【作者】 陈泉；

【导师】 祁庆生；

【作者基本信息】 山东大学 ， 微生物学， 2011， 博士

【摘要】 聚羟基脂肪酸酯(PHA)是微生物天然合成的一类功能性生物聚酯,由于PHA具有与传统石油化工类塑料相似的材料性质,因此PHA一直以来都被认为是传统石化类塑料的最佳替代品。另外,PHA还具有传统塑料所不具备的性质,如生物可降解性、生物相容性、压电性、光学活性等特殊性质。PHA众多的优良性能,决定了其在应用中具有明显的优势。在过去的几十年里,PHA,尤其是聚-3-羟基丁酸酯(PHB)、3-羟基丁酸和3-羟基戊酸共聚物(PHBV)材料的产品已经被广泛应用于包装、涂层、医药以及组织工程学等众多领域。随着全球石油资源的日益缺乏,降低PHA材料的生产成本,推动PHA的工业化生产和商业应用开发对促进塑料产业的可持续发展具有非常重要的意义。根据单体碳原子数的不同,PHA可以分为两类：短链PHA (scl-PHA)和中长链PHA (mcl-PHA)。短链PHA单体由3-5个碳原子组成,中长链PHA单体由6-14个碳原子组成。单体组成的多样性决定了PHA物理性能的多样性。PHB是短链PHA的典型代表,也是研究最早最透彻的一类PHA均聚物,它的组成单位是单一的R型3-羟基丁酸的重复单元。PHB的材料性质既脆又硬,这大大限制了它的应用范围。微生物天然合成的共聚物是当前PHA生产最受瞩目的类群,共聚物中除3HB外其他单体(如3HV、3HHx或3HO)的掺入对PHA的物理性能有较大的改善。由于现有生产菌株的局限性,人们开始致力于利用基因工程技术对现有菌株进行改造,或者在新宿主中构建PHA的合成途径从而实现PHA的高效生产。目前,对微生物合成PHA的研究已经取得了显著的成果。但是,过高的生产成本仍然是限制PHA产业化的主要障碍。本论文的目的就是通过在重组大肠杆菌中构建短链聚合物PHB及PHBV的合成途径,并对其相关代谢途径进行系统研究和基因工程改造,从而降低这两种短链PHA的生产成本,为推动PHA的产业化生产奠定理论基础。PHB是PHA家族中研究的最为透彻的聚合物,一直都作为研究PHA发酵生产策略的模式产物。在降低PHB生产成本的方法中,使用廉价底物和选用高产菌株(如重组大肠杆菌)成为目前PHB生产研究中最普遍的方法。大肠杆菌作为PHB的生产菌株具有一系列的优点：遗传背景清楚,底物利用范围广,生长迅速,体内不存在PHA降解系统以及产物易于提取等。在本论文中,我们比较了不同大肠杆菌作为PHB生产出发菌株的潜力,并最终选择大肠杆菌DH5α作为PHB生产的出发菌株。同时,我们考察了溶氧对重组大肠杆菌发酵生产PHB的影响,充分证实了大肠杆菌生产PHB溶氧变化的“二阶段发酵”模式：(Ⅰ)细胞生长阶段,该阶段需要保证充足的供氧以维持细胞生长；(Ⅱ)PHB积累阶段,该阶段胞内PHB大量积累,需要在低溶氧状态下完成。PHB生产的经济性在很大程度上(最大可达整个生产成本的50%)由原材料成本,特别是碳源成本决定。因此,我们首先考察了重组大肠杆菌分别以葡萄糖、木糖、阿拉伯糖、乳糖和果糖作为碳源生产PHB的情况。我们发现,重组大肠杆菌以上述单糖作为碳源时,都能够高效的生产PHB。但是,当前存在的很多廉价碳源(如半纤维素水解液)其成分都是多种糖的混合物。由于碳源代谢阻遏现象的存在,大肠杆菌在利用多种碳源生长时,通常会优先选择利用其中某一种碳源,而其它碳源的代谢将被这种碳源所抑制,直至这种碳源被耗尽其它碳源才能被利用,这将大大降低发酵生产的效率。为了解决大肠杆菌多底物利用的问题,我们利用Red同源重组技术敲除了大肠杆菌PTS系统中的关键酶基因-ptsG基因。我们构建的大肠杆菌DH5α△ptsG/pBHR68能够同时利用多种糖的混合物高效生产PHB。利用5L发酵罐培养时,大肠杆菌能够利用“玉米芯糖蜜”生产占细胞干重84.6%的PHB。随后,我们对重组大肠杆菌利用菊芋糖发酵生产PHB进行了初步的研究,发现果糖是一种可用于PHB生产非常有价值的碳源。我们利用摇瓶发酵,生产的PHB占细胞干重的54.2%。通过对重组大肠杆菌PHB合成途径的改造和发酵策略的优化,我们解决了大肠杆菌多种碳源同时利用的问题,对模式产物PHB的生产有了较深入的了解,为下一步的研究提供了工作基础。与PHB相比,共聚物PHBV的硬度和结晶度都有所降低,耐冲击能力大幅增强,加工性能明显改善,其机械性能更接近于石化来源的热塑性塑料聚乙烯,是一种具有巨大潜在价值的生物可降解“绿色材料”。PHBV的优良性能决定了它比均聚物PHB具有更为广泛的用途。影响PHBV机械性能的一个重要因素就是PHBV中3HV的摩尔含量。因此,控制共聚物中的单体组成在PHBV的生产中至关重要。到目前为止,利用罗氏真养菌以及重组大肠杆菌生产PHBV都是通过控制加入丙酸的浓度来控制聚合物的单体组成。但是,多数生产方法中利用丙酸生产PHBV的效率和转化率(Y3HV/Prop)都很低。这主要是由于丙酸活化效率不高以及丙酰辅酶A的竞争性支路途径存在造成的。针对PHBV生产中丙酸活化效率不高的问题,我们对大肠杆菌的丙酸代谢途径进行了研究。大肠杆菌以丙酸为唯一碳源生长时,丙酸代谢的主要途径为2-甲基柠檬酸途径,该途径中的关键酶是由prpBCDE操纵子编码的。文献报道,由于PTS系统的存在,大肠杆菌prpBCDE操纵子的转录会因为葡萄糖或甘油等“PTS糖”的存在而受到抑制。为了考察利用共底物丙酸生产PHBV时碳源对大肠杆菌丙酸代谢和PHBV合成的影响,我们利用大肠杆菌DH5a/pBHR68分别利用葡萄糖、甘油和木糖为碳源发酵生产PHBV。通过比较,我们发现大肠杆菌DH5a/pBHR68利用木糖生长时,其代谢丙酸和生产PHBV的能力明显高于以葡萄糖或甘油为碳源时的情况。为了消除PTS系统的影响,我们利用ptsG缺陷型菌株DH5α△ptsG/pBHR68进行了发酵实验。结果发现ptsG基因缺失后,葡萄糖对丙酸代谢不再有抑制作用了。这进一步地证实了PTS系统对大肠杆菌丙酸活化及PHBV合成的不利影响。一些研究者通过外源表达编码丙酰辅酶A合成酶的prpE基因消除PTS系统对丙酸活化的影响,并取得了显著的效果。在本论文中,我们发现编码乙酰辅酶A合成酶的acs基因同样能够促进葡萄糖存在时丙酸的活化。而且,我们的实验结果表明：过量表达acs基因后,大肠杆菌能够利用丙酸合成3HV摩尔分数更高的PHBV共聚物。尽管上述方法能够提高大肠杆菌合成的PHBV中3HV的摩尔分数,但是丙酸转化为3HV的转化率Y3HV/Prop仍然只有0.15 g g-1。该数值远低于丙酸转化为3HV的理论转化率1.35 g g-1。