将添加代谢途径后的产羟基-L-脯氨酸和葫芦巴碱模型,进行细胞生长表型的模拟,并且结合使用FVA、OptKnock、GDLS、IdealKnock等不同方法进行基因敲除策略的预测,比较各模拟结果的差异。4.根据模拟结果设计五组改良后的M9培养基:葡萄糖组、甘油组、葡萄糖铁组、葡萄糖甘油组和倍量葡萄糖组。制备悬菌液后各接种0.1 mL至新鲜的各种改良M9培养基中培养。研究菌落数生长差异时,在培养12 h后,接种至LB琼脂培养基中,18 h后计数。研究对数生长期菌落数差异时,则分别在培养12 h、18 h、24 h后,接种至各种相应的M9琼脂培养基中,72 h后计数。5.通过基因库里的葫芦巴碱合成酶CTgS2(BAC43759.1)的信息,合成基因片段,利用限制性内切酶NdeⅠ、XhoⅠ在pET24a(+)的酶切位点,将基因片段与载体pET24a(+)连接。将连接产物转化进大肠杆菌DH5α,挑取转化子进行扩增培养,并提取质粒进行酶切和测序验证。将验证正确的重组质粒进行复制扩增后转化进大肠杆菌BL21(DE3)表达,对转化成功的细菌在相同的培养条件下,以终浓度为0.3 mM、0.6 mM、1 mM的IPTG以及终浓度为0.8 g/L的TNDA-1蛋白促进剂为诱导剂,分别诱导6h、8h、10h。诱导结束后将细菌体破碎并进行SDS-PAGE电泳,观察重组蛋白的表达情况。成果1.成功实践了代谢网络模型的改造和修改,FBA各分析方法基本都能执行成功,并与文献实验数据进行比对。2.用代谢网络模型预测了促进羟基-L-脯氨酸产量的基因敲除策略:可通过敲除酮戊二酸脱氢酶、果糖6-磷酸醛缩酶、异柠檬酸裂合酶、磷酸甘油酸酯脱氢酶实现合成酶的过表达,该策略结合比对其他分析结果后发现其具有一定可信度。3.用代谢网络模型预测了促进葫芦巴碱产量的基因敲除策略:可通过敲除乙醛脱氢酶、苹果酸酶、丙酮酸激酶、转氢酶实现合成酶的过表达。4.M9培养基优化实验方面,就碳源的选择而言,葡萄糖的增菌效果是较甘油明显的。培养基中添加Fe2+或增加同类碳源浓度,都能促进细菌生长,且增菌效果没有明显的差异,只有在培养24 h后出现差异,基本符合模拟结果。5.重组葫芦巴碱合成酶研究方面,重组质粒经过酶切和测序验证后转化进大肠杆菌BL21(DE3),进行不同浓度诱导剂和不同诱导时间的蛋白诱导,经SDS-PAGE电泳,发现各条件下都没有顺利诱导出目标蛋白。但是通过本实验,对重组蛋白技术有了一定的认识,并在连接载体的选择、基因重组验证等技术方面积累了相关的经验。结论1.代谢工程可以便利地结合基因组规模代谢网络模型进行模拟和分析,而FBA分析则在预测最大产量和预估可提升空间的方面有很大的指导作用。2.用IdealKnock方法筛选的备选敲除反应比用FVA方法筛选的要更有效,更适合用于OptKnock进行基因敲除预测。3.OptKnock方法虽然在预测基因敲除方面耗时较长,但是只要结合适合的模型反应预处理方法,计算成功率比GDLS方法高。4.经过模拟预测,对于重组大肠杆菌生产羟基-L-脯氨酸的代谢网络模型,敲除酮戊二酸脱氢酶(AKGDH)、果糖6-磷酸醛缩酶(F6PA)、异柠檬酸裂合酶(ICL)、磷酸甘油酸酯脱氢酶(PGCD),可以有利于脯氨酸4-羟化酶的过表达。5.对于重组大肠杆菌生产葫芦巴碱的代谢网络模型,敲除乙醛脱氢酶(ALDD2y)、苹果酸酶(ME2)、丙酮酸激酶(PYK)、NAD(P)转氢酶(THD2pp),可以有利于葫芦巴碱合成酶的过表达。6.代谢网络FBA分析的模拟结果,对M9培养基的优化具有很强指导意义,对于细胞生长,加入适当量的Fe2+,可以有效地促进生长,无需靠提高葡萄糖的浓度,碳源方面,葡萄糖在促进细胞生长方面比甘油稍优,但是如果生产中需要使用甘油,需要适当地提高甘油浓度,或者搭配葡萄糖,才能达到单纯葡萄糖做碳源时的生长效果。7.本研究利用基因库里的葫芦巴碱合成酶CTgS2(BAC43759.1)的信息,合成基因片段,利用限制性内切酶NdeⅠ、XhoⅠ,将基因片段与载体pET24a(+)连接。重组质粒pET24a-CTgS2转化进大肠杆菌DH5α中扩增,转化大肠杆菌BL21(DE3)中表达,没有成功诱导出目的蛋白,考虑诱导失败有可能与载体的选择,和基因重组过程中验证体系欠缺完善有关。
[Abstract]:Objective To combine genome-scale metabolic network model with metabolic engineering (recombinant E.coli producing hydroxy-L-proline and cucurbitacin) to modify the relevant pathways of E.coli model, compare the simulation results of different analytical methods, and predict the feasible gene knockout strategy, and optimize the conditions of E.coli culture in M9 medium. Methods 1. Download the metabolic network model of Escherichia coli BL21 (DE3): iB21_1397, and add the synthetic routes of hydroxy-L-proline and cucurbitacin to form two new models. Gene prediction is needed to guide the optimization of M9 medium. 3. Hydroxyl-L-proline and cucurbitacin production models with metabolic pathways were used to simulate cell growth phenotype. FVA, OptKnock, GDLS, IdealKnock were used to predict gene knockout strategies, and the differences of simulation results were compared. Five groups of improved M9 media were prepared and inoculated in 0.1 mL to fresh M9 medium respectively. The difference of colony number was studied when cultured in LB agar medium 12 hours later and counted after 18 hours. When the long-term colonies were different, they were inoculated into various M9 agar media 12, 18 and 24 hours after culture, and counted after 72 hours. 5. Gene fragments were synthesized by the cucurbitacin synthase CTgS2 (BAC43759.1) in the gene library, and the gene fragments and vectors were digested by restriction endonucleases Nde I and Xho I at the site of pET24a (+). PET24a (+) ligation. The conjugated product was transformed into E. coli DH5a, and the transformant was selected for amplification and culture, and the plasmid was extracted for enzyme digestion and sequencing verification. 