【作者】 姜春艳；

【导师】 白林泉；

【作者基本信息】 上海交通大学 ， 微生物学， 2013， 博士

【摘要】 盐霉素（salinomycin）是由白色链霉菌产生的聚醚类抗生素，作为球虫抑制药和生长促进剂被广泛地应用于畜牧业和家禽饲养业。最新研究发现盐霉素能够高效抑制上皮肿瘤肝细胞的生长。但是前期盐霉素生物合成的研究仅停滞于同位素标记的前体的喂养实验。通过大片段缺失实验，我们发现之前文献报道的4.5-kb盐霉素基因簇序列与盐霉素的生物合成并不相关。随后，我们根据目前报道的聚醚类抗生素生物合成基因簇中的关键酶环氧化酶的高度保守性，基于CODEHOP和PaBaliS原理首次设计并成功筛选到适于挖掘潜在的聚醚类抗生素基因簇的简并性引物。利用这对简并性引物，我们在白色链霉菌XM211的基因组中克隆到了可能与盐霉素生物合成相关的环氧化酶基因slnC。通过分析slnC基因置换突变株JCY16及其回补株的发酵产物，我们确定slnC是盐霉素生物合成所必需的。最终，我们通过染色体步移获得了127-kb的DNA区域，进一步序列测定和分析发现其中包括I型聚酮合酶基因、环氧化酶基因、环氧水解酶基因、细胞色素P450单氧化酶基因以及调节、转运基因等。通过序列分析我们也推测了盐霉素各聚酮合酶（SlnA1-A9）中的模块和结构域组成。由于模块14和模块6的结构域组成中均缺少脱水酶（DH）结构域，所以合成的盐霉素聚酮链在C3及C19为羟基结构，而并不是预测的C2-C3位（与该位置四氢呋喃环形成相关）和C18-C19位双键。我们初步推测这两步脱水反应是由独立的脱水酶催化完成，但在克隆的127-kb序列中我们并未发现脱水酶基因。这两个位置的双键形成原因成为一个令人困惑而又值得深究的问题。盐霉素生物合成基因簇的核心区域含有一个甲基转移酶基因slnM，其编码的蛋白与羧甲基转移酶TcmP具有较高的同源性。但盐霉素生物合成推测没有任何甲基化反应且HPLC-MS检测白色链霉菌XM211的发酵产物中并不含有盐霉素甲基化衍生物。有趣的是，体内失活和回补实验表明slnM是盐霉素生物合成的必需基因，且slnM突变株积累了一个新的化合物1。核磁共振谱显示化合物1是盐霉素的结构类似物，不同之处在于C19位为羟基，C17-C18为双键结构，并且化合物1中三元螺环尚未形成，为二元螺环结构。将化合物1喂养到丧失盐霉素产生能力但含有完整的slnM基因的突变株中，发酵结果显示部分化合物1被转化为盐霉素，说明1可能是盐霉素生物合成的中间产物。在大肠杆菌中异源表达的SlnM能够催化化合物1转化为盐霉素，表明SlnM催化了盐霉素生物合成的最后一步反应，即负责C18-C19位双键和三元螺环的同时形成。通过监测S-腺苷甲硫氨酸（SAM）在反应体系中的浓度，我们发现SAM在反应中并不被消耗。进一步的实验结果表明甲基化抑制剂sinefungin（SAM的结构类似物）能够替代SAM作为SlnM催化反应的辅因子。我们还发现化合物1中C1位羧基被甲酯化封闭后得到的化合物不能被SlnM催化利用。根据这些实验结果，我们推断了SlnM的反应机理，即SAM或sinefungin通过静电作用使化合物1稳定的处于类似于盐霉素的构型，有利于SlnM中的活性氨基酸催化C18-C19位双键的形成和三元螺环的形成，得到盐霉素。SlnM的发现不但扩展了我们对SAM在生化反应中作用的认识，而且揭示了聚酮类化合物中双键形成的一类新机制。为了探索盐霉素的生物合成机制，我们通过基因置换手段对基因簇中另外12个基因进行了失活。突变株发酵产物经HPLC-MS检测显示，其中5个基因是盐霉素生物合成的必需基因，可能参与盐霉素聚酮链释放（slnBI）、PKS后修饰（slnF）、环氧化水解（slnBII、slnBIII）以及正调控（slnR）等；另外5个基因与盐霉素的生物合成相关但并非必需基因，可能与四碳前体的合成（orf11、orf12）、聚酮链延伸过程非正常底物的清除（slnDII）以及调控（orf15、orf16）等相关。与其它已报道的聚醚类抗生素基因簇不同，盐霉素生物合成基因簇中包含三个环氧水解酶基因：slnBI、slnBII、slnBIII。slnBII和slnBIII突变株都丧失了盐霉素产生能力，积累了两个新的化合物A和B，暗示SlnBII和SlnBIII与环氧水解的级联反应相关。而slnBI突变株中盐霉素产量不到野生型产量的5%，暗示SlnBI可能与盐霉素聚酮链的释放相关。盐霉素生物合成基因簇的克隆与基因功能分析为探明盐霉素的生物合成机理提供了可能，这不但扩大了我们对聚醚类抗生素生物合成的认识，并且为定向设计具有更高活性的盐霉素衍生物和提高盐霉素产量奠定了基础。更多还原

