Research On The Biosynthetic Mechanism Of Streptonigrin, An Antimor Antibiotic

Streptonigrin(STN,1) is an aminoquinone antitumor antibioticproduced by Streptomyces flocculus. Structurally, STN，streptonigrone(2), and lavendamycin (7) constitute a family of nature products called“streptonigrinoids”. STN is comprised of four aromatic rings. Amongthem, the aminoquinolinone and pyridine rings are nearly co-planar, andthe multi-substituted phenyl ring faces them perpendicularly.STN is active against a broad range of tumors, including breast, lung,head, and neck cancer, lymphomas and melanomas. The antitumormechanism studies have identified that STN induces DNA single-anddouble-strand breaks, unscheduled DNA synthesis, DNA adductformation, inhibits topoisomerase II, blocks synthesis of DNA and RNA,induces sister-chromatid exchanges and chromosomal aberrations. Inaddition, STN also shows in vivo and in vitro antiviral properties, andpotent, broad spectrum antibacterial activities against bacteria andfungi.However, its clinic use is limited because of its severe andprolonged bone narrow depression. Therefore, STN attracted much attention from chemists and biologists.Thechemists have made manyefforts to modulate its pharmaceutical properties by total synthesis toobtain the STN analogs. Moreover, the early biosynthetic studies wereperformed by elaborate feeding experiments with isotopically labeledprecursors, suggesting the biosynthetic origin of STN, but the detailedbiosynthetic mechanism in molecular level still remains unclear.For cloning the biosynthetic gene cluster of STN, we obtained thegenome sequence of the STN producing strain by454sequencingtechnology. The putative biosynthetic gene cluster was identified by usinga reported C-methyltransferase MppJ catalyzing C-methylation ofphenylanaline as a target, and confirmed by a large deletion resulting inabolishing STN production. The boundaries of the STN gene cluster wereverified by a series of gene replacement experiments. This clustercontains48genes with stnA as its upstream boundary and the stnT4as thedownstream boundary. Bioinformatic analysis revealed that the clusterinclude four genes encoding SAM-dependent methyltansferase, fourgenes for leucine carboxyl methyltransferase (LCM family), thirteenoxidoreductase genes, four genes involved in3-hydroxyanthranilic acidbiosynthesis, regulator, resistance genes and unknown genes. Wegenerated21gene inactivation mutants, from8of which we identified11compounds. The mutant derived from inactivation of stnA accumulatedcompounds3,4and5, which are determined to be STN derivatives by NMR and HR-ESIMS analysis. These three compounds were determinedto be the biosynthetic intermediates by feeding the compounds3and5tothe non-producing mutant. Compound6was accumulated in the mutantsincluding stnT4, stnF3, and stnF4, and comfirmed to be anintermediate of STN biosynthesis also by a feeding experiment. Weproposed that StnA may be responsible for the hydrolysis of3to yield6.stnB1encodes the α-subunit of the aromatic ring dioxygenase.Inactivation stnB1generated a mutant which accumulated lavendamycin7and its methyl ester product8, and the yield of-carbolineoxapropaline D15was increased compared with the wild type strainaccording to the HPLC analysis. Feeding compound7and8s eparativelyto the nonproducing mutant stnP2could completely restore theproduction of STN while compound15couldn’t do it. These resultsindicated that compounds7and8were intermediates but15might be ashunt product produced by an unscheduled decarboxylation.We expressed and purified the StnB1-2proteins from E.coli andcarried out the in vitro experiment with compound7and8as substrates,and the expected products10and9were detected in the presence ofsodium dithionate (Na2S2O4) as a source of electron, respectively. Thebiochemical results prelimarily comfirmed that the StnBs wereresponsible for the cleavage of the N-C8bond of indole ring and theformation of two hydroxyl groups. Furthermore, the assays with compounds7and8as co-substrates indicated that the compound8mayserve a real substrate of StnBs. These results allow us to propose that themethyl esterification of lavendamycin7generates8which is oxidized byStnBs to form an intermediate of STN biosynthesis.stnF1-4encode leucine carboxyl methyltransferases. Among themutants derived from these four genes, only stnF1abolished STNproduction, while stnF2still produced reduced amounts of STN. ThestnF3-4mutants both accumulated compound6. Thus we proposedStnF1or StnF2may be responsible for the methyl esterification of7toproduce8, with the in vitro experiment we identified StnF2rather thanStnF1was responsible for the methyl esterification of7to produce8. Wealso inactivated four SAM-dependent methyltransferases. From themutant strain which stnQ1was inactivated, we obtained3-demethylatedSTN13. Inactivation of stnQ2completely stopped STN production. ThestnQ3inactivation mutant produced compound11, the6-desmethoxylated10. From the stnQ4fermentation extracts, we identified compound12,10-demethylated STN. Feeding compound11to the nonproducingstnP2mutantindicated the compound11might be a shunt product.Moreover, we performed other four genes inactivation. These mutantsdon’t either affect the production of STN or abolished the STNproduction without any analog or intermediate accumulated. We proposedthese genes may participate in the streptonigrin biosynthesis at an early stage. Based on bioinformatics analysis, structure elucidation of theobtained compounds, feeding experiment and biochemical assays ofthekey step, we proposed the biosynthetic pathway of STN. This workprovides opportunities to illuminate the enzymology of novel reactionsinvolved in this pathway and to create, using genetic andchemo-enzymatic methods, new streptonigrinoid analogues as potentialtherapeutical agents.