6 mM, 1 mM IPTG and 0.8 g/L TNDA-1 protein promoter were induced for 6, 8 and 10 hours respectively. After induction, the bacterial bodies were broken and SDS-PAGE electrophoresis was performed to observe the expression of recombinant proteins. Comparing with the experimental data in the literature. 2. Using metabolic network model, we predicted the gene knockout strategy to promote the production of hydroxy-L-proline: synthase overexpression could be achieved by knocking out ketoglutarate dehydrogenase, fructose-6-phosphate aldolase, isocitrate lyase, and phosphoglyceride dehydrogenase, which combined with other analysis results. 3. Gene knockout strategies to promote cucurbitacin production were predicted by metabolic network model: synthase overexpression could be achieved by knocking out acetaldehyde dehydrogenase, malic acid enzyme, pyruvate kinase, and transhydrogenase. 4. M9 medium optimization experiment showed that glucose was more effective than glycerol in the selection of carbon source. Adding Fe2+ to the medium or increasing the concentration of the same carbon source can promote the growth of bacteria, and there is no significant difference in the growth of bacteria. Only after 24 hours of culture, there is a difference, basically in line with the simulation results. 5. In the study of recombinant cucurbitacin synthase, the recombinant plasmid was transformed into E. coli BL21 (DE3) after digestion and sequencing verification. Through SDS-PAGE electrophoresis, it was found that the target protein was not successfully induced under all conditions. However, through this experiment, we have a certain understanding of the recombinant protein technology, and accumulated relevant experience in connection vector selection, gene recombination verification and other technologies. Conclusion 1. Xie can easily simulate and analyze the genome-scale metabolic network model, while FBA analysis has a great guiding role in predicting the maximum yield and predicting the upgradable space. 2. The alternative knockout response screened by Ideal Knock method is more effective than that screened by FVA method and is more suitable for OptKnock gene analysis. Knock-out prediction. 3. Although the OptKnock method is time-consuming in predicting gene knockout, the calculation success rate is higher than that of GDLS method as long as the appropriate model reaction pretreatment method is combined. 4. After simulation prediction, for the metabolic network model of recombinant E. coli producing hydroxy-L-proline, the knockout ketoglutarate dehydrogenase (AKGDH), fructose 6-phosphorus are predicted. Acetaldehyde dehydrogenase (F6PA), isocitrate lyase (ICL), phosphoglyceride dehydrogenase (PGCD) can facilitate the over-expression of proline 4-hydroxylase. 5. It is advantageous to knock out acetaldehyde dehydrogenase (ALDD2y), malate enzyme (ME2), pyruvate kinase (PYK), NAD (P) transhydrogenase (THD2pp) in the metabolic network model of cucurbitacin production by recombinant E. coli. The simulated results of FBA analysis of metabolic network are of great guiding significance to the optimization of M9 medium. Fe2+ can effectively promote cell growth without increasing glucose concentration. In terms of carbon source, glucose is slightly superior to glycerol in promoting cell growth, but it can improve cell growth. If glycerol is needed in production, it is necessary to increase the glycerol concentration properly, or to mix with glucose to achieve the growth effect of pure glucose as carbon source. 