【作者】 赵鑫锐；

【导师】 堵国成；

【作者基本信息】 江南大学 ， 发酵工程， 2014， 博士

【摘要】 黄酒中潜在致癌物质氨基甲酸乙酯的存在严重影响了我国传统发酵食品的安全性，降低黄酒中的氨基甲酸乙酯含量的研究具有重要的学术意义和应用前景。本论文以黄酒生产菌株酿酒酵母Saccharomyces cerevisiae N85为研究对象，通过对其偏好型氮源的研究，提出了造成菌株发酵过程中尿素积累的调控机理。在此基础上，在便于代谢改造的酿酒酵母单倍体模式菌株Saccharomyces cerevisiae CEN.PK2-1C中对尿素代谢相关调控途径进行了改造，并对改造效果进行了验证。结果表明新尝试的代谢改造策略可降低模式菌株发酵过程中的尿素积累量。将在模式菌株中效果最好的代谢改造策略应用到酿酒酵母N85菌株的黄酒模拟发酵中，结果表明也可显著降低发酵液中的氨基甲酸乙酯含量。说明本论文所采用的代谢改造策略不仅有效，而且适合于真实的黄酒生产。本文主要研究成果如下：(1)在21种常见氮源中寻找到了酿酒酵母N85菌株的7种偏好型氮源，并从中发现了造成菌株发酵过程中尿素积累的三种关键氮源(谷氨酰胺、谷氨酸和精氨酸)；通过定量PCR检测，发现了参与尿素代谢抑制调控的关键调控因子(Gln3p、Gat1p、Tor1p等)，并在此基础上提出了偏好型氮源对菌株尿素代谢产生抑制作用的机理。(2)以Gln3p的磷酸化调控方式为基础，对另外的三个GATA调控因子(Gat1p、Dal80p和Gzf3p)的胞内定位进行研究。通过序列比对和荧光蛋白融合表达技术，发现Dal80p和Gzf3p的胞内定位不受氮源调控的影响；而Gat1p的调控方式则与Gln3p相似。除此之外还发现了Gat1p的核定位序列、核定位调控序列以及核定位序列上的磷酸化位点。(3)在Gln3p和Gat1p核定位研究成果的基础上，尝试采用不同的策略对Gln3p和Gat1p进行了一系列的代谢改造(磷酸化位点的定点突变、核定位调控序列的缺失等)。实验结果表明，这些策略都可以在一定程度上缓解Gln3p和Gat1p所受的抑制调控，而联合使用这些策略能获得最佳的改造效果。效果最好的两株基因工程菌发酵48h后的尿素积累量比对照菌株分别下降了37.4%和44.3%，说明对Gln3p和Gat1p的改造策略有继续用于改造N85生产菌株的潜力。除此之外，进一步尝试敲除负调控因子Ure2p来增强代谢改造的效果。实验结果表明，虽然敲除URE2对增强Gln3p和Gat1p活性的效果显著，但对菌株的生长会造成强烈的抑制作用，因此这种策略无法应用于生产菌株的代谢改造中。(4)尝试通过对尿素代谢相关调控途径的代谢改造(去磷酸化调控、泛素化调控和Dal81p和Dal82p的激活作用)，降低模式菌株发酵过程中的尿素积累量。实验结果表明，强化胞内的去磷酸化调控作用的效果有限；而强化Dal81p和Dal82p的激活作用可以增强菌株尿素代谢的能力。此外，以尿素代谢相关调控因子的泛素化检测结果为基础，对Dal81p上的泛素化位点进行了定点突变。突变后Dal81p的激活作用更强，和Dal82p共同表达后能获得最佳的代谢改造效果。Dal81p和Dal82p共表达的基因工程菌株在48h发酵的过程中尿素的积累量比对照菌株下降了55.7%。(5)将在模式菌株中获得良好效果的代谢改造策略应用于生产菌株N85，并在黄酒模拟体系中检测菌株的发酵特性和氨基甲酸乙酯产生量。实验结果表明，采用经代谢改造后的N85菌株，发酵过程中的尿素积累量和EC产生量都大幅下降(分别降低了63%和72%)，但发酵特性并未发生明显的改变。更多还原

【Abstract】 In this dissertation, based on the research on the preferred nitrogen sources for a diploidSaccharomyces cerevisiae strain N85, the inhibitory mechanism of normal nitrogen sourceson urea utilization could be proposed. With the guidance of urea metabolism regulation,several metabolic engineering strategies were applied in a model haploid S. cerevisiae strainCEN.PK2-1C. The results showed that the concentration of urea could be significantlyreduced by combining metabolic engineering strategies during the fermentation tests.Furthermore, the optimal strategy used in model strain also had better effect on the haploid S.cerevisiae strain N85. The main results were described as follows:(1) Although diploid S. cerevisiae strain N85is able to use many nitrogen sources for growth,the utilization rates of these components are different. In this study, seven nitrogen sourceswere considered as preferred nitrogen sources for S. cerevisiae strain N85. In addition, it wasfound that there were mainly two kinds of inhibitory effects on urea metabolism by preferrednitrogen sources. Furthermore, regulators of nitrogen catabolite repression (NCR) and targetof rapamycin (TOR) pathway were identified as being involved in urea accumulation byreal-time quantitative PCR. Based on these results, preferred nitrogen sources were found torepress urea utilization by converting them to glutamine or glutamate. Gln3p can be retainedin the cytoplasm by glutamine, while Gat1p can be retained by glutamine and glutamate.(2) Based on the nuclear localization signal (NLS) and nuclear localization regulation signal(NLRS) in Gln3p, the localization of Gat1p, Dal80p and Gzf3p were studied. The residues348–375and366–510were identified as the NLS and NLRS of Gat1p, respectively, and theresidues at positions360(serine) and361(serine) are likely to be the phosphorylation sites inGat1p. Dal80p and Gzf3p are not regulated by phosphorylation, although Gzf3p has an NLSat its C-terminus.(3) In order to increase the nuclear localization of Gln3p and Gat1p, the phosphorylation siteson NLS were mutated and the NLRS was truncated. By combining these strategies, the genes(DUR1,2and DUR3) involved in urea utilization could be significantly activated in thepresence of glutamine. During shake-flask fermentations of the genetically modified strains,little urea accumulated in the media. Furthermore, the disruption of URE2provided anadditional method of reducing urea accumulation. However, this method could not be appliedin S. cerevisiae strain N85because the disruption of URE2would repress the growth of strain.(4) In order to further enhance the expression of DUR1,2and DUR3, several metabolicengineering strategies were attempted on the dephosphorylation regulation, ubiquitylationregulation and the activation of Dal81p and Dal82p. The results showed that the strengtheningeffect of dephosphorylation were limited; whereas overexpression of Dal81p and Dal82pcould enhance the expression of DUR1,2and DUR3. Furthermore, based on the ubiquitylationdetection of regulators, the effect of combining metabolic engineering strategies ondeubiquitylated Dal81p and Dal82p was the best.(5) Based on the former metabolic engineering results, the best strategy used in model strainwas selected. The concentration of urea and EC in a model rice wine system were examined to confirm the effect of metabolic engineering. The results showed that the concentration ofurea and EC could be reduced by63%and72%, respectively. In addition, the examination ofthe normal nutrients and flavour compounds in rice wine indicated that there were fewdifferences in fermentation characteristics between the wild-type S. cerevisiae N85strain andthe genetically modified strain. Therefore, the metabolic engineering strategies attempted inthis study have great potential as the methods for eliminating EC during rice wine production.更多还原