1555J. CHEM. SOC. PERKIN TRANS. I 1992 Mycinamicin Biosynthesis: Intact Incorporation of an Intermediate by a Chai n -el ongat i on Process i n Micromonospora griseorubida Hideaki Suzuki,*Sa Satoshi Takenaka,b Kenji Kinoshita,= Yuji Kogami,a Tatsuro Fujiwaraa and Toshiro Morohoshia a Research Laboratories, Toyo Jozo Co., Ltd., Ohito-cho, Shizuoka 410-23, Japan Development Division of Fermentation Technology, To yo Jozo Co., Ltd., Ohito-cho, Shizuoka 41 0-23, Japan The N-acetylcysteamine thioester t of a hypothetical intermediate in the biosynthesis of mycinamicin was incorporated intact into mycinamicin aglycone in cooperation with dotriacolides in Micro-monospora griseorubida. Mycinamicin I1 is a 16-membered macrolide antibiotic com- i0 posed of a branched lactone and two sugars, desosamine and mycinose.It is produced by Micronlonospora griseoruhida, and has strong antibacterial activity against Gram-positive -o,sObacteria. I Macrolide compounds, related to mycinamicin I1 were isolated from the culture filtrate of M. griseorubidu and Fig. 1 Structure of dotriacolide characterized by physicochemical method^.^.^ Bioconversion analysis using a blocked mutant identified these compounds as OH 0intermediates in the biosynthesis of mycinamicin, and thus PHWPHV OH 0 ii40the biosynthetic steps leading from protomycinolide IV to Biopdymer doEt mycinamicin I1 were delineated.6 Protomycinolide IV,4 the first macrolide intermediate in the biosynthesis of mycinamicin TI, is assembled from three acetates and five propionates, as identified 21iiiby incorporation analysis with ' 3C-labelled precursor^.^ 1It is currently believed that the biosynthetic pathway of macrolactone formation resembles that of fatty acids. Cane and *OCOCH2CHMe2 * iv OR 0 Yang' and Yeu et aL9 independently demonstrated that I3C-40Et labelled N-acetylcysteamine thioesterst (Scheme l), the activated forms of hypothetical intermediates in the biosynthesis of macrolide antibiotics, were incorporated into the macro- V -lactone rings in their intact forms when they were fed to producing cultures of the relevant microorganism.These results support the hypothesis that, in macrolactone formation, the thioester of a P-keto fatty acid undergoes a series of reduction- dehydration-reduction steps before the next chain elongation reaction. Mycinoic acids, considered to be biosynthetic Me'intermediates in the formation of the macrolactone, have been isolated from M.griseoruhida,"*' and similar compounds have 1 also been obtained from Srreptomyces.fradiac.' Scheme 1 Synthesis of a N-acetylcysteamine thioester of intermedi-Recently, we discovered that M. griseorubida produces ate; = I3C, R = tetrahydropyranyl, NAC = N-acetylcysteamine.large amounts of dotrialcolides l3 (Fig. l), accompanied by R~ugcwrs:i, H,SO,, EtOH; ii, lithium diisopropylamide, tetrahydro- mycinamicin production (data not shown). In this communi- furan, 13CH31; iii, dihydropyran, pyridinium toluene-p-sulfonate, cation, we report the intact incorporation of the N-acetyl- CH,CI,; iv, CICO,CH,CHMe,; v, HSCH,CH,NHAc; vi, CF3C02H, cysteamine thioester? of (2R,3R)-2-['3C]methyl-3-hydroxy-MeOH.pentanoate 1, an elaborated branched-chain fatty acid, into mycinolide IV 34 (Scheme 2), a mycinamicin aglycone, in combination with dotriacolides. Although compound 1 has not been isolated from M. griseorubida, the data reported here strongly suggest that the macrolide carbon skeleton is also assem bled by a chain-elongation process in Micronzotzospora H3' Or: M. griseorubidaSPP. The 100 MHz I3C NMR spectroscopic data for the bio- conversion experiments are summarized in Table 1. Compound Me' 1 did not I3C-enrich the C-21 of compound 3 to a greater extent than four methyl groups (C-17, C-18, C-19 and C-20) without 1 the addition of dotriacolides, indicating that the precursor had most likely been degraded to [3-"C]propionate followed by mycinolide IV 3 Scheme 2 Intact incorporation of a chain-elongation intermediate into t S(2-Aminoethyl) thioester.mycinamicin aglycone in M. griseoruhida; NAC = N-acetylcysteamine 1556 Table 1 Effect of dotriacolides on incorporation of 1 into 3 estimated by 100 MHz I3CNMR spectroscopy (in CDC!,) Relative peak heightsb Compound added C-17 C-18 C-19 C-20 C-21 1 3.65 2.99 2.95 3.15 3.18 1 + dotriacolides 2.19 2.10 2.45 2.18 3.1 1 I' The values were the mean obtained from three independent experiments. The relative peak heights with I3C-enrichment were calculated as the proportions to natural abundance. its general incorporation into 3.However, 1, after feeding to strain B-120-9 with the addition of dotriacolides, resulted in compound 3 specifically I3C-enriched at C-21, based on the decrease in the height of the signals for C-17, C-18, C-19 and C-20. M. gt-iscwrubiciaproduces a large amount of dotriacolides under antibiotic fermentation conditions (data not shown). These results suggest that the macrolide carbon skeleton in M. griscwuhitlr is assembled in a stepwise manner by a chain- elongation process as shown in s.fradiae' and s. erythreus,* and that dotriacolides would play an important role in mycinamicin biosynthesis in M. griseorubida. There has not been reported a compound such as the dotriacolides in Srwptonzjws.This suggests that the nature of macrolactone biosynthesis and/or the structure of cell surface between Micronionospot-a spp. and Streptontyces spp. might have some differences. Further investigation to grasp the biosynthesis of mycinamicin and to elucidate the role of dotriacolides in M. gt-iscwt-uhiduare in progress. Experimental S~ntliesisof' 1.