7. In this study, the cucurbitacin synthase CTgS2 (BAC43759.1) in the gene library was used to synthesize gene fragments, and the restriction endonuclease Nde I, Xho I was used to synthesize gene fragments. The recombinant plasmid pET24a-CTgS2 was transformed into E. coli DH5a and expressed in E. coli BL21 (DE3). The target protein was not successfully induced. The failure of induction might be related to the choice of vector and the lack of validation system in the process of gene recombination.
【学位授予单位】：广州中医药大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R91;Q78
[Abstract]:Objective To combine genome-scale metabolic network model with metabolic engineering (recombinant E.coli producing hydroxy-L-proline and cucurbitacin) to modify the relevant pathways of E.coli model, compare the simulation results of different analytical methods, and predict the feasible gene knockout strategy, and optimize the conditions of E.coli culture in M9 medium. Methods 1. Download the metabolic network model of Escherichia coli BL21 (DE3): iB21_1397, and add the synthetic routes of hydroxy-L-proline and cucurbitacin to form two new models. Gene prediction is needed to guide the optimization of M9 medium. 3. Hydroxyl-L-proline and cucurbitacin production models with metabolic pathways were used to simulate cell growth phenotype. FVA, OptKnock, GDLS, IdealKnock were used to predict gene knockout strategies, and the differences of simulation results were compared. Five groups of improved M9 media were prepared and inoculated in 0.1 mL to fresh M9 medium respectively. The difference of colony number was studied when cultured in LB agar medium 12 hours later and counted after 18 hours. When the long-term colonies were different, they were inoculated into various M9 agar media 12, 18 and 24 hours after culture, and counted after 72 hours. 5. Gene fragments were synthesized by the cucurbitacin synthase CTgS2 (BAC43759.1) in the gene library, and the gene fragments and vectors were digested by restriction endonucleases Nde I and Xho I at the site of pET24a (+). PET24a (+) ligation. The conjugated product was transformed into E. coli DH5a, and the transformant was selected for amplification and culture, and the plasmid was extracted for enzyme digestion and sequencing verification. 6 mM, 1 mM IPTG and 0.8 g/L TNDA-1 protein promoter were induced for 6, 8 and 10 hours respectively. After induction, the bacterial bodies were broken and SDS-PAGE electrophoresis was performed to observe the expression of recombinant proteins. Comparing with the experimental data in the literature. 2. Using metabolic network model, we predicted the gene knockout strategy to promote the production of hydroxy-L-proline: synthase overexpression could be achieved by knocking out ketoglutarate dehydrogenase, fructose-6-phosphate aldolase, isocitrate lyase, and phosphoglyceride dehydrogenase, which combined with other analysis results. 3. Gene knockout strategies to promote cucurbitacin production were predicted by metabolic network model: synthase overexpression could be achieved by knocking out acetaldehyde dehydrogenase, malic acid enzyme, pyruvate kinase, and transhydrogenase. 