针对这一问题,我们首先阻断了丙酰辅酶A流向MCC途径(敲除prpC)和TCA循环(敲除scpC)中的途径以减少重组大肠杆菌中丙酰辅酶A的流失。出乎意料地是,敲除prpC基因并提高3HV的摩尔分数效果并不明显,反而是敲除scpC基因大大提高了PHBV中3HV的比例。这说明,大肠杆菌利用丙酸生产PHBV时,流向TCA循环的代谢通量要远高于流向MCC途径的代谢通量。为了防止丙酰辅酶A转化为丙酸,我们又敲除了负责丙酸产生的pta基因。但是,敲除pta基因并没有使大肠杆菌产生PHBV的3HV摩尔分数提高。另外,值得注意的是,同时敲除prpC和scpC基因的菌株QW102/pBHR68的Y3HV/Prop可以达到0.64 g g-1,大约是基因敲除前生产菌株的5倍。本论文对重组大肠杆菌利用共底物生产PHBV过程中丙酸代谢相关途径进行了深入的研究,并利用基因敲除技术改变了以往PHBV生产“只开源不节流”的生产模式,为PHBV的商业化生产提供了理论依据。生产成本过高是制约PHBV工业化发展的主要因素,而PHBV生产成本高的主要原因就是丙酸的加入。丙酸价格较贵,并且对菌体生长具有一定毒性,所以需要较为复杂的控制策略来减少丙酸的抑制。但是,在传统的生产方法中,丙酸是合成3HV的不可缺少的必需前体物。而目前关于利用非相关碳源合成PHBV的报道由于种种限制,无法满足PHBV大规模生产的要求。因此,寻找并构建新的以廉价的单一碳源合成PHBV的途径,对于推动PHBV的商业化具有非常重要的意义。在前期的实验中,我们发现：大肠杆菌DH5a/pBHR68即使在不加入丙酸的情况下也能够产生极少量的3HV。这说明在大肠杆菌中存在能够利用葡萄糖内源产生丙酰辅酶A或丙酸的途径。在本论文中,我们通过实验证实了大肠杆菌中丙酰辅酶A来源于其自身合成的苏氨酸,并且通过基因工程改造在大肠杆菌中构建了一条能够利用非相关碳源高效合成PHBV的途径。首先,我们发现苏氨酸脱氨酶是限制苏氨酸转化为丙酰辅酶A的关键酶。于是,我们在大肠杆菌过量表达了编码苏氨酸脱氨酶的ilvA基因。并且,我们通过比较不同宿主来源的ilvA基因,发现来源于谷氨酸棒杆菌中的苏氨酸脱氨酶具有更高的催化能力。在添加苏氨酸的情况下,过量表达ilvACG基因的重组大肠杆菌产生的PHBV共聚物中3HV的摩尔分数为5.09%,是对照菌株3HV摩尔分数的10倍以上。随后,我们过量表达了编码苏氨酸合成相关酶的thrABC操纵子,同时通过定点突变消除了苏氨酸对thrA基因编码的天冬氨酸激酶的反馈抑制,实现了大肠杆菌利用木糖或葡萄糖为单一碳源合成PHBV的途径构建。最后,我们利用前面得到的基因缺陷型菌株得到了一系列能够产生不同3HV摩尔分数的PHBV的生产菌株。菌株QW103/pHB-ilvA/pCL-thrABC能够利用木糖为单一碳源生产3HV摩尔分数达到17.5%的PHBV共聚物。利用非相关碳源合成不同组成的PHBV,不仅解决了PHBV生产成本过高的问题,同时避免了添加丙酸时复杂的控制过程。我们相信,随着对菌株的进一步驯化,该生产策略可以在将来的PHBV工业化生产中得到广泛的应用。本论文初步研究并优化了模式产物PHB的发酵条件,解决了大肠杆菌多底物同时利用的问题；系统分析了大肠杆菌PTS系统对丙酸代谢的影响,为改善共底物生产PHBV的策略奠定了理论基础；通过对大肠杆菌苏氨酸合成与代谢的研究,创造性地构建了一条重组大肠杆菌利用非相关碳源合成PHBV的途径,开创了重组大肠杆菌利用单一碳源生产PHBV的新思路。更多还原

【Abstract】 Polyhydroxyalkanoates (PHAs), macromolecule-polyesters naturally produced by many species of microorganisms, are being considered as a replacement for conventional plastics. PHAs not only possess the properties as well as traditional petrochemical plastics but also are biodegradable, biocompatible, piezoelectric, optically active etc. Over the past years, PHAs, particularly poly-3-hydroxybutyrate (PHB) and copolymers of 3-hydroxybutyrate and 3-hydroxyvalerate (PHBV) have been used for packaging, coating, tissue engineering applications. In conclusion, efforts in biopolymer research must be made to develop and enhance PHAs production processes at low cost levels, and the development of PHA with better properties is necessary.PHAs can be divided into two broad classes based on the size of monomers incorporated into the polymer. Short-chain-length PHAs (scl-PHAs) consist of monomer units of C3 to C5; medium-chain-length PHAs (mcl-PHAs) consist of monomer units of C6 to C14. The large diversity of monomers found in PHAs provides a wide spectrum of polymers with varying physical properties. The homopolymer PHB is a relatively stiff and brittle bioplastic, which is of limited use. PHA copolymers composed of primarily 3HB with a fraction of longer chain monomers, such