【Abstract】 Salinomycin, a polyether antibiotic, is produced by Streptomycesalbus. It is widely used as an anticoccidial and growth-promoting agent inanimal husbandry. Recently, salinomycin has also been identified as anagent to kill epithelial cancer stem cells. However, the biosynthesis ofsalinomycin have not been studied, except the feeding experiments withisotope-labeled precursors.The previously reported4.5-kb region was proven to be irrelevant tosalinomycin biosynthesis by deletion of this portion. Therefore, a strategywith degenerate epoxidase-specific PCR primers was developed to clonethe polyether biosynthetic gene clusters. A putative salinomycin-specificepoxidase gene slnC was cloned from the producer S. albus XM211usingthis strategy. The targeted replacement of slnC and subsequenttrans-complementation proved its involvement in salinomycinbiosynthesis. A127-kb region was sequenced, including genes encodingtype I PKS (slnA1-slnA9), epoxidase, epoxide hydrolase, regulator andtransporter. The domain structure of distributed modules of salinomycinPKS were also deduced. Intriguingly, the lack of dehydratase (DH)domains in modules14and6implies the presence of hydroxyl groups at C3and C19, which does not match the predicted double bonds at C2-C3and C18-C19of salinomycin. This discrepancy suggests the possibilitythat the required dehydration reactions may be catalyzed by discretedehydratases after polyketide chain assembly. There is no putativedehydratase gene, however, in the cloned region.Within the salinomycin gene cluster, immediately downstream ofslnDI resides slnM. In silico analysis reveals that SlnM shows moderatehomology to TcmP in tetracenomycin biosynthesis, which catalyzesmethylation of a terminal carboxyl group. However, there is nomethylated salinomycin detected in XM211fermentation extract throughHPLC-MS analysis. Through gene replacement andtrans-complementation, the involvement of slnM in salinomycinbiosynthesis was confirmed. Besides, slnM mutant accumulates a novelcompound1. The NMR data indicated C18-C19bond saturation and thepresence of a C19hydroxyl group in1. Different from our expectation,no tricyclic spiroacetal was found in1, instead, a hydroxyl group ispresent at C13. Thus, it was very interesting to study the mechanism andtiming of the formation of the tricyclic spiroacetal. Feeding of1to JCY16(ΔslnC) or JCY34(ΔslnA9), both of which abolished the production ofsalinomycin but harbored the complete slnM gene, resulted in therestoration of salinomycin, suggesting that1is a possible intermediate ofsalinomycin biosynthesis. Heterologously expressed SlnM catalyzes the conversion of1tosalinomycin. This result suggests SlnM is responsible for the last reactionin salinomycin biosynthesis. To test whether S-adenosyl-methionine(SAM) serves as a methyl donor, we monitored the concentrations ofSAM and S-adenosyl-L-homocysteine (SAH) in the reaction mixture.Unexpectedly, neither consumption of SAM nor accumulation of SAHoccurred in the reaction mixture as detected by HPLC. Furthermore,sinefungin, a methyltransferase inhibitor and an SAM analogue unable todonate a methyl group, was found to stimulate the activity of SlnM ascomparable to SAM. In additional, the carboxyl group of1was proven tobe essential since esterified compound1could not be catalyzed by SlnM.Taking all this information into account, we have proposed a mechanismfor SlnM-catalyzed reaction. With a positive charge at neutral pH, SAMor sinefungin stabilizes the conformation of compound1similar tosalinomycin by electrostatic interactions in SlnM, which helps thedehydration of C18-C19and subsequent formation of the tricyclicspiroacetal. The biochemical characterization of SlnM as aSAM-dependent enzyme not only provides new insight into the catalyticroles of SAM, but also reveals an unprecedented mechanism forpolyketide double bond formation.In order to gain insight into the salinomycin biosynthetic mechanism,other12gene replacements were conducted. Among these genes,5genes were found to be essential for salinomycin biosynthesis and possiblyresponsible for polyketide chain release (slnBI), modification (slnF),oxidative cyclization (slnBII, slnBIII) and regulation (slnR). Moreover,5genes were identified as relevant to salinomycin biosynthesis andputatively involved in removal of aberrant extender units (slnDII),precursor supply (orf11, orf12) and regulation (orf15, orf16). Differentfrom other reported polyether gene clusters, salinomycin biosyntheticgene cluster contains three epoxide hydrolase genes: slnBI, slnBII andslnBIII. Gene replacements suggest that SlnBII and SlnBIII areresponsible for epoxide-opening cascades while SlnBI is involved inpolyketide chain release.The data presented here expand our understanding of polyetherbiosynthetic mechanisms and pave the way for targeted engineering ofsalinomycin bioactivity and productivity.更多还原