4. M9 medium optimization experiment showed that glucose was more effective than glycerol in the selection of carbon source. Adding Fe2+ to the medium or increasing the concentration of the same carbon source can promote the growth of bacteria, and there is no significant difference in the growth of bacteria. Only after 24 hours of culture, there is a difference, basically in line with the simulation results. 5. In the study of recombinant cucurbitacin synthase, the recombinant plasmid was transformed into E. coli BL21 (DE3) after digestion and sequencing verification. Through SDS-PAGE electrophoresis, it was found that the target protein was not successfully induced under all conditions. However, through this experiment, we have a certain understanding of the recombinant protein technology, and accumulated relevant experience in connection vector selection, gene recombination verification and other technologies. Conclusion 1. Xie can easily simulate and analyze the genome-scale metabolic network model, while FBA analysis has a great guiding role in predicting the maximum yield and predicting the upgradable space. 2. The alternative knockout response screened by Ideal Knock method is more effective than that screened by FVA method and is more suitable for OptKnock gene analysis. Knock-out prediction. 3. Although the OptKnock method is time-consuming in predicting gene knockout, the calculation success rate is higher than that of GDLS method as long as the appropriate model reaction pretreatment method is combined. 4. After simulation prediction, for the metabolic network model of recombinant E. coli producing hydroxy-L-proline, the knockout ketoglutarate dehydrogenase (AKGDH), fructose 6-phosphorus are predicted. Acetaldehyde dehydrogenase (F6PA), isocitrate lyase (ICL), phosphoglyceride dehydrogenase (PGCD) can facilitate the over-expression of proline 4-hydroxylase. 5. It is advantageous to knock out acetaldehyde dehydrogenase (ALDD2y), malate enzyme (ME2), pyruvate kinase (PYK), NAD (P) transhydrogenase (THD2pp) in the metabolic network model of cucurbitacin production by recombinant E. coli. The simulated results of FBA analysis of metabolic network are of great guiding significance to the optimization of M9 medium. Fe2+ can effectively promote cell growth without increasing glucose concentration. In terms of carbon source, glucose is slightly superior to glycerol in promoting cell growth, but it can improve cell growth. If glycerol is needed in production, it is necessary to increase the glycerol concentration properly, or to mix with glucose to achieve the growth effect of pure glucose as carbon source. 7. In this study, the cucurbitacin synthase CTgS2 (BAC43759.1) in the gene library was used to synthesize gene fragments, and the restriction endonuclease Nde I, Xho I was used to synthesize gene fragments. The recombinant plasmid pET24a-CTgS2 was transformed into E. coli DH5a and expressed in E. coli BL21 (DE3). The target protein was not successfully induced. The failure of induction might be related to the choice of vector and the lack of validation system in the process of gene recombination.
【学位授予单位】：广州中医药大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R91;Q78
【相似文献】
相关期刊论文 前10条
1 赵静;俞鸿;骆建华;曹志伟;李亦学;;应用复杂网络理论研究代谢网络的进展[J];科学通报;2006年11期
2 赵静;陶林;俞鸿;骆建华;曹志伟;李亦学;;代谢网络的蝴蝶结结构特征及其功能意义[J];科学通报;2007年01期
3 王卓;陈琦;刘雷;;代谢网络进化过程中拓扑结构与功能之间的关联[J];科学通报;2009年02期
4 陈琦;王卓;魏冬青;;代谢网络流分析进展及应用[J];科学通报;2010年14期
5 丁德武;丁彦蕊;陆克中;须文波;黄海生;;重建代谢网络及其结构与功能的分析[J];计算机与应用化学;2010年05期
6 丁德武;黄海生;陆克中;何小青;计博婧;;代谢网络结构与功能的复杂网络分析[J];计算机与应用化学;2011年03期
7 陈t，
本文编号：2209390
本文链接：https://www.wllwen.com/yixuelunwen/yiyaoxuelunwen/2209390.html
最近更新
教材专著