as 3HV,3HHx or 3HO, are more flexible and tougher plastics.PHB and PHBV have been produced on a commercial scale since 1970s. Promising strategies involve genetic engineering of microorganisms to introduce production pathways. This challenge requires the expression of several genes along with optimization of PHA synthesis in the host. Although excellent progress has been made in recombinant hosts, the barriers to obtaining high quantities of PHAs at low cost still remain to be solved. The aim of this paper is to reduce cost of PHB and PHBV production through pathway construction and metabolic engineering for scl-PHAs synthetic pathway in recombinant Escherichia coli.PHB is the best known PHA and has been studied intensively as a model product in the development of fermentation strategies. PHB production cost can be reduced by several means, including the use of cheap substrates or the enhancement of product yield, e.g., by using recombinant E. coli. E. coli has a number of advantages as a host for PHB production. These include a wide range of utilizable carbon sources, fast growth with a high level of productivity, the absence of an intracellular depolymerization system, easier purification due to fragile cells and large accumulated granules, as well as well-understood genetics and biochemistry. In this study, we compared PHB accumulation of different E. coli strain and chose recombinant E. coli DH5a as the PHB producer. We investigated the effect of oxygen supply on PHB synthesis in recombinant E.coli, and confirmed that PHB fermentation process in E. coli could be divided into two phage:(Ⅰ) an active growth phase during which PHB content is kept relatively constant at a low level and (Ⅱ) an active PHB synthesis phase during which PHB is actively accumulated with a concomitant increase of PHB content.The cost of the carbon source contributes significantly to the overall production cost of PHB. Then we investigated the PHB accumulation from glucose, xylose, arabinose, lactose and fructose in recombinant E. coli. We found that recombinant E. coli could produce PHB efficiently with each of these sugars as carbon source. While, Many cheap renewable carbon sources, such as corn cob hydrolysates, are mixture of sugars. Due to CCR, E. coli displays sequential sugar consumption when it is grown in media derived from mixture of sugars. Based on the fermentation analysis of E. coli strains and cheap renewable resources suitable for PHB production, we constructed aptsG mutant of E. coli DH5a AptsG. Application of E. coli DH5a AptsG/pBHR6 & mutant, we could product PHB efficiently from cheap renewable sugar mixture by the simultaneous consumption of different sugars. Batch fermentation at lab scale showed that E. coli DH5a