Abstract:

Natural products play a crucial role as sources of therapeutic agents for human health and agricultural pesticides. With the development of sequencing technologies, genome mining employing various bioinformatic tools has become an important approach to discovering novel natural products. Due to a staggering amount of cryptic natural product biosynthetic gene clusters available, prioritizing those capable of generating the most potent bioactive molecules has gained paramount significance. To avoid self-damage, some bioactive molecule producers employ self-resistance enzymes, which are slightly mutated versions of the original metabolic enzymes. These mutated enzymes maintain the original functions but are insensitive to the bioactive compound. The presence of a self-resistance enzyme in a natural product biosynthetic gene cluster serves as a predictive indicator of the biological activity of the compound it generates. On the other hand, the biosynthetic gene cluster of a natural product could be located using both structure and activity information as probes. Meanwhile, the accumulating knowledge of antibiotic resistance mechanisms has facilitated the discovery of novel antibiotics. Dereplication of natural products with known resistance mechanisms has been achieved using indicator strains expressing the resistance genes. While these approaches successfully utilized self-resistance genes to connect molecules to biological activities, a more impactful application is to accurately link biological activity and genomic information through target-guided mining of natural products. The concept is to use a self-resistance gene as a predictive tool to prioritize and identify biosynthetic gene clusters encoding compounds that inhibit specific targets. Recent breakthroughs in self-resistance gene identification have bridged the gap between activity-guided and genome-driven approaches for natural product discovery and functional assignment. This review summarizes recent representative studies on bioactive natural product discovery guided by self-resistance genes, as well as future trends. These include the following five points: 1) locating biosynthetic gene clusters based on self-resistance genes, 2) predicting the target of secondary metabolites through self-resistance genes, 3) rapid dereplication of bioactive compounds with known reaction mechanisms from self-resistance mechanisms, 4) genome mining of bioactive natural products guided by the target and the internal relationship with self-resistance genes, and 5) the development of genome data mining tools directed by self-resistance genes.

Key words: natural products, self-resistance gene, genome mining, biosynthesis, biosynthetic gene clusters

摘要：

天然产物是医药与农药的重要来源。基因组测序和生物信息学分析技术的飞速发展，揭示了大量功能未知的天然产物生物合成基因簇，利用生物信息学工具，从这些庞大的基因簇数据中挖掘活性天然产物已经成为发现新型天然药物的重要途径。天然产物的生产者们利用自抗性基因所表达的自抗性酶来保护自身，这种自抗性酶是体内一些初级代谢途径中管家酶的变体，不但对于活性天然产物具有较好的耐受性，还可以在生产活性天然产物的同时确保宿主体内代谢的正常进行。因而，自抗性基因指导的天然产物研究有效地将活性导向和基因组导向的天然产物发掘策略桥连起来，为精准发掘具有目标活性的新型天然产物提供了有效策略。本综述就利用自抗性基因作为探针进行天然产物发掘的代表性研究工作进行了整理和总结，并对研究趋势进行了展望，主要包括：①对于活性已知的天然产物，利用其自抗性基因来定位生物合成基因簇的研究；②以天然产物生物合成基因簇中的自抗性基因为线索，预测产物的作用靶点的研究；③利用天然产物自抗性机制，将具有已知作用机制的活性分子进行快速排重的研究；④利用自抗性基因与天然产物及其活性的内在联系，以目标靶点导向的活性天然产物基因组挖掘；⑤自抗性基因导向的基因组数据挖掘工具的发展情况。

关键词: 天然产物, 自抗性基因, 基因组挖掘, 生物合成, 基因簇