-Compound 1 was synthesized as shown in Scheme 1. The microbial polyester containing 7040% of 3-hydroxybutanoate and 2&30% of 3-hydroxyvalerate (PHBIPHV) was hydrolysed in ethanol with a catalytic amount of sulfuric acid to give monomeric esters.I4 Lithium di-isopropylamide (3.0 cm3 of a 2.0 mol dm-3 solution in hexane, 6.0 mmol) was added to tetrahydrofuran (THF) (15 cm3) and stirred at -75 'C for 10 min. A solution of ethyl (3R)-3-hydroxypentanoate (monomeric ester of PHV) (0.73 g, 5.0 mmol) in THF (0.5 cm3) added dropwisely at -75 "C.After warming the mixture to -20 "C and stirring for 0.5 h, it was cooled again to -75 'C and ['3C]-methyl iodide (0.31 cm3, 5.0 mmol) was added dropwisely. The mixture was stirred at -75 -C for 2 h then at -5 -C for 16 h. The reaction mixture was concentrated in LWC'UO to remove most of THF. The residue was dissolved in ethyl acetate (50 cm3) and washed with water and brine. The organic layer was dried (MgS04) and the filtrate was concentrated in ziucuo after filtration. The residue was purified by silica gel chromatography, eluting with hexane- ethyl acetate (30:l), to give 0.62 g (77%) of ethyl (2R,3R)-2-[' 3C]-methyl-3-hydroxypentanoate 2.The ,C-enrichment level of 2 was more than 95%. Compound 2 was converted into 1 using the mixed anhydride method.' Spectral data of 1 including [XI,,were coincident with that of compound ob-tained during bioconversion experiments in the biosynthesis of t y lactone.' J. CHEM. SOC. PERKIN TRANS. 1 1992 Culture jbr Microbial Conversion of 1.-M. griseorubida B-120-9, a blocked mutant of a high mycinamicin I1 producing industrial strain, produces 3 singly under antibiotic fermen- tation conditions. Strain B-120-9 was inoculated into a 150 cm3 Erlenmeyer flask containing seed medium (soluble starch, 2.0%; yeast extract, 0.5%; NZ Amine type A, 0.5%: tryptose, 0.5%; CaCO,, 0.1%; FeS04*7H,0, 0.002%; pH 7.2) (20 cm3), and cultivated on rotary shaker at 28 "C for 48 h.This seed culture was then inoculated (10% by volume) into a 150 cm3 Erlen- meyer flask containing production medium (glucose, 2.0%; Casitone, 0.5%; MgS0,-7H20, 0.6%; CaCo,, 0.3%; K2HP0,, 0.2%; FeS04.7H,0, 0.001%; CoC12-6H,0, 0.0002%; Zn-SO4-7H2O,O.OOOl%; pH 7.0) (20 cm3). After 48 h at the same culture conditions, 1 (0.002 g) or 1 (0.002 g) with dotriacolides (0.001 g) were added to the flask and further cultivation was continued for 120 h. I3C-Labelled 3 was extracted from the culture fitrate with ethyl acetate at pH 9.0 and the organic extract was concentrated to afford a crude powder. By repeated crystallization from acetone-hexane, 3C-labelled 3 (0.002 g) was obtained as colourless crystals. Its purity was more than 90% (HPLC).Acknowledgements We are grateful to Dr. M. Hayashi for his support and encouragement. References 1 S. Satoi, N. Muto, M. Hayashi, T. Fujii and M. Otani, J. Antibiot., 1980,33, 364. 2 M. Hayashi, M. Ohno and S. Satoi, J. Chem. Soc., Chem. Commun., 1980, 1 19. 3 M. Hayashi, M. Ohno, S. Katsumata, S. Satoi, K. Harada and M. Suzuki, J. Antibiot., 1981,34, 276. 4 M. Hayashi, H. Ohara, M. Ohno, H. Sakakibara, S. Satoi, K. Harada and M. Suzuki, J. Antibiot., 1981,34, 1075. 5 K. Kinoshita, Y. Imura, S. Takenaka and M. Hayashi, J. Antibiot., 1989,42,1869. 6 H. Suzuki, S. Takenaka, K. Kinoshita and T. Morohoshi, J. Antibiot., 1990,43, 1508. 7 M. S. Puar, B. K. Lee, H. Munayyer, R. Brambilla and J. A. Weitz, J. Antibiol., 1981, 34, 6 19. 8 D. E. Cane and C.-C.Yang, J. Am. Chem. Soc., 1987, 109, 1255. 9 S. Yeu, J. S. Duncan, Y. Yamamoto and C. R. Hutcinson, J. Am. Cliem. Soc., 1987, 109, 1255. 10 K. Kinoshita, S. Takenaka and M. Hayashi, J. Chem. Soc., Chem. Coniniun., 1988,943. I1 K. Kinoshita, S. Takenaka and M. Hayashi, J. Chem. Soc., Perkin Trans. I, 1991, 2547. 12 M. L. Huber, J. W. Paschal, J. P.Leeds, H. A. Kirst, J. A. Wind, F. D. Miller and J. R. Turner, Antiniicroh. Agents Chemother., 1990, 20, 214. 13 Y. Ikeda, S. Kondo, T. Suwa, M. Tsuchiya, D. Ikeda, M. Hamada, T. Takeuchi and H. Umezawa, J. Antibiot., 1981,34, 1628. 14 D. Seebach and M. F. Zuger, Tetrahedron Lett., 1984,25,2747. Paper 2/02539F Received 15th May 1992 Accepted 18th May 1992
展开▼