AptsG/pBHR68 was able to produced PHB from corn cob molesses up to 84.6% of cell dry weight in 32 hours; meanwhile, the cell dry weight reached 8.24 g/L. Subsequently, we tried to produce PHB from Jerusalem artichoke sugar in recombinant E. coli, and yielded a high concentration of cell with 54.2% PHB content.PHBV is less brittle and less crystalline than PHB homopolymer, making it more suitable for commercial applications. The control of copolymer composition is important because the physical and mechanical material properties of the copolymer depend on the fraction of 3-hydroxyvalerate(3HV). To date, most attempts at producing PHBV copolymer with different compositions in recombinant E. coli have used the same strategy as that used with R. eutropha; that is, varying the propionate concentration to vary the 3HV fraction. However, many bacteria showed a low 3HV yield with propionate(Y3HV/Prop).The low yield may result from the activation of external propionate and the alternative pathways in propionate metabolism.In E. coli, prpBCDE operon encodes the enzymes for propionate activation and metabolism (also known as 2-methylcitrate pathway), allowing growth on propionate as a sole carbon and energy source. Transcription of the prpBCDE operon is down-regulated by glucose or glycerol due to the catabolite repression caused by phosphoenolpyruvate-dependent phosphotransferase system (PTS). To determine the influence of carbon sources on propionate metabolism and PHBV synthesis, we used glucose, glycerol and xylose as carbon source respectively to produce PHBV in recombinant E. coli DH5a/pBHR68. We found that E. coli showed higher ability to metabolite propionate and produce PHBV with xylose as carbon source than that with glucose or glycerol as carbon source. Furthermore, the experimental results of DH5α△ptsG/pBHR68 also confirmed the catabolite repression in E. coli. To avoid the catabolite repression in PHBV production, a few researchers overexpressed prpE gene, which encodes propionyl-CoA synthetase. In this study, we found acetyl-CoA synthetase, encoded by acs, could also activate propionate and E. coli produced PHBV copolymer with higher 3HV fraction when it was overexpressed. Although E. coli produce PHBV copolymer with higher 3HV fraction, E. coli showed poor Y3HV/Prop about 0.15 g g-1, which is a small percentage of the maximum theoretical value of 1.35 g g-1. The low Y3HV/ProP resulted from the alternative pathways initiating complete oxidation of propionate in E. coli. To block the endogenous propionyl-CoA catabolism, we deleted the metabolic pathways diverted to MCC cycle and/or TCA cycle. To our surprise, deletion of prpC gene did not improve the 3HV fraction significantly, while deletion of scpC gene greatly improved the 3HV fraction in the copolymer. This result suggested that the metabolic flux from propionyl-CoA to methylmalonyl-CoA pathway was much more than that to MCC cycle in E. coli. Phosphate acetyltransferase (encoded by pta) catalyzes both acetyl-CoA and propionyl-CoA to acetate and propionate, respectively. To reduce the propionate formation, pta was deleted in mutant QW102. However, deletion of pta did not improve the 3HV fraction in the copolymer. It was notable that Y3HV/Prop of QW102/pBHR68 reached up to 0.64 g g-1, approximately five times higher than the Y3HV/Prop reached by the wild type DH5a/pBHR68.The prohibitively high price of PHBV hinders industrialization of the copolymer. A major factor of this condition was that propionate, which was activated to form the propionyl-CoA precursor of 3HV, is expensive to produce industrially and is considerably more costly than glucose. A more economical alternative is to produce propionyl-CoA from an inexpensive, unrelated carbon source.In previous experiments, we found that recombinant E. coli DH5a/pBHR68 produced PHBV with small amount of 3HV fraction even when it was not supplied with the precursor substrate, propionate. This suggested that propionyl-CoA and/or propionate can be generated in vivo through certain metabolic pathways from glucose. In this study, based on the analysis major propionyl-CoA origin in recombinant E. coli, we developed a PHBV biosynthesis pathway from single unrelated carbon source via threonine biosynthesis in E. coli. Firstly, we found that the limit step of propionyl-CoA formation was threonine deamination. Overexpression of ilvA effectively draws carbon flux towards the synthesis of 2-ketobutyrate under aerobic conditions. Therefore, we first overexpressed native ilvA from E. coli. Overexpression of ilvAEC achieved PHBV production with doubled 3HV fraction in the copolymer. To further improve the deamination efficiency, we tried the threonine deaminase from other bacteria and found overexpression of ilvACG could improve the 3HV fraction in the copolymer more than 10 times, from 0.43 mol% to 5.09 mol%.Afterwards, we removed the feedback inhibition of threonine by mutating and overexpressing the thrABC operon in E. coli. Finally, we constructed a series of strains and mutants, which were able to produce PHBV copolymer with varied monomer compositions. The highest 3-hydroxyvalerate fraction of 17.5 mol% in the copolymer was obtained by the mutant QW103/pHB-ilvA/pCL-thrABC. As a result, the PHBV production via this strategy not only provided PHBV copolymer with different properties at low cost, but also avoided complex control strategy when propionate was co-fed. Further improvement of the host strains should lead to a perspective and practical